From Axel.Riese@risc.uni-linz.ac.at Mon Dec 1 02:03:00 1997 From: Axel.Riese@risc.uni-linz.ac.at (Axel Riese) Date: Mon, 01 Dec 1997 02:03:00 -0000 Subject: Petzold and Mingw32 References: <34806755.262747815@smtp.netzone.com> Message-ID: <199712010819.AA12709@melmac.risc.uni-linz.ac.at> Hi ! > >Another choice that works fine with dialog boxes etc. is to use rc.exe > >from Microsoft and resrc.exe from RSXNT which "includes" the .res file created > > I think you mean rsrc.exe from RSXNTDJ1.ZIP available from djgpp mirrors. > > >by rc into the executable. Of course, you are right. You find rsrc.exe either in RSXNTDJ1.ZIP or RSXNT131.ZIP (Winsite mirrors or CTAN (TeX-archive)). - Axel =========================================================================== Axel Riese Research Institute for Symbolic Computation J. Kepler University Linz A-4040 Linz e-Mail: Axel.Riese@risc.uni-linz.ac.at Austria URL: http://www.risc.uni-linz.ac.at/people/ariese/home/ =========================================================================== - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Mon Dec 1 02:03:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Mon, 01 Dec 1997 02:03:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: I have recently provided a patch to Cygnus to provide control of various environment variable settings on a program by program basis. Geoffrey has asked that I get feedback on my changes to see if they are something that is useful to the list as a whole. My changes cause a program to read the registry for options settings prior to reading settings from the environment. So, for instance, you can do something like this: cygwin -o 'c:\bin\bash.exe' binmode notitle That will set options in the registry which will cause bash to always use 'binmode' as its default and to never modify the title bar. The available options are: [no]binmode - Set default open mode [no]title - Display program names on the console title bar [no]glob - [Do not] interpret wild cards on commands run from CMD. [no]strip_title - [Do not] strip the path component of files names displayed in the title bar [no]tty - Attach a UNIX-style tty to the program All of the above are available as normal CYGWIN_* environment variables: CYGWIN_BINMODE, CYGWIN_NOBINMODE CYGWIN_TITLE, CYGWIN_NOTITLE etc. These settings are similar to those found in Sergey's recent coolview release except that it is possible to say CYGWIN_NOTITLE as well as CYGWIN_TITLE. Which leads into another source of discussion. I would prefer something like this for setting these options: set CYGWIN32=binmode notitle tty rather than: set CYGWIN_BINMODE=1 set CYGWIN_TTY=1 set CYGWIN_NOTITLE=1 My current patch allows both uses but it would probably better to stick with one. Here are the pros and cons as we see them so far: 1) Registry Pros: a) Don't need to mess with system environment variables. b) Allows pinpoint control of what program gets what options, e.g. setting `tty' on for bash means that bash always starts with a true /dev/tty. Other programs run from the command line will not also use tty handling as seen when using the CYGWIN_TTY environment variable. c) "Solves" the problem of programs, ported from UNIX, which need to use `binmode' by default. Cons: a) May slow down process startup slightly (I haven't noticed this in simple tests, though). b) Contributes to creeping featuritis. c) The BINMODE option is nice, but may delay true port of a product in favor of a "SET CYGWIN_BINMODE" band-aid. 2) CYGWIN32 Pros: a) Minimizes environment variable name space pollution. b) Mirrors proposed registry options syntax. c) More intuitive? Cons: a) May be harder to test for in shell scripts. b) Less intuitive? I'd appreciate any and all feedback on this. Please send comments to the list in general. Thanks. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From justins@iname.com Mon Dec 1 02:03:00 1997 From: justins@iname.com (Justin Smith) Date: Mon, 01 Dec 1997 02:03:00 -0000 Subject: More on Less (pardon the pun) References: <34814A75.A162719D@iname.com> Message-ID: <348205D8.4BE64C51@iname.com> Justin Smith wrote: > I just started playing with GNUWin32, and ran into a problem with > Less. I am compiling Less-332, and it compiles ok, but it gives me the > error ': No such file or directory' when attempting to view > certain files. > Never mind; I figured it out. My mistake for not mounting "/" as binary. Less works fine now. -- Justin Smith, Systems Engineer with the law office of Dewey, Cheetam and Howe. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Mon Dec 1 02:03:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Mon, 01 Dec 1997 02:03:00 -0000 Subject: cygwin32 (mingw32) news update Message-ID: <199712010401.EAA43478@out1.ibm.net> > > cygwin32 (mingw32) + egcs -- news update > > Mumit Khan (khan@xraylith.wisc.edu) > >Just a heads up for the egcs users -- I just ran the testsuite on the >latest snapshot (+ extra fixes for template handling + use sjlj for >exception handling), and the results for i386-cygwin32 are almost as >good as on i386-linux-gnulibc1! > Only one or two problems: mingw32 make emphatically does not work. It needs sh.exe, and, when it gets cygwin sh.exe, it has directory addressing conflicts. It definitely does not understand GNU makefiles, even when individual command-line compiles work. For some odd reason, compiling a program or library that references sleep, _sleep, __sleep, or Sleep fails during link, no matter what `extra' libraries are referenced. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From PATRIK@Oslo4GW.Norway.NCR.COM Mon Dec 1 02:03:00 1997 From: PATRIK@Oslo4GW.Norway.NCR.COM (Patrik Wennberg - NCR) Date: Mon, 01 Dec 1997 02:03:00 -0000 Subject: cygwin32 NT4.0 NTFS - problem? Message-ID: <34827681@Oslo4GW.Norway.NCR.COM> It was when I compiled perl5.004_04 (many thanks to Chris Faylor for supplying the patch to it) that I stumbled upon a problem that I can't find any solution to. I've looked through the mailing-list archive without being able to find any related info, so I need some help. It seems that the default behavior for permissions in a NTFS differ very much from UNIX. It looks like they have mixed the concept of "user" and "group" so that nt-groups appear as owners to objects in the fs. The result from this is that e.g. the test command fails when it test for write-permission. Please see the following script that explains the situation: file=$$_test_file test_perm() { if test -w $file then echo "write permission" else echo "NO write permission" fi } #My uid and gid: id >$file # Since I've just created the file, # I do have "writepermission" to it. ls -l $file test_perm echo "fuubar" >>$file #Just to convince myself ;-) chown `id -u -n` $file ls -l $file test_perm rm -f $file My perl5.004_04 is up and running, only because I made an ugly hack to the "-w test" in the source, but I am not happy with the situation. Is this something that should/could be fixed in cygwin32 or is it something that we just have to live with? PS. Of course I've tried everything both with and without Sergey's patches. Best Regards Patrik Wennberg _______________________________________________________________ | | Patrik Ericson Wennberg | NCR Norway, PS Division |--- | Email: Patrik.Wennberg@Norway.NCR.COM |--- | Phone: +47-22953770 | Cellphone (inc. SMS): +47-91321474 | MobileFax: +47-91240069 |--- | NCR Norge A.S. | Box 24, N-0409 OSLO | NORWAY |______________________________________________________________ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Ian@kiwiplan.co.nz Mon Dec 1 02:03:00 1997 From: Ian@kiwiplan.co.nz (Ian Collins) Date: Mon, 01 Dec 1997 02:03:00 -0000 Subject: Bash problems with #!... Message-ID: <07551A47EAD6D01186430060B025C6F41B6F63@ntserver.kiwiplan2.co.nz> Now I'm pulling my hair out. I'm having a problem with a script. It's a big script, and I know it should work OK (dangerous comment to make). The first line of the script is #!/bin/ksh /bin/ksh is a copy of the bash.exe in the b18 release directory (it is the latest from Sergey). When I delete the #!/bin/ksh line, the script works fine (my SHELL env variable is also /bin/ksh). The script problem is weird to say the least. It says there are syntax errors on lines with no syntax errors, and if you do a slight change to the script the errors move to a completely different place. The script in question is quite a large one, and unsurprisingly I can't reproduce it on a small script. I have CYGWIN_TTY set to 1. The filesystem is not mounted as binary (that makes it even worse). It is mounted as, e: / native no-mixed,text!=binary I am running on NT4 SP3. (OK, the obvious solution is to just take the #!/bin/ksh from the script, but the script works on 6 other Unix machines, and I hate changing things just for one machine. Especially when hundreds of scripts are involved). Ian Collins. KIWIPLAN NZ. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Eric.Gauthier@imag.fr Mon Dec 1 02:03:00 1997 From: Eric.Gauthier@imag.fr (Eric Gauthier) Date: Mon, 01 Dec 1997 02:03:00 -0000 Subject: ncurses configuration Message-ID: <34828B4B.5529@imag.fr> Hello, does anybody know how to configure terminal to properly work when using a screen buffer bigger than the window size (with a scroll bar). I can't use less with this feature. thanks Gauthier eric -- ------------------------------------------------------------------- Eric GAUTHIER email:Eric.Gauthier@imag.fr Equipe Reseaux Neuronaux LEIBNIZ / SHARP INRIA Rhones-Alpes 46 Av. Felix Viallet 38031 Grenoble Cedex FRANCE Tel : (33) 04 76 61 52 46 ou (33) 04 76 57 48 14 fax : 76 57 46 02 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From HolcomB@sauk.com Mon Dec 1 03:37:00 1997 From: HolcomB@sauk.com (Donald A. Holcomb Jr.) Date: Mon, 01 Dec 1997 03:37:00 -0000 Subject: Install problems Message-ID: <3481B59C.45A@sauk.com> I am running Win 3.1 on a 486. The install seems to run because I get the window and the icons, but when I execute GUNZIP, the screen goes black for a second then comes right back (the same happens when I do the WINZIP install). What am I missing to do an install so I have a compiler? Have gone through the README and FAQ files a number of times to no avail. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 1 04:26:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 01 Dec 1997 04:26:00 -0000 Subject: Bash problems with #!... Message-ID: <19971201122625.21025.qmail@hotmail.com> >From: Ian Collins >To: "'Gnu Mailing list'" >Subject: Bash problems with #!... >Date: Mon, 1 Dec 1997 17:38:11 +1300 > >Now I'm pulling my hair out. >I'm having a problem with a script. It's a big script, and I know it >should work OK (dangerous comment to make). >The first line of the script is #!/bin/ksh > >/bin/ksh is a copy of the bash.exe in the b18 release directory (it is >the latest from Sergey). >When I delete the #!/bin/ksh line, the script works fine (my SHELL env >variable is also /bin/ksh). > Did you try to use #!/bin/sh to see if the problems still exist. I'm under the impression that the executable name may make a difference in mode of operation. [snip] >Ian Collins. >KIWIPLAN NZ. > > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 1 05:09:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 01 Dec 1997 05:09:00 -0000 Subject: Getting bash working under Emacs Message-ID: <19971201130911.3224.qmail@hotmail.com> I'm not claiming this will work but it is worth a try. In a memo from Chris Faylor taking a vote for some modifications he mentions some environment variable switches two of which were CYGWIN_BINMODE and CYGWIN_NOBINMODE. I suggest for your emacs experiment to set CYGWIN_NOBINMODE=1 before calling emacs. Note that you will have needed to install Sergey's patches for this to work. What do you think Chris, would this allow bash to interpret the ^M from emacs as part of the line termination? What would the other side effects be from CYGWIN_NOBINMODE being set on? - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Fri, 28 Nov 1997 18:38:46 -0800 >To: gnu-win32@cygnus.com, ntemacs-users@cs.washington.edu >From: "Guy G. Piggford" >Subject: Getting bash working under Emacs > >Sorry about sending this to both lists, but I'm not sure exactly where the >fix is so I hoping that the wider audience helps. > >I'm using the cygnus tools along with the latest coolview patches and emacs >19.34.6. I've just finished getting bash 2.01.1 compiled and working and >patched it so that bash can handle the case-insensitivity of NT correctly. >Things looked pretty good, I'd had some mucking around to do since >configure ran but came up with a pretty hopeless configuration so I had to >spend most of last night night finding all of the errors in config.h, but >after tidying those up, it compiled and worked just fine. > >Then I tried to run bash as an inferior shell in emacs, it runs, but after >every command is typed it comes back with an error about command^M not >being recognised. Now I know that the coolview patches seem to have >removed a lot of useful compatability options to do with text!=binary so >I'm assuming that this is another one. > >Has anyone seen this behaviour? I don't know whether the best bet is to >start delving into emacs lisp to find the problem (not my best language) or >try to patch bash to work differently. > >Any suggestions? > >Guy > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 1 05:09:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 01 Dec 1997 05:09:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <19971201124852.4781.qmail@hotmail.com> Sounds like a great idea. >From: cgf@bbc.com (Chris Faylor) >Subject: Feedback needed on proposed cygwin feature >Date: Mon, 1 Dec 1997 04:12:00 GMT >To: gnu-win32@cygnus.com >Reply-To: cgf@bbc.com > [snip] > >Which leads into another source of discussion. I would prefer something >like this for setting these options: > > set CYGWIN32=binmode notitle tty Less environment clutter. Nice idea. > >rather than: > > set CYGWIN_BINMODE=1 > set CYGWIN_TTY=1 > set CYGWIN_NOTITLE=1 > >My current patch allows both uses but it would probably better to stick >with one. > >Here are the pros and cons as we see them so far: > >1) Registry > > Pros: > > a) Don't need to mess with system environment variables. > > b) Allows pinpoint control of what program gets what options, e.g. > setting `tty' on for bash means that bash always starts with > a true /dev/tty. Other programs run from the command line will > not also use tty handling as seen when using the CYGWIN_TTY > environment variable. > > c) "Solves" the problem of programs, ported from UNIX, which need > to use `binmode' by default. > text=binary on by default would be a definite benefit. > Cons: > > a) May slow down process startup slightly (I haven't noticed this > in simple tests, though). > > b) Contributes to creeping featuritis. > > c) The BINMODE option is nice, but may delay true port of a > product in favor of a "SET CYGWIN_BINMODE" band-aid. I see nothing wrong with "What ain't broke don't fix" attitudes. > >2) CYGWIN32 > > Pros: > > a) Minimizes environment variable name space pollution. > This one gets my vote! > b) Mirrors proposed registry options syntax. > > c) More intuitive? Less hassle. > > Cons: > > a) May be harder to test for in shell scripts. > To aid in testing a switch or environment variable could be added to turn on debug mode to display the values of the registry variables on stderr. Or, a standalone program to list and modify registry elements by a given key? > b) Less intuitive? Only for those that aren't intuitive. > >I'd appreciate any and all feedback on this. Please send comments to the list >in general. > >Thanks. >-- > http://www.bbc.com/ cgf@bbc.com "Strange how unreal >VMS=>UNIX Solutions Boston Business Computing the real can be." >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From norm@comanche.dyn.ml.org Mon Dec 1 09:09:00 1997 From: norm@comanche.dyn.ml.org (Norm Peterson) Date: Mon, 01 Dec 1997 09:09:00 -0000 Subject: Too Much Mail Volume References: <01bcfdf8$56371b20$8a82400c@hypercow> Message-ID: On Sun, 30 Nov 1997, Peter Yeh wrote: > I second the idea of moving this mailing list to a newsgroup. > > > > > >This mailing list has so much volume that I think we would be better > >served with a newsgroup to move the traffic and to provide a FAQ. > > > >How about comp.unix.gnu-win32? How about just dropping into one of the alt.* hierarchy? From what I understand, you don't need to have any of the voting (or any particular reason) to start the group. This could serve as a testbed before moving into the comp.* area. If it fails, ah well back to the mail list. If not, great. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 1 09:09:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 01 Dec 1997 09:09:00 -0000 Subject: egcs, C++ and GnuWin32 References: <3.0.3.32.19971127001234.00a7b5c0@ng.netgate.net> Message-ID: On Thu, 27 Nov 1997, Theodore W. Leung wrote: > I am able to compile and link the program, but when I invoke a.exe, I get an > error saying that the dll 'll' cannot be found in the path. As far as I > can see, there is no ll.dll in the cygwin distribution. If I do an objdump > -p on a.exe, I get the stuff below. Can anyone get me started on how to > track this down? > Make sure that you're linking with export libraries (eg., libkernel32.a etc) that were distributed/compiled with egcs gcc, not gnu-win32 b18 gcc. Could you send me the output of the verbose linking step? % gcc -o blah.exe *.o -v -Wl,--verbose Best if you can share the source code of course, but the above might be sufficient to see where the problem lies. Thanks Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From norm@comanche.dyn.ml.org Mon Dec 1 09:09:00 1997 From: norm@comanche.dyn.ml.org (Norm Peterson) Date: Mon, 01 Dec 1997 09:09:00 -0000 Subject: Feedback needed on proposed cygwin feature References: Message-ID: On Mon, 1 Dec 1997, Chris Faylor wrote: > The available options are: > > [no]binmode - Set default open mode > [no]title - Display program names on the console > title bar > [no]glob - [Do not] interpret wild cards on commands > run from CMD. > [no]strip_title - [Do not] strip the path component of files > names displayed in the title bar > [no]tty - Attach a UNIX-style tty to the program How about adding one more entry for the root dir? This would help with the chroot() problems. If the registry key (or env variable, or whatever) is set, use it's value as the root dir. Otherwise use the global mount as usual. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 1 09:09:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 01 Dec 1997 09:09:00 -0000 Subject: cc1 got fatal signal 33 References: <199712010020.LAA11403@mali.cd.comdyn.com.au> Message-ID: On Mon, 1 Dec 1997, raf wrote: > >I am using b18 and get the following error. > > >"gcc: Internal compiler error: program cc1 got fatal signal 33" > > i think this is just b18's way of saying 'syntax error'. > at least that's what it's meant when i've seen it. > It means exactly what is says -- there is an "internal error" in cc1, the C compiler driver. Typically if there is an internal inconsistency (read: bug!), the compiler driver aborts with one of the signals. These should be reported, since in theory, you should never see these messages. Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ggp@informix.com Mon Dec 1 11:43:00 1997 From: ggp@informix.com (Guy Gascoigne - Piggford) Date: Mon, 01 Dec 1997 11:43:00 -0000 Subject: More on getting Bash 2 working under Emacs Message-ID: <3.0.3.32.19971201114258.00bac390@pop.pdx.informix.com> On getting to work today I find that I don't get the strange ^M behaviour using my bash and emacs here that I get at home. As far as I know these two systems should be the same as far as Emacs and the Cygnus stuff is concerned so I obviously have some more delving to do. Very strange, either way it does mean that as fas as this machine is concerned bash 2, with case-insensitive completion an globing is working under emacs. Now all I have to do is get it working at home. Guy - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Mon Dec 1 11:43:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Mon, 01 Dec 1997 11:43:00 -0000 Subject: More on mingw32, make, sleep, _sleep, __sleep, and Sleep Message-ID: <199712011919.TAA34370@out2.ibm.net> My apologies about the problems with mingw32. If you have your mingmake.bat set your environment variables exactly right (following the README), ming make will actually generate make.bat files from cygwin ./configured Makefiles, and, in rare instances, will actually make GNU makefiles without using bash or sh (but don't use a makefile that changes directories while compiling). If you are less than exactly right, ming will still compile individual .c files, but will not make, and will not find sleep, _sleep, __sleep, and Sleep. The following is ming gdb running an experimental version of the ming slrn news reader: Command name abbreviations are allowed if unambiguous. (gdb) run run Starting program: D:\mingw32\src\slrn\src/slrn.exe bff30000:C:\WINDOWS\SYSTEM\GDI32.DLL bff60000:C:\WINDOWS\SYSTEM\USER32.DLL bfed0000:C:\WINDOWS\SYSTEM\ADVAPI32.DLL 7e2e0000:C:\WINDOWS\SYSTEM\WSOCK32.DLL bff70000:C:\WINDOWS\SYSTEM\KERNEL32.DLL 7fd50000:C:\WINDOWS\SYSTEM\CRTDLL.DLL [failed reading symbols from DLL] "C:\WINDOWS\SYSTEM\COMCTL32.DLL": error reading line numbers 7eb40000:C:\WINDOWS\SYSTEM\SHELL32.DLL [failed reading symbols from DLL] "C:\WINDOWS\SYSTEM\SHLWAPI.DLL": error reading line numbers 61000000:C:\WINDOWS\SYSTEM\URL.DLL slrn 0.9.4.3 (Dec 1 1997 14:06:42) Program received signal SIGSEGV, Segmentation fault. 0x438c51 in _image_base__ () (gdb)  - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Mon Dec 1 11:43:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Mon, 01 Dec 1997 11:43:00 -0000 Subject: Getting bash working under Emacs Message-ID: <199712011335.IAA00750@hardy.bbc.com> Sorry, it won't work. My previous mail did not make it clear that the CYGWIN_BINMODE environment variable is not currently implemented in any version of B18. Ditto for CYGWIN_GLOB. The other CYGWIN_* environment variables are already implemented in coolview, I think. >I'm not claiming this will work but it is worth a try. In a memo from >Chris Faylor taking a vote for some modifications he mentions some >environment variable switches two of which were CYGWIN_BINMODE and >CYGWIN_NOBINMODE. I suggest for your emacs experiment to set >CYGWIN_NOBINMODE=1 before calling emacs. Note that you will have needed >to install Sergey's patches for this to work. > >What do you think Chris, would this allow bash to interpret the ^M from >emacs as part of the line termination? What would the other side >effects be from CYGWIN_NOBINMODE being set on? > >- \\||// >---o0O0--Earnie--0O0o---- >-earnie_boyd@hotmail.com- >------ooo0O--O0ooo------- > >>Date: Fri, 28 Nov 1997 18:38:46 -0800 >>To: gnu-win32@cygnus.com, ntemacs-users@cs.washington.edu >>From: "Guy G. Piggford" >>Subject: Getting bash working under Emacs >> >>Sorry about sending this to both lists, but I'm not sure exactly where >the >>fix is so I hoping that the wider audience helps. >> >>I'm using the cygnus tools along with the latest coolview patches and >emacs >>19.34.6. I've just finished getting bash 2.01.1 compiled and working >and >>patched it so that bash can handle the case-insensitivity of NT >correctly. >>Things looked pretty good, I'd had some mucking around to do since >>configure ran but came up with a pretty hopeless configuration so I had >to >>spend most of last night night finding all of the errors in config.h, >but >>after tidying those up, it compiled and worked just fine. >> >>Then I tried to run bash as an inferior shell in emacs, it runs, but >after >>every command is typed it comes back with an error about command^M not >>being recognised. Now I know that the coolview patches seem to have >>removed a lot of useful compatability options to do with text!=binary >so >>I'm assuming that this is another one. >> >>Has anyone seen this behaviour? I don't know whether the best bet is >to >>start delving into emacs lisp to find the problem (not my best >language) or >>try to patch bash to work differently. >> >>Any suggestions? >> >>Guy >> >>- >>For help on using this list (especially unsubscribing), send a message >to >>"gnu-win32-request@cygnus.com" with one line of text: "help". >> > > > >______________________________________________________ >Get Your Private, Free Email at http://www.hotmail.com > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fermin.bueno@vanmail.amd.com Mon Dec 1 11:43:00 1997 From: fermin.bueno@vanmail.amd.com (Fermin Bueno) Date: Mon, 01 Dec 1997 11:43:00 -0000 Subject: less or more, where are them? Message-ID: <01BCFE6C.42CBB220@stoxo18.amd.com> Hello all, Do anyone knows where can I find more or less? I have installed gnu-win32 in my NT machine, I can only find zmore which seem to be a shell-script not related to more and less.exe, but when I execute it nothing happens. Can anyone help me? Thanks in advanced Fermin Fermin.Bueno@vantis.com Nordic Field Application Engineer V A N T I S The Programmable Logic Company From AMD Adolfsbergsvagen 31 Phone +46-(0)8-629 28 50 Box 20008 Fax +46-(0)8-98 09 06 161 02 BROMMA, SWEDEN Mobile +46-(0)708-74 55 53 http://www.vantis.com Audix - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mello@lamont.ldgo.columbia.edu Mon Dec 1 11:43:00 1997 From: mello@lamont.ldgo.columbia.edu (Ulisses T. Mello) Date: Mon, 01 Dec 1997 11:43:00 -0000 Subject: g++, opengl, and STDCALL, * __attribute__ ((stdcall)) problem. References: <01bcfe01$09ed4ba0$fa173185@gbird0.fu.is.saga-u.ac.jp> Message-ID: Thanks Colin. The recommended sintax looks really strange and I wonder what would be the solution for ponters to functions. For example: typedef void APIENTRY (APIENTRY * PFNGLARRAYELEMENTEXTPROC) (GLint i); In this case I changed to: typedef void APIENTRY (* PFNGLARRAYELEMENTEXTPROC) (GLint i); and it compiles but in some applications i get a warning message complaining that the type specifier will be omitted for parameter. Is this OK? Cheers -- ulisses. On Mon, 1 Dec 1997, Colin Peters wrote: > > From: ulisses mello > Date: Saturday, November 29, 1997 10:55 PM > > >I am trying to compile a simple program (cube.c) with g++. I have no > >problem to compile it using gcc. However, GL/gl.h does not compile in > >g++. For example, I get the error: > > > >sintax error before '__attribute__' on the function > > > >WINGDIAP const GLubyte * APIENTRY glGetString( GLenum name); > > > >making the substitutions from windows.h it becomes: > > > >const GLubyte * __attribute ((stdcall)) glGetString( GLenum name); > > > >I am aware the g++ flags an error for things like: > > > >int * __stdcall foo(); // but has no problems with > >int __stdcall * foo(); > > > >But changing the order of * in gl.h has not solved the problem. > > I think if you move APIENTRY to the *end* of the function prototype then > it will work with g++ (and gcc), but, of course, it will stop working with > other compilers I suspect. If you are just using gcc and g++ then no > problem, otherwise make separate copies of those headers for gcc/g++. > > WINGDIAP const GLubyte * glGetString( GLenum name) APIENTRY; > > Looks strange, but I think it will work. > > This is, I think, a bug in the C++ parser, and I have no idea if/when it > will be fixed. > > Good luck, > Colin. > > -- Colin Peters -- colin at fu.is.saga-u.ac.jp > -- Saga University Dept. of Information Science > -- http://www.fu.is.saga-u.ac.jp/~colin > -- http://www.geocities.com/Tokyo/Towers/6162 > > > > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tomas.fasth@twinspot.net Mon Dec 1 11:43:00 1997 From: tomas.fasth@twinspot.net (Tomas Fasth) Date: Mon, 01 Dec 1997 11:43:00 -0000 Subject: Do we need commercial advertisements in the gnu-win32 list? References: <009BDFEE.EA443220.29140@ifk20.mach.uni-karlsruhe.de> <3481B9AE.C48FDD48@opennt.com> Message-ID: <3482F866.1C470A4F@twinspot.net> Jason, I joined this mailing list because of my interest in the freely availible gnu-win32 distribution. I am personally not a bit interested in commercial binary-only alternatives, especially not yours since I concider it far to expensive. I do not consider this mailing list a proper place for you to advertise about features in your products. Remember that you are using cygnus resources for doing this. Do you have their consent? If you have not done it already, I suggest you to set up your own mailing list for issues related to your own products. Thanks, Tomas - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 1 11:43:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 01 Dec 1997 11:43:00 -0000 Subject: cygwin32 (mingw32) news update References: <199712010401.EAA43478@out1.ibm.net> Message-ID: <9712011702.AA18011@modi.xraylith.wisc.edu> vischne@ibm.net writes: > > Only one or two problems: > > mingw32 make emphatically does not work. It needs sh.exe, and, when it > gets cygwin sh.exe, it has directory addressing conflicts. It definitely > does not understand GNU makefiles, even when individual command-line > compiles work. Yeah, this is certainly a problem. I'll check with the other make port (which I should've known about ;-) to see if it fixes the problem. btw, mingw32 make *does not* need sh.exe; it uses sh.exe if it finds in the path. Of course, if there is no sh.exe in path, then sh specific commands will fail (eg., 'cd blah; $(MAKE) all'). To make things worse, Cygwin32 bash seems to have a problem with the following construct: sh -c "c:/path/to/executable.exe args" Where the pathname is not converted to //c/path/to/executable.exe. my make port has another bug which might be due to the "quote eating" habit of MS CRTDLL spawn functions. This is a low priority for me, but I'll of course accept patches from you to fix this behaviour. > > For some odd reason, compiling a program or library that references sleep, > _sleep, __sleep, or Sleep fails during link, no matter what `extra' > libraries are referenced. > It does not fail for me. I've asked you for more information, and I'd appreciate a response. Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From JONCKHEERE@d0tng.fnal.gov Mon Dec 1 11:43:00 1997 From: JONCKHEERE@d0tng.fnal.gov (JONCKHEERE@d0tng.fnal.gov) Date: Mon, 01 Dec 1997 11:43:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <971201101031.2680056c@d0tng.fnal.gov> D0TNG (D0 AXP cluster) @FNAL, 1-DEC-1997 --------------------------------------------------------------------------- > From: SMTP%"cgf@bbc.com" 1-DEC-1997 06:08:43.18 > To: JONCKHEERE > CC: > Subj: Feedback needed on proposed cygwin feature > > I have recently provided a patch to Cygnus to provide control of various > environment variable settings on a program by program basis. Is this patch generally available yet? If so where? If not when can we expect to see it? > Geoffrey has asked that I get feedback on my changes to see if they are > something that is useful to the list as a whole. >From your discription it looks like *just* the sort of thing we need. But probably needs some generalization. We are porting a large body of code from Unix to NT, including the code development system used on Unix (CVS and a home brewed set of scripts and makefiles). It all works very well in the DOS command line window. But we'd like to integrate it into the windows environment. The biggest problem we face is that the various exes require different, sometimes conflicting environment variable values. At the moment we invoke each exe via a bash shell script which sets the variables. But this is something of a kludge. It sounds like your patch, or a generalization of it, might allow a much more elegant solution to our problem. > My changes cause a program to read the registry for options settings prior > to reading settings from the environment. So, for instance, you can do > something like this: > > cygwin -o 'c:\bin\bash.exe' binmode notitle > > That will set options in the registry which will cause bash to always > use 'binmode' as its default and to never modify the title bar. > > The available options are: > > [no]binmode - Set default open mode > [no]title - Display program names on the console > title bar > [no]glob - [Do not] interpret wild cards on commands > run from CMD. > [no]strip_title - [Do not] strip the path component of files > names displayed in the title bar > [no]tty - Attach a UNIX-style tty to the program > > All of the above are available as normal CYGWIN_* environment variables: > > CYGWIN_BINMODE, CYGWIN_NOBINMODE > CYGWIN_TITLE, CYGWIN_NOTITLE > etc. > > These settings are similar to those found in Sergey's recent coolview > release except that it is possible to say CYGWIN_NOTITLE as well as > CYGWIN_TITLE. Can the list of variables be expanded by the user? I'd like to be able to do things like setting PYTHONPATH (or just PATH) to an arbitrary value depending on the exe, and the version of the exe, I'm running. > Which leads into another source of discussion. I would prefer something > like this for setting these options: > > set CYGWIN32=binmode notitle tty > > rather than: > > set CYGWIN_BINMODE=1 > set CYGWIN_TTY=1 > set CYGWIN_NOTITLE=1 > > My current patch allows both uses but it would probably better to stick > with one. > > Here are the pros and cons as we see them so far: > > 1) Registry > > Pros: > > a) Don't need to mess with system environment variables. > > b) Allows pinpoint control of what program gets what options, e.g. > setting `tty' on for bash means that bash always starts with > a true /dev/tty. Other programs run from the command line will > not also use tty handling as seen when using the CYGWIN_TTY > environment variable. > > c) "Solves" the problem of programs, ported from UNIX, which need > to use `binmode' by default. > > Cons: > > a) May slow down process startup slightly (I haven't noticed this > in simple tests, though). > > b) Contributes to creeping featuritis. > > c) The BINMODE option is nice, but may delay true port of a > product in favor of a "SET CYGWIN_BINMODE" band-aid. > > 2) CYGWIN32 > > Pros: > > a) Minimizes environment variable name space pollution. > > b) Mirrors proposed registry options syntax. > > c) More intuitive? > > Cons: > > a) May be harder to test for in shell scripts. > > b) Less intuitive? > > I'd appreciate any and all feedback on this. Please send comments to the list > in general. > > Thanks. Naively, the cygwin32 form seems most appealing. But that depends on the expandability of the list and if the list can be different for each exe. Hope this helps, Alan > -- > http://www.bbc.com/ cgf@bbc.com "Strange how unreal > VMS=>UNIX Solutions Boston Business Computing the real can be." > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jazz@softway.com Mon Dec 1 11:47:00 1997 From: jazz@softway.com (Jason Zions) Date: Mon, 01 Dec 1997 11:47:00 -0000 Subject: Feedback needed on proposed cygwin feature References: Message-ID: <348313F1.45C1076E@opennt.com> It strikes me that the best place for per-binary attributed like this would be in the binary itself. I don't know how closely cygwin's exec() looks at the program being exec'd; if it looks at the program header, there should be a way to squirrel the attributes away there someplace. Many "real Unix" systems do just this sort of thing; e.g. chatr in HP-UX, which is used to alter a.out attributes. Pro: a) attributes are *really* a part of the binary; symlinks and hardlinks all pointing to the same bits on the disk will behave the same, while name-based approaches may be inconsistent b) makes it easier for attributes to work over network mounts (when filesystems can be mounted in different places) Con: a) extracting the attributes from the binary may be slower b) may be hard to find a "safe place" to stash the bits Jason - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jazz@softway.com Mon Dec 1 11:53:00 1997 From: jazz@softway.com (Jason Zions) Date: Mon, 01 Dec 1997 11:53:00 -0000 Subject: cygwin32 NT4.0 NTFS - problem? References: <34827681@Oslo4GW.Norway.NCR.COM> Message-ID: <3483157E.17B64A67@opennt.com> The NTFS security model doesn't distinguish between users and groups; each ACE in an ACL identifies a security principal (either user or group) and lists the principal's access rights. A file has an owner, which is any security principal (again, can be a group). A file also has a primary group, but I think NTFS enforces this to actually be a group and not a user. By default, Win32 frequently makes the Administrators *group* the owner of any file created by any member of the Administrators group. Typical ACLs would have two entries: an ACE for the Administrators group and an ACE for Everyone. Cygwin has to somehow map those two ACEs into the classical Unix mode bits; something someplace is *always* going to appear odd when an ACL doesn't have exactly an ACE for the owner, an ACE for a group, and an ACE for everyone/authenticated users/whatever cygwin thinks "Unix Other" maps to. Jason - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From lizt@synopsys.com Mon Dec 1 15:49:00 1997 From: lizt@synopsys.com (B. Elizabeth Trojan) Date: Mon, 01 Dec 1997 15:49:00 -0000 Subject: cygnus' notion of symbolic links Message-ID: <34834DFF.7566F4CF@synopsys.com> Can someone explain how symbolic links work using the Cygnus ln -s command? -- Liz Trojan Synopsys (503)531-2297 Office Beaverton (503)690-6906 FAX - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Mon Dec 1 18:07:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Mon, 01 Dec 1997 18:07:00 -0000 Subject: ncurses configuration References: <34828B4B.5529@imag.fr> Message-ID: In article < 34828B4B.5529@imag.fr >, Eric Gauthier wrote: >Hello, >does anybody know how to configure terminal to properly work when using >a screen buffer bigger than the window size (with a scroll bar). >I can't use less with this feature. There is a bug in the screen handling code which screws up scrolling regions when the screen buffer is bigger than the window size. It should be fixed in B19. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From bbhattac@hotmail.com Mon Dec 1 19:39:00 1997 From: bbhattac@hotmail.com (Bhaskar Bhattacharya) Date: Mon, 01 Dec 1997 19:39:00 -0000 Subject: fread() Message-ID: <19971202033829.11379.qmail@hotmail.com> Thanks for many people pointing out the problem. I will try using "rb" in the fopen() and see if the problem goes away (I am sure now that it will). Since I don't need to use the "rb" option in the emx port (rsxnt for Win32), I completely forgot about it . Bhaskar Bhattacharya ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From John.Turner@pobox.com Mon Dec 1 20:18:00 1997 From: John.Turner@pobox.com (John A. Turner) Date: Mon, 01 Dec 1997 20:18:00 -0000 Subject: Problems with 'bash' References: <19971128152134.7905.qmail@hotmail.com> Message-ID: <348380AF.D0F23845@POBox.com> Earnie Boyd wrote: > After posting the results of the poll I got _many_ more positives than I > had negatives. OK, I guess I misread the results. My apologies. > Besides, if I don't post a response how will anyone know > if that person was ever helped. Know, that I have to see and delete > these posts again also. True. > Please accept my apologies for any inconvience. What I will do > (assuming, I remember before sending the mail) is to add (NBY) to the > subject. If your mail program allows you to set up rules, then you set > up a rule to just delete a subject containing (NBY) from > earnie_boyd@hotmail.com. This will begin after 11/28/1997 10:30am EST. That's an astoundingly good idea, and one I wish I'd thought of. Thanks very much, both for posting the pointer messages and for now labelling them to enable filtering. Excellent idea. -John - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Ian@kiwiplan.co.nz Mon Dec 1 20:18:00 1997 From: Ian@kiwiplan.co.nz (Ian Collins) Date: Mon, 01 Dec 1997 20:18:00 -0000 Subject: Bash problems with #!... Message-ID: <07551A47EAD6D01186430060B025C6F41B6F66@ntserver.kiwiplan2.co.nz> Yes, I'd tried that as well. /bin/sh is also a copy/link (I've tried both) to the latest bash. > -----Original Message----- > From: Earnie Boyd [SMTP:earnie_boyd@hotmail.com] > Sent: Tuesday, December 02, 1997 1:26 AM > To: Ian@kiwiplan.co.nz > Cc: gnu-win32@cygnus.com > Subject: Re: Bash problems with #!... > > > > >From: Ian Collins > >To: "'Gnu Mailing list'" > >Subject: Bash problems with #!... > >Date: Mon, 1 Dec 1997 17:38:11 +1300 > > > >Now I'm pulling my hair out. > >I'm having a problem with a script. It's a big script, and I know it > >should work OK (dangerous comment to make). > >The first line of the script is #!/bin/ksh > > > >/bin/ksh is a copy of the bash.exe in the b18 release directory (it > is > >the latest from Sergey). > >When I delete the #!/bin/ksh line, the script works fine (my SHELL > env > >variable is also /bin/ksh). > > > > Did you try to use #!/bin/sh to see if the problems still exist. I'm > under the impression that the executable name may make a difference in > > mode of operation. > > [snip] > >Ian Collins. > >KIWIPLAN NZ. > > > > > >- > >For help on using this list (especially unsubscribing), send a > message > to > >"gnu-win32-request@cygnus.com" with one line of text: "help". > > > > > - \\||// > ---o0O0--Earnie--0O0o---- > -earnie_boyd@hotmail.com- > ------ooo0O--O0ooo------- > > > ______________________________________________________ > Get Your Private, Free Email at http://www.hotmail.com > - > For help on using this list (especially unsubscribing), send a message > to > "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Lavis@sheltonbbs.com Mon Dec 1 20:18:00 1997 From: Lavis@sheltonbbs.com (Loyd Avis III) Date: Mon, 01 Dec 1997 20:18:00 -0000 Subject: Fw: Problems with setup on cdk-src Message-ID: <01bcfed8$7ba60480$267ac4ce@host> -----Original Message----- From: Kathy Powers To: Loyd Avis III Date: Monday, December 01, 1997 12:20 PM Subject: Re: Problems with setup on cdk-src >Thanks for your email; however Cygwin32 is an unsupported product. Sorry. >The best I can suggest is to send email to: gnu-win32@cygnus.com. Our web >site gives this information as well http://www.cygnus.com/misc/gnu-win32/ . > >We are looking at porting our full GNUPro C/C++ development toolchain to >Windows NT/95 Native in the future. Let me know if you would like me to >keep your name on file for updated information on this availability. > >Additionally, we do support our GNUPro toolchain on over 150 different >platform combinations. Please see our web site for details: > http://www.cygnus.com . Let me know if you might have interest in GNUPro on >any of our supported platforms. > >Good luck, >Kathy > > > >At 12:43 AM 11/30/97 -0600, you wrote: >>>>> > >Hello, I have downloaded the program cdk.exe and installed it.... I then read that I needed these files size=2>xca,xcb,xcc,xcd,xce,xcf,xcg,xch,xci,xcj Then I followed the instructions > > Connect to our ftp server and cd to the following directory: > > ` ftp://ftp.cygnus.com/pub/gnu-win32/latest '. If you want the user > > tools source code, `cd' into the `usertools-src' subdirectory. > > Download the files there. If you want the Cygnus Developer's Kit > > sources, `cd' into the `cdk-src' subdirectory. Download the files > > there. > > For the compiler tools source: > > copy /b xca + xcb + xcc + xcd + ... + xck + xcl cdksrc.tar.gz > > del xc*.* > > gunzip -d cdksrc.tar.gz > > tar xvf cdksrc.tar > > > > When I unzip the files it comes up UNEXPECTED END OF FILE..... > > and then I cannot proceed with the other commands... Should I reload the files again? or what? > > > > The zip file is corrupted.... and I cannot find which file is the culprit...is there anyway you can help me? > > > > Thank You > > < mailto:Lavis@sheltonlink.com >Lavis@sheltonlink.com > > > > > > > Attachment Converted: "c:\program files\eudora\attach\smime.p7s" > > > > > > > > > >Kathy Powers >Cygnus Solutions >Western Region Account Manager >Phone: +1-360-379-3680 >Fax: +1-360-379-3678 >Email: kpowers@cygnus.com >Web site: http://www.cygnus.com > From iverson@lionheart.com Mon Dec 1 20:18:00 1997 From: iverson@lionheart.com (Tim Iverson) Date: Mon, 01 Dec 1997 20:18:00 -0000 Subject: Too Much Mail Volume References: Message-ID: I don't like the alt.* idea. I don't get (and don't want) any of the alt.* groups, so putting it there puts it out of my reach. IMHO, the only stuff that belongs in alt.* is that which no one wants anyway. If this list falls into that category, then it should not be moved to a newsgroup. - Tim Iverson iverson@lionheart-com +---------------- | Date: Mon, 1 Dec 1997 08:28:38 -0600 (CST) | From: Norm Peterson | To: gnu-win32@cygnus.com | Subject: Re: Too Much Mail Volume | | On Sun, 30 Nov 1997, Peter Yeh wrote: | > I second the idea of moving this mailing list to a newsgroup. | > | > >This mailing list has so much volume that I think we would be better | > >served with a newsgroup to move the traffic and to provide a FAQ. | | How about just dropping into one of the alt.* hierarchy? From what I | understand, you don't need to have any of the voting (or any particular - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Ian@kiwiplan.co.nz Mon Dec 1 20:18:00 1997 From: Ian@kiwiplan.co.nz (Ian Collins) Date: Mon, 01 Dec 1997 20:18:00 -0000 Subject: Problems with bash #! Message-ID: <07551A47EAD6D01186430060B025C6F41B6F65@ntserver.kiwiplan2.co.nz> Now I'm pulling my hair out. I'm having a problem with a script. It's a big script, and I know it should work OK (dangerous comment to make). The first line of the script is #!/bin/ksh /bin/ksh is a copy of the bash.exe in the b18 release directory (it is the latest from Sergey). When I delete the #!/bin/ksh line, the script works fine (my SHELL env variable is also /bin/ksh). The script problem is weird to say the least. It says there are syntax errors on lines with no syntax errors, and if you do a slight change to the script the errors move to a completely different place. The script in question is quite a large one, and unsurprisingly I can't reproduce it on a small script. I have CYGWIN_TTY set to 1. The filesystem is not mounted as binary (that makes it even worse). It is mounted as, e: / native no-mixed,text!=binary I am running on NT4 SP3. (OK, the obvious solution is to just take the #!/bin/ksh from the script, but the script works on 6 other Unix machines, and I hate changing things just for one machine. Especially when hundreds of scripts are involved). Ian Collins. KIWIPLAN NZ. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 1 20:18:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 01 Dec 1997 20:18:00 -0000 Subject: More on mingw32, make, sleep, _sleep, __sleep, and Sleep References: <199712011919.TAA34370@out2.ibm.net> Message-ID: <9712020355.AA18907@modi.xraylith.wisc.edu> vischne@ibm.net writes: > My apologies about the problems with mingw32. If you have your > mingmake.bat set your environment variables exactly right (following the > README), ming make will actually generate make.bat files from cygwin > ./configured Makefiles, and, in rare instances, will actually make GNU > makefiles without using bash or sh (but don't use a makefile that changes > directories while compiling). Exactly. However, if you replace all your "cd subdir; $(MAKE)" with "$(MAKE) -C subdir", then at least it works with .bat files. > > Command name abbreviations are allowed if unambiguous. > (gdb) run > run > Starting program: D:\mingw32\src\slrn\src/slrn.exe > bff30000:C:\WINDOWS\SYSTEM\GDI32.DLL > bff60000:C:\WINDOWS\SYSTEM\USER32.DLL > bfed0000:C:\WINDOWS\SYSTEM\ADVAPI32.DLL > 7e2e0000:C:\WINDOWS\SYSTEM\WSOCK32.DLL > bff70000:C:\WINDOWS\SYSTEM\KERNEL32.DLL > 7fd50000:C:\WINDOWS\SYSTEM\CRTDLL.DLL > > [failed reading symbols from DLL] > "C:\WINDOWS\SYSTEM\COMCTL32.DLL": error reading line numbers > > 7eb40000:C:\WINDOWS\SYSTEM\SHELL32.DLL > > [failed reading symbols from DLL] > "C:\WINDOWS\SYSTEM\SHLWAPI.DLL": error reading line numbers > > 61000000:C:\WINDOWS\SYSTEM\URL.DLL > slrn 0.9.4.3 (Dec 1 1997 14:06:42) > > Program received signal SIGSEGV, Segmentation fault. > 0x438c51 in _image_base__ () > (gdb) > This unfortunately doesn't tell me anything at all as to where it died and why. The gdb _image_base problem is well known; whenever you try to step over a function that is exported by one of the Win32 DLLs, gdb for some reason gets into _image_base instead and you have use "finish" to get out of it. From the looks of it, it might dying in a call to a routine exported by URL.DLL, but that's just wild guess. In this case, my suggestion is to move up one frame ("up") and then examine your code. I'm assuming that you've built everything with debugging on, so it will give you a better idea what's going on. At the very least, you'll see the parameters the to routine that's dying and maybe the Win32 API docs will tell you why it's dying. Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Ian@kiwiplan.co.nz Mon Dec 1 20:18:00 1997 From: Ian@kiwiplan.co.nz (Ian Collins) Date: Mon, 01 Dec 1997 20:18:00 -0000 Subject: gdb and isatty call Message-ID: <07551A47EAD6D01186430060B025C6F41B6F67@ntserver.kiwiplan2.co.nz> When I run gdb, isatty always says I do not have a tty. For example, #include main() { extern int isatty(); FILE *lf; lf = fopen("/usr3/ian/fubar", "w+"); fprintf(lf, "started\n"); if (isatty(fileno(stdin))) { fprintf(lf, "interactive\n"); } else { fprintf(lf, "batch\n"); } } That works fine (or at least as expected) when run interactively or in batch, but if I run it in gdb (with or without the -nw flag), isatty thinks I am in batch. Any ideas? (I'm running NT 4 SP3 with Gnu B18 with latest Sergeys patches). Ian Collins. KIWIPLAN NZ. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From justins@iName.com Mon Dec 1 22:10:00 1997 From: justins@iName.com (Justin Smith) Date: Mon, 01 Dec 1997 22:10:00 -0000 Subject: Feedback needed on proposed cygwin feature References: Message-ID: <3483A664.395F4BD8@iname.com> Chris Faylor wrote: > > I have recently provided a patch to Cygnus to provide control of various > environment variable settings on a program by program basis. > > Geoffrey has asked that I get feedback on my changes to see if they are > something that is useful to the list as a whole. > > My changes cause a program to read the registry for options settings prior > to reading settings from the environment. So, for instance, you can do > something like this: > > cygwin -o 'c:\bin\bash.exe' binmode notitle > > That will set options in the registry which will cause bash to always > use 'binmode' as its default and to never modify the title bar. > I can see a possible problem: the change is limited to the copy of the program in that directory. If you use the above example, but have another copy of the program in a different directory, the changes will not carry over, which in some, but not all, eyes could be a bad thing, since you would have to have an entry for each instance of the program you want to run. If, however, you just use the program name without the path, the changes would be global for all instances of that program, which is a little better in my opinion. But that just brings up another question: what about instances run from symlinks? -- Justin Smith, Systems Engineer with the law office of Dewey, Cheetam and Howe. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Mon Dec 1 22:51:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Mon, 01 Dec 1997 22:51:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <199712012028.PAA05036@hardy.bbc.com> >Jason Zions wrote: >It strikes me that the best place for per-binary attributed like this >would be in the binary itself. I don't know how closely cygwin's exec() >looks at the program being exec'd; if it looks at the program header, >there should be a way to squirrel the attributes away there someplace. > >Many "real Unix" systems do just this sort of thing; e.g. chatr in >HP-UX, which is used to alter a.out attributes. > >Pro: > a) attributes are *really* a part of the binary; symlinks and >hardlinks all pointing to the same bits on the disk will behave the >same, while name-based approaches may be inconsistent > > b) makes it easier for attributes to work over network mounts (when >filesystems can be mounted in different places) > >Con: > a) extracting the attributes from the binary may be slower > > b) may be hard to find a "safe place" to stash the bits Actually, Geoffrey had already mentioned this scenario in private email. I should have mentioned it in my original message. It is certainly doable using extended attributes on Windows NT. Extended attributes are not an option for Windows 95, however. I think that it would end up being slightly slower than using the registry since the registry options stand a better chance of being cached than a file's attributes. I agree that it probably would be possible to modify the binary header. I don't have the expertise to do so, however. If anyone else can volunteer a way to store arbitrary data in PE binary I'll be happy to incorporate it. Also, in actuality, the binary does not know that it has been invoked via a symbolic link since this is a cygwin affectation. If you have two files hard linked together you might actually *want* to have separate settings for them. cgf - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jazz@softway.com Mon Dec 1 22:51:00 1997 From: jazz@softway.com (Jason Zions) Date: Mon, 01 Dec 1997 22:51:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <199712012028.PAA05036@hardy.bbc.com> Message-ID: <34833D90.3C6DB7B9@opennt.com> > Extended attributes are not an option for Windows 95, however. Didn't Sergey's Coolview extensions provide a mechanism for storing extended attributes on FAT filesystems and under Win95? Or am I misremembering things? Jason - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Mon Dec 1 22:51:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Mon, 01 Dec 1997 22:51:00 -0000 Subject: Feedback needed on proposed cygwin feature References: Message-ID: <6j3hs3X$pfB@mike.franken.de> Hi Chris, [...] > cygwin -o 'c:\bin\bash.exe' binmode notitle Where and how will these values get stored in the registry ? > [...] >Which leads into another source of discussion. I would prefer something >like this for setting these options: > > set CYGWIN32=binmode notitle tty I'd like this way, too. [...] >Here are the pros and cons as we see them so far: > >1) Registry [...] > >2) CYGWIN32 Uhm, perhaps I didn't get the point, but why is it necessary to list pros and cons for this ? I'd like to see both possibilities in the next version of cygwin and whoever wants can use one or both of them. If one leaves the registry alone, only the env vars get used. Perhaps it would be useful to have another env var available like CYGWIN_REGISTRY=0|1 to indicate, whether one wants to use the registry or not - this could speed up things and would allow to temporarily disable the registry entries. Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Mon Dec 1 22:51:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Mon, 01 Dec 1997 22:51:00 -0000 Subject: Too Much Mail Volume References: <01bcfdf8$56371b20$8a82400c@hypercow> Message-ID: <6j3hsQAppfB@mike.franken.de> Hi, >I second the idea of moving this mailing list to a newsgroup. No. This isn't a high traffic mailing list compared to a lot of other lists I am on. Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tjump@cais.com Mon Dec 1 22:51:00 1997 From: tjump@cais.com (Theodore Jump) Date: Mon, 01 Dec 1997 22:51:00 -0000 Subject: Feedback needed on proposed cygwin feature References: Message-ID: <34861520.493395265@smtp> (First off, my apologies to Chris as he's going to get this text twice as I forgot to send it to the list only.) Reflexively (meaning that I haven't coded or ported anything or done extensive enough work inside the cygwin environment to have a real strong opinion) I would say that I vote for your contribution be added in to the "next" official release and I vote for the options to be done in this way: 1. "cygwin-wide" Defaults are stored in the registry, and should be stored in the HKEY_LOCAL_MACHINE or HKEY_CURRENT_CONFIG hives to facilitate sharing among all "users" of the machine. 2. "program-specific" Defaults are stored in the registry underneath same hive as #1 above. This would be analagous to the "App Paths" registry keys that are found in: HKEY_LOCAL_MACHINE/Software/Microsoft/Windows/App Paths/ Which specify modified paths to use for specific applications, if they need something other than the default system-wide path. 3. The environment variable CYGWIN32 operate as Chris documented, and if present overrides the defaults are stated in 1 and 2 above. 4. The 'cygwin' command operated as Chris documented, any options set there overriding anything set by the previous three options. In all cases it could result in the existance of the CYGWIN32 env-var as Chris documented so that applications that are cygwin-ized could check that for safety and for runtime options. It may be desireable to allow an option at the registry level to disable usage of the 2nd two methods - for paranoia or security reasons (e.g.: to prevent something launched from inetd or some other daemon from altering a setting for a specific program as mandated by the registry). I do recognize that this is not the simplest of approches, however it does facilitate both site-mandated configurations (particularly if there is a registry-stored option for disabling 3 & 4) and user- or program-specified additions/overrides. By maintaining the support for the CYGWIN32 var it makes it trivial to set/use that via scripts and batch files without modifying the system registry (a not-quite-fast option) and without potentially huge command lines. Just my 2c. -Ted - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Mon Dec 1 22:51:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Mon, 01 Dec 1997 22:51:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <199712012246.RAA05721@hardy.bbc.com> >> Extended attributes are not an option for Windows 95, however. > >Didn't Sergey's Coolview extensions provide a mechanism for storing >extended attributes on FAT filesystems and under Win95? Or am I >misremembering things? Just under Windows NT, I think. I do remember that people using FAT on Windows NT complained because cygwin was creating a large file to hold the extended attributes, also. That may be another mark against using them. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 1 22:51:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 01 Dec 1997 22:51:00 -0000 Subject: ash (small sh-like shell)? Message-ID: <9712020651.AA19083@modi.xraylith.wisc.edu> Has anyone done a serious port of ash to gnu-win32? I did quick hack to get it to build and it seems to work ok for using it as shell for make and such, but there are tons of problems when it comes to win32<-->Posix pathname handling and such, which I'm scared to death of. I'm looking for something tiny and light that can act as a shell for doing simple things, nothing as grand as bash. Suggestions welcome. Thanks Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From anthonm@tams.com.au Mon Dec 1 23:05:00 1997 From: anthonm@tams.com.au (Michael Anthon) Date: Mon, 01 Dec 1997 23:05:00 -0000 Subject: Access to SCSI DAT Tape Drive Message-ID: <3120D53DC9D5D011A59A00C0F0163833036C11@server1> Does anyone know if it is possible to access a SCSI tape drive from within Cygwin. I found a thread regardiing this in the archives but it degenerated into a discussion of the virtues of being multi-lingual without telling me what I need to know. I have a Sony DAT Drive. It is set up as follows BUS = 0 PORT = 1 LUN = 0 ID = 3 I am not sure what device this would be under Cygwin (assuming that there is one) I also have a generic ASPI driver installed for the tape as well ( I think ). The SCSI controller is an Advansys ABP-9X0. I have a requirement to read a single (very large) file from a tape written on a unix system using dd and a 512 byte block size. I can't seem to find any way to do this under NT so am looking to Cygwin as an alternative. Regards Michael Anthon - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From noer@cygnus.com Tue Dec 2 00:17:00 1997 From: noer@cygnus.com (Geoffrey Noer) Date: Tue, 02 Dec 1997 00:17:00 -0000 Subject: Gnu-Win32 Central is up... References: <3.0.5.32.19971126094520.007a3a10@208.221.108.11> Message-ID: <199712020816.AAA25639@rtl.cygnus.com> jman wrote: > > http://www.deninc.com/~gnu-win32 > ftp://ftp.deninc.com (Dedicated to Gnu-Win32) Thanks for hosting the mirror and informational pages. I added links on the main project WWW page. I am still in need of additional mirrors, especially within the USA. Please let me know if anyone else can start one going. The beta 18 release completely swamped Cygnus' ftp server for 2 days when that one came out (97% utilization of our T1 for 36 hours straight!). It will make several people at Cygnus less unhappy if the eventual release of 19 has a lighter impact than 18 did. BTW, for those people who have been asking, I will announce a date for beta 19 when I have one. I am still busy fixing bugs as well as working on customer releases right now. Please be patient with me -- for now at least Sergey has his coolview release out which includes most of the latest Cygwin32 features/patches. Regards, Geoffrey Noer noer@cygnus.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fox@cat.nyu.edu Tue Dec 2 04:47:00 1997 From: fox@cat.nyu.edu (David Fox) Date: Tue, 02 Dec 1997 04:47:00 -0000 Subject: Do we need commercial advertisements in the gnu-win32 list? References: <009BDFEE.EA443220.29140@ifk20.mach.uni-karlsruhe.de> <3481B9AE.C48FDD48@opennt.com> <3482F866.1C470A4F@twinspot.net> Message-ID: Tomas Fasth writes: > Jason, > > I joined this mailing list because of my interest in the freely > availible gnu-win32 distribution. I am personally not a bit interested > in commercial binary-only alternatives, especially not yours since I > concider it far to expensive. I do not consider this mailing list a > proper place for you to advertise about features in your products. > Remember that you are using cygnus resources for doing this. Do you have > their consent? If you have not done it already, I suggest you to set up > your own mailing list for issues related to your own products. Well, I find the information in Jason's posts (and in the follow-ups) quite interesting and informative, even if I'm not considering using a commercial product. Just a second opinion. It sure beats "Fast Cash"! -- David Fox http://www.cat.nyu.edu/fox xoF divaD NYU Media Research Lab fox@cat.nyu.edu baL hcraeseR aideM UYN - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kkont@estia.iit.nrcps.ariadne-t.gr Tue Dec 2 05:17:00 1997 From: kkont@estia.iit.nrcps.ariadne-t.gr (Kimon Kontovasilis) Date: Tue, 02 Dec 1997 05:17:00 -0000 Subject: Feedback needed on proposed cygwin feature References: Message-ID: <9712021319.AA02630@estia.iit.nrcps.ariadne-t.gr> Dear Chris, > > I have recently provided a patch to Cygnus to provide control of various > environment variable settings on a program by program basis. > > Geoffrey has asked that I get feedback on my changes to see if they are > something that is useful to the list as a whole. > Thank you for your efforts on this. Here is my feedback, as requested: > My changes cause a program to read the registry for options settings prior > to reading settings from the environment. So, for instance, you can do > something like this: > > cygwin -o 'c:\bin\bash.exe' binmode notitle > > That will set options in the registry which will cause bash to always > use 'binmode' as its default and to never modify the title bar. > I find the feature a good idea. I think, however, that there should be a `default' registry entry, applying to programs not explicitly registered; this could be arranged, for example, by `cygwin -d binmode title'. (Without the default entry, a user will have to register each and every program individually, if environment variables are to be dispensed with.) Another issue I see important in connection with registry entries for specific programs (like the `bash' example you gave) is they way the linkage between the program and the settings is specified. If the program is identified through its full pathname (as the example on bash above indicates) then dangerous things may happen. For example, if the program is moved to a different directory, the association will be broken and the program will start using the default settings, something obviously unsatisfactory. (This problem does not arise with environment variables.) Should the linkage between program and settings use only the program's basename? Well, this would remove the side-effects following relocation of a file, but would force two programs of the same name, in different directories, to use the same settings (and no way to override this in the registry would be possible). This is less of an issue to me. Even this arrangement may not be entirely adequate; for, if the program reads the settings using argv[0] as the key to search (speculation here, since I don't know any implementation details) there will be problems in Win95, when the standard shell (COMMAND.COM) is used. This shell sets argv[0] so that the individual path components have their 8.3 `short-form' names (although the cygnus startup code replaces /-separators for \-ones). For example, on my system the `find' program gets an argv[0] equal to /GNUWIN32/H-I386~1/BIN/FIND.EXE when run under COMMAND.COM. Then, two, otherwise identical registry entries, one for find.exe and one for FIND.EXE might be required. This is, of course, unsatisfactory. > Here are the pros and cons as we see them so far: > > 1) Registry > > Pros: > > a) Don't need to mess with system environment variables. Something nice indeed. Furthermore, the variables may always be used to override things, on a per-session or a per-run basis. > > b) Allows pinpoint control of what program gets what options, e.g. > setting `tty' on for bash means that bash always starts with > a true /dev/tty. Other programs run from the command line will > not also use tty handling as seen when using the CYGWIN_TTY > environment variable. > Something also nice and not possible with env. vars. See, however, my considerations on this feature, expressed above. > c) "Solves" the problem of programs, ported from UNIX, which need > to use `binmode' by default. > A small issue here arises: what are the semantics of both using [no]binmode and the mount-settings for various directories? A possible (and preferred to me) choice is to have a program disregard mount-settings if a program-specific registry entry exists. A suggestion for precedence is harder in connection with the `default' registry entry I mentioned above. Any suggestions? Note, however, that the proper mode should also cover standard input/output, when these streams are not directed to the console. Sergey, has solved that in recent coolview versions (I believe by checking the mount entry for /, for a binary setting--is it so?). I definitely wouldn't want to see this aspect broken again. Things like `gunzip .. | tar xf -' should work, regardless of the shell used for invocation. > Cons: > > a) May slow down process startup slightly (I haven't noticed this > in simple tests, though). > There are so many other major sources of slowness, that I don't think this one will make a significant difference. > b) Contributes to creeping featuritis. > Something that we should try to keep to a minimum. I definitely wouldn't want to see twenty possible options being settable, a year from now. Discipline and discretion must be applied when using any powerful tool or mechanism. In particular, the scheme proposed should NOT be used for `sweeping under the rug' problems that should be fixed by updating the source code of individual programs. > c) The BINMODE option is nice, but may delay true port of a > product in favor of a "SET CYGWIN_BINMODE" band-aid. > For programs that their natural mode of operation is `binary' (cmp is an obvious example), I believe the program's code should enforce it explicitly; registry or environment settings are, in my opinion, just a temporary patch. (I admit that there is a `grey zone' of programs that could benefit from both file modes.) > The available options are: > > [no]binmode - Set default open mode > [no]title - Display program names on the console > title bar > [no]glob - [Do not] interpret wild cards on commands > run from CMD. I hope this will affect only invocations from within CMD or COMMAND.COM, or any (and all) non-Unix shell. While at it, let me note that I would like to see a setting like dosglob, which, when activated, would perform dos-like globbing (the sort of thing you get when you link with an appropriate module in Microsoft compilers). Again this should apply only in case the program is invoked from within CMD/COMMAND.COM. This would make globbing more transparent from within Microsoft shells (whose design is responsible for this mess in the first place), when using a mixture of cygwin and non-cygwin tools. > [no]strip_title - [Do not] strip the path component of files > names displayed in the title bar > [no]tty - Attach a UNIX-style tty to the program > > All of the above are available as normal CYGWIN_* environment variables: > > CYGWIN_BINMODE, CYGWIN_NOBINMODE > CYGWIN_TITLE, CYGWIN_NOTITLE > etc. > > These settings are similar to those found in Sergey's recent coolview > release except that it is possible to say CYGWIN_NOTITLE as well as > CYGWIN_TITLE. > What happens if one has set both CYGWIN_TITLE and CYGWIN_NOTITLE? > Which leads into another source of discussion. I would prefer something > like this for setting these options: > > set CYGWIN32=binmode notitle tty > > rather than: > > set CYGWIN_BINMODE=1 > set CYGWIN_TTY=1 > set CYGWIN_NOTITLE=1 > > My current patch allows both uses but it would probably better to stick > with one. > I definitely prefer the CYGWIN32 approach. > 2) CYGWIN32 > > Pros: > > a) Minimizes environment variable name space pollution. > > b) Mirrors proposed registry options syntax. > > c) More intuitive? > I agree with all > Cons: > > a) May be harder to test for in shell scripts. > Still easy (and possible using built-in sh features), for any competent shell programmer. An example script may be provided for helping users with little Unix experience. > b) Less intuitive? Not for me. It depends on the individual user's frame of mind. The resemblance of registry and environment syntax is, in my opinion, enough justification. Best Regards, Kimon Kontovasilis. ------------------------ Dr. Kimon Kontovasilis, National Center for Scientific Research, Institute for Informatics & Telecommunications, e-mail: kimon@cyclades.nrcps.ariadne-t.gr - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mgrant@ici.net Tue Dec 2 05:57:00 1997 From: mgrant@ici.net (Martin Carl Grant) Date: Tue, 02 Dec 1997 05:57:00 -0000 Subject: Too Much Mail Volume References: <6j3hsQAppfB@mike.franken.de> Message-ID: > >I second the idea of moving this mailing list to a newsgroup. > No. This isn't a high traffic mailing list compared to a lot of other > lists I am on. How about having someone autopost, (using some kind of elm filter), all mail which originates on this mailing list. Likewise any posts on newsgroup that haven't been sent on the mailing list could be forwarded. That way anyone could subscribe to the mailing list or read the newsgroup and not miss anything. All we need is a person to write these two scripts, who is also responsible enough to monitor them. --------------------------------------------------------------- Martin Grant "Fight me!" :-) -Martin www.ici.net/cust_pages/mgrant "Just call me The Doctor." seer@earthling.net "Fizban! Where?" --------------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 2 06:11:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Tue, 02 Dec 1997 06:11:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <199712021409.JAA00868@hardy.bbc.com> >Chris Faylor wrote: >> >> I have recently provided a patch to Cygnus to provide control of various >> environment variable settings on a program by program basis. >> >> Geoffrey has asked that I get feedback on my changes to see if they are >> something that is useful to the list as a whole. >> >> My changes cause a program to read the registry for options settings prior >> to reading settings from the environment. So, for instance, you can do >> something like this: >> >> cygwin -o 'c:\bin\bash.exe' binmode notitle >> >> That will set options in the registry which will cause bash to always >> use 'binmode' as its default and to never modify the title bar. >> > > I can see a possible problem: the change is limited to the copy of the >program in that directory. If you use the above example, but have >another copy of the program in a different directory, the changes will >not carry over, which in some, but not all, eyes could be a bad thing, >since you would have to have an entry for each instance of the program >you want to run. If, however, you just use the program name without the >path, the changes would be global for all instances of that program, >which is a little better in my opinion. I'm not sure that I agree. Actually, it could be made to check a default location, the program without the path, and the program with the path but that would certainly slow down program startup for, IMHO, minimal gain. I'm also not sure what the gain would be in keeping multiple copies of a program around. If you have the same program sitting in multiple directories it seems to me that you are a candidate for running CleanSweep. > But that just brings up another question: what about instances run from >symlinks? Symlinks are a cygwin construction. The actual program being run is always the program that they point to. GetModuleFilename returns the actual path name of the program and that is what I (currently) use. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 2 07:16:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Tue, 02 Dec 1997 07:16:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <6j3hs3X$pfB@mike.franken.de> Message-ID: In article < 6j3hs3X$pfB@mike.franken.de >, Michael Hirmke wrote: >Hi Chris, > >[...] >> cygwin -o 'c:\bin\bash.exe' binmode notitle > >Where and how will these values get stored in the registry ? > >> >[...] >>Which leads into another source of discussion. I would prefer something >>like this for setting these options: >> >> set CYGWIN32=binmode notitle tty > >I'd like this way, too. > >[...] >>Here are the pros and cons as we see them so far: >> >>1) Registry >[...] >> >>2) CYGWIN32 > >Uhm, perhaps I didn't get the point, but why is it necessary to list >pros and cons for this ? I'd like to see both possibilities in the next >version of cygwin and whoever wants can use one or both of them. If one >leaves the registry alone, only the env vars get used. >Perhaps it would be useful to have another env var available like >CYGWIN_REGISTRY=0|1 to indicate, whether one wants to use the registry >or not - this could speed up things and would allow to temporarily >disable the registry entries. I listed pros and cons because the features are not currently in any version of cygwin.dll besides mine. If the consensus is that the Cons are more serious than the Pros then I'm sure that Geoffrey will decide not to include the features in cygwin.dll. And, I'd agree with that decision. Btw, since I don't seem to have made myself sufficiently clear, I was not talking about removing environment variable support. I was talking about changing CYGWIN_* environment variables to CYGWIN32=*. I don't think there is any reason to keep both style of environment variables. That would be needlessly confusing especially since it would never be clear which had precedence. CYGWIN_REGISTRY is an interesting idea. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 2 07:22:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Tue, 02 Dec 1997 07:22:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <199712021453.JAA01002@hardy.bbc.com> >> My changes cause a program to read the registry for options settings prior >> to reading settings from the environment. So, for instance, you can do >> something like this: >> >> cygwin -o 'c:\bin\bash.exe' binmode notitle >> >> That will set options in the registry which will cause bash to always >> use 'binmode' as its default and to never modify the title bar. >> > >I find the feature a good idea. I think, however, that there should be >a `default' registry entry, applying to programs not explicitly >registered; this could be arranged, for example, by > > `cygwin -d binmode title'. > >(Without the default entry, a user will have to register each and every >program individually, if environment variables are to be dispensed with.) Actually, I have implemented a registry default (should have mentioned it) but I'm not talking about dispensing with environment variables. They are still very useful in certain situations. Environment variables take precedence over registry options. >Another issue I see important in connection with registry entries for >specific programs (like the `bash' example you gave) is they way the >linkage between the program and the settings is specified. If the program >is identified through its full pathname (as the example on bash above >indicates) then dangerous things may happen. For example, if the program is >moved to a different directory, the association will be broken and the >program will start using the default settings, something obviously >unsatisfactory. (This problem does not arise with environment variables.) I don't see any way around this other than to either use NT extended attributes (which don't work on Windows 95) or to modify the binary header which I don't have the expertise to do. Or, continue to use environment variables... >Should the linkage between program and settings use only the program's >basename? Well, this would remove the side-effects following relocation >of a file, but would force two programs of the same name, in different >directories, to use the same settings (and no way to override this in the >registry would be possible). This is less of an issue to me. Forgoing the the path part of the program name is a possibility. I don't like it because it makes debugging a little program harder. >Even this arrangement may not be entirely adequate; for, if the program reads >the settings using argv[0] as the key to search (speculation here, since >I don't know any implementation details) there will be problems in Win95, when >the standard shell (COMMAND.COM) is used. This shell sets argv[0] so that the >individual path components have their 8.3 `short-form' names (although the >cygnus startup code replaces /-separators for \-ones). For example, on my >system the `find' program gets an argv[0] equal to > > /GNUWIN32/H-I386~1/BIN/FIND.EXE > >when run under COMMAND.COM. Then, two, otherwise identical registry entries, >one for find.exe and one for FIND.EXE might be required. This is, of course, >unsatisfactory. I use GetModuleFilename to find the name of the running program. This should always return a consistent name. >> c) "Solves" the problem of programs, ported from UNIX, which need >> to use `binmode' by default. >> >A small issue here arises: what are the semantics of both using [no]binmode >and the mount-settings for various directories? A possible (and preferred >to me) choice is to have a program disregard mount-settings if a >program-specific registry entry exists. A suggestion for precedence is >harder in connection with the `default' registry entry I mentioned above. >Any suggestions? Note, however, that the proper mode should also cover >standard input/output, when these streams are not directed to the console. >Sergey, has solved that in recent coolview versions >(I believe by checking the mount entry for /, for a binary setting--is it so?). >I definitely wouldn't want to see this aspect broken again. Things like >`gunzip .. | tar xf -' should work, regardless of the shell used for invocation. This is tricky, I agree. For the way that it is currently implemented, the 'binmode' setting merely changes the program's default setting for opens which do not specify binmode or textmode. That means that if you specifically set O_TEXTMODE (or whatever) in an open it will override the binmode setting. If you have a disk mounted as textmode but have `binmode' set, then a file opened on that disk will be opened as textmode. >> Cons: >> >> a) May slow down process startup slightly (I haven't noticed this >> in simple tests, though). >> >There are so many other major sources of slowness, that I don't think this >one will make a significant difference. Yeah, but I don't really want to contribute to slowing down cygwin. Every nifty feature added to cygwin may have a price. >> b) Contributes to creeping featuritis. >> >Something that we should try to keep to a minimum. I definitely wouldn't want >to see twenty possible options being settable, a year from now. >Discipline and discretion must be applied when using any powerful tool or >mechanism. In particular, the scheme proposed should NOT be used for >`sweeping under the rug' problems that should be fixed by updating the source >code of individual programs. > >> c) The BINMODE option is nice, but may delay true port of a >> product in favor of a "SET CYGWIN_BINMODE" band-aid. >> >For programs that their natural mode of operation is `binary' (cmp is an >obvious example), I believe the program's code should enforce it >explicitly; registry or environment settings are, in my opinion, >just a temporary patch. (I admit that there is a `grey zone' of programs >that could benefit from both file modes.) > >> The available options are: >> >> [no]binmode - Set default open mode >> [no]title - Display program names on the console >> title bar >> [no]glob - [Do not] interpret wild cards on commands >> run from CMD. >I hope this will affect only invocations from within CMD or COMMAND.COM, >or any (and all) non-Unix shell. Yup. >While at it, let me note that I would like to see a setting like >dosglob, which, when activated, would perform dos-like globbing (the >sort of thing you get when you link with an appropriate module in >Microsoft compilers). Again this should apply only in case the program >is invoked from within CMD/COMMAND.COM. This would make globbing more >transparent from within Microsoft shells (whose design is responsible >for this mess in the first place), when using a mixture of cygwin and >non-cygwin tools. I'm not sure how this would work. Do you mean that you'd actually like to have things like c:\bin\*foo.exe be equivalent to c:\bin\*.exe and you'd like to disallow c:\*\foo.exe? >> [no]strip_title - [Do not] strip the path component of files >> names displayed in the title bar >> [no]tty - Attach a UNIX-style tty to the program >> >> All of the above are available as normal CYGWIN_* environment variables: >> >> CYGWIN_BINMODE, CYGWIN_NOBINMODE >> CYGWIN_TITLE, CYGWIN_NOTITLE >> etc. >> >> These settings are similar to those found in Sergey's recent coolview >> release except that it is possible to say CYGWIN_NOTITLE as well as >> CYGWIN_TITLE. >> > >What happens if one has set both CYGWIN_TITLE and CYGWIN_NOTITLE? Currently, CYGWIN_TITLE would win because the Win32 environment is maintained in sorted order and CYGWIN_TITLE occurs after CYGWIN_NOTITLE. If this is an issue, I could disable the CYGWIN_NO* environment variables. Thanks very much for the feedback. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mka@redes.int.com.mx Tue Dec 2 07:22:00 1997 From: mka@redes.int.com.mx (Michael Anderson) Date: Tue, 02 Dec 1997 07:22:00 -0000 Subject: Too Much Mail Volume References: Message-ID: <348427C4.7EE8E65E@redes.int.com.mx> If this mailing list does get moved to a newsgroup, the mailing list should continue also. Many ISPs outside the USA don't offer newsgroups, but I suspect that all have email. -- Mike Anderson mka@redes.int.com.mx Guanajuato, GTO, Mexico - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mark.buckingham@us.nestle.com Tue Dec 2 07:22:00 1997 From: mark.buckingham@us.nestle.com (Buckingham, Mark, SOLON, NUSA) Date: Tue, 02 Dec 1997 07:22:00 -0000 Subject: bash problem. (Am I missing something?) Message-ID: I downloaded the coolview patch, and have run into a strange problem with bash: let's say I have a file called test.sh: ------------------------------ #!/bin/sh export TESTVAR=1 ------------------------------ If I try to run it, (after chmod-ing and all that...) like this: sh-2.01$ . ./test.sh I get this error: sh: ./test.sh: error 0 And the TESTVAR variable isn't defined. Can anyone tell me what I'm doing wrong? Thanks. -------------- next part -------------- A non-text attachment was scrubbed... Name: bin00000.bin Type: application/vnd.ms-tnef Size: 1709 bytes Desc: not available URL: From cgf@bbc.com Tue Dec 2 07:23:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Tue, 02 Dec 1997 07:23:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <971201101031.2680056c@d0tng.fnal.gov> Message-ID: In article < 971201101031.2680056c@d0tng.fnal.gov >, wrote: > D0TNG (D0 AXP cluster) @FNAL, 1-DEC-1997 >--------------------------------------------------------------------------- >> From: SMTP%"cgf@bbc.com" 1-DEC-1997 06:08:43.18 >> To: JONCKHEERE >> CC: >> Subj: Feedback needed on proposed cygwin feature >> >> I have recently provided a patch to Cygnus to provide control of various >> environment variable settings on a program by program basis. > >Is this patch generally available yet? If so where? If not when can we >expect to see it? I submitted the patch for B19 but it was "rejected" due to the radical nature of the change. Geoffrey asked that I get feedback from the group before he committed to adding it. >> My changes cause a program to read the registry for options settings prior >> to reading settings from the environment. So, for instance, you can do >> something like this: >> >> cygwin -o 'c:\bin\bash.exe' binmode notitle >> >> That will set options in the registry which will cause bash to always >> use 'binmode' as its default and to never modify the title bar. >> >> The available options are: >> >> [no]binmode - Set default open mode >> [no]title - Display program names on the console >> title bar >> [no]glob - [Do not] interpret wild cards on commands >> run from CMD. >> [no]strip_title - [Do not] strip the path component of files >> names displayed in the title bar >> [no]tty - Attach a UNIX-style tty to the program >> >> All of the above are available as normal CYGWIN_* environment variables: >> >> CYGWIN_BINMODE, CYGWIN_NOBINMODE >> CYGWIN_TITLE, CYGWIN_NOTITLE >> etc. >> >> These settings are similar to those found in Sergey's recent coolview >> release except that it is possible to say CYGWIN_NOTITLE as well as >> CYGWIN_TITLE. > >Can the list of variables be expanded by the user? I'd like to be able >to do things like setting PYTHONPATH (or just PATH) to an arbitrary >value depending on the exe, and the version of the exe, I'm running. Currently the list of options is fixed. If you want to add arbitrary environment variables from the registry then possibly that would be another mechanism entirely. Maybe something like: cygwin -e c:\usr\local\bin\python.exe PATH='/python/bin:/bin' LIB='/python/lib' I think it's a good idea, though. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 2 07:41:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Tue, 02 Dec 1997 07:41:00 -0000 Subject: Too Much Mail Volume References: <01bcfdf8$56371b20$8a82400c@hypercow> <6j3hsQAppfB@mike.franken.de> Message-ID: In article < 6j3hsQAppfB@mike.franken.de >, Michael Hirmke wrote: >Hi, > >>I second the idea of moving this mailing list to a newsgroup. > >No. This isn't a high traffic mailing list compared to a lot of other >lists I am on. I vote emphatically against moving this to a newsgroup. I don't think that the traffic justifies it. Also, we'll all be subject to much more spam both from the newsgroup spammers and from the spammers who harvest mail addresses from newsgroup headers. If cygwin or some well-connected site wanted to set up a private NNTP server to host a private newsgroup, however, I think that would be a great idea. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 2 07:41:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Tue, 02 Dec 1997 07:41:00 -0000 Subject: less or more, where are them? References: <01BCFE6C.42CBB220@stoxo18.amd.com> Message-ID: In article < 01BCFE6C.42CBB220@stoxo18.amd.com >, Fermin Bueno wrote: >Hello all, > >Do anyone knows where can I find more or less? I have installed gnu-win32 in my NT machine, I can only find zmore which seem to be a shell-script not related to more and less.exe, but when I execute it nothing happens. Can anyone help me? > >Thanks in advanced I have a port of less on my web page: http://www.tiac.net/users/cgf It needs a coolview version of cygwin.dll to work correctly: http://www.lexa.ru/sos -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 2 07:41:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Tue, 02 Dec 1997 07:41:00 -0000 Subject: cygnus' notion of symbolic links References: <34834DFF.7566F4CF@synopsys.com> Message-ID: In article < 34834DFF.7566F4CF@synopsys.com >, B. Elizabeth Trojan wrote: >Can someone explain how symbolic links work using the Cygnus >ln -s command? A file is written with a special 'cookie' at the beginning followed by a path spec. The path spec is the location of the file to which the link refers. In coolview and B19 versions of cygwin, symbolic links also have the 'system' attribute set. The file is not read at all if this attribute is not set. This speeds up checking for symbolic links. As you may be able to see from this brief description, symbolic links are entirely a cygwin construction. You can't use them with native windows programs. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From lhall@rfk.com Tue Dec 2 09:41:00 1997 From: lhall@rfk.com (Larry Hall (RFK Partners Inc)) Date: Tue, 02 Dec 1997 09:41:00 -0000 Subject: Too Much Mail Volume References: <6j3hsQAppfB@mike.franken.de> Message-ID: <3.0.3.32.19971202115812.009da9b0@ma.ultranet.com> At 08:56 AM 12/2/97 -0500, Martin Carl Grant wrote: >> >I second the idea of moving this mailing list to a newsgroup. > >> No. This isn't a high traffic mailing list compared to a lot of other >> lists I am on. > >How about having someone autopost, (using some kind of elm filter), all >mail which originates on this mailing list. > >Likewise any posts on newsgroup that haven't been sent on the mailing >list could be forwarded. > >That way anyone could subscribe to the mailing list or read the newsgroup >and not miss anything. > >All we need is a person to write these two scripts, who is also responsible >enough to monitor them. > >--------------------------------------------------------------- >Martin Grant "Fight me!" :-) -Martin >www.ici.net/cust_pages/mgrant "Just call me The Doctor." >seer@earthling.net "Fizban! Where?" >--------------------------------------------------------------- This is a good idea. May I suggest someone who is interested in seeing this mail list in a newsgroup pick up on this idea? I, for one, have no problem with the concept of this "group" as a mail list but I think it would be beneficial to those who like the newsgroup forum better to be able to access it in this form too. This idea seems to provide that nicely without forcing a complete change of forum for those that are happy with the mail list. Larry Hall lhall@rfk.com RFK Partners, Inc. (781) 239-1053 8 Grove Street (781) 239-1655 - FAX Wellesley, MA 02181 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Tue Dec 2 09:41:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Tue, 02 Dec 1997 09:41:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <6j3hs3X$pfB@mike.franken.de> Message-ID: <199712021741.EAA13772@mundook.cs.mu.OZ.AU> cgf@bbc.com (Chris Faylor) writes: >Btw, since I don't seem to have made myself sufficiently clear, I was >not talking about removing environment variable support. I was talking >about changing CYGWIN_* environment variables to CYGWIN32=*. I don't see any good reason to hard-code the `32'. Using `CYGWIN32' won't look very good on say the 64-bit NT Alpha port. -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From carvalho@penelope.dss.inpe.br Tue Dec 2 09:41:00 1997 From: carvalho@penelope.dss.inpe.br (Himilcon de Castro Carvalho) Date: Tue, 02 Dec 1997 09:41:00 -0000 Subject: Problems running gcc: cpp not found Message-ID: <01bcff43$149daa20$2010a396@carvalho1.dss.inpe.br> Hello all, I'm having problems trying to put gcc to work. I've read the FAQ and Mailing List archives and no answer was found. After having installed, created c:\bin (and copied sh.exe to it) and c:\tmp, and set the PATH and environment variables in autoexec.bat (as described in the README file), gcc continued issuing the message about CPP. Using the SET command (from dos), I could see that the GCC_EXEC_PREFIX variable was set to c:\gnuwin32\b18\H-i386~1\lib\gcc-lib\ . Is it correct to have the long name H-i386-cygwin32 cut like that? I've tryed also to use normal slashes "/", and substituting c:\ with //c/, with no result. Any idea of what is going on? Thanks a lot, Himilcon Carvalho - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From marcus@bighorn.dr.lucent.com Tue Dec 2 09:41:00 1997 From: marcus@bighorn.dr.lucent.com (marcus@bighorn.dr.lucent.com) Date: Tue, 02 Dec 1997 09:41:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <199712021608.JAA04416@chorus.dr.lucent.com> Jason Zions wrote: >It strikes me that the best place for per-binary attributed like this >would be in the binary itself. I don't know how closely cygwin's exec() >looks at the program being exec'd; if it looks at the program header, >there should be a way to squirrel the attributes away there someplace. If there isn't anyplace in the file header that can be used to store this info away in that doesn't upset NT/W95, or if it is costly to open the object file whenever a program starts execution, perhaps it would be possible to provide some global variables in the crt0.o module for controlling various attributes. The tool to set the attributes would have to go through the name list to find the variables, then find the section to modify. Shouldn't be too bad, and the performance of the attribute setting program isn't critical anyway, but it would require that the file not be fully stripped. On the other hand, perhaps a data section could be defined that contains an array of attribute flags, or even space for ASCII attribute names. This section (perhaps .cygwin_attributes) could be initially provided by the crt0.o file, then a pointer to it is passed to the DLL on process startup. Using an array of flags allows easy allowance for significant expansion, but at the cost of having a fixed structure (i.e. the first flag corresponds to the force-binary attribute, the second flag...) while the ASCII attribute name requires a "big enough" fixed size buffer be pre-allocated to hold the list of attributes, and what happens when an attribute name is not recognized by the DLL? Of course, perhaps the attribute setting program could change the section size, etc.... cgf replies to Jason: > Also, in actuality, the binary does not know that it has been invoked via > a symbolic link since this is a cygwin affectation. If you have two > files hard linked together you might actually *want* to have separate > settings for them. You might want separate settings in some cases, but you might also get very confused by separate settings as well. If this is supposed to be "Unix-like", then it would be least surprising to attach the attribute to the file instead of the name. Other issues with using name: The full path name must be unambiguously stored with the attribute and expanded for execution. Otherwise, some peculiar things may not work correctly, such as invoking the program as "../bin/program", using a symbolic link to a directory "symlinktobin/program", etc. If the attributes are not consistently found whenever a program is run, the program will have inconsistent behaviour. If attributes are set for a program, then the program is moved to another location or given another name, it looses the attributes. Furthermore, if a new program is created (perhaps months later) with the same name and location, it will mysteriously gain the attributes. These problems do apply to symbolic links currently, so they aren't new, but they can be confusing particularly since they may have subtle effects on the execution of the program instead of a hard failure to follow a link. marcus hall Lucent Technologies - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From len.langsdorf@bentley.com Tue Dec 2 10:27:00 1997 From: len.langsdorf@bentley.com (Len Langsdorf) Date: Tue, 02 Dec 1997 10:27:00 -0000 Subject: CVS and SSH Message-ID: <0DEA4AD566C0CF11911400001B16BB80029B4CF6@extonpo.bentley.com> Yes, but does it handle key database and if not how couild you use with cvs? > -----Original Message----- > From: Martin Hamilton [SMTP:martin@mrrl.lut.ac.uk] > Sent: Friday, November 28, 1997 3:59 AM > To: gnu-win32@cygnus.com > Subject: Re: CVS and SSH > > -----BEGIN PGP SIGNED MESSAGE----- > > > Chris Cowan writes: > > | Here's a URL you may be interested in > | ftp://ftp.net.lut.ac.uk/martin/ssh-win32/ > > Even with b18 there seem to be enough missing bits that ssh only kind > of works, sometimes... If anyone's interested in (a) hacking on ssh > to make it work under gnu-win32 or (b) hacking on a GPL'd version of > ssh implemented from scratch from the IETF's SSH group's Internet > Drafts, you might want to join the 'psst' list - mail to > psst-request@net.lut.ac.uk with a message body of 'subscribe'. > > Cheerio, > > Martin > > > > -----BEGIN PGP SIGNATURE----- > Version: 2.6.3i > Charset: noconv > > iQCVAwUBNH6HztZdpXZXTSjhAQGuJAQAgFNcHEYFfL9WqhD8JLL9XP20benA7CyE > Ejt2TAgRvDw+Qt6y6xqRMaSNMFz9bhaz9sJXjVR8y6GYoBpO61Cj4jfgsRx1ZdUI > GFe02Z2nURpj7mi0j7F9S351/jydYJ7TJ1ySPP9zUCkvPdxpTlsmEz+6ODBGiK+B > 37+z2DHYnbc= > =EAjR > -----END PGP SIGNATURE----- > > - > For help on using this list (especially unsubscribing), send a message > to > "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ggp@informix.com Tue Dec 2 10:27:00 1997 From: ggp@informix.com (Guy Gascoigne - Piggford) Date: Tue, 02 Dec 1997 10:27:00 -0000 Subject: Shell-Here Explorer context menu Message-ID: <3.0.3.32.19971202102708.009bc4b0@pop.pdx.informix.com> Have you ever wanted to simply click on a directory in the Windows explorer and open a bash shell in that directory, well now you can. Save the attached file as ShellHere.inf, then right click on it and select Install. This then makes a few registry changes and the next time that you right click on a directory or a drive on explorer you will see a "Shell Prompt Here" option. Voila. I took the concept from the Microsoft Dos-Here powertoy and then adapted it to work with bash, though the idea could easily be extended to work with any shell. Hopefully someone else will find this useful, if anyone wants to put this on a web site so that it remains generally available then please go ahead. Guy From vischne@ibm.net-nospam Tue Dec 2 10:27:00 1997 From: vischne@ibm.net-nospam (root) Date: Tue, 02 Dec 1997 10:27:00 -0000 Subject: P. S. on using mingw32 with cygwin32 and doing libslang properly - slw32tty.c [0/1] Message-ID: <199712021803.SAA35120@out2.ibm.net> Content-Type: multipart/mixed; Boundary="*- Boundary FhypIN3xBe5kvrZF/hr" --*- Boundary FhypIN3xBe5kvrZF/hr Content-Type: Text/Plain; charset=US-ASCII Attached is a simplified version of slw32tty.c for use with the slang library. P. S. on the simplified slw32tty.c: That `getchar()' in slw32tty.c should read `_getch()', and people using slang on Win32 should get some advice: 1. If you compile slang with libtermcap, libtermcap has to have termcap.c edited so that the code has the _exact_ address of the `termcap' file in file notation that the particular system can follow. Otherwise, slang programs linked using libtermcap will crash. 2. You need to load ansi.sys using your system config.h file. Without ansi.sys, your dos window will have lots of escape sequences printed in it, but no graphics. With ansi.sys, you may get a full-screen window with double-wide characters when you run the worm.exe test program, but it will be in color, and it will do the wormy things. This probably means that autoexec.bat has to have a `set TERM=ansi' and an appropriate `c:\etc\termcap' file with an ansi entry 3. You could get rid of sldostty.c and tell people to use slw32tty.c instead. You could also get rid of the other dos file, and use the file list in modules.unx, minus slunix.c and slutty.c, plus slw32tty.c --*- Boundary FhypIN3xBe5kvrZF/hr - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ryl@innovix.com Tue Dec 2 10:27:00 1997 From: ryl@innovix.com (Raymond Y. Lillard) Date: Tue, 02 Dec 1997 10:27:00 -0000 Subject: Newsgroup proposal (was: Too Much Mail Volume) References: Message-ID: > I don't like the alt.* idea. I don't get (and don't want) any of the > alt.* groups, so putting it there puts it out of my reach. IMHO, the only > stuff that belongs in alt.* is that which no one wants anyway. If this > list falls into that category, then it should not be moved to a newsgroup. I heartliy agree with the above opinion regarding alt.* newsgroups. I see two solutions to the (problem?) of too much mail which could be implemented by those who feel they have such a problem. These are solutions which they can implement themselves or with assistance from their local systems administrator. 1. Get a better mail reader. There are several readers which can sort incoming mail. Setup the incoming mail filters to drop all mail from this list into a gnuwin32 mailbox. 2. Gateway this list to a locally distributed newsgroup. You might need help from your local sysadm. Those who still feel they get too much mail from this group should consider unsubscribing. There are two logical places where support (as in general agreement) for a newsgroup might be found. They are: * comp.os.ms-windows.gnu-win32 * gnu.gnu-win32 An additional possiblilty for those who enjoy bearding the lion in his den is: * microsoft.public.gnu-win32 BTW, most of the Microsoft Windows related traffic is now in the "microsoft.public" heirachy rather than anything under "comp.os". If anybody cares, my opinion is that the world will care about gnu-win32 for about 3 more years. If a newsgroup is made it will then be as popular as "comp.sys.att" or "comp.os.cpm". After that time those of us who know how the world could have been will an insignificant minority. Face it, the raison d'etre for gnu-win32 is that there presently is a significantly sized group of poeple who are struggling to make the transition from Unix and its work-a-likes. In about three years time, we will all have gone over to the dark side or started selling real estate. Those of us who leave to sell real estate will have a finely tuned Linux box up and cooking at home. It will be secreted in a closet or in the basement where none of our acquaintances will see it, just a few special and trusted old friends. May all beings be happy, Ray - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From shebs@cygnus.com Tue Dec 2 12:16:00 1997 From: shebs@cygnus.com (Stan Shebs) Date: Tue, 02 Dec 1997 12:16:00 -0000 Subject: GNU debugger References: Message-ID: smith@sunserv.kfki.hu (Kovacs Jozsef) writes: > I'd like to know whether the GNU debugger works with > Microsoft Compiled COFF object file format. > > I tried and it didn't work. It was my fault or GDB > really can't debug a program which was compiled by > MS compiler. GDB can read the files, but not the MS-specific debug information. Some MS compilers can generate standard COFF debug info, but that only works for C, not C++. > If not, do you plan to make GDB-WIN32 compatible with > MS compiler? At present, we don't know of anyone working on this addition. It would be very desirable however. Stan Shebs Cygnus Solutions shebs@cygnus.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From wcaskey@cmh.edu Tue Dec 2 12:16:00 1997 From: wcaskey@cmh.edu (Caskey, William) Date: Tue, 02 Dec 1997 12:16:00 -0000 Subject: GNU-Win32 Installation Problem Message-ID: <119D55D6D69FD011AB9B000092966BB411A173@EXCHANGE2> > I downloaded Beta 18 and installed it on my Windows NT sytem yesterday > along with Ming32 (unfortunately without reading the instructions > completely). Today, I deleted everything and started over. I installed > GNU Win32 and adjusted the environment variables as follows (per your > readme.txt in the Minimalist version): > > PATH = (added c:\gnuwin32\b18\H-i386-cygwin32\bin; > c:\gnuwin32\b18\tcl\bin to the PATH line) > TCL_LIBRARY = c:\gnuwin32\b18\tcl\lib\tcl7.6 > GDBTK_LIBRARY = c:\gnuwin32\b18\share\gdbtcl > GCC_EXEC_PREFIX = c:\gnuwin32\b18\H-i386-cygwin32\lib\gcc-lib > LIBRARY_PATH = c:\gnuwin32\b18\H-i386-cygwin32\lib > > I created and compiled the 'hello' program but compilation was > interrupted with the following message: > > gcc: installation problem, cannot exec 'cpp': no such file or > directory. > > Is there more configuration I need to do? I browsed the subdirectories > and couldn't find a cpp executable, but did find c++.exe and > c++filt.exe. I've been using Visual C++ and Visual Basic for > development, but now that I have a Linux installation to support, I > need a new compiler. Thanks for your help. > > *************************************** > * Bill Caskey > * > * Director, Research & Grants Administration * > * The Children's Mercy Hospital * > * 2401 Gillham Road * > * Kansas City, MO 64108 * > * Voice: (816) 234-3879 * > * FAX: (816) 855-1982 * > * E-mail: wcaskey@cmh.edu * > *************************************** > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net-nospam Tue Dec 2 12:23:00 1997 From: vischne@ibm.net-nospam (root) Date: Tue, 02 Dec 1997 12:23:00 -0000 Subject: P. S. on using mingw32 with cygwin32 and doing libslang properly - slw32tty.c [1/1] Message-ID: <199712021803.SAA15952@out2.ibm.net> Content-Type: Application/octet-stream; name=slw32tty.c Content-Transfer-Encoding: x-uue begin 755 slw32tty.c M+RH@0V]P>7)I9VAT("AC*2`Q.3DR+"`Q.3DU($IO:&X@12X@1&%V:7,-"B`J M($%L;"!R:6=H=',@PT*?0T*#0H-"B\J+2TM M+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM M+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2I<#0H@*B`@1G5N8W1I;VXZ"6EN="!3 M3&%N9U]I;FET7W1T>2`H:6YT(&-A;F-E;%]C:&%R+"!I;G0@;F]?9FQO=U]C M;VYT2`H=F]I9"D[#0H@*@T*("H@=V%I="!F M;W(@86YD(&=E="!T:&4@;F5X="!A=F%I;&%B;&4@:V5Y7-T2"D)/3X)1&5L("@Q,C0%XS("T@82!P8R!.54P@8VAA7,@ M87)E('!R969I>&5D(&)Y(&$@;G5L;"!C:&%R86-T97(-"EPJ+2TM+2TM+2TM M+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM M+2TM+2TM+2TM+2TM+2TM+2HO#0IU;G-I9VYE9"!I;G0@4TQS>7-?9V5T:V5Y M("AV;VED*0T*>PT*"7)E='5R;B!?9V5T8V@H*3L-"GT-"@T*+RHM+2TM+2TM M+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM M+2TM+2TM+2TM+2TM+2TM+2TM*EP-"B`J("!&=6YC=&EO;CH)=F]I9"!33&%N M9U]S971?86)O I'm having trouble porting an application from UNIX to NT 4.0, and I've been trying to use the cygwin32.dll to emulate the UNIX layers. My problem is that I also need to use some of the .LIB libraries provided by Microsoft (MSVC++ 4.0), and they don't seem to want to play nice together. I've gone through this list's archives and tried several things, but haven't been able to get everything to work together. I've seen several places that say that if you want help, you need to be specific, so this may be a little long. It says in the documentation, that to be able to use both kinds of modules, I will have to use link.exe instead of ld. I've found this to be true. So to start, I wrote a small dummy program with some calls to cygwin functions that I need and try to link it with link.exe. So I write a program (end of this message), and link it with link.exe (and crt0.o, libcygwin.a, libkernel32.a libgcc.a ) and I get some undefined externals, namely __data_start__, __data_end__, __bss_start__, __bss_end__. Looking through the mailing list archives, I find that as long as I'm not using fork(), I don't need these, so I dummy them out, link fine with link.exe, but when I run, I never get to my main(). I look at the mailing list archives and see that a small #ifndef to some gas code (obj-coff.c) might help, but does not seem to work for me. So I figure that since my main isn't getting called, I'd try /ENTRY:mainCRTStartup. Still no call to my main. Ok, maybe Microsoft's startup code is easier to deal with, so after checking the mailing list archives, I find that I can create a .LIB from a .DLL using dumpbin and lib, so I apply that process to cygwin.dll and create a cygwin.lib. I link my program with cygwin.lib, kernel32.lib, user32.lib, and libcd.lib, and I find that my main() does get called. getenv() works, gettimeofday() works, but when I try fileno() on stdin and/or stdout, I get the same result, 0. When I take that result and pass it to tcgetattr() (with a pointer to a termios struct), I get an access violation : The Instruction at 0x10023893 referenced memory at 0x00000054. The memory could not be "read". So it looks to me like stdout and/or stdin is being set up differently by the MS startup code than cygwin expects. So if I use cygwin startup code, I can't find my main(), and if I use MS startup code, stdin and stdout are unavailable. Has anyone succesfully used .LIBs and cygwin calls in an application? Which startup code did you use? How did you link? Anyone? Thanks, Tony ----- #include #include #include #include char _data_start__, _data_end__, _bss_start__, _bss_end__; int main() { int fd; register char *termcap_name; struct termios ttybuff; struct timeval et0; int tty; printf("Starting\n"); tty = fileno(stdin); printf("fileno for stdin = %d\n",tty); tty = fileno(stdin); printf("fileno for stdout = %d\n",tty); gettimeofday(&et0,NULL); printf("%d sec %d usec\n",et0.tv_sec, et0.tv_usec); termcap_name = getenv("TERMCAP"); tcgetattr(tty,&ttybuff); printf("c_iflag = 0x%x, c_oflag= 0x%x c_cflag = 0x%x\n",ttybuff.c_iflag,ttybuff.c_oflag,ttybuff.c_cflag); } - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kasper@altisimo.com Tue Dec 2 14:04:00 1997 From: kasper@altisimo.com (Richard Kasperowski) Date: Tue, 02 Dec 1997 14:04:00 -0000 Subject: GNU indent? References: <8525654E.006842FD.00@notes.ba.hcia.com> Message-ID: <348484DE.F2FB2CC@altisimo.com> cmeis@hcia.com wrote: > > Chris Meisenzahl > 11/13/97 02:00 PM > Does anyone know where I can find a C code beautifier like GNU Indent > compiled for Win32 console mode? Or even DOS? > The only compiler I have right now is VC++ 5.0, so any source code would > have to work w/ that. > > I inherited some C code to look through that has no indenting? So my reply is a little late (half a month)... I have an 'indent' that I compiled with cygwin32. Works great. I'd be happy to post the binary on my FTP server if there's any interest out there. -- Richard Kasperowski Altisimo Computing http://www.altisimo.com/ kasper@altisimo.com Cambridge, MA, USA Tel: (617)576-1552 Fax: (617)576-2441 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jeffdbREMOVETHIS@netzone.com Tue Dec 2 15:46:00 1997 From: jeffdbREMOVETHIS@netzone.com (Mikey) Date: Tue, 02 Dec 1997 15:46:00 -0000 Subject: bash problem. (Am I missing something?) References: Message-ID: <34879d39.87374718@smtp.netzone.com> try source ./test.sh #!/bin/sh causes the script to run in a subshell, so any exports are lost when the toplevel shell resumes. at least I think that's right;^) no idea where the error comes from unless #! aren't the very first 2 chars in the file. to test shell scripts do set -xv either as the first command in the script, or at the command line, and this will echo all commands. set -axv will export the -xv setting to subshells On Tue, 2 Dec 1997 05:35:57 -0800, you wrote: >I downloaded the coolview patch, and have run into a strange problem >with bash: > >let's say I have a file called test.sh: >------------------------------ >#!/bin/sh > >export TESTVAR=1 >------------------------------ > >If I try to run it, (after chmod-ing and all that...) > >like this: > >sh-2.01$ . ./test.sh > >I get this error: > >sh: ./test.sh: error 0 > >And the TESTVAR variable isn't defined. Can anyone tell me what I'm >doing wrong? > >Thanks. > > > (jeffdbREMOVETHIS@netzone.com) delete REMOVETHIS from the above to reply Mikey Windows vs Linux is a no-win situation. Windows 95: -- 32-bit extensions and a GUI shell for a 16-bit patch to an 8-bit OS originally coded for a 4-bit processor written by a 2-bit company that can't produce 1 bit of quality. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From nadam@CAM.ORG Tue Dec 2 16:18:00 1997 From: nadam@CAM.ORG (Adam Nowak) Date: Tue, 02 Dec 1997 16:18:00 -0000 Subject: GDB crash Message-ID: <3484A565.5B69@cam.org> Hi, Here is the problem wiht GDB on Win95. This is my system and part of AUTOEXEC.BAT: Windows 95, v4.00.950 B Pentium 166 MMX, 32Mb RAM Part of my AUTOEXEC.BAT setting GNUWIN32 environment variables: SET GCC_EXEC_PREFIX=C:\gnuwin32\b18\H-i386-cygwin32\lib\gcc-lib\ SET PATH=%PATH%;C:\gnuwin32\b18\H-i386-cygwin32\bin;C:\gnuwin32\b18\tcl\bin SET PATH=%PATH%;C:\tools\vim46 SET TCL_LIBRARY=C:/gnuwin32/b18/tcl/lib/tcl7.6 SET GDBTK_LIBRARY=C:/gnuwin32/b18/share/gdbtcl As you can see above this is GNUWIN32 version B18. C++ compiler, bash shell, other tools seem to work correctly. See below GDB: can not stop at the break point. The problem does not depend on whether I use GUI or command line interface. Best Regards, Adam Nowak nadam@cam.org ------------------------------------------- C:\users\adam>gdb -nw t.exe GNU gdb 4.16.1 Copyright 1997 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i386-cygwin32"... (gdb) list 1 #include 2 3 main() 4 { 5 cout << "Hallo world" << endl; 6 } (gdb) stop at 5 (gdb) run Starting program: /users/adam/t.exe bff30000:/WINDOWS/SYSTEM/GDI32.DLL bff60000:/WINDOWS/SYSTEM/USER32.DLL 7c6c0000:/WINDOWS/SYSTEM/WSOCK32.DLL bfec0000:/WINDOWS/SYSTEM/ADVAPI32.DLL bff70000:/WINDOWS/SYSTEM/KERNEL32.DLL [failed reading symbols from DLL] "/WINDOWS/SYSTEM/COMCTL32.DLL": error reading line numbers 7fdc0000:/WINDOWS/SYSTEM/SHELL32.DLL [failed reading symbols from DLL] "/WINDOWS/SYSTEM/SHLWAPI.DLL": error reading line numbers 7cbb0000:/WINDOWS/SYSTEM/URL.DLL Hallo world Program exited normally. (gdb) DB caused an invalid page fault in module CW3215.DLL at 0137:005e3f66. Registers: EAX=0280fffc CS=0137 EIP=005e3f66 EFLGS=00010203 EBX=0492fc08 SS=013f ESP=0492fbfc EBP=0492fc28 ECX=00000001 DS=013f ESI=00000000 FS=2b6f EDX=00000020 ES=013f EDI=0060002c GS=0000 Bytes at CS:EIP: 8b 00 89 43 1c 64 67 a1 00 00 89 03 64 67 89 1e Stack dump: 02820bd8 005fbccc 00602e58 00602f71 005e32df 006039c0 0492fc04 00000000 00000000 00602f71 0492fc3c 0492fc3c 005e2547 02820bd8 00000003 005f6b8c - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From stanton@haas.berkeley.edu Tue Dec 2 16:54:00 1997 From: stanton@haas.berkeley.edu (Richard Stanton) Date: Tue, 02 Dec 1997 16:54:00 -0000 Subject: Strange error messages: Message-ID: <199712022202.OAA08242@grether.haas.berkeley.edu> Does anyone have any ideas what the following error messages mean? Richard Stanton [f:\4nt300]find e:/mail -type f -print0 | xargs -0 -e grep -n -i -e "Lati" NUL //d/gnuwin32/b18/H-i386-cygwin32/bin/xargs.exe: can not fit single argument within argument list size limit [f:\4nt300]find e:/mail -type f -print | xargs -e grep -n -i -e "Lati" NUL //d/gnuwin32/b18/H-i386-cygwin32/bin/xargs.exe: environment is too large for exec [f:\4nt300]find e:/mail -type f -print | xargs grep -n -i -e "Lati" NUL //d/gnuwin32/b18/H-i386-cygwin32/bin/xargs.exe: environment is too large for exec - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dahms@ifk20.mach.uni-karlsruhe.de Tue Dec 2 16:54:00 1997 From: dahms@ifk20.mach.uni-karlsruhe.de (dahms@ifk20.mach.uni-karlsruhe.de) Date: Tue, 02 Dec 1997 16:54:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <009BE319.3FA64320.29436@ifk20.mach.uni-karlsruhe.de> Hi Fergus, you wrote: : I don't see any good reason to hard-code the `32'. : : Using `CYGWIN32' won't look very good on say the 64-bit NT Alpha port. MS should have made the jump to >32bit at NT4, at a time several other unices already were 64bit, and provide a real useful POSIX before forcing all of us to NT, saving all this hassle! Bye, Heribert (dahms@ifk20.mach.uni-karlsruhe.de) - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dahms@ifk20.mach.uni-karlsruhe.de Tue Dec 2 16:54:00 1997 From: dahms@ifk20.mach.uni-karlsruhe.de (dahms@ifk20.mach.uni-karlsruhe.de) Date: Tue, 02 Dec 1997 16:54:00 -0000 Subject: Do we need commercial advertisements in the gnu-win32 list? Message-ID: <009BE318.17F1E240.29434@ifk20.mach.uni-karlsruhe.de> Hi David, you wrote: : Well, I find the information in Jason's posts (and in the follow-ups) : quite interesting and informative, even if I'm not considering using a : commercial product. Just a second opinion. It sure beats "Fast Cash"! My opinion, too, except I already used OpenNT. So far I remember only technical discussions, but no commercial advertisement from Softway. Bye, Heribert (dahms@ifk20.mach.uni-karlsruhe.de) - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From noer@cygnus.com Tue Dec 2 18:26:00 1997 From: noer@cygnus.com (Geoffrey Noer) Date: Tue, 02 Dec 1997 18:26:00 -0000 Subject: Updated the WWW site Message-ID: <199712030225.SAA28076@cirdan.cygnus.com> Although there isn't a lot of new information yet, I rewrote the project WWW pages last night. I think they look a bit better now and there is now room for more info... -- Geoffrey Noer noer@cygnus.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From iverson@cisco.com Tue Dec 2 20:23:00 1997 From: iverson@cisco.com (Tim Iverson) Date: Tue, 02 Dec 1997 20:23:00 -0000 Subject: Newsgroup proposal (was: Too Much Mail Volume) References: Message-ID: <199712030337.TAA05455@stars.cisco.com> Hmmm. Most of the newsgroups I read have less traffic than this mailing list, so those that say they're on bigger lists just makes me wonder about the wisdom of not moving those to newsgroups also. IMHO, mailing lists are best when the information is either confidential or of value to only a select population. This is an open list, so the former doesn't apply and judging from the volume, neither does the latter. A filtering MUA is not the answer -- we already use those. The problem is the volume itself. As it stands now, *I* am basically the expire process for my own little gnu-win32 newsgroup. I would far prefer that this was offloaded onto a real newsgroup. Anyway, I'm just whining here. I don't have the time or resources to setup a gateway, or I'd vote the old-fashioned way, and just do it. I'm hoping someone else out there does have the time. ;) There are more comments inline ... - Tim Iverson +---------------- | From: ryl@innovix.com (Raymond Y. Lillard) | Subject: Newsgroup proposal (was: Too Much Mail Volume) | To: iverson@lionheart.com (Tim Iverson) | Date: Tue, 2 Dec 1997 09:46:38 -0800 (PST) | Cc: gnu-win32@cygnus.com ... | 1. Get a better mail reader. There are several readers which can ... | 2. Gateway this list to a locally distributed newsgroup. You might ... | Those who still feel they get too much mail from this group should | consider unsubscribing. I've thought about it; 99% of this list is useless, but every once in a while, someone says something interesting. | will an insignificant minority. Face it, the raison d'etre for | gnu-win32 is that there presently is a significantly sized group | of poeple who are struggling to make the transition from Unix and | its work-a-likes. Bzzzt. Not for me. The reason I'm interested in gnu-win32 is not for the Unixen stuff -- when I want Unix, I use the real thing. No, it's for the compiler. There are no quality compilers for the windows platform. There are absolutely no quality cross-compilers. | real estate will have a finely tuned Linux box up and cooking at home. | It will be secreted in a closet or in the basement where none of our | acquaintances will see it, just a few special and trusted old friends. Heh. Not to start a religious war, but why on earth would anyone want to run Linux when they can get the real thing for free (*BSD)? Yeah, MS/NT or variants will always win the popularity contest, but it's gonna take a long time (more than a puny 3 years) before it's anywhere near as robust as a Unix system. Even Linux beats NT. - Tim Iverson iverson@lionheart-com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Tue Dec 2 20:23:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Tue, 02 Dec 1997 20:23:00 -0000 Subject: Problems with 'bash' Message-ID: <19971203042259.362.qmail@hotmail.com> Hi Henri, I wanted to give you the URL for the coolview updates. It is http://www.lexa.ru/sos ; this sight is managed by Sergey Okhapkin. As for your problem; I suspect the build of perl that you have. You might want to check the mail archives at http://www.cygnus.com/ml/gnu-win32 Goodluck, - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- PS: Hope you don't mind that I also posted this to the list. >Date: Tue, 2 Dec 1997 13:43:42 +0100 >From: periat@ens.ascom.ch (Periat Henri) >To: earnie_boyd@hotmail.com >Subject: Re: Problems with 'bash' > >Hi Earnie > >> >> > >> >> >1. Why does perl -e 'print "Hello world\n";' doesn't work ? >> >> > I't works fine on in my 'bash' on my UNIX system. >> >> > I also know that a workaround exists with 'qq', thanks to Perl, >> but >> >> > this is just a simple example. It seems that 'bash' hates '"'. >> >> > >> >> I just tried the perl example you gave using virgin b18 and the perl >> package I downloaded from the gnu mirrors and built with the gnu-win32 >> gcc package and it worked just fine. > >I use the binary package of b18 and the Perl binary package >perl5.00402-bindist04-bc. The error message i get is very strange: >Can't locate method "hello" via package "world" at -e line 1. > >But it works fine from DOS command line. > >> >> I don't recall that perl comes with the cdk.exe package. >> >> >> > >> >> >2. The 'make' can't handle DOS paths like 'C:/foo' in 'VPATH' >> >> > In the FAQ i found the workaround for this problem: '//C/foo'. >> >> > But programs i call from the Makefile can't deal with such paths. >> >> > Who can i make things like this working: >> >> > >> >> > %.o: %.c >> >> > program $< >> >> > >> >> Are your directory mounts in text=binary mode? Do you have the coolview >> updates from Sergey? Are you sure that the make you are executing is >> the one from gnu-win32? To find out which program is being executed >> from bash do "type -all make". > >- Perl and all the gnu tools reside in 'C:/usr' and this directory is mounted > with 'mount -b C:/usr /usr'. > >- No i don't have any updates. What are these for and where do i get them ? > >- Yes i'm sure. 'make -v' shows version 3.75, but mention nothing about > cygnus. > >- 'bash%type -all make' tells me, that it found make in /usr/local/bin. > Where i expect it. > >- Maybe i was not precise enough what kind o programs i meant: no gnu programs, > like a cross compiler from CAD-UL > > >Thanks a lot for your help so far >Henri > > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From lhall@rfk.com Tue Dec 2 20:23:00 1997 From: lhall@rfk.com (Larry Hall) Date: Tue, 02 Dec 1997 20:23:00 -0000 Subject: building cygwin.dll Message-ID: <3.0.3.32.19971202230420.009ba100@ma.ultranet.com> Perhaps someone can help me. I'm trying to build the cygwin DLL. I'm using the source with the original b18 distribution plus the source for winsup and newlib from Sergey's coolview-src_tar.gz, circa 10/30. I'm building on NT 4.0 + sp3. I have substituted the original b18 signal.h for the one in Sergey's newlib directory in addition to removing #ifdef OLDWAY statements from code in this directory. The configure went fine. The source is in /usr/src/cdk and the objects in /usr/obj/cdk. All mounts are binary. I start the compile with make all-target-winsup in the /usr/obj/cdk directory. This proceeds uneventfully until it reaches the /usr/obj/cdk/gcc directory. When it goes to compile /usr/src/cdk/newlib/libm/math/s_ilogb.c, it complains: /usr/obj/cdk/gcc/xgcc -B/usr/obj/cdk/gcc/ -idirafter /usr/obj/cdk/./newlib/targ-include -idirafter /usr/src/cdk/newlib/libc/include -nostdinc -O2 -O2 -DHAVE_OPENDIR -DHAVE_RENAME -DSIGNAL_PROVIDED -fno-builtin -I/usr/obj/cdk/newlib/./targ-include -I/usr/src/cdk/newlib/./libc/include -c /usr/src/cdk/newlib/libm/math/s_ilogb.c In file included from /usr/obj/cdk/gcc/include/syslimits.h:7, from /usr/obj/cdk/gcc/include/limits.h:11, from /usr/src/cdk/newlib/libm/math/s_ilogb.c:62: /usr/obj/cdk/gcc/include/limits.h:116: limits.h: No such file or directory make[3]: *** [s_ilogb.o] Error 1 make[2]: *** [math/lib.a] Error 1 make[1]: *** [libm/libm.a] Error 2 make: *** [all-target-newlib] Error 2 I've attached the limits.h and syslimits.h file the above error references. My CPLUS_INCLUDE_PATH and C_INCLUDE_PATH read as follows: CPLUS_INCLUDE_PATH=/usr/include:/usr/include/g++ C_INCLUDE_PATH=/usr/include These paths are valid. Obviously, a limits.h file exists since its the one complaining that it cannot find limits.h. The line that generates the complaint reads: #include_next /* recurse down to the real one */ I have not been able to find anything in man pages that tell me what #include_next should do. Can anyone tell me why I get this complaint and what I might try to rectify it? TIA, From earnie_boyd@hotmail.com Tue Dec 2 20:23:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Tue, 02 Dec 1997 20:23:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <19971203040857.9644.qmail@hotmail.com> >From: Chris Faylor >Date: Mon, 1 Dec 1997 17:46:58 -0500 (EST) >To: jazz@softway.com >Subject: Re: Feedback needed on proposed cygwin feature >Cc: gnu-win32@cygnus.com > >>> Extended attributes are not an option for Windows 95, however. >> >>Didn't Sergey's Coolview extensions provide a mechanism for storing >>extended attributes on FAT filesystems and under Win95? Or am I >>misremembering things? > >Just under Windows NT, I think. > >I do remember that people using FAT on Windows NT complained because >cygwin was creating a large file to hold the extended attributes, also. >That may be another mark against using them. Yes, the file grew to be about 150M before I found out what was eating up the disk space. I removed the coolview patches on my Windows NT but now use the latest coolview on my W95. My WinNT has a FAT mount. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From pfife@cisco.com Tue Dec 2 22:06:00 1997 From: pfife@cisco.com (Peter R Fife) Date: Tue, 02 Dec 1997 22:06:00 -0000 Subject: Too Much Mail Volume References: <01bcfdf8$56371b20$8a82400c@hypercow> <6j3hsQAppfB@mike.franken.de> Message-ID: <3.0.3.32.19971203164417.0074bae0@ringer.cisco.com> I don't see what the fuss is about with spam email and the like. Just set up the appropriate filters to put spam etc into a spam mailbox. Then, quickly scan the contents of this mailbox, read anything that looks interesting and delete the rest. That's what I do. The only problem I see with spam email is wasting Internet bandwidth. So, I filter this mailing list to a particular mailbox. I only read it when I want to. Regards, Peter At 03:38 PM 12/2/97 GMT, Chris Faylor wrote: >In article < 6j3hsQAppfB@mike.franken.de >, >Michael Hirmke wrote: >>Hi, >> >>>I second the idea of moving this mailing list to a newsgroup. >> >>No. This isn't a high traffic mailing list compared to a lot of other >>lists I am on. > >I vote emphatically against moving this to a newsgroup. I don't think that >the traffic justifies it. Also, we'll all be subject to much more spam >both from the newsgroup spammers and from the spammers who harvest mail >addresses from newsgroup headers. > >If cygwin or some well-connected site wanted to set up a private NNTP >server to host a private newsgroup, however, I think that would be a >great idea. > > >-- > http://www.bbc.com/ cgf@bbc.com "Strange how unreal >VMS=>UNIX Solutions Boston Business Computing the real can be." >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From anthonm@tams.com.au Wed Dec 3 00:01:00 1997 From: anthonm@tams.com.au (Michael Anthon) Date: Wed, 03 Dec 1997 00:01:00 -0000 Subject: Too Much Mail Volume Message-ID: <3120D53DC9D5D011A59A00C0F016383303A302@server1> About 10 messages in one day is not very high volume. I am behind a firewall and accessing news groups is a problem for me. I have no objection to a dual system, but some thought might be put into removing or distorting mail addresses since junk mailers that harvest news groups are a real problem. On Wednesday, 3 December 1997 2:58, Larry Hall (RFK Partners Inc) [SMTP:lhall@rfk.com] wrote: > At 08:56 AM 12/2/97 -0500, Martin Carl Grant wrote: > >> >I second the idea of moving this mailing list to a newsgroup. > > > >> No. This isn't a high traffic mailing list compared to a lot of other > >> lists I am on. > > > >How about having someone autopost, (using some kind of elm filter), all > >mail which originates on this mailing list. > > > >Likewise any posts on newsgroup that haven't been sent on the mailing > >list could be forwarded. > > > >That way anyone could subscribe to the mailing list or read the newsgroup > >and not miss anything. > > > >All we need is a person to write these two scripts, who is also responsible > >enough to monitor them. > > > >--------------------------------------------------------------- > >Martin Grant "Fight me!" :-) -Martin > >www.ici.net/cust_pages/mgrant "Just call me The Doctor." > >seer@earthling.net "Fizban! Where?" > >--------------------------------------------------------------- > > This is a good idea. May I suggest someone who is interested in seeing this > mail list in a newsgroup pick up on this idea? I, for one, have no problem > with the concept of this "group" as a mail list but I think it would be > beneficial to those who like the newsgroup forum better to be able to access > it in this form too. This idea seems to provide that nicely without forcing > a complete change of forum for those that are happy with the mail list. > > > > > Larry Hall lhall@rfk.com > RFK Partners, Inc. (781) 239-1053 > 8 Grove Street (781) 239-1655 - FAX > Wellesley, MA 02181 > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ply@perform.fr Wed Dec 3 00:22:00 1997 From: ply@perform.fr (ply) Date: Wed, 03 Dec 1997 00:22:00 -0000 Subject: install on e: HELP NEEDED Message-ID: <01BCFFCD.0BB82F90@pcply.perform.fr> Goal : I want to compile kaffe (java virtual machine) on windows Nt. I follow directives and download cygwin b18. Having no more space on C: i install Cygwin on E: configure script won't work, bash don't want to launch it. The only way to do is to launch it with a point before : . ./configure instead of a standard ./configure Inside the script execution another script is called, the call fails. In fact every shell call fails. It's not possible to unmount /, then it is not possible to mount root as e: I mount root / as e: directly in registery. I create a /bin directory in e: where i put a link of sh.exe. it didn't work better I put a copy of sh.exe, it didn't work more I put a link of all exe in /bin. I lauch bash in a command box after calling this .bat : SET GCC_EXEC_PREFIX=E:\gnuwin32\b18\H-i386-cygwin32\lib\gcc-lib\ PATH=%PATH%;\bin;E:\gnuwin32\b18\H-i386-cygwin32\bin;E:\gnuwin32\b18\tcl\bin SET TCL_LIBRARY=E:/gnuwin32/b18/tcl/lib/tcl7.6 SET GDBTK_LIBRARY=E:/gnuwin32/b18/share/gdbtcl CD /D E:\gnuwin32\b18\H-i386-cygwin32\bin when i call cygnwin -V, answer is CYGWIN version ??? Where am i wrong ? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From martin@mrrl.lut.ac.uk Wed Dec 3 01:28:00 1997 From: martin@mrrl.lut.ac.uk (Martin Hamilton) Date: Wed, 03 Dec 1997 01:28:00 -0000 Subject: CVS and SSH References: <0DEA4AD566C0CF11911400001B16BB80029B4CF6@extonpo.bentley.com> Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Len Langsdorf writes: | Yes, but does it handle key database and if not how couild you use with | cvs? It doesn't do very much of anything! Cheerio, Martin -----BEGIN PGP SIGNATURE----- Version: 2.6.3i Charset: noconv iQCVAwUBNIUmENZdpXZXTSjhAQErhAP8CQ9N5K4wl/M0chLuu80MoWyx2KLBJ28j pTfu6rCGZafjg2IV00X0HlsRFDMP/RUWbWfBfP/hP9CgY+Re1jIhzfBTBw4uwIJZ h+m+d54a7SxaaMls7IlzTOEUoeImz8DCCipwUDMD5uQDqH2keGJM//CArVlthRl9 2++sO8OnLPQ= =g2/f -----END PGP SIGNATURE----- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cvn@interchain.nl Wed Dec 3 02:48:00 1997 From: cvn@interchain.nl (Kees van Veen) Date: Wed, 03 Dec 1997 02:48:00 -0000 Subject: Too Much Mail Volume References: <348427C4.7EE8E65E@redes.int.com.mx> Message-ID: <3485394E.4C46F71D@interchain.nl> Michael Anderson wrote: > > If this mailing list does get moved to a newsgroup, the mailing list > should continue also. Many ISPs outside the USA don't offer newsgroups, > but I suspect that all have email. > -- That goes for me, yes Kees - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tixadou@rungis.cisi.fr Wed Dec 3 03:05:00 1997 From: tixadou@rungis.cisi.fr (JEAN YVES TIXADOU) Date: Wed, 03 Dec 1997 03:05:00 -0000 Subject: [Q] install problem + what make the installer ? Message-ID: <199712031104.MAA19465@euclide.cisi.fr> Hi, I'm new in this list and enjoy to use unix program on my PC. I have downloading cdk and usertools and as the first step I have trying to install usertools on an OSR2 W95. the installation was made with success on one PC but on the other the (known) message "decompression of %s failed, there may be not enough space in TEMP dir" (or something like that). Browsing the entire mailing archive, one suggest to delete all the file currently in the temp dir and to set var TMP, TEMP and TMPDIR to the now empty dir. The message also states that install shield need also place in the windows/temp dir. I have 100M free on disk where w95 is installed and 230M on the partition of the second disk where I have created the dir holds by the vars TMP... . After cleaning dir and setting these var the error message remains unchanged. Know I want to know what make the install process exactly or in other words can I copy the successfully installed hierarchy from the first PC to the other from a tar gzipped archive ? Does I have to set some Hkey-XXX in the registry database ? It appears that no file were installed in the w95 dirs, is this true ? I'm afraid for not being able to install the cdk distrib. Thanks in advance. -- Best Regards, jy. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Robert_Seeger@baynetworks.com Wed Dec 3 03:29:00 1997 From: Robert_Seeger@baynetworks.com (Robert Seeger) Date: Wed, 03 Dec 1997 03:29:00 -0000 Subject: Shell-Here Explorer context menu Message-ID: <3.0.32.19971203045255.00ab6630@bl-mail2.corpeast.baynetworks.com> Very nice. I like this. I've wished for this type of thing for a while, but didn't know how to do it. Thanks. . . Robert Seeger At 10:27 AM 12/2/97 -0800, you wrote: >Have you ever wanted to simply click on a directory in the Windows explorer >and open a bash shell in that directory, well now you can. Save the >attached file as ShellHere.inf, then right click on it and select Install. >This then makes a few registry changes and the next time that you right >click on a directory or a drive on explorer you will see a "Shell Prompt >Here" option. Voila. > >I took the concept from the Microsoft Dos-Here powertoy and then adapted it >to work with bash, though the idea could easily be extended to work with >any shell. > >Hopefully someone else will find this useful, if anyone wants to put this >on a web site so that it remains generally available then please go ahead. > >Guy > > >Attachment Converted: "C:\Eudora\ATTACH\ShellHere.inf" > >-- >'The time has come,' the Walrus said, 'To talk of many things: >Of shoes - and ships - and sealing wax - Of cabbages - and kings - >And why the sea is boiling hot - And whether pigs have wings.' >Lewis Carroll, The Walrus and the Carpenter > ======================================== Robert Seeger Network Engineer Bay Networks Telephone: (518)237-2087 Pager: (800)SKY-8888 Pin#1264792 Fax: (518)237-4190 Email: rseeger@baynetworks.com Address: 224 5th Ave, Apt#2 Lansingburgh, NY 12182 ======================================== - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tomas.fasth@twinspot.net Wed Dec 3 03:29:00 1997 From: tomas.fasth@twinspot.net (Tomas Fasth) Date: Wed, 03 Dec 1997 03:29:00 -0000 Subject: Do we need commercial advertisements in the gnu-win32 list? References: <009BDFEE.EA443220.29140@ifk20.mach.uni-karlsruhe.de> <3481B9AE.C48FDD48@opennt.com> <3482F866.1C470A4F@twinspot.net> Message-ID: <3485429E.39FD13DC@twinspot.net> Maybe the subject should have read "commercial statements" instead. I am not native english speaking and may have used the wrong word. David Fox wrote: > Well, I find the information in Jason's posts (and in the follow-ups) > quite interesting and informative, even if I'm not considering using a > commercial product. Just a second opinion. It sure beats "Fast Cash"! I already sent a letter directly to Jason appologizing for my harsh tone. My complaint was not about him personally or his contributions on this list, really. I myself find many postings from Jason interesting and informative. My complaint is about the overall tendency. When someone asks about fork() in this list I would expect a discussion on how to implement it for gnu-win32 in a best possible way, not "Hey, use our costware instead, it's much better". When someone mention copying this component from that costware as a solution to someone's problem, I don't want to know about that costware's new coming features, ports and releases. Not as a reply to me and not on this list. Is that too much to ask for? If such information are of interest for some people but not the majority on this list then simply create a separate list and carry on! I bet the real issue here is about a maturing gnu-win32 starting to threat, or at least fear to threat, some commercial business out there. As with most freeware collaborations on the net, because of the mere number of contributors and testers the freeware tend to get high quality by time, which is not as easily achievable (and much more costly) when doing it the "other" way. Of course there are good reasons for Microsoft to not reveal all details lingering under the hood of win32 for example. They are afraid of competing freeware! ;-) Tomas - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 3 04:55:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 03 Dec 1997 04:55:00 -0000 Subject: bash problem. (Am I missing something?) Message-ID: <19971203125457.24549.qmail@hotmail.com> >From: jeffdbREMOVETHIS@netzone.com (Mikey) >To: gnu-win32@cygnus.com >Subject: Re: bash problem. (Am I missing something?) >Date: Tue, 02 Dec 1997 23:46:09 GMT >Reply-To: jeffdbREMOVETHIS@netzone.com > >try > >source ./test.sh > >#!/bin/sh causes the script to run in a subshell, so any exports >are lost when the toplevel shell resumes. at least I think that's right;^) wrong. source is equates to Bourne's . command which will read the given FILENAME and execute the commands in the current shell context. This means that any environment variable set will remain at the end of the script. Any cd performed will remain in effect as well. >no idea where the error comes from unless #! aren't the very first 2 chars in the file. > I believe the problem to be \r in the script file. And possibly text!=binary. [snip] - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 3 04:55:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 03 Dec 1997 04:55:00 -0000 Subject: bash problem. (Am I missing something?) Message-ID: <19971203125444.14018.qmail@hotmail.com> >From: jeffdbREMOVETHIS@netzone.com (Mikey) >To: gnu-win32@cygnus.com >Subject: Re: bash problem. (Am I missing something?) >Date: Tue, 02 Dec 1997 23:46:09 GMT >Reply-To: jeffdbREMOVETHIS@netzone.com > >try > >source ./test.sh > >#!/bin/sh causes the script to run in a subshell, so any exports >are lost when the toplevel shell resumes. at least I think that's right;^) wrong. source is equates to Bourne's . command which will read the given FILENAME and execute the commands in the current shell context. This means that any environment variable set will remain at the end of the script. Any cd performed will remain in effect as well. >no idea where the error comes from unless #! aren't the very first 2 chars in the file. > I believe the problem to be \r in the script file. And possibly text!=binary. [snip] - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 3 04:58:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 03 Dec 1997 04:58:00 -0000 Subject: GNU-Win32 Installation Problem (NBY) Message-ID: <19971203125805.22007.qmail@hotmail.com> >From: "Caskey, William" >To: "'gnu-win32@cygnus.com'" >Subject: GNU-Win32 Installation Problem >Date: Tue, 2 Dec 1997 14:04:35 -0600 > >> I downloaded Beta 18 and installed it on my Windows NT sytem yesterday >> along with Ming32 (unfortunately without reading the instructions >> completely). Today, I deleted everything and started over. I installed >> GNU Win32 and adjusted the environment variables as follows (per your >> readme.txt in the Minimalist version): >> >> PATH = (added c:\gnuwin32\b18\H-i386-cygwin32\bin; >> c:\gnuwin32\b18\tcl\bin to the PATH line) >> TCL_LIBRARY = c:\gnuwin32\b18\tcl\lib\tcl7.6 >> GDBTK_LIBRARY = c:\gnuwin32\b18\share\gdbtcl >> GCC_EXEC_PREFIX = c:\gnuwin32\b18\H-i386-cygwin32\lib\gcc-lib Add an ending \ to GCC_EXEC_PREFIX. IE: set GCC_EXEC_PREFIX=c:\gnuwin32\b18\H-i386-cygwin32\lib\gcc-lib\ [snip] - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tomas.fasth@twinspot.net Wed Dec 3 06:00:00 1997 From: tomas.fasth@twinspot.net (Tomas Fasth) Date: Wed, 03 Dec 1997 06:00:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <199712021608.JAA04416@chorus.dr.lucent.com> Message-ID: <34856623.C5226DBC@twinspot.net> Modifying the exe file "on-the-fly" for environmental settings is not to recommend. Two of the reasons I can think of is: * Security considerations in a multiuser / multiprocess environment. * Will cause problem in environments with active virus protection. Tomas - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 3 07:38:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 03 Dec 1997 07:38:00 -0000 Subject: bash problem. (Am I missing something?) Message-ID: <19971203131924.13096.qmail@hotmail.com> I suspect that you haven't changed your mounts to text=binary. To do this: bash% cd /gnuwin32/b18/H-i386-cygwin32/bin bash% ./umount / bash% mount -b c:\\ / You need to umount and mount -b all directories in mount. After you have done this then you need to remove all \r (^M) from your script files especially the bash startup files. To do this carefully use the following (don't do this on a true binary file): bash% cp ./scriptfile /tmp/scriptfile bash% tr -d "\r" < /tmp/scriptfile > ./scriptfile - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- PS: for further information check the mail archives at http://www.cygnus.com/ml/gnu-win32 >From: "Buckingham,Mark,SOLON,NUSA" >To: "gnu-win32@cygnus.com" >Subject: bash problem. (Am I missing something?) >Date: Tue, 2 Dec 1997 05:35:57 -0800 > >I downloaded the coolview patch, and have run into a strange problem >with bash: > >let's say I have a file called test.sh: >------------------------------ >#!/bin/sh > >export TESTVAR=1 >------------------------------ > >If I try to run it, (after chmod-ing and all that...) > >like this: > >sh-2.01$ . ./test.sh > >I get this error: > >sh: ./test.sh: error 0 > >And the TESTVAR variable isn't defined. Can anyone tell me what I'm >doing wrong? > >Thanks. > > > > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Wed Dec 3 07:38:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Wed, 03 Dec 1997 07:38:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <199712021608.JAA04416@chorus.dr.lucent.com> Message-ID: In article < 199712021608.JAA04416@chorus.dr.lucent.com >, wrote: >Jason Zions wrote: >>It strikes me that the best place for per-binary attributed like this >>would be in the binary itself. I don't know how closely cygwin's exec() >>looks at the program being exec'd; if it looks at the program header, >>there should be a way to squirrel the attributes away there someplace. > >If there isn't anyplace in the file header that can be used to store this >info away in that doesn't upset NT/W95, or if it is costly to open the object >file whenever a program starts execution, perhaps it would be possible to >provide some global variables in the crt0.o module for controlling various >attributes. The tool to set the attributes would have to go through the >name list to find the variables, then find the section to modify. Shouldn't >be too bad, and the performance of the attribute setting program isn't >critical anyway, but it would require that the file not be fully stripped. Hmm. You're right. It should be possible to store an identifying string in the binary for the options setting program to look for and then modify the settings in the .exe. I feel like I've been arguing out of both sides of my mouth, but the only problem I see with this is for the people who want to set up various symbolic links to the file and then set different registry options for each link. If we modify the binary, then this wouldn't work. Thanks for this suggestion. >Other issues with using name: > >The full path name must be unambiguously stored with the attribute and expanded >for execution. Otherwise, some peculiar things may not work correctly, such >as invoking the program as "../bin/program", using a symbolic link to a >directory "symlinktobin/program", etc. If the attributes are not consistently >found whenever a program is run, the program will have inconsistent behaviour. This shouldn't be an issue since, in the current scheme, only the MS-DOS path spec retrieved from GetModuleFilename is used. >If attributes are set for a program, then the program is moved to >another location or given another name, it looses the attributes. >Furthermore, if a new program is created (perhaps months later) with >the same name and location, it will mysteriously gain the attributes. >These problems do apply to symbolic links currently, so they aren't >new, but they can be confusing particularly since they may have subtle >effects on the execution of the program instead of a hard failure to >follow a link. Well put. You're right. I hate to set something up like this which requires that you "remember" to remove the registry option when you remove the file. Of course, you could have cygwin recognize when a program has been deleted and delete it from the registry, I guess. I think I like the binary modification idea better, though. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jt@boxhill.com Wed Dec 3 07:38:00 1997 From: jt@boxhill.com (Jason Tishler) Date: Wed, 03 Dec 1997 07:38:00 -0000 Subject: Piped Commands Loss Of Output Message-ID: <34856BB1.68CC2642@boxhill.com> Has anyone noticed a transient loss of output with piped commands similar to the following: $ fgrep export .bashrc | wc 12 30 339 $ fgrep export .bashrc | wc $ Here is another one: $ grep . .bash_history | fgrep ls ls ls ... ls //k ls //k/vim-5.0h/vim .bashrc ls //k/vim-5.0h/ ls //k/vim-5.0h/bin/vim .bashrc ls ls $ grep . .bash_history | fgrep ls $ grep . .bash_history | fgrep ls ls ls $ >From the above examples, I conclude that sometimes I get all output, no output, and some output. I am using b18 with Sergey's coolview 25 Nov 97. I seem to remember that I began to observe this problem when I upgraded to one of Sergey's coolview ?? Sept 97 versions. Has anyone observed this too? Anyone know whats wrong? Thanks, Jason -- Jason Tishler Phone: +1 (212) 989-4455 ext. 2120 Box Hill Systems Corporation Fax: +1 (212) 989-6817 161 Avenue of the Americas Email: jt@boxhill.com New York, NY 10013 USA WWW: http://www.boxhill.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Wed Dec 3 07:38:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Wed, 03 Dec 1997 07:38:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <6j3hs3X$pfB@mike.franken.de> <199712021741.EAA13772@mundook.cs.mu.OZ.AU> Message-ID: In article < 199712021741.EAA13772@mundook.cs.mu.OZ.AU >, Fergus Henderson wrote: >cgf@bbc.com (Chris Faylor) writes: > >>Btw, since I don't seem to have made myself sufficiently clear, I was >>not talking about removing environment variable support. I was talking >>about changing CYGWIN_* environment variables to CYGWIN32=*. > >I don't see any good reason to hard-code the `32'. > >Using `CYGWIN32' won't look very good on say the 64-bit NT Alpha port. Good point. Hmm. How about: CYGWIN32ORGREATER ? :-) Only kidding. Now that I think of it, I can't remember why I added the 32 to begin with. Probably just to distinguish it from the CYGWIN_* style variables a little more. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tjump@cais.com Wed Dec 3 07:38:00 1997 From: tjump@cais.com (Theodore Jump) Date: Wed, 03 Dec 1997 07:38:00 -0000 Subject: Reply-To: message field ? Message-ID: <348a7b1a.2194437@smtp> I beg forgiveness from the consituency for this off topic message, however it does affect us all... How many times have you unthinkingly hit Reply and not sent the message to the list but to the message author instead? Maybe I'm silly but I tend to do this way too much, sooo ... List-manager-person-out-there, I have a question for you ... Would it be possible to get the cygwin email list messages as they are sent from Majordomo to set a "Reply-To:" field in the messages so that all us subscribers could just hit "reply" and it would go back to the listserv rather than needing to remember to change the "To:" field or add the listserv to the CC: list? -Ted PS: I *did* send this to the admin address first but it bounced of majordomo, thus this repeat ... - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mark.buckingham@us.nestle.com Wed Dec 3 07:38:00 1997 From: mark.buckingham@us.nestle.com (Buckingham, Mark, SOLON, NUSA) Date: Wed, 03 Dec 1997 07:38:00 -0000 Subject: bash problem. (Am I missing something?) Message-ID: Sure do. I didn't have C: mounted binary. (Thanks Earnie) > > >Are you sure you have a /bin/sh.exe? > > >Larry Hall lhall@rfk.com >RFK Partners, Inc. (781) 239-1053 >8 Grove Street (781) 239-1655 - FAX >Wellesley, MA 02181 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tjump@cais.com Wed Dec 3 07:38:00 1997 From: tjump@cais.com (Theodore Jump) Date: Wed, 03 Dec 1997 07:38:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <009BE319.3FA64320.29436@ifk20.mach.uni-karlsruhe.de> Message-ID: <348877ba.1329734@smtp> >: Using `CYGWIN32' won't look very good on say the 64-bit NT Alpha port. It's still banging against the "Win32" API even though it's running on a 64-bit machine - so "CYGWIN32" is still accurate, IMHO. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dkb@mitre.org Wed Dec 3 08:05:00 1997 From: dkb@mitre.org (Daryl K. Baker) Date: Wed, 03 Dec 1997 08:05:00 -0000 Subject: Stop the mail on too much mail References: <3120D53DC9D5D011A59A00C0F016383303A302@server1> Message-ID: <3.0.3.32.19971203110651.00e61100@skidmark.mitre.org> The subject says it all. This thread has generated more traffic than any of the messages of significance. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 3 08:26:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 03 Dec 1997 08:26:00 -0000 Subject: Problems running gcc: cpp not found (NBY) Message-ID: <19971203131835.12883.qmail@hotmail.com> >From: "Himilcon de Castro Carvalho" >To: >Subject: Problems running gcc: cpp not found >Date: Tue, 2 Dec 1997 14:55:23 -0200 > >Hello all, > >I'm having problems trying to put gcc to work. I've read the FAQ and Mailing >List archives and no answer was found. > >After having installed, created c:\bin (and copied sh.exe to it) and c:\tmp, >and set the PATH and environment variables in autoexec.bat (as described in >the README file), gcc continued issuing the message about CPP. > >Using the SET command (from dos), I could see that the GCC_EXEC_PREFIX >variable was set to c:\gnuwin32\b18\H-i386~1\lib\gcc-lib\ . Is it correct to >have the long name H-i386-cygwin32 cut like that? I've tryed also to [snip] No. You need the long name set in the GCC_EXEC_PREFIX variable. You should be able to do a SET GCC_EXEC_PREFIX=C:\gnuwin32\b18\H-i386-cygwin32\lib\gcc-lib\ from DOS or in the autoexec.bat file. or from bash do: export GCC_EXEC_PREFIX=/gnuwin32/b18/H-i386-cygwin32/lib/gcc-lib/ ---- While I'm at it let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 3 08:26:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 03 Dec 1997 08:26:00 -0000 Subject: building cygwin.dll Message-ID: <19971203162531.22332.qmail@hotmail.com> >Date: Tue, 02 Dec 1997 23:04:20 -0500 >To: gnu-win32@cygnus.com >From: Larry Hall >Subject: building cygwin.dll > >Perhaps someone can help me. [snip] >#include_next /* recurse down to the real one */ > >I have not been able to find anything in man pages that tell me what >#include_next should do. Can anyone tell me why I get this complaint and >what I might try to rectify it? > The #include_next is expecting to find another limits.h file based upon the include paths. The documentation for this can be found in the cpp.info file. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jazz@softway.com Wed Dec 3 08:45:00 1997 From: jazz@softway.com (Jason Zions) Date: Wed, 03 Dec 1997 08:45:00 -0000 Subject: GNU debugger References: Message-ID: <34858CCE.657022E8@opennt.com> > GDB can read the files, but not the MS-specific debug information. > Some MS compilers can generate standard COFF debug info, but that > only works for C, not C++. > > > If not, do you plan to make GDB-WIN32 compatible with > > MS compiler? > > At present, we don't know of anyone working on this addition. It would be > very desirable however. We looked at it for a while; there are parts of the MS debug information that are either undocumented or incorrectly documented, and the whole thing turned into a larger mess than was worth it to us. Jason Zions Softway Systems Makers of OpenNT, POSIX for NT - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Wed Dec 3 08:45:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Wed, 03 Dec 1997 08:45:00 -0000 Subject: Too Much Mail Volume Message-ID: <199712031631.LAA01113@hardy.bbc.com> >I don't see what the fuss is about with spam email and the like. Just set >up the appropriate filters to put spam etc into a spam mailbox. Then, >quickly scan the contents of this mailbox, read anything that looks >interesting and delete the rest. That's what I do. The only problem I see >with spam email is wasting Internet bandwidth. This particular argument crops up every time someone mentions spam. What "set up the appropriate filters" means is: Take a stab at filtering the spam. Then when something slips through, tinker with the filters again. Another one comes through? No problem. Just add another rule to the filter. Hmm. How does that work again? man procmailex. Oh yeah. Not getting any email? Oops. The filter was too general. Someone complains that they sent you mail and didn't respond? Oh yeah. I haven't checked my spam mailbox in a while. I don't care for this scenario. YMMV. >So, I filter this mailing list to a particular mailbox. I only read it when >I want to. I don't have the option of only reading email when I want to. And, as you mentioned, I don't particularly like any of my limited internet bandwidth being eaten by junk email. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jkolen@typhoon.coginst.uwf.edu Wed Dec 3 15:20:00 1997 From: jkolen@typhoon.coginst.uwf.edu (John F. Kolen) Date: Wed, 03 Dec 1997 15:20:00 -0000 Subject: help for template woes with g++ Message-ID: <9712031721.ZM14602@typhoon.coginst.uwf.edu> Those of you who are struggling with linker errors caused by templates (such as those found in STL) might like to know that -frepo works much better when combined with -O3. It turns out that repo stumbles on inlined methods declared outside the class. The -O3 forces the compiler to generate inline code, otherwise the compiler _decides_ what to inline and to generate. See g++ info page for more details. Also, if you use the String class, you should explicitly link with -lg++ as well. Some critical data definitions reside there. John -- John F. Kolen voice: (850)474-3075 Assistant Professor fax: (850)474-3023 Dept. of Computer Science University of West Florida Pensacola, FL 32514 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From rvb@gluck.coda.cs.cmu.edu Wed Dec 3 15:26:00 1997 From: rvb@gluck.coda.cs.cmu.edu (Robert.V.Baron) Date: Wed, 03 Dec 1997 15:26:00 -0000 Subject: Too Much Mail Volume Message-ID: <881165089/rvb@gluck.coda.cs.cmu.edu> I think the traffic is more like 20 to 30 pieces a day. For me, the issue of volume is less important than quality. I'd like to see several groups (news or mail), for: newbies programming questions interactions with microsoft DLL's, tools technical discussions perhaps more. This would allow us to sort and categorize the volume better. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Wed Dec 3 15:26:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Wed, 03 Dec 1997 15:26:00 -0000 Subject: building cygwin.dll References: <3.0.3.32.19971202230420.009ba100@ma.ultranet.com> Message-ID: > In file included from /usr/obj/cdk/gcc/include/syslimits.h:7, > from /usr/obj/cdk/gcc/include/limits.h:11, > from /usr/src/cdk/newlib/libm/math/s_ilogb.c:62: > /usr/obj/cdk/gcc/include/limits.h:116: limits.h: No such file or directory Classic include_next problem. As a quick workaround, edit the file /usr/obj/cdk/gcc/gcc/include/limits.h, and remove the beginning and ending extra stuff to include a system specific limits file if it exists. Basically, delete the initial lines until your first line is the following: #ifndef _LIMITS_H__ now delete backwards from the end until your last line is: #endif /* _LIMITS_H__ */ Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From wcaskey@cmh.edu Wed Dec 3 15:26:00 1997 From: wcaskey@cmh.edu (Caskey, William) Date: Wed, 03 Dec 1997 15:26:00 -0000 Subject: GNU Win32 Installation Message-ID: <119D55D6D69FD011AB9B000092966BB411A177@EXCHANGE2> Thanks for your reply. I fixed the environment variable by adding a \, but still encountered the same problem. I then explored the directories and noticed a cp.exe executable. I renamed this to cpp.exe and everything works fine. Could this just have been a typo as the zip was created or something? Thought I'd let you know. Again, thanks for your reponse. Now, playing with the compiler...Bill *************************************** * Bill Caskey * * Director, Research & Grants Administration * * The Children's Mercy Hospital * * 2401 Gillham Road * * Kansas City, MO 64108 * * Voice: (816) 234-3879 * * FAX: (816) 855-1982 * * E-mail: wcaskey@cmh.edu * *************************************** - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From rdparker@butlermfg.org Wed Dec 3 15:26:00 1997 From: rdparker@butlermfg.org (Parker, Ron) Date: Wed, 03 Dec 1997 15:26:00 -0000 Subject: *** can't find configure.in... with coolview Message-ID: Given this directory hierarchy, ~ +--- src | +--- cdk <-- Original b18 cdk source w/o patches +--- obj +--- cdk and these commands, cd ~/obj/cdk time ~/src/cdk/configure -v 2>&1|tee config.log I receive the following output message Configuring for a i386-pc-cygwin32 host. *** Can't find configure.in. Try using --srcdir=some_dir Yet, if I specify the source directory I receive an "Invalid source directory" message. I am using B18 with nothing but the 03-Nov-1997 coolview installed on top of it. I have only the environment variables specified in the readme, HOME=/home/rdparker and CYGWIN_TTY=1. I have tried Windows NT 4.0, 4.0 sp 3, and NT 5.0 Beta 1. Everything works as advertised without coolview. I have seen a number of messages indicating a successful compilation of the CDK with coolview and Sergey's coolview-src.tar.gz installed. This is all that I am trying to do. I have been cross compiling from Linux, but need to be able to configure and build from Windows 9x or NT [45]. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From eyesore01@hotmail.com Wed Dec 3 15:26:00 1997 From: eyesore01@hotmail.com (Gilgamesh -) Date: Wed, 03 Dec 1997 15:26:00 -0000 Subject: Argh! Message-ID: <19971203183406.18773.qmail@hotmail.com> Hi all, I am running Windows NT 4.0 workstation on a p166 with 96 megs of ram & a 2gb hard drive ( ~700mb free). I have installed gnuwin32 b18 and set up all the environment variables as listed in the FAQ. Now, whenever I run gcc I get the error(s) below, but bash and the other "unix" tools work fine.. I've tried everything I can think of and bugged everyone I know, but still no luck ;) If anyone knows how to fix this, I would really appreciate the information. Thanks! (unknown) heap_init: unable to allocate heap, win32 error 87 cygwin: terminating (unknown) In cygwin_except_handler (unknown) Exception trapped! (unknown) exception C0000005 at 1001E4B2 (unknown) exception: ax 0 bx 0 cx 6 dx 7 (unknown) exception: si 6779632D di 36383369 bp 241FB30 sp 241FB30 (unknown) exception is: STATUS_ACCESS_VIOLATION (unknown) Stack trace: (unknown) frame 0: sp = 0x241F964, pc = 0x1001282A (unknown) frame 1: sp = 0x241F980, pc = 0x77F94512 (unknown) frame 2: sp = 0x241F9A4, pc = 0x77F88EEB (unknown) frame 3: sp = 0x241FA30, pc = 0x77F76266 (unknown) frame 4: sp = 0x241FB30, pc = 0x10011D4D (unknown) frame 5: sp = 0x241FB44, pc = 0x10018972 (unknown) frame 6: sp = 0x241FB5C, pc = 0x10011615 (unknown) frame 7: sp = 0x241FF94, pc = 0x10011BFF (unknown) frame 8: sp = 0x241FFA0, pc = 0x4118D5 (unknown) frame 9: sp = 0x241FFB0, pc = 0x40103B (unknown) frame 10: sp = 0x241FFC0, pc = 0x77F1B304 (unknown) frame 11: sp = 0x241FFF0, pc = 0x0 (unknown) End of stack trace (unknown) In cygwin_except_handler (unknown) Error while dumping state (probably corrupted stack) *note that these last 2 lines will repeat until I break out of the program or shut down the DOS shell. ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Wed Dec 3 15:26:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Wed, 03 Dec 1997 15:26:00 -0000 Subject: egcs-1.0 is released Message-ID: <9712032111.AA21703@modi.xraylith.wisc.edu> EGCS 1.0 is now out! The good news for Gnu-Win32 users is that almost *all* the changes that affect cygwin32 have made it in, except for 2 small patches to the C++ front-end. See Changes below. I'm currently travelling, so can't really make a new distribution and test it before unleashing it onto the unsuspecting users. Hopefully you will see a egcs-1.0 release for cygwin32 (mingw32 will take a few extra days) this weekend. Checking the last FSF snapshot, I see that Kenner has changed some of the flags (eg., -windows to -mswindows or something like that). If you follow FSF snapshots and know what's going on, please drop me a line with a short explanation of what needs to be changed, so I can make things consistent between the two. Also, it's very important that I find a mirror site(s). I can't afford to overload our teeny server, which is meant to serve *research* tools. Let me know if you're willing to help. Changes from egcs-1.0: 1. C++ front end fix for vtable and rtti of explicitly instantiated templates. This is from Jason Merrill, and is in the mainline sources already. 2. Cygwin32 config fixes: - Use collect2 (this makes -frepo work) - Use sjlj as the default exception mechanism since DWARF2 unwinding doesn't work with exceptions yet. Caveats for b18 users: 1. Can't use export libraries that are built with b18 toolchain. This also applies to libcygwin.a distributed by Sergey. You can safely use Sergey's DLL and my libcygwin.a (until some interface change occurs), so please don't replace mine with Sergey's. I'll distribute the last publicly available source tree on Sergey's web page (Nov. 3 I believe). 2. There might be binary incompatibility due to the use of exception and RTTI info that are on by default. Most of the simple stuff seems to work just fine, but at least for a large application that I have (it's a loadable DLL), I had to rebuild from source. Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Wed Dec 3 15:26:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Wed, 03 Dec 1997 15:26:00 -0000 Subject: Too much mail volume Message-ID: <199712032134.VAA51864@out4.ibm.net> Look, if you guys had any smarts, you would scrap the mail version of this list entirely and go with the archive instead. That way, people interested in knowing what is going on in the discussion can tune in to the archive and not bother with their e-mail server. You would also add a URL for posting followups to archive articles. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mark.buckingham@us.nestle.com Wed Dec 3 15:26:00 1997 From: mark.buckingham@us.nestle.com (Buckingham, Mark, SOLON, NUSA) Date: Wed, 03 Dec 1997 15:26:00 -0000 Subject: Problems with ncurses Message-ID: I'm trying to get ncurses compiled & running. (I downloaded ncurses 1.9.9e from the Slackware linux distribution.) I remounted by C: drive as text = binary. No problems. I ran configure. No problems. Ran make. No problems. When I try to run any of the tests, I get something like the following: (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) In cygwin_except_handler exc C0000005 at 10042894 sp 242E5B0 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) Exception trapped! (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) exception C0000005 at 10042894 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) exception: ax 242E5E0 bx 242E5E0 cx F9 dx 0 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) exception: si 0 di 242E5E0 bp 242E5BC sp 242E5B0 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) exception is: STATUS_ACCESS_VIOLATION (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) Stack trace: (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 0: sp = 0x242E3E4, pc = 0x10007BB2 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 1: sp = 0x242E400, pc = 0x77F94072 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 2: sp = 0x242E424, pc = 0x77F88A53 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 3: sp = 0x242E4B0, pc = 0x77F75E82 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 4: sp = 0x242E5BC, pc = 0x4098B3 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 5: sp = 0x242EF64, pc = 0x402B86 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 6: sp = 0x242F080, pc = 0x402D5F (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 7: sp = 0x242F0A0, pc = 0x403CDA (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 8: sp = 0x242F0C4, pc = 0x401361 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 9: sp = 0x242F0DC, pc = 0x4010DF (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 10: sp = 0x242F504, pc = 0x100063BA (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 11: sp = 0x242FF94, pc = 0x100063CB (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 12: sp = 0x242FFA0, pc = 0x41D735 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 13: sp = 0x242FFB0, pc = 0x40103B (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 14: sp = 0x242FFC0, pc = 0x77F1AFC1 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) frame 15: sp = 0x242FFF0, pc = 0x0 (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) End of stack trace I get this on both my NT 4.0 machine with NTFS, and my windows 95 machine. Here is the environment on my NT 4.0 machine: !C:=C:\gnuwin32\b18\H-i386-cygwin32\bin COMSPEC=C:\WINNT\system32\cmd.exe GCC_EXEC_PREFIX=C:\gnuwin32\b18\H-i386-cygwin32\lib\gcc-lib\ GDBTK_LIBRARY=C:/gnuwin32/b18/share/gdbtcl LIBRARY_PATH=C:\gnuwin32\b18\H-i386-cygwin32\lib;C:\gnuwin32\b18\H-i386- cygwin32\i386-cygwin32\lib; OSTYPE=cygwin32 PATH=/gnuwin32/b18/H-i386-cygwin32/bin:/gnuwin32/b18/tcl/bin:.:/winnt:/w innt/system32:. PATHEXT=.COM;.EXE;.BAT;.CMD PID=3448 SHELL=/bin/sh SHLVL=1 TCL_LIBRARY=C:/gnuwin32/b18/tcl/lib/tcl7.6 TERM=dumb _=/gnuwin32/b18/H-i386-cygwin32/bin/env I know people have gotten ncurses to work ok. Any ideas? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From periat@ens.ascom.ch Wed Dec 3 15:26:00 1997 From: periat@ens.ascom.ch (Periat Henri) Date: Wed, 03 Dec 1997 15:26:00 -0000 Subject: Definition of VPATH Message-ID: <199712031715.SAA04768@stva89.ens.ascom.ch> Hi everybody I'm very new to cygnu and have to make a UNIX gmake file to run also on MS-NT boxes. Now the problem which stops me in the moment is the ':' problem: - with 'VPATH:=/home/periat/src:D:/home/periat/src' make can't find the source files if they are on drive 'D:' only. - with 'VPATH:=/home/periat/src;D:/home/periat/src' or 'VPATH:=/home/periat/src://D/home/periat/src' '$<' contains '//D/home/periat/src/foo.c' and non cygnus programs can't deal with such file names. Can anyone tell me how i can solve this problem? Thanks Henri I have b18 and the 'coolview' patches. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jazz@softway.com Wed Dec 3 15:26:00 1997 From: jazz@softway.com (Jason Zions) Date: Wed, 03 Dec 1997 15:26:00 -0000 Subject: Do we need commercial advertisements in the gnu-win32 list? References: <009BDFEE.EA443220.29140@ifk20.mach.uni-karlsruhe.de> <3481B9AE.C48FDD48@opennt.com> <3482F866.1C470A4F@twinspot.net> <3485429E.39FD13DC@twinspot.net> Message-ID: <3485EAC1.C47E3828@opennt.com> > I bet the real issue here is about a maturing gnu-win32 starting to > threat, or at least fear to threat, some commercial business out there. > As with most freeware collaborations on the net, because of the mere > number of contributors and testers the freeware tend to get high quality > by time, which is not as easily achievable (and much more costly) when > doing it the "other" way. According to your position, Tomas, I'm not even allowed to respond to this set of assertions. How sad. Jason Zions Softway Systems Inc. Makers of OpenNT, POSIX for Windows NT - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From petdr@students.cs.mu.oz.au Wed Dec 3 16:49:00 1997 From: petdr@students.cs.mu.oz.au (Peter David ROSS) Date: Wed, 03 Dec 1997 16:49:00 -0000 Subject: BUG: stdin not a constant Message-ID: <199712040048.LAA26596@rimmer.cs.mu.OZ.AU> Hi, The following code doesn't compile. ------ #include static FILE *out = stdout; void main(void) { /* out = stdout; */ fprintf(out, "hello world\n"); } ------ It gets the following error message: test.c:3: initializer element is not constant The fix is to remove the initialiser and add the line which is commented out. However the code I am trying to port has these sort of variables spread all over the place, and I would like to know if there is someway I can get the compiler to treat it as a constant. Pete. ---- +----------------------------------------------------------------------+ | Peter Ross Sci/Eng Melbourne Uni | | email: petdr@cs.mu.oz.au WWW: http://www.cs.mu.oz.au/~petdr | +----------------------------------------------------------------------+ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From pgarceau@submergedsoftware.com Wed Dec 3 17:53:00 1997 From: pgarceau@submergedsoftware.com (Paul Garceau) Date: Wed, 03 Dec 1997 17:53:00 -0000 Subject: Too Much Mail Volume References: <199711301745.MAA02725@entropy.math.fsu.edu> Message-ID: <199712040152.UAA14702@bw.webex.net> Greetings, On 30 Nov 97 at 12:45, the Illustrious Steven Bellenot wrote: > > This mailing list has so much volume that I think we would be better > served with a newsgroup to move the traffic and to provide a FAQ. I tend to agree with this...in two days I've received nearly 100 emails from the list. > > How about comp.unix.gnu-win32? My suggestion would be comp.programming.gnu-win32. Problem is that usenet groups are notorious when it comes to spam/UCE and mailing lists are not when you consider the fact that in the month or so that I've been subscribed I have only seen two UCEs come through based on the mailing list itself. The biggest danger with a usenet group is the spam or UCE that such usenet groups generate. If the group were completely moderated that might be different. As we are all probably aware, however, there is not much time for anyone who regularly subscribes to this mailing list to moderate such a group. What about a mailing list specifically defined as a digest of the gnu-win32 responses? This way, if someone didn't want to deal with the email, they would only have to deal with a single (weekly) digest that came out. If they desired to reply to something from the digest, then their email would be cc: to the list and any replies to their email would automatically generate a cc: to the original author of the "thread" itself. The announcement side of the mailing list does something similar except that, as far as I know, there are no replies whatsoever generated by the mailing list to the originator of the announcement in question. Comments? Feedback? Paul G. Chronicles of Aurum: Book I A Journey Into the Fantastic http://www.submergedsoftware.com/ ICQ#977283 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From pgarceau@submergedsoftware.com Wed Dec 3 17:53:00 1997 From: pgarceau@submergedsoftware.com (Paul Garceau) Date: Wed, 03 Dec 1997 17:53:00 -0000 Subject: cygwin32 (mingw32) news update References: <199712010401.EAA43478@out1.ibm.net> Message-ID: <199712040152.UAA15026@bw.webex.net> On 1 Dec 97 at 4:01, the Illustrious vischne@ibm.net wrote: > > > > cygwin32 (mingw32) + egcs -- news update > > > > Mumit Khan (khan@xraylith.wisc.edu) > > > >Just a heads up for the egcs users -- I just ran the testsuite on the > >latest snapshot (+ extra fixes for template handling + use sjlj for > >exception handling), and the results for i386-cygwin32 are almost as > >good as on i386-linux-gnulibc1! > > > > Only one or two problems: > > mingw32 make emphatically does not work. It needs sh.exe, and, when it > gets cygwin sh.exe, it has directory addressing conflicts. It definitely > does not understand GNU makefiles, even when individual command-line > compiles work. > For some odd reason, compiling a program or library that references > sleep, _sleep, __sleep, or Sleep fails during link, no matter what > `extra' libraries are referenced. Largely correct...from a minimalist standpoint, however, it was noted that such things wouldn't work in the documentation that comes with mingw32 as far as I recall. Peace, Paul G. Chronicles of Aurum: Book I A Journey Into the Fantastic http://www.submergedsoftware.com/ ICQ#977283 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From pfife@cisco.com Wed Dec 3 17:53:00 1997 From: pfife@cisco.com (Peter R Fife) Date: Wed, 03 Dec 1997 17:53:00 -0000 Subject: Too Much Mail Volume References: <199712031631.LAA01113@hardy.bbc.com> Message-ID: <3.0.3.32.19971204063207.00d95dbc@ringer.cisco.com> At 11:31 AM 12/3/97 -0500, Chris Faylor wrote: >>I don't see what the fuss is about with spam email and the like. Just set >>up the appropriate filters to put spam etc into a spam mailbox. Then, >>quickly scan the contents of this mailbox, read anything that looks >>interesting and delete the rest. That's what I do. The only problem I see >>with spam email is wasting Internet bandwidth. > >This particular argument crops up every time someone mentions spam. >What "set up the appropriate filters" means is: Take a stab at filtering >the spam. Then when something slips through, tinker with the filters >again. Another one comes through? No problem. Just add another rule >to the filter. Hmm. How does that work again? man procmailex. Oh yeah. > >Not getting any email? Oops. The filter was too general. Someone >complains that they sent you mail and didn't respond? Oh yeah. I >haven't checked my spam mailbox in a while. Interesting; I have 3 mailboxes for my incoming mail - MUST Answer, Should Answer & May Answer. To date, I have not received any spam sent direct To: pfife@cisco.com; so, all my spam ends up in May Answer. > >I don't care for this scenario. YMMV. > >>So, I filter this mailing list to a particular mailbox. I only read it when >>I want to. > >I don't have the option of only reading email when I want to. And, as you >mentioned, I don't particularly like any of my limited internet bandwidth >being eaten by junk email. > I read my email approx. 6 hours each day. When I read my May Answer mailbox, I quickly scan it, read what I want and delete the rest. Maybe the spammers don't mail me direct or something, but all my spam ends up in May Answer. I rarely read any of it. So, it works for me so far, that's all I can say. Peter > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From pgarceau@submergedsoftware.com Wed Dec 3 17:53:00 1997 From: pgarceau@submergedsoftware.com (Paul Garceau) Date: Wed, 03 Dec 1997 17:53:00 -0000 Subject: Too Much Mail Volume References: <01bcfdf8$56371b20$8a82400c@hypercow> Message-ID: <199712040152.UAA14775@bw.webex.net> On 1 Dec 97 at 8:28, the Illustrious Norm Peterson wrote: > > On Sun, 30 Nov 1997, Peter Yeh wrote: > > > I second the idea of moving this mailing list to a newsgroup. > > > > > > > > > >This mailing list has so much volume that I think we would be better > > >served with a newsgroup to move the traffic and to provide a FAQ. > > > > > >How about comp.unix.gnu-win32? > > How about just dropping into one of the alt.* hierarchy? From what I > understand, you don't need to have any of the voting (or any particular > reason) to start the group. This could serve as a testbed before moving > into the comp.* area. If it fails, ah well back to the mail list. If > not, great. alt.* groups, by their nature, are completely vulnerable to UCE/Spam. Many of us barely have time to reply as we do the mailing list...to open ourselves up to mindless spam or UCE seems to be the wrong answer, in my humble opinion. Peace, Paul G. > > > > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". > Chronicles of Aurum: Book I A Journey Into the Fantastic http://www.submergedsoftware.com/ ICQ#977283 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From pgarceau@submergedsoftware.com Wed Dec 3 17:53:00 1997 From: pgarceau@submergedsoftware.com (Paul Garceau) Date: Wed, 03 Dec 1997 17:53:00 -0000 Subject: Too Much Mail Volume References: Message-ID: <199712040152.UAA15020@bw.webex.net> On 2 Dec 97 at 15:38, the Illustrious Chris Faylor wrote: > In article < 6j3hsQAppfB@mike.franken.de >, > Michael Hirmke wrote: > >Hi, > > > >>I second the idea of moving this mailing list to a newsgroup. > > > >No. This isn't a high traffic mailing list compared to a lot of other > >lists I am on. > > I vote emphatically against moving this to a newsgroup. I don't think > that the traffic justifies it. Also, we'll all be subject to much more > spam both from the newsgroup spammers and from the spammers who harvest > mail addresses from newsgroup headers. > > If cygwin or some well-connected site wanted to set up a private NNTP > server to host a private newsgroup, however, I think that would be a > great idea. I can agree to this. It means, of course, that whoever does set up the NNTP server will need to make doubly-certain that they can't be attacked by any email moles that are capable of rapeing an NNTP server by stripping out all of the email addresses stored therein. Peace, Paul G. Chronicles of Aurum: Book I A Journey Into the Fantastic http://www.submergedsoftware.com/ ICQ#977283 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From weiku@utkux.utcc.utk.edu Wed Dec 3 17:53:00 1997 From: weiku@utkux.utcc.utk.edu (Wei Ku) Date: Wed, 03 Dec 1997 17:53:00 -0000 Subject: Newsgroup proposal (was: Too Much Mail Volume) Message-ID: <01bd001c$a3fe53c0$9a12a980@wku1.utk.edu> I agree with you ! If there will be a vote. I'd prefer to make it a newsgroup. Wei Ku -----Original Message----- From: Tim Iverson To: gnu-win32@cygnus.com Date: Wednesday, December 03, 1997 11:56 AM Subject: Re: Newsgroup proposal (was: Too Much Mail Volume) >Hmmm. Most of the newsgroups I read have less traffic than this mailing >list, so those that say they're on bigger lists just makes me wonder about >the wisdom of not moving those to newsgroups also. > >IMHO, mailing lists are best when the information is either confidential or >of value to only a select population. This is an open list, so the former >doesn't apply and judging from the volume, neither does the latter. > >A filtering MUA is not the answer -- we already use those. The problem is >the volume itself. As it stands now, *I* am basically the expire process >for my own little gnu-win32 newsgroup. I would far prefer that this was >offloaded onto a real newsgroup. > >Anyway, I'm just whining here. I don't have the time or resources to setup >a gateway, or I'd vote the old-fashioned way, and just do it. I'm hoping >someone else out there does have the time. ;) > >There are more comments inline ... > >- Tim Iverson > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Wed Dec 3 18:36:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Wed, 03 Dec 1997 18:36:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <199712021608.JAA04416@chorus.dr.lucent.com> <34856623.C5226DBC@twinspot.net> Message-ID: In article < 34856623.C5226DBC@twinspot.net >, Tomas Fasth wrote: >Modifying the exe file "on-the-fly" for environmental settings is not to >recommend. Two of the reasons I can think of is: > >* Security considerations in a multiuser / multiprocess environment. What security considerations are there that are not also present with any other scheme, whether it is using extended attributes or setting options in the registry? You would have to have the right privileges to change the binary. >* Will cause problem in environments with active virus protection. How does a virus detection program detect the difference between installing a new version of bash or changing a byte in the existing file? -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jeff@dtv.net Wed Dec 3 23:14:00 1997 From: jeff@dtv.net (Jeff Smart) Date: Wed, 03 Dec 1997 23:14:00 -0000 Subject: gethostby... broken when forked Message-ID: <34862B09.3FFE@dtv.net> I have found the BSD calls gethostbyname and gethostbyaddr break when they are used in a child process that was forked. I compiled this snippet on Win95 using gnu-win32 b18 with the latest coolview dll and lib dated 26Nov97. The code will always return an exception violation: STATUS_ACCESS_VIOLATION once gethostbyname is called. If fork() is removed, then gethostbyname works.. It appears fork() is screwing these address lookup calls. I do not have the latest sources to find this bug. Has anyone else found this nasty ? Cheers, Jeff [----------------- begin -----------------------] #include #include #include #include int main(argc, argv) int argc; char *argv[]; { struct hostent *hostptr; /* Internet host struct from netdb.h */ char hostname[256]; int childpid; strcpy(hostname, argv[1]); printf("PARENT STARTED PID=%d\n", getpid()); if ((childpid = fork()) < 0) { printf("fork failed\n"); exit(0); } else if (childpid == 0) { /* start child process */ printf("CHILD STARTED PID=%d\n", getpid()); hostptr = gethostbyname(hostname); /* BOMBS HERE */ printf("hostptr = %p\n", hostptr); exit(0); } } [--------------------- end --------------------] - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From petdr@students.cs.mu.oz.au Wed Dec 3 23:14:00 1997 From: petdr@students.cs.mu.oz.au (Peter David ROSS) Date: Wed, 03 Dec 1997 23:14:00 -0000 Subject: BUG: stdin not a constant (fwd) Message-ID: <199712040054.LAA26717@rimmer.cs.mu.OZ.AU> I forgot to mention that I am using release b18 under win95. ----- Forwarded message from petdr ----- Hi, The following code doesn't compile. ------ #include static FILE *out = stdout; void main(void) { /* out = stdout; */ fprintf(out, "hello world\n"); } ------ It gets the following error message: test.c:3: initializer element is not constant The fix is to remove the initialiser and add the line which is commented out. However the code I am trying to port has these sort of variables spread all over the place, and I would like to know if there is someway I can get the compiler to treat it as a constant. Pete. ---- +----------------------------------------------------------------------+ | Peter Ross Sci/Eng Melbourne Uni | | email: petdr@cs.mu.oz.au WWW: http://www.cs.mu.oz.au/~petdr | +----------------------------------------------------------------------+ ----- End of forwarded message from petdr ----- ---- +----------------------------------------------------------------------+ | Peter Ross Sci/Eng Melbourne Uni | | email: petdr@cs.mu.oz.au WWW: http://www.cs.mu.oz.au/~petdr | +----------------------------------------------------------------------+ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Wed Dec 3 23:14:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Wed, 03 Dec 1997 23:14:00 -0000 Subject: help for template woes with g++ References: <9712031721.ZM14602@typhoon.coginst.uwf.edu> Message-ID: > Those of you who are struggling with linker errors caused by templates (such > as those found in STL) might like to know that -frepo works much better when > combined with -O3. It turns out that repo stumbles on inlined methods declared > outside the class. The -O3 forces the compiler to generate inline code, > otherwise the compiler _decides_ what to inline and to generate. See g++ info > page for more details. > Yes, this is a well-known problem. I believe I did document a few of these cases a few years in my STL Newbie Guide (now hopelessly outdated). -frepo has had problems as well, most of which have been fixed in egcs. fwiw, with egcs-1.0 (and fsf 2.8 when released), this problem goes away, at least on ELF and PE-COFF (i386). You can just let the compiler instantiate all the templates it needs, and the linker will throw away the duplicates. Basically, the template instances go in named sections with ".linkonce" attribute, and then linker does the rest. Sometime this weekend, I should have egcs-1.0 distribution for gnu-win32. Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Thu Dec 4 00:20:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Thu, 04 Dec 1997 00:20:00 -0000 Subject: BUG: stdin not a constant References: <199712040048.LAA26596@rimmer.cs.mu.OZ.AU> Message-ID: <199712040820.TAA17063@mundook.cs.mu.OZ.AU> Peter David ROSS writes: >The following code doesn't compile. > >------ >#include > >static FILE *out = stdout; > >void main(void) >{ > /* out = stdout; */ > fprintf(out, "hello world\n"); >} That code is not ANSI/ISO-conformant C. The ANSI/ISO C standard does not require `stdout' to be a constant suitable for use in a static initializer. >The fix is to remove the initialiser and add the line which is commented out. Correct. >However the code I am trying to port has these sort of variables spread >all over the place, You have my sympathy ;-) >and I would like to know if there is someway I can >get the compiler to treat it as a constant. I'm fairly sure there's no simple way. I think you'll just have to fix your non-conformant source code. -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From eckert@netcom.com Thu Dec 4 00:21:00 1997 From: eckert@netcom.com (SRE) Date: Thu, 04 Dec 1997 00:21:00 -0000 Subject: Shell-Here Explorer context menu References: <3.0.32.19971203045255.00ab6630@bl-mail2.corpeast.baynetworks.com> Message-ID: <3.0.1.32.19971203233232.00986430@10.0.2.2> At 04:52 AM 12/03/97 -0500, Robert Seeger wrote: >Very nice. I like this. I've wished for this type of thing for a while, but >didn't know how to do it. Thanks. . . I'm fairly new to this list, so hopefully I'm not asking a stoopid questiont that's been asked and answered recently... Opening a shell is nice, now how about drag-and-drop paths? Is there a semi-convenient way to drag a file out of Explorer and drop it into a bash window? The default behavior is to have paths like "D:\temp\file.txt" which bash (of course) interprets as "D:tempfile.txt" or worse since the backslashes are not seen as directory separators. It should be seen as "//d/temp/file.txt", right? This same thing is a problem when trying to use utilities like diff from a (NT) command shell, or double-clicking on an icon that's supposed to start a script, or dragging a file onto an icon. Anything with file names as arguments. SRE mailto:eckert@netcom.com -or- mailto:list_owner@juno.com "The only possible interpretation of any research whatever in the `social sciences' is: some do, some don't." -- Ernest Rutherford - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From anazawa@hic-soft.co.jp Thu Dec 4 00:21:00 1997 From: anazawa@hic-soft.co.jp (ANAZAWA Akio) Date: Thu, 04 Dec 1997 00:21:00 -0000 Subject: BUG: stdin not a constant (fwd) References: <199712040054.LAA26717@rimmer.cs.mu.OZ.AU> Message-ID: <199712040821.RAA19631@mail.hic-soft.co.jp> Hi, Peter. In message "BUG: stdin not a constant (fwd)" on 97/12/04, Peter David ROSS writes: Peter> The fix is to remove the initialiser and add the line which is Peter> commented out. Peter> However the code I am trying to port has these sort of Peter> variables spread all over the place, and I would like to know Peter> if there is someway I can get the compiler to treat it as a Peter> constant. One of very limited solution could be... #include /* static FILE *out = stdout; */ #define out stdout void main(void) { /* out = stdout; */ fprintf(out, "hello world\n"); } Anyway, I'm worndering why Cygnus's gcc couldn't accept Peter's code. All of other portings of gcc (and others) that I can use around here accepted it (gcc 2.7.0 on SPARC, 2.6.3 on FreeBSD, Sun's workshop cc, VC++ 4.2 and 5.0). -- @___ anazawa@hic-soft.co.jp c>//_ --- Human Interface Communications Co.,Ltd.(H.I.C.) (_)\(_) --- Phone +81 3 5467 2401 / Fax +81 3 5467 2402 ------------------------------------------------------------------------ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kh@wg.icl.co.uk Thu Dec 4 01:44:00 1997 From: kh@wg.icl.co.uk (Kevin Hughes) Date: Thu, 04 Dec 1997 01:44:00 -0000 Subject: case sensitive directory names Message-ID: <01BD0098.CFFDAEE0@rodney.wg.icl.co.uk> I have a bad habit of using upper and lower case names in my directory names - not mixing them but having a directory called UPPERCASE rather than uppercase. This is no problem on Unix which copes as it handles upper and lower case for the names. NT does not - all files are essentially lower case. Fine I can cope with that but gnuwin appears to remember the name I typed and then uses that in the answer to pwd. Hence can do the following bash$ pwd /devel/kh/CC/SC bash$ cd .. bash$ pwd /devel/kh/CC bash$ cd sc bash$ pwd /devel/kh/CC/sc Since I have scripts that check "where" I am in a directory structure before they will do anything this gets a bit tedious. I also have scripts which store away the answer to "pwd" and check it next time I try to do things. I Is there any real reason for this behaviour.? Can I turn it off and force lower case always? Kevin - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From iverson@lionheart.com Thu Dec 4 02:13:00 1997 From: iverson@lionheart.com (Tim Iverson) Date: Thu, 04 Dec 1997 02:13:00 -0000 Subject: Too much mail volume References: <199712032134.VAA51864@out4.ibm.net> Message-ID: Bzzzt. Web? Just say no. In fact if the list went there, I'd drop it. Web is fine for initial research, but beyond that it's a waste of time. Certainly, browser interfaces for reading and sending messages on the Web are at least a decade behind what's available for mail and news on Unixen. - Tim Iverson iverson@lionheart-com +---------------- | From: vischne@ibm.net | Date: Wed, 3 Dec 1997 21:34:09 GMT | Subject: Re: Too much mail volume | To: gnu-win32@cygnus.com | | Look, if you guys had any smarts, you would scrap the mail version of this | list entirely and go with the archive instead. | | That way, people interested in knowing what is going on in the discussion | can tune in to the archive and not bother with their e-mail server. You | would also add a URL for posting followups to archive articles. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kkont@estia.iit.nrcps.ariadne-t.gr Thu Dec 4 04:36:00 1997 From: kkont@estia.iit.nrcps.ariadne-t.gr (Kimon Kontovasilis) Date: Thu, 04 Dec 1997 04:36:00 -0000 Subject: Piped Commands Loss Of Output References: <34856BB1.68CC2642@boxhill.com> Message-ID: <9712041237.AA09143@estia.iit.nrcps.ariadne-t.gr> > > Has anyone noticed a transient loss of output with piped commands > similar to the following: > > $ fgrep export .bashrc | wc > 12 30 339 > $ fgrep export .bashrc | wc > $ > Yes, I 've observed it too. > >From the above examples, I conclude that sometimes I get all output, > no output, and some output. > I experience the same random behavior; it happens only from within bash. (When I use 4DOS.COM, I don't get problems; I cannot be decisive about COMMAND.COM, since I don't use it frequently.) > I am using b18 with Sergey's coolview 25 Nov 97. I seem to remember > that I began to observe this problem when I upgraded to one of Sergey's > coolview ?? Sept 97 versions. Has anyone observed this too? Anyone > know whats wrong? > I use Win95 + b18 + coolview of Nov. 03, 1997. (Is your Nov. 25 version different? I didn't notice an announcement for anything newer from what I have. If yes, what are the differences?) Bye, Kimon. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tomas.fasth@twinspot.net Thu Dec 4 04:39:00 1997 From: tomas.fasth@twinspot.net (Tomas Fasth) Date: Thu, 04 Dec 1997 04:39:00 -0000 Subject: Do we need commercial advertisements in the gnu-win32 list? References: <009BDFEE.EA443220.29140@ifk20.mach.uni-karlsruhe.de> <3481B9AE.C48FDD48@opennt.com> <3482F866.1C470A4F@twinspot.net> <3485429E.39FD13DC@twinspot.net> <3485EAC1.C47E3828@opennt.com> Message-ID: <3486A21A.AE6B396E@twinspot.net> Jason Zions wrote: > According to your position, Tomas, I'm not even allowed to respond to > this set of assertions. How sad. Well Jason, I hate censorship. So please respond as you like, as long as you use a common sense of what's within the scope of this list and what's not. I honor free speach and all such that goes with basic human rights. And I don't feel good about making this into a fight between you and me. But I thought there was a reason for naming this list GNU-win32. GNU stands for certain fundamental principles. Commercial binary-only software is not part of it as I understand. I respect you as a frequent contributor to this list and consider you an individual among individuals, but not a representative for your company. And that's the whole point. Please don't trade your well deserved reputation in this community of freeware consumers and contributors for some possible marginal gains in sales of your company's commercial products. Not doing this might affect your personal earnings in some way, but that goes for everybody on this list I guess. I consider keeping a low commercial profile a proper behavior on a list like this. Sorry for making you feeling sad. Tomas - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 4 05:07:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 04 Dec 1997 05:07:00 -0000 Subject: GNU Win32 Installation Message-ID: <19971204130711.19767.qmail@hotmail.com> >From: "Caskey, William" >To: "'gnu-win32@cygnus.com'" >Subject: GNU Win32 Installation >Date: Wed, 3 Dec 1997 11:22:37 -0600 > >Thanks for your reply. I fixed the environment variable by adding a \, >but still encountered the same problem. I then explored the directories >and noticed a cp.exe executable. I renamed this to cpp.exe and cp.exe located in the /gnuwin32/b18/H-i386-cygwin32/bin directory is the copy command. IE cp fromhere tothere cpp.exe is located in /gnuwin32/b18/H-i386-cygwin32/lib/gcc-lib/i386-cygwin32/cygnus-2.7.2-970404 - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 4 06:19:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 04 Dec 1997 06:19:00 -0000 Subject: case sensitive directory names Message-ID: <19971204141828.3649.qmail@hotmail.com> >From: Kevin Hughes >To: "Gnuwin95 (E-mail)" >Subject: case sensitive directory names >Date: Thu, 4 Dec 1997 09:41:33 -0800 > >I have a bad habit of using upper and lower case names in my directory names - not mixing them but having a directory called UPPERCASE rather than uppercase. > >This is no problem on Unix which copes as it handles upper and lower case for the names. NT does not - all files are essentially lower case. Fine I can cope with that but gnuwin appears to remember the name I typed and then uses that in the answer to pwd. Hence can do the following > > >bash$ pwd >/devel/kh/CC/SC >bash$ cd .. >bash$ pwd >/devel/kh/CC >bash$ cd sc >bash$ pwd >/devel/kh/CC/sc > >Since I have scripts that check "where" I am in a directory structure before they will do anything this gets a bit tedious. I also have scripts which store away the answer to "pwd" and check it next time I try to do things. I > >Is there any real reason for this behaviour.? Can I turn it off and force lower case always? > The real problem is that the underlying file system is case insensitive. When you cd, an environment variable is set that the pwd command reads and echos. Because lowercase and LOWERCASE resolve to finding the directory then the environment variable will be set to what you keyed on the command line. UNIX's file system on the otherhand is case sensitive. So if lowercase exists and LOWERCASE doesn't then when you cd LOWERCASE you would get an error. What problems is this nature causing? From your examples I don't see a real problem as the directory names should be found in either case. If it is something you really want you could modify the result from the pwd command to translate to lowercase. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 4 06:19:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 04 Dec 1997 06:19:00 -0000 Subject: Too Much Mail Volume Message-ID: <19971204133521.16333.qmail@hotmail.com> >From: "Paul Garceau" >To: gnu-win32@cygnus.com >Date: Wed, 3 Dec 1997 17:54:18 -0800 >Subject: Re: Too Much Mail Volume >Reply-to: pgarceau@submergedsoftware.com > >Greetings, > >On 30 Nov 97 at 12:45, the Illustrious Steven Bellenot wrote: > >> >> This mailing list has so much volume that I think we would be better >> served with a newsgroup to move the traffic and to provide a FAQ. > > I tend to agree with this...in two days I've received nearly 100 emails >from the list. > Most of my mail from this list for the past 3 or 4 days has been this thread. Please, put an end to it. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cvn@interchain.nl Thu Dec 4 07:19:00 1997 From: cvn@interchain.nl (Kees van Veen) Date: Thu, 04 Dec 1997 07:19:00 -0000 Subject: gethostby... broken when forked References: <34862B09.3FFE@dtv.net> Message-ID: <3486CA4B.24CA477A@interchain.nl> Jeff Smart wrote: > > I have found the BSD calls gethostbyname and gethostbyaddr break when > they are used in a child process that was forked. > ... > Has anyone else found this nasty ? Yes I have experienced this on W95 too, I moved to NT4 SP3 where your example program works fine. Kees - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From GBradfor@fhssmtp.redstone.army.mil Thu Dec 4 07:19:00 1997 From: GBradfor@fhssmtp.redstone.army.mil (GBradfor) Date: Thu, 04 Dec 1997 07:19:00 -0000 Subject: Too Much Mail Volume Message-ID: <0009FC54.3272@fhssmtp.redstone.army.mil> On 12/3/97, Michael Anthon at Internet wrote: >About 10 messages in one day is not very high volume. I am behind a >firewall and accessing news groups is a problem for me. I have no >objection to a dual system, but some thought might be put into removing >or distorting mail addresses since junk mailers that harvest news >groups are a real problem. 10 a day! I was gone a week and had well over 200 messages to clear out of my box from this list alone. I was out yesterday, and this morning I had 75 messages in my box! Personally, I (and my mail admin) like the idea of a newsgroup. /s/ Pat Bradford Respond to: gbradfor@redstone.army.mil AND/OR p_n_brad@traveller.com Received: from michp758.redstone.army.mil (136.205.10.78) by FHSSMTP.REDSTONE.ARMY.MIL with SMTP (IMA Internet Exchange 2.11 Enterprise) id 00009990; Wed, 3 Dec 97 15:16:04 -0600 Received: from mic14.redstone.army.mil by michp758.redstone.army.mil id aa09358; 3 Dec 97 15:10 CST Received: from runyon.cygnus.com by mic14.redstone.army.mil id aa19032; 3 Dec 97 15:10 CST by runyon.cygnus.com (8.8.7-cygnus/8.8.7) id AAA26028; Wed, 3 Dec 1997 00:01:36 -0800 (PST) Received: from server1.tams (tams.com.au [139.130.64.206]) by runyon.cygnus.com (8.8.7-cygnus/8.8.7) with ESMTP id AAA26024 for ; Wed, 3 Dec 1997 00:01:32 -0800 (PST) Received: by server1 with Internet Mail Service (5.0.1458.49) id ; Wed, 3 Dec 1997 18:01:21 +1000 Message-ID: < 3120D53DC9D5D011A59A00C0F016383303A302@server1 > From: Michael Anthon To: gnu-win32@cygnus.com Subject: RE: Too Much Mail Volume Date: Wed, 3 Dec 1997 18:01:19 +1000 X-Priority: 3 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.0.1458.49) Content-Type: text/plain Sender: owner-gnu-win32@cygnus.com Precedence: bulk - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From banders@ECD.Rockwell.COM Thu Dec 4 07:28:00 1997 From: banders@ECD.Rockwell.COM (Bartlee A. Anderson) Date: Thu, 04 Dec 1997 07:28:00 -0000 Subject: Too much mail volume References: <199712032134.VAA51864@out4.ibm.net> Message-ID: <3486CC8A.2E5C@switch.rockwell.com> vischne@ibm.net wrote: > > Look, if you guys had any smarts, you would scrap the mail version of this While I __don't__ agree with this slam at our collective cognitive abilities, I somewhat like the ideas proposed. Anyone else think there is some merit here? Bart > list entirely and go with the archive instead. > > That way, people interested in knowing what is going on in the discussion > can tune in to the archive and not bother with their e-mail server. You > would also add a URL for posting followups to archive articles. > > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". -- Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) Rockwell International Electronic Commerce Division 300 Bauman Ct. banders@ecd.rockwell.com MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 Wood Dale, IL 60191 FAX (630) 227-9771 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Thu Dec 4 07:28:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Thu, 04 Dec 1997 07:28:00 -0000 Subject: Piped Commands Loss Of Output References: <34856BB1.68CC2642@boxhill.com> Message-ID: In article < 34856BB1.68CC2642@boxhill.com >, Jason Tishler wrote: >Has anyone noticed a transient loss of output with piped commands >similar to the following: > > $ fgrep export .bashrc | wc > 12 30 339 > $ fgrep export .bashrc | wc > $ > >[snip] > >I am using b18 with Sergey's coolview 25 Nov 97. I seem to remember >that I began to observe this problem when I upgraded to one of Sergey's >coolview ?? Sept 97 versions. Has anyone observed this too? Anyone >know whats wrong? I've observed this. I added this to my .profile to "fix" the problem. Maybe it will work for you: trap '' 5 In cygwin B18, signal 5 is SIGHUP. I think there is a problem in recent coolviews with the way SIGHUP is propagated to members of a process group. The `trap' command merely causes the shell to ignore hangup signals. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 4 07:28:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 04 Dec 1997 07:28:00 -0000 Subject: Shell-Here Explorer context menu (NBY) Message-ID: <19971204135649.22359.qmail@hotmail.com> BASH is command line only; just like MSDOS only a little smarter. The rest of the .exe's in the set are to be executed in a command shell. No GUI. However you should be able to create GUI programs with this tool set. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Wed, 03 Dec 1997 23:32:32 -0800 >To: gnu-win32@cygnus.com >From: SRE >Subject: Re: Shell-Here Explorer context menu > >At 04:52 AM 12/03/97 -0500, Robert Seeger wrote: >>Very nice. I like this. I've wished for this type of thing for a while, but >>didn't know how to do it. Thanks. . . > >I'm fairly new to this list, so hopefully I'm not asking a stoopid >questiont that's been asked and answered recently... > >Opening a shell is nice, now how about drag-and-drop paths? Is there a >semi-convenient way to drag a file out of Explorer and drop it into a >bash window? The default behavior is to have paths like "D:\temp\file.txt" >which bash (of course) interprets as "D:tempfile.txt" or worse since the >backslashes are not seen as directory separators. It should be seen as >"//d/temp/file.txt", right? > >This same thing is a problem when trying to use utilities like diff from a >(NT) command shell, or double-clicking on an icon that's supposed to start >a script, or dragging a file onto an icon. Anything with file names as arguments. > >SRE > > mailto:eckert@netcom.com -or- mailto:list_owner@juno.com > >"The only possible interpretation of any research whatever > in the `social sciences' is: some do, some don't." > -- Ernest Rutherford > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Thu Dec 4 07:28:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Thu, 04 Dec 1997 07:28:00 -0000 Subject: Too Much Mail Volume References: <199712040152.UAA15020@bw.webex.net> Message-ID: In article < 199712040152.UAA15020@bw.webex.net >, Paul Garceau wrote: >On 2 Dec 97 at 15:38, the Illustrious Chris Faylor wrote: >> >>I second the idea of moving this mailing list to a newsgroup. >> > >> >No. This isn't a high traffic mailing list compared to a lot of other >> >lists I am on. >> >> I vote emphatically against moving this to a newsgroup. I don't think >> that the traffic justifies it. Also, we'll all be subject to much more >> spam both from the newsgroup spammers and from the spammers who harvest >> mail addresses from newsgroup headers. >> >> If cygwin or some well-connected site wanted to set up a private NNTP >> server to host a private newsgroup, however, I think that would be a >> great idea. > > I can agree to this. It means, of course, that whoever does set up the >NNTP server will need to make doubly-certain that they can't be attacked >by any email moles that are capable of rapeing an NNTP server by stripping >out all of the email addresses stored therein. God, I hate spam. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Thu Dec 4 07:33:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Thu, 04 Dec 1997 07:33:00 -0000 Subject: BUG: stdin not a constant (fwd) References: <199712040054.LAA26717@rimmer.cs.mu.OZ.AU> <199712040821.RAA19631@mail.hic-soft.co.jp> Message-ID: <199712041532.CAA05115@mundook.cs.mu.OZ.AU> ANAZAWA Akio wrote: >Anyway, I'm worndering why Cygnus's gcc couldn't accept Peter's code. >All of other portings of gcc (and others) that I can use around here >accepted it (gcc 2.7.0 on SPARC, 2.6.3 on FreeBSD, Sun's workshop cc, >VC++ 4.2 and 5.0). The important thing is not the version of gcc you're using, it's the C library you're using. Cygwin32 uses Cygnus's "newlib" C library. Cygnus chose to build a new version of the C library (mostly using existing BSD-licence freeware sources), rather than using the probably better-quality GNU C library. This was apparently for financial reasons. -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ert@cit.org.by Thu Dec 4 08:19:00 1997 From: ert@cit.org.by (Andrew Lipnitsky) Date: Thu, 04 Dec 1997 08:19:00 -0000 Subject: gethostby... broken when forked References: <34862B09.3FFE@dtv.net> Message-ID: <3486D7E4.2EB64B2A@cit.org.by> Hi Jeff!!! Your program work fine for me [WinNT4.0 SP3 + gnuwin b18] but it does nothing because the parent process exit before child can do something. The following source determines host properly (at leas for me). #include #include #include #include int main(argc, argv) int argc; char *argv[]; { struct hostent *hostptr; /* Internet host struct from netdb.h */ char hostname[256]; int childpid; strcpy(hostname, argv[1]); printf("PARENT STARTED PID=%d\n", getpid()); if ((childpid = fork()) < 0) { printf("fork failed\n"); exit(0); } else if (childpid == 0) { /* start child process */ printf("CHILD STARTED PID=%d\n", getpid()); hostptr = gethostbyname(hostname); /* BOMBS HERE */ printf("hostptr = %p\n hostname=%s\n", hostptr,hostptr->h_name); exit(0); } else sleep(100); // !!! parent process wait for child } - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jes23@cus.cam.ac.uk Thu Dec 4 09:38:00 1997 From: jes23@cus.cam.ac.uk (J.E. Scott) Date: Thu, 04 Dec 1997 09:38:00 -0000 Subject: should samba drive be mounted binary or text Message-ID: Hi, I'm in total confusion about the text=binary thing. I just got bash to read my shell scripts by mounting my local drive text=binary. Should I mount my unix filesystem (via samba) text=binary or text!=binary? James _________________________________________________________________________ James Scott Tel: +44-1223-339651 Judge Institute of Management Fax: -339652 University of Cambridge Eml: jes23@cam.ac.uk Trumpington Street Web: www-cfr.jims.cam.ac.uk/ Cambridge, England CB2 1AG - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jazz@softway.com Thu Dec 4 09:47:00 1997 From: jazz@softway.com (Jason Zions) Date: Thu, 04 Dec 1997 09:47:00 -0000 Subject: case sensitive directory names References: <01BD0098.CFFDAEE0@rodney.wg.icl.co.uk> Message-ID: <3486ECB1.5C616684@opennt.com> It's not gnuwin that "remembers" the original case you typed - it's the filesystem. NTFS and FAT16 (under NT, anyway) are case-storing filesystems; Win32 is case-insensitive when looking at the stored filenames. Instead of using the bash built-in pwd, use /bin/pwd to get the "real" working directory in a case-consistent way. /bin/pwd walks the filesystem to find out where you are, while the bash built-in tracks it by assuming a starting point and watching the cd commands fly by. (If you cd through a symlink, I think you'll get wildly different answers from the builtin pwd and /bin/pwd; I don't know which is more useful to your scripts.) Jason - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jes23@cus.cam.ac.uk Thu Dec 4 10:09:00 1997 From: jes23@cus.cam.ac.uk (J.E. Scott) Date: Thu, 04 Dec 1997 10:09:00 -0000 Subject: should samba drive be mounted binary or text References: Message-ID: also, does anyone know why more and less don't work in the coolview 'rxvt'. They don't stop at the end of each page, and just emulate cat. James _________________________________________________________________________ James Scott Tel: +44-1223-339651 Judge Institute of Management Fax: -339652 University of Cambridge Eml: jes23@cam.ac.uk Trumpington Street Web: www-cfr.jims.cam.ac.uk/ Cambridge, England CB2 1AG - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From lhall@rfk.com Thu Dec 4 10:29:00 1997 From: lhall@rfk.com (Larry Hall) Date: Thu, 04 Dec 1997 10:29:00 -0000 Subject: Too Much Mail Volume References: <0009FC54.3272@fhssmtp.redstone.army.mil> Message-ID: <3.0.3.32.19971204132345.009a2190@ma.ultranet.com> At 08:49 AM 12/4/97 -0600, GBradfor wrote: > > >On 12/3/97, Michael Anthon at Internet wrote: > >>About 10 messages in one day is not very high volume. I am behind a >>firewall and accessing news groups is a problem for me. I have no >>objection to a dual system, but some thought might be put into removing >>or distorting mail addresses since junk mailers that harvest news >>groups are a real problem. > >10 a day! I was gone a week and had well over 200 messages to clear out of my >box from this list alone. I was out yesterday, and this morning I had 75 >messages in my box! Personally, I (and my mail admin) like the idea of a >newsgroup. > >/s/ >Pat Bradford > > Respond to: > gbradfor@redstone.army.mil > AND/OR > p_n_brad@traveller.com > Whenever this subject comes up on this list (and believe me, it ain't the first time!), I hesitate to contribute to the flow. I think people should have access to this list's information in the form that is best for them. However, it seems everytime this comes up, there is allot of discussion on this list about the hows and wheres and never the whos and whens. If people want a newsgroup, it seems to me it is more productive to figure out the latter rather than the former first. Since NOONE ever seems to volunteer to take this ball and running with it, I'm left with the impression that either noone is REALLY that interested or everybody simply wants someone else to do it. Let me make a suggestion. If there is someone out there that is really interested in getting this list into a newsgroup, send 1 mail message to this list asking that other interested parties to send email back to him or her. Some or all of those parties can formulate what they think is a reasonable plan off-line from this list and then present it back to the list when they are done. It will be painful to have yet another round of these discussions on this list at that time but at least there will be a cohesive (I should hope) proposal to discuss which will address many of the concerns and provide a structure, mechanism, and responsible body for moving this forward or letting it drop. More random discussions on this topic which bring up the same issues with no conclusion every three months or so is not productive for anyone. While I have noticed at times that this list generates mail on the order of 50-75 messages daily, this seems to happen most often when somewhat off-topic threads like this one get proposed or another pertinent topic drifts off-center. Normally I don't find the traffic that unwieldy myself. But to each his/her own. However, as a non-rhetorical question, can we either address this topic in a logical manner off-line from this list or let it die? I'm sure all of our mailboxes will appreciate the reduced traffic either way. I've said my piece. Any flames can be directed specifically to me. No need to clutter the list even more with responses from people who feel I've stepped on their toes! Larry Hall lhall@rfk.com RFK Partners, Inc. (781) 239-1053 8 Grove Street (781) 239-1655 - FAX Wellesley, MA 02181 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From diegom@pts.mot.com Thu Dec 4 10:29:00 1997 From: diegom@pts.mot.com (Diego Montanez) Date: Thu, 04 Dec 1997 10:29:00 -0000 Subject: .bashrc problem Message-ID: <9712041353.AA18188@motprux.pr.paging.mot.com> Hi, I have a problem. The default bash shell that comes with b18 correctly interpreted /.bashrc. But after installing the latest coolvew programs, bash reports an error 0. What's this? Thanks for any help. Diego --- D I E G O A M O N T A N E Z - D I E G O M @ PTS . M O T . C O M - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From gmartin@insidenet.com Thu Dec 4 10:55:00 1997 From: gmartin@insidenet.com (Greg Martin) Date: Thu, 04 Dec 1997 10:55:00 -0000 Subject: Too Much Mail Volume References: <881165089/rvb@gluck.coda.cs.cmu.edu> Message-ID: <3486FC3B.8DE53724@insidenet.com> As a newbie I must say I love Robert's suggestion! Greg Martin Robert.V.Baron wrote: > I think the traffic is more like 20 to 30 pieces a day. For me, the > issue of volume is less important than quality. I'd like to see several > groups (news or mail), for: > newbies > programming questions > interactions with microsoft DLL's, tools > technical discussions > perhaps more. This would allow us to sort and categorize the volume better. > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tomas.fasth@twinspot.net Thu Dec 4 13:02:00 1997 From: tomas.fasth@twinspot.net (Tomas Fasth) Date: Thu, 04 Dec 1997 13:02:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <199712021608.JAA04416@chorus.dr.lucent.com> <34856623.C5226DBC@twinspot.net> Message-ID: <34870542.A0933226@twinspot.net> Chris Faylor wrote: > What security considerations are there that are not also present with > any other scheme, whether it is using extended attributes or setting options > in the registry? You would have to have the right privileges to change > the binary. The binary is normally a single entity, shared among users. Configuring a certain behavior at compile time is just fine, having it modified after installment is not. It will simply introduce all kinds of nightmares. If a user wants to change the behavior of a certain binary, it has to be done within that particular user's environment only. Otherwise, you will end up with a situation where no-one can trust current settings and being forced to check/reset the settings at each and every point of use. If I remember right, the registry allow user specific entries. Also, it's nothing new in the Unix environment to have configuration files for binaries stored within the file system space controlled by current user. We just have to figure out a viable structure to store such information into. > How does a virus detection program detect the difference between installing > a new version of bash or changing a byte in the existing file? It does not. At both occations the virus tripwire will be sprung. But a binary installation is normally a system level activity, or at least done with an intention to share the binary among some or all of the users on that system. A change of a binary's runtime behavior should not require a change to the binary itself. I'm quite surprised that this option came up in the discussion in the first place. Everybody having worked in the Unix environment should realize the obvious security breach such solution would introduce. NT is certainly not an exception. I strongly recommend to leave the binary alone. Tomas - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sreuss@sunwise.uwaterloo.ca Thu Dec 4 13:02:00 1997 From: sreuss@sunwise.uwaterloo.ca (Stephen Reuss) Date: Thu, 04 Dec 1997 13:02:00 -0000 Subject: malloc.h file needed... Message-ID: <3486F0A4.5C19@sunwise.uwaterloo.ca> I'm compiling some C code (meschach libs) and need the malloc.h header file. I looked through the cygwin32 1.8b distribution and noted that there is documentation about a mmalloc library. 1) Is this meant to replace the malloc stuff? 2) How do I do it? - suggestions in the info file are to replance defininions with mmalloc etc... - is there no header for this implementaiotn of malloc 3) Could I get the standard malloc stuff anywhere for cygwin32? Thanks in advance -- --------------------------------------------------------------------- Stephen Reuss ===== ====== ===== M.A.Sc. student, Mechanical Engineering == == == == University of Waterloo, Canada = ===== == = E-mail: sreuss@sunwise.uwaterloo.ca == == == == Phone : (519)888-4567 X3859 ===== == ===== Fax : (519)888-6197 *** G R O U P *** --------------------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ggp@informix.com Thu Dec 4 13:02:00 1997 From: ggp@informix.com (Guy Gascoigne - Piggford) Date: Thu, 04 Dec 1997 13:02:00 -0000 Subject: case sensitive directory names References: <19971204141828.3649.qmail@hotmail.com> Message-ID: <3.0.3.32.19971204120703.00c375b0@pop.pdx.informix.com> At 06:18 AM 12/4/97 PST, you wrote: >What problems is this nature causing? From your examples I don't see a >real problem as the directory names should be found in either case. If >it is something you really want you could modify the result from the pwd >command to translate to lowercase. I can give you one poblem that I've noticed. We use Clearcase for source control, many clearcase commands require knowledge about the Current Working Directory, now take the following scenario. I have a directory: s:\source\Stuff\Project1 if, in bash, I cd //s/source/stuff/project1, it works and appears to set the current directory correctly, however it actually sets the dir to be what was typed case and all), future pwd calls return the lowercase name and not the real mixed case name, this causes Clearcase all kinds of problems. Now I realise that part of the problem here is that Clearcase's handling of a case insensitive filesystem is somewhat lacking, but the behaviour of the Cygwin cd command is also somewhat lacking. Having not looked at it, I can't say exactly where the problem lies, but it would be nice to fix it one day. Guy - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From diegom@pts.mot.com Thu Dec 4 13:02:00 1997 From: diegom@pts.mot.com (Diego Montanez) Date: Thu, 04 Dec 1997 13:02:00 -0000 Subject: compiling termutils and termcap libs Message-ID: <9712041656.AA25282@motprux.pr.paging.mot.com> Hi, I managed to compile the termcap library and termutils with b18, but I had to add -D_STRICT_ANSI on the Makefile because string.h was giving an error with a declaration on line 55. Just to let you all know. Diego --- D I E G O A M O N T A N E Z - D I E G O M @ PTS . M O T . C O M --------------------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From lbaxter@smtp.primavera.com Thu Dec 4 13:02:00 1997 From: lbaxter@smtp.primavera.com (Lincoln Baxter) Date: Thu, 04 Dec 1997 13:02:00 -0000 Subject: drag/drop args (was Re: Shell-Here Explorer) Message-ID: <9712048812.AA881263589@primavera.com> From: Lincoln Baxter on 12/04/97 01:39 PM >>> I've been using perl for quite some time from a CMD script, but if I call it from BASH it can't find any of the files! Not true Look at the bashrc file I sent you! (and attached again) I use perl ALL THE TIME. (from bash) on NT Do you have the PERL5LIB and other environment vars set? ********* Do you have #!perl at the top of all your scripts?***************** ********* Are you calling the script with a pathname the INCLUDES the .pl extenstion ? *********** BASH does not use the the PATHEXT winNT EV... (it would be nice if it did) but what the hell >>>The pathnames used by BASH cannot be understood by perl... If you are talking about the virtual mount paths OF COURSE... these are stored in the registry under a cynwyn key. No other tools know about them. If you are talk about the symlinks... OF COURSE... it would have been nice of WinNT has supported these, (They are JUST like unix symlinks in their contents)... bash understands them... it is too bad M$ didn't put this in. If you look at the .bashrc file you will notice that I create an environment var (EV) CDISK with represents the current disk that the virtual mount represents. When I use paths as arguements to winNT programs I use the $CDISK EV. Set the SetLocation and GoTo functions in the bashrc file I sent you. I strongly urge you to understand this .bashrc file... I spend ALOT of time on it... still am, it makes bash VERY usable in a NT environment >>>so does that mean I have to get a new version of perl which was compiled for use with gnu-win32 That is an interesting Idea... but it would be a generic unix port, without support the WINNT stuff like OLE automatic etc that NT perl does today. What I'd like is: for bash to understand how to convert ANSI terminal escape sequences for bold and reverse video, to the NT output formats so that scripts that use these would work like they do on unix. Also I'd also like, to know why it gets 'Out of queue slots!' messages during rm -rf and it would be nice to see the 'setgrent not implemented' messages go away... Lincoln (See attached file: lincoln.bashrc) (See attached file: .bashrc) From jeffdbREMOVETHIS@netzone.com Thu Dec 4 13:02:00 1997 From: jeffdbREMOVETHIS@netzone.com (Mikey) Date: Thu, 04 Dec 1997 13:02:00 -0000 Subject: bash problem. (Am I missing something?) References: <19971203125444.14018.qmail@hotmail.com> Message-ID: <34870cc6.70818231@smtp.netzone.com> Thank you for the information ;^) However executing the script via source ./test.sh or source test.sh or . test.sh (^ note space) has the intended result, the var gets set in the current shell the reason I said to use source instead of . is that . is ambiguous in a mail message. see. .-) executing via a $PATH search like this test.sh or by ./test.sh like he did the variable remains unset. which would indicate to me that the script is executed in a subshell, as I stated. On Wed, 03 Dec 1997 04:54:44 PST, you wrote: > > >> >>try >> >>source ./test.sh >> >>#!/bin/sh causes the script to run in a subshell, so any exports >>are lost when the toplevel shell resumes. at least I think that's >right;^) > >wrong. source is equates to Bourne's . command which will read the >given FILENAME and execute the commands in the current shell context. >This means that any environment variable set will remain at the end of >the script. Any cd performed will remain in effect as well. > >>no idea where the error comes from unless #! aren't the very first 2 >chars in the file. >> > >I believe the problem to be \r in the script file. And possibly >text!=binary. > >[snip] > >- \\||// >---o0O0--Earnie--0O0o---- >-earnie_boyd@hotmail.com- >------ooo0O--O0ooo------- > > >______________________________________________________ >Get Your Private, Free Email at http://www.hotmail.com >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > (jeffdbREMOVETHIS@netzone.com) delete REMOVETHIS from the above to reply Mikey Windows vs Linux is a no-win situation. Windows 95: -- 32-bit extensions and a GUI shell for a 16-bit patch to an 8-bit OS originally coded for a 4-bit processor written by a 2-bit company that can't produce 1 bit of quality. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From eckert@netcom.com Thu Dec 4 13:02:00 1997 From: eckert@netcom.com (SRE) Date: Thu, 04 Dec 1997 13:02:00 -0000 Subject: drag/drop args (was Re: Shell-Here Explorer) References: <19971204135649.22359.qmail@hotmail.com> Message-ID: <3.0.1.32.19971204092711.009cf9d0@10.0.2.2> At 05:56 AM 12/04/97 PST, Earnie Boyd wrote: >BASH is command line only; just like MSDOS only a little smarter. Right, but I've got lots of reasons to drag files into command line applications. I do it with FTP, for instance, instead of "cd path" and then "put file". I just type "put", and drag the file off explorer. Dragging into a BASH window works also, it's just that none of the tools know how to deal with the text that shows up on the command line. >The rest of the .exe's in the set are to be executed in a command shell. >No GUI. I don't really want a GUI shell. My drag/drop icons typically start either a NT CMD script or a console application... neither of which know about windows, but both of which need to be able to feed command line args into a program or set of programs. I'm just looking for a way to easily create arguments, rather than typing them in by hand. It is "windows", after all, not "vt100". One of the reasons I use my NT box more than my Sun and HP unix boxes is precisely because I don't have to type filenames or set up lots of clever aliases. But what I'm hearing is that you can only use gnu-win32 tools in isolation from other tools, right? For instance, I've been using perl for quite some time from a CMD script, but if I call it from BASH it can't find any of the files! The pathnames used by BASH cannot be understood by perl... so does that mean I have to get a new version of perl which was compiled for use with gnu-win32 ? Ugh. SRE mailto:eckert@netcom.com -or- mailto:list_owner@juno.com "The only possible interpretation of any research whatever in the `social sciences' is: some do, some don't." -- Ernest Rutherford - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mike_lottridge@mentorg.com Thu Dec 4 16:29:00 1997 From: mike_lottridge@mentorg.com (Mike Lottridge) Date: Thu, 04 Dec 1997 16:29:00 -0000 Subject: coolview error with bashrc Message-ID: <01BD00C3.128264C0.mike_lottridge@mentorg.com> I just installed the latest coolview patch, and find the following error message when I invoke bash: bash.exe: C:/usr/.bashrc: error 0 I did not get this error prior to the coolview patch, and haven't changed anything in the bashrc file. Regards, Mike Lottridge Technical Program Manager, WorkXpert Product Group Mike_Lottridge@mentorg.com (503)685-1024 fax: (503)685-1950 http://www.workxpert.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ycdtosa@icsa.es Thu Dec 4 16:29:00 1997 From: ycdtosa@icsa.es (ycdtosa) Date: Thu, 04 Dec 1997 16:29:00 -0000 Subject: Help on groff config Message-ID: <199712042329.XAA07949@icsa.es> this is my second time around, I tried gnu-win32 once with no success, but I'm back. I just did install gnu-win32 on my NT40 and guess what! I made it work. I did create the /tmp, set the enviroment, the TERM, mount the /bin, get the passwd and group files on /etc , configure and build the gnu hello-1.3 (out of the box), got vim binary (not cygnus) working, got less binary (a cygnus one) working, got lynx (a cygnus one) no working yet. I do even got the green exes and blue dirs... etc. And wanted to *thank you all*, cause I been reading the archives for a couple of days before installing and got most of the answers I needed from there, and cause I got you binaries from some where near BUT... ;-) ... here is the hard stuff I'm porting (out of the box: sorry, doing my best) groff-1.11a and got it to configure and build, and even test-groff almost right (ascii, latin1) but I cant make Install the make install doesn't get the extension neither the path correctly. Well, I dont pretend to fix the configure (dont know who to do so: remember the hello stuff abobe, it was my first time on configure) but kind of may move the files to the right directories.... I just hapens I dont know how to handle the /devlj4 /devascii /devlatin1 files or what should I put under /whatever/groff/fonts do I have to make any weirdo mapping? or something like that Oh, I would send the Makefile or the config or whatever to anyone who requires it, even the binaries if i do get them working. Thanks in advantage -------------- ycdtosa@icsa.es -------------- The present day composer refuses to die - - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ycdtosa@icsa.es Thu Dec 4 16:29:00 1997 From: ycdtosa@icsa.es (ycdtosa) Date: Thu, 04 Dec 1997 16:29:00 -0000 Subject: Piped Commands Loss Of Output Message-ID: <199712042329.XAA07951@icsa.es> Since I did install gnu-win32 from b18 with no coolview neither patches I cant tell: that doesn't happen to me. same output on $ ls | fgrep z 10 times $ grep . .bash_history | fgrep ls | wc with same output 20 times since this is my first gnu-win32 instalation I dont fell sure enought to upgrade with the coolview yet, (don't know watta hell is that) but maybe I do so in a couple of days. I'll post ! ;-) ---------- > De: Jason Tishler > A: gnu-win32 > Asunto: Piped Commands Loss Of Output > Fecha: mi????rcoles 3 de diciembre de 1997 15:24 > > Has anyone noticed a transient loss of output with piped commands > similar to the following: > > $ fgrep export .bashrc | wc > 12 30 339 > $ fgrep export .bashrc | wc > $ > > Here is another one: > > $ grep . .bash_history | fgrep ls > ls > ls > ... > ls //k > ls //k/vim-5.0h/vim .bashrc > ls //k/vim-5.0h/ > ls //k/vim-5.0h/bin/vim .bashrc > ls > ls > $ grep . .bash_history | fgrep ls > $ grep . .bash_history | fgrep ls > ls > ls > $ > > From the above examples, I conclude that sometimes I get all output, > no output, and some output. > > I am using b18 with Sergey's coolview 25 Nov 97. I seem to remember > that I began to observe this problem when I upgraded to one of Sergey's > coolview ?? Sept 97 versions. Has anyone observed this too? Anyone > know whats wrong? > > Thanks, > Jason > > -- > Jason Tishler Phone: +1 (212) 989-4455 ext. 2120 > Box Hill Systems Corporation Fax: +1 (212) 989-6817 > 161 Avenue of the Americas Email: jt@boxhill.com > New York, NY 10013 USA WWW: http://www.boxhill.com > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net-nospam Thu Dec 4 16:29:00 1997 From: vischne@ibm.net-nospam (vischne@ibm.net-nospam) Date: Thu, 04 Dec 1997 16:29:00 -0000 Subject: ansi.sys seems to help Linux terminal apps Message-ID: <199712050029.AAA04844@out5.ibm.net> The built-in cygnus version of ansi.sys that every gcc console application inherits from cygwin.dll seems to work differently, and slightly better, when you have devicehigh=ansi.sys in your config.sys file. Now, if only the cygwin.dll version of ansi.sys did a better job of keeping up with fast typists and with filling in portions of the termcap screen, there would be no need to talk about going over to CRT.DLL. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jfmiller@polymail.cpunix.calpoly.edu Thu Dec 4 16:29:00 1997 From: jfmiller@polymail.cpunix.calpoly.edu (John Miller) Date: Thu, 04 Dec 1997 16:29:00 -0000 Subject: _beginthread Problems with win95 and mingw32 using C++ Message-ID: <3487421B.1637356E@polymail.calpoly.edu> Dear Friends, I have encountered a problem using C++ with the _beginthread(VOID (*)(PVOID), unsigned, void *) function which is part of the windows95 operating system that allows concurrent execution of multiple threads. the following program compiles and runs fine as a C program but dies at the compilation proper stage for C++ as shown below. I am using gnu-win32b18 /w Mingw32. I am running windows95 (not that this really changes much). I was able to solve the error below by commenting out the "__attribute__((__cdecl__))" portion, but then when I attempt to use the _beginthread() finction I get an ld message that says "undefined reference to _beginthread()" ld fatal signal 1" this also doesn't happen in "C". Please help. thanks, John Miller test.c/test.cc: #include #include #include int main(void) { printf("hello world"); return 0; } D:\mingw32\st20b\test>gcc -c test.cc -o test.o In file included from test.cc:2: d:\\mingw32\\include/nonansi/process.h:86: warning: `__cdecl__' attribute ignored d:\\mingw32\\include/nonansi/process.h:87: `type name' declared as function returning a function d:\\mingw32\\include/nonansi/process.h:87: warning: ANSI C++ forbids declaration`pfuncStart' with no type or storage class GCC.EXE: Internal compiler error: program cc1plus got fatal signal 33 D:\mingw32\st20b\test> - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sopchak@aoainc.com Thu Dec 4 16:29:00 1997 From: sopchak@aoainc.com (Jim Sopchak) Date: Thu, 04 Dec 1997 16:29:00 -0000 Subject: BUG: stdin not a constant References: <199712040048.LAA26596@rimmer.cs.mu.OZ.AU> <199712040820.TAA17063@mundook.cs.mu.OZ.AU> Message-ID: <348734F0.4C28@aoainc.com> Petdr, Having dealt with the same problem in a port before, I've found the best way to do it (IMHO), is to quickly identify all such globals (static or otherwise) from the compiler/linker errors and initialize them first thing in main() before any use. Think of it as being the same responsibility you have when initializing member variables in a constructor. #include static FILE *out = (FILE *)NULL ; void my_out_func(char *szMsg) { fprintf(out, "%s", szMsg); } void main(void) { out = stdout ; fprintf(out, "hello world\n"); my_out_func("...so cold 'n cruel.\n") ; } Bon apetite, Jim - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mark.n@usa.net Thu Dec 4 16:29:00 1997 From: mark.n@usa.net (Mark Newnham) Date: Thu, 04 Dec 1997 16:29:00 -0000 Subject: Too Much Mail Message-ID: <95DE24D1FCFED0118B910060940A334004B8@mailhub.adacorp.com> Just another opinion..... I have internet mail delivery to my desktop. I have no access to newsgroups I have a tediously slow Internet connection so, for those of us (and there are many) without news, moving the group to news basically kills our involvement with the list. Moving the system to an archive server seems a retrograde step, as it means we have to continually make the effort to check what's new - and just seeing new topics is for many a useful learning experience. My vote would be to retain the status quo, but if we have to go to a newsgroup, can somebody put a mail feed in that would forward all postings to those of with no access, and also allow us to post the group that way. Regards Mark - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ycdtosa@icsa.es Thu Dec 4 16:29:00 1997 From: ycdtosa@icsa.es (ycdtosa) Date: Thu, 04 Dec 1997 16:29:00 -0000 Subject: no pipe Message-ID: <199712042328.XAA07947@icsa.es> A small problem: I cant get the pipeline on the shell :( I have set different TERMs, and got the same result... Actually I should tell, I'm from Spain so got a weird keyb layout (4 you) and wanna keepit that way. (or I would find no keys like.. ???? this is a spanish letter that... you do not have neither need that we (spanish) *do* need. anyway to write '|' I must follow this secuence: hit [esc] hit [alt-gr] + the 1 key or to get the # key... hit [esc] hit [alt-gr] + the 3 key. Actually is the hit esc stuff what is driving me crazy is there anything I can do to fix this #????"2! stuff note.- I got the vim (from cygnus web site link) and got the | # @ etc stuff working properly.. (just inside the vim, no on bash) - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From anthonm@tams.com.au Thu Dec 4 17:07:00 1997 From: anthonm@tams.com.au (Michael Anthon) Date: Thu, 04 Dec 1997 17:07:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <3120D53DC9D5D011A59A00C0F016383303A323@server1> On Thursday, 4 December 1997 12:31, cgf@bbc.com [SMTP:cgf@bbc.com] wrote: > In article < 34856623.C5226DBC@twinspot.net >, > Tomas Fasth wrote: > >Modifying the exe file "on-the-fly" for environmental settings is not to > >recommend. Two of the reasons I can think of is: > > > >* Security considerations in a multiuser / multiprocess environment. > > What security considerations are there that are not also present with > any other scheme, whether it is using extended attributes or setting options > in the registry? You would have to have the right privileges to change > the binary. Modification of binary files in a multi-user environment is not a good thing. The registry is there, you may as well use it. Something I have not yet seen mentioned is the fact that each user can have a separate profile in the registry, making it easy for each user to have different settings. Modifying the binary might get just a little frustrating if someone else keeps changing it to the way THEY like things (unless you want to fill your hdd with multiple copies of executables) Also, for a user to be able to modify a binary, they would need write access to that binary. I suspect this would cause nightmares for sys admins, not to mention the possibility of adding back doors directly into the binaries themselves. I bet a normal user can't modify any of the shell binaries in a standard Unix setup. > > >* Will cause problem in environments with active virus protection. > > How does a virus detection program detect the difference between installing > a new version of bash or changing a byte in the existing file? SOME virus protection schemes are now including a mechansim whereby any changes to an executable file will raise a warning. Norman Anitvirus is one in particular that I have used that does this. It's a bit of a pain if you forget to disable that feature before installing any new software. If that installation tries to upgrade, for example, a dll file, as most of them seem to do these days, the Virus detection kicks in and gives you a warning. The same thing would apply to modification of the executables as suggested here. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Thu Dec 4 17:09:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Thu, 04 Dec 1997 17:09:00 -0000 Subject: Feedback needed on proposed cygwin feature Message-ID: <199712050109.UAA05569@hardy.bbc.com> >>My changes cause a program to read the registry for options settings prior >>to reading settings from the environment. So, for instance, you can do >>something like this: > >> cygwin -o 'c:\bin\bash.exe' binmode notitle > >I nearly sent a rave about how pointless this is because even more >functionality can already be achieved with environment variables alone >(at the cost of extra short-lived bash processes to tailor the environment). So use environment variables then. I'm not advocating getting rid of environment variables. I don't see how reading of options from the registry can remove functionality. What the registry can do, that your bash scripts only approach can't, is allow things like setting up inetd to default to CYGWIN_TTY. Running inetd from a shell script that sets CYGWIN_TTY=1 is not easy. The other alternative is to set the global environment variable CYGWIN_TTY to 1 and then unset it on a per-session basis. Yup. That's pretty flexible. On top of this, you have to add the not-inconsiderable speed improvement from reading the registry versus firing up bash (or ash or whatever) to start up a shell script which sets an environment variable and runs another program. >My most serious concern was the loss of functionality: > > Your proposition only seems to cater for per-application configuration, > not per-*process* configuration. > >Someone else made this point and referred to multiple copies of binaries >which you thought was a bad thing. But your proposition would *require* >multiple binaries if users wanted per-process configuration! Use environment variables, then. You haven't lost anything. >Without this extension, the proposition only speeds up per-application >configuration and users would then have to fiddle the registry *and* the >environment to achieve per-process configuration. Don't fiddle with the registry. Use environment variables. >With this extension, it speeds up per-process configuration as well. I'm not so sure, since symbolic links mean nothing to the normal command shell. So, you could have: -rwxr-xr-x 1 cgf everyone 1716020 Oct 28 09:02 /bin/bash.exe lrwxr-xr-x 1 cgf everyone 19 Dec 03 09:40 bashbin.exe -> bash.exe and have bash default to "text" mode opens and bashbin default to "binmode" opens. But running bashbin would never work from the command shell: C:\bin>bashbin The name specified is not recognized as an internal or external command, operable program or batch file. Of course, this would work: C:\>bash $ CYGWIN32=nobinmode bash $ Personally, I don't think that that is extremely odious. >So my suggestion is not to use dos-style file paths in the registry, >but rather cygwin-style file paths (thus including binaries, scripts >and symbolic links). This shouldn't be a problem since it only applies >to cygwin programs anyway(?). When you ask a process what name it was invoked by, it returns a MS-DOS style path name. That's what I'm currently using. It is pretty foolproof. I don't remember saying that the registry stuff wouldn't work with scripts but neither scripts nor symbolic links are impossible to do, and if the consensus is that it is crucial, then I'll look into implementing it. My feeling from reading this newsgroup is that 99% of the problems that occur are when a user ports the nifty UNIX XYZ program to cygwin32. Everything compiles ok, but when the user tries to run the program it seems to screw up reading data files. In this case, setting a 'binmode' registry option for the program would solve the problem. I haven't seen much traffic on the list from people complaining that they have to set up shell scripts to turn off 'TITLE' or 'TTY' for a program some of the time but leave them on for others. I also don't see why a program would *want* to sometimes default to opening files in O_BINARY and sometimes default to O_TEXT. Of course, since CYGWIN_BINMODE is not available yet perhaps this is a requirement that is waiting in the wings. Perhaps someone could, in a non-raving manner, enlighten me with some real-world examples. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kunglao@prairienet.org Thu Dec 4 18:54:00 1997 From: kunglao@prairienet.org (Norm Peterson) Date: Thu, 04 Dec 1997 18:54:00 -0000 Subject: Too Much Mail Volume References: <348427C4.7EE8E65E@redes.int.com.mx> Message-ID: <34876CEB.6A4B7C22@prairienet.org> Michael Anderson wrote: > > If this mailing list does get moved to a newsgroup, the mailing list > should continue also. Many ISPs outside the USA don't offer newsgroups, > but I suspect that all have email. Certainly, no one's suggested doing away with the mail list. Also, the reason I mentioned just making an alt.* group is for speed, doing away with the comp.* required voting, waiting, etc. Scripts to do the proper forwarding to/from the list/group wouldn't be that difficult to make (there may even be some out there already). Anyone who doesn't get (or just doesn't like) reading/posting via the group would still get everything via the mail list. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From anazawa@hic-soft.co.jp Thu Dec 4 20:25:00 1997 From: anazawa@hic-soft.co.jp (ANAZAWA Akio) Date: Thu, 04 Dec 1997 20:25:00 -0000 Subject: BUG: stdin not a constant (fwd) References: <199712040054.LAA26717@rimmer.cs.mu.OZ.AU> <199712040821.RAA19631@mail.hic-soft.co.jp> <199712041532.CAA05115@mundook.cs.mu.OZ.AU> Message-ID: <199712050424.NAA25778@mail.hic-soft.co.jp> Hi, Fergus. In message "Re: BUG: stdin not a constant (fwd)" on 97/12/05, Fergus Henderson writes: Fergus> ANAZAWA Akio wrote: >> Anyway, I'm worndering why Cygnus's gcc couldn't accept Peter's >> code. All of other portings of gcc (and others) that I can use >> around here accepted it (gcc 2.7.0 on SPARC, 2.6.3 on FreeBSD, >> Sun's workshop cc, VC++ 4.2 and 5.0). Fergus> The important thing is not the version of gcc you're using, Fergus> it's the C library you're using. Cygwin32 uses Cygnus's Fergus> "newlib" C library. Fergus> Cygnus chose to build a new version of the C library (mostly Fergus> using existing BSD-licence freeware sources), rather than Fergus> using the probably better-quality GNU C library. This was Fergus> apparently for financial reasons. This morning, I checked header files to see what's differ. In FreeBSD 2.1.0 (yes, I'm using this old version), stdout was defined as; #define stdout (&__sF[1]) In this traditional way, stdout could be treated as constant (fixed address) and be assinable by compiler's intializer. In the other hand, in Cygnus', stdout was defined as // in stdio.h #define stdout (_impure_ptr->_stdout) // in sys/reent.h extern struct _reent *_impure_ptr __ATTRIBUTE_IMPURE_PTR__; I haven't seen them so much in detail, but these staff may be orienting to make C library into re-entrant and thread-safe. -- @___ anazawa@hic-soft.co.jp c>//_ --- Human Interface Communications Co.,Ltd.(H.I.C.) (_)\(_) --- Phone +81 3 5467 2401 / Fax +81 3 5467 2402 ------------------------------------------------------------------------ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From svarten@bank.posten.se Fri Dec 5 00:48:00 1997 From: svarten@bank.posten.se (Björn Svartengren) Date: Fri, 05 Dec 1997 00:48:00 -0000 Subject: [Q]: A quick question... Message-ID: <3487BFCC.739076F@bank.posten.se> I know this has probably been asked 1mil times and that it's in some FAQ someware, BUT I can't find it so I ask here anyhow :) I wonna compile a binary staticly, so that I can install the binary on a box without the gnuwin32 installed in anypart. How do I do this? Your's Bj????rne -- "Very funny Scotty, now beam down my cloths!" - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Darren.Starsmore@ska.com Fri Dec 5 00:53:00 1997 From: Darren.Starsmore@ska.com (Darren Starsmore) Date: Fri, 05 Dec 1997 00:53:00 -0000 Subject: RCS integration into MS VC++ ? Message-ID: <3487A56E.25A3@ska.com> Hi All, I know this is straying a little but... Does anyone out there in cyberland know how to integrate RCS (5.x) into MS Visual C++ ? Thanx, Darren. -- Darren Starsmore | email: Darren.Starsmore@ska.com FTIS 1, | tel: +41 1 33 24961 Credit Suisse First Boston, | fax: +41 1 33 22529 Zurich, Switzerland | private email: darren@InternetAddress.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kh@wg.icl.co.uk Fri Dec 5 01:25:00 1997 From: kh@wg.icl.co.uk (Kevin Hughes) Date: Fri, 05 Dec 1997 01:25:00 -0000 Subject: case sensitive directory names Message-ID: <01BD0154.83D95160@rodney.wg.icl.co.uk> -----Original Message----- From: Jason Zions [SMTP:jazz@softway.com] Sent: Thursday, December 04, 1997 5:47 PM To: Kevin Hughes Cc: Gnuwin95 (E-mail) Subject: Re: case sensitive directory names It's not gnuwin that "remembers" the original case you typed - it's the filesystem. NTFS and FAT16 (under NT, anyway) are case-storing filesystems; Win32 is case-insensitive when looking at the stored filenames. Instead of using the bash built-in pwd, use /bin/pwd to get the "real" working directory in a case-consistent way. /bin/pwd walks the filesystem to find out where you are, while the bash built-in tracks it by assuming a starting point and watching the cd commands fly by. (If you cd through a symlink, I think you'll get wildly different answers from the builtin pwd and /bin/pwd; I don't know which is more useful to your scripts.) Jason Jason, Thanks for the suggestion but you have left me confused. The only pwd I know about is the pwd in the bash shell which I thought was the same as the pwd.exe in /gnuwin32/b18/H-i386-cygwin32/bin. There is no /bin/pwd I tried calling the /gnuwin32/b18/H-i386-cygwin32/bin/pwd directly and get exactly the same results as doing pwd What is the /bin/pwd you are refering to Kevin - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 5 05:56:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 05 Dec 1997 05:56:00 -0000 Subject: Help on groff config Message-ID: <19971205130329.7868.qmail@hotmail.com> I created and use the following script. Note the instructions included in the script. #!/bin/sh # file: install # author: earnie email: earnie_boyd@hotmail.com # Install Preprocessor for the gnu-win32 "make install" command. # # To use: # mv /bin/install.exe /bin/ginstall # copy this file to /bin/install # # Note: # No warranty implicit or explicit is given. # You may copy, distribute and use as you see fit. # USE AT YOU OWN RISK. # tstr="" while test $# -gt 0; \ do \ if test -e $1.exe; \ then \ tstr="$tstr$1.exe $2.exe"; \ shift; \ else \ tstr="$tstr$1 "; \ fi; \ shift; \ done echo "/bin/ginstall $tstr" eval "/bin/ginstall $tstr"; \ exit - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From colin@bird.fu.is.saga-u.ac.jp Fri Dec 5 05:56:00 1997 From: colin@bird.fu.is.saga-u.ac.jp (Colin Peters) Date: Fri, 05 Dec 1997 05:56:00 -0000 Subject: _beginthread Problems with win95 and mingw32 using C++ Message-ID: <01bd0141$5f8510a0$fa173185@gbird0.fu.is.saga-u.ac.jp> From: John Miller To: gnu-win32@cygnus.com Date: Friday, December 05, 1997 2:25 PM Subject: _beginthread Problems with win95 and mingw32 using C++ >I was able to solve the error below by >commenting out the "__attribute__((__cdecl__))" portion, but then when I >attempt to use the _beginthread() finction I get an ld message that says >"undefined reference to _beginthread()" ld fatal signal 1" this also >doesn't happen in "C". Please help. The reason the function doesn't work without __cdecl__ is that it is getting name mangled like a C++ function (I think, see below). The header file should feature a pair of sequences like this around the _beginthread and other prototypes: #ifdef __cplusplus extern "C" { #endif ... function prototypes ... #ifdef __cplusplus } #endif The version of Mingw32 you are using might have a bug in that header. You could try downloading version 0.1.4 from my web page, or the latest snapshot if you feel brave (on second thought, if you download the latest snapshot only pull out the process.h header file, the new specs won't work with b18). Now that it occurs to me, what does that prototype look like? It should be unsigned long _beginthread (void (*pfuncStart)(void *), unsigned unStackSize, void* pArgList); It occurs to me that older versions of Mingw32 might have had an attempt to cdecl pfuncStart, which is unnecessary and may cause problems in some cases. Grab process.h out of 0.1.4 (I think) or the latest snapshot. Or just replace the prototype with the one given above. If none of this fixes things it might be a bug in the C++ parser (I doubt it though). It works fine for me using Mumit Khan's Mingw32-native version of EGCS. I can't test against b18 right now (since it isn't on my disk drive). Good luck, Colin. -- Colin Peters -- colin at fu.is.saga-u.ac.jp -- Saga University Dept. of Information Science -- http://www.fu.is.saga-u.ac.jp/~colin -- http://www.geocities.com/Tokyo/Towers/6162 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jt@boxhill.com Fri Dec 5 05:56:00 1997 From: jt@boxhill.com (Jason Tishler) Date: Fri, 05 Dec 1997 05:56:00 -0000 Subject: Piped Commands Loss Of Output References: <34856BB1.68CC2642@boxhill.com> Message-ID: <348802AB.DCA1C07C@boxhill.com> Chris Faylor wrote: > I've observed this. I added this to my .profile to "fix" the problem. > Maybe it will work for you: > > trap '' 5 Thanks Chris -- the above fixed my problem. > In cygwin B18, signal 5 is SIGHUP. I think there is a problem in recent > coolviews with the way SIGHUP is propagated to members of a process group. Shouldn't the fix really be in coolview? Thanks, Jason -- Jason Tishler Phone: +1 (212) 989-4455 ext. 2120 Box Hill Systems Corporation Fax: +1 (212) 989-6817 161 Avenue of the Americas Email: jt@boxhill.com New York, NY 10013 USA WWW: http://www.boxhill.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Fri Dec 5 05:56:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Fri, 05 Dec 1997 05:56:00 -0000 Subject: Piped Commands Loss Of Output Message-ID: <199712051355.IAA00621@hardy.bbc.com> >Chris Faylor wrote: >> I've observed this. I added this to my .profile to "fix" the problem. >> Maybe it will work for you: >> >> trap '' 5 > >Thanks Chris -- the above fixed my problem. > >> In cygwin B18, signal 5 is SIGHUP. I think there is a problem in recent >> coolviews with the way SIGHUP is propagated to members of a process group. > >Shouldn't the fix really be in coolview? Yup. I'll look into what is causing this today. Maybe *I* broke something. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Fri Dec 5 06:13:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Fri, 05 Dec 1997 06:13:00 -0000 Subject: Piped Commands Loss Of Output Message-ID: <01BD0167.C56DD250@gater.krystalbank.msk.ru> Kimon Kontovasilis wrote: > I use Win95 + b18 + coolview of Nov. 03, 1997. (Is your Nov. 25 version > different? I didn't notice an announcement for anything newer from what > I have. If yes, what are the differences?) The last coolview version is nov 03, the last web page update was nov 25 :-) I just fixed the names of user's rights required to run inetd on NT. -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mark.young@unn.ac.uk Fri Dec 5 06:13:00 1997 From: mark.young@unn.ac.uk (Mark Young (in6)) Date: Fri, 05 Dec 1997 06:13:00 -0000 Subject: gcc Message-ID: <199712051413.GAA25953@cygnus.com> Hi I just installed the gnu-win32 on my win95 system but when I try to run gcc I get the error that cpp cannot be found. I checked through the mail archives and have installed it according to the FAQ (i think) and can't find any answers. Any suggestions? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ert@cit.org.by Fri Dec 5 08:07:00 1997 From: ert@cit.org.by (Andrew Lipnitsky) Date: Fri, 05 Dec 1997 08:07:00 -0000 Subject: inetd problem. References: <01BCF5CA.EA2148A0@gater.krystalbank.msk.ru> Message-ID: <34882668.CCD1A1@cit.org.by> Hi Sergey! A few weeks ago I send message about "telnetd:All network ports in use" error. After setting _wide_ system variables CYGWIN_TTY and CYGWIN_TITLE this error disappeared. But when I set this variable for me and inetd accounts only [WindowsNT 4.0 SP3] this error arise again. Why ? At second, I am still can not connect via telnet. When I try to enter my login name and password I always see "login incorrect". I am user of Windows NT domain so I try \+ and + - in all cases this fail. Thanks in advance. ERT. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Fri Dec 5 08:07:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Fri, 05 Dec 1997 08:07:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <199712021608.JAA04416@chorus.dr.lucent.com> <34856623.C5226DBC@twinspot.net> <34870542.A0933226@twinspot.net> Message-ID: In article < 34870542.A0933226@twinspot.net >, Tomas Fasth wrote: >Chris Faylor wrote: > >> What security considerations are there that are not also present with >> any other scheme, whether it is using extended attributes or setting options >> in the registry? You would have to have the right privileges to change >> the binary. > >The binary is normally a single entity, shared among users. Configuring >a certain behavior at compile time is just fine, having it modified >after installment is not. It will simply introduce all kinds of >nightmares. > >If a user wants to change the behavior of a certain binary, it has to be >done within that particular user's environment only. Otherwise, you will >end up with a situation where no-one can trust current settings and >being forced to check/reset the settings at each and every point of use. > >If I remember right, the registry allow user specific entries. Also, >it's nothing new in the Unix environment to have configuration files for >binaries stored within the file system space controlled by current user. >We just have to figure out a viable structure to store such information >into. > >> How does a virus detection program detect the difference between installing >> a new version of bash or changing a byte in the existing file? > >It does not. At both occations the virus tripwire will be sprung. Can you point me to some specific virus software that will complain given the above scenario? >But a binary installation is normally a system level activity, or at >least done with an intention to share the binary among some or all of >the users on that system. > >A change of a binary's runtime behavior should not require a change to >the binary itself. I'm quite surprised that this option came up in the >discussion in the first place. Everybody having worked in the Unix >environment should realize the obvious security breach such solution >would introduce. NT is certainly not an exception. I guess I need more explanation about why it is "obvious" that modifying the binary results in a security breach. A user either has the right to modify the binary or they don't. If they don't, it is not a security problem. If they do then they can replace 'ls' with 'rm' if they want to. They can also edit the binary with 'vi'. So what? Your point about individual users not being able to set their own per-binary defaults is a good one, though. So, I guess that means that we're back to the registry, with all its attendent quirks. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From samkho@osam.com Fri Dec 5 08:07:00 1997 From: samkho@osam.com (Sam Kho) Date: Fri, 05 Dec 1997 08:07:00 -0000 Subject: ftp source Message-ID: <01BD0146.D39C0D40@iCarpool.ci.mtnview.ca.us> hi. i have a set of cygnus win32 binaries which includes ftp (different from the win95 ftp). i just downloaded the cygnus gnu-win32 usertools source, but it does not include the source for this. where might i find this? thank you! - sam - - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Fri Dec 5 08:07:00 1997 From: cgf@bbc.com (Chris Faylor) Date: Fri, 05 Dec 1997 08:07:00 -0000 Subject: Feedback needed on proposed cygwin feature References: <3120D53DC9D5D011A59A00C0F016383303A323@server1> Message-ID: In article < 3120D53DC9D5D011A59A00C0F016383303A323@server1 >, Michael Anthon wrote: >On Thursday, 4 December 1997 12:31, cgf@bbc.com [SMTP:cgf@bbc.com] >wrote: >> In article < 34856623.C5226DBC@twinspot.net >, >> Tomas Fasth wrote: >> >Modifying the exe file "on-the-fly" for environmental settings is not >to >> >recommend. Two of the reasons I can think of is: >> > >> >* Security considerations in a multiuser / multiprocess environment. >> >> What security considerations are there that are not also present with >> any other scheme, whether it is using extended attributes or setting >options >> in the registry? You would have to have the right privileges to >change >> the binary. > >Modification of binary files in a multi-user environment is not a good >thing. The registry is there, you may as well use it. Something I have >not yet seen mentioned is the fact that each user can have a separate >profile in the registry, making it easy for each user to have different >settings. Modifying the binary might get just a little frustrating if >someone else keeps changing it to the way THEY like things (unless you >want to fill your hdd with multiple copies of executables) >Also, for a user to be able to modify a binary, they would need write >access to that binary. I suspect this would cause nightmares for sys >admins, not to mention the possibility of adding back doors directly >into the binaries themselves. I bet a normal user can't modify any of >the shell binaries in a standard Unix setup. No, you're right. A normal user would use environment variables, which are still available. I had envisioned the use of the registry as a way to set defaults for specific programs on a system, not as a way for each user to modify their needs for a specific program. It is a subtle difference. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@buggy.prospect.com.ru Fri Dec 5 08:35:00 1997 From: sos@buggy.prospect.com.ru (Sergey Okhapkin) Date: Fri, 05 Dec 1997 08:35:00 -0000 Subject: Help on groff config Message-ID: <01BD01B5.26647CB0@sos> ycdtosa wrote: > I'm porting (out of the box: sorry, doing my best) > groff-1.11a > and got it to configure and build, and even test-groff almost right (ascii, > latin1) > but I cant make Install > the make install doesn't get the extension neither the path correctly. Modify Makefile.comm in the root groff directory: install_prog: -test -d $(bindir) || mkdir $(bindir) rm -f $(bindir)/$(NAMEPREFIX)$(PROG) $(INSTALL_PROGRAM) $(PROG).exe $(bindir)/$(NAMEPREFIX)$(PROG).exe ^^^^ ^^^^ -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From carvalho@penelope.dss.inpe.br Fri Dec 5 09:13:00 1997 From: carvalho@penelope.dss.inpe.br (Himilcon de Castro Carvalho) Date: Fri, 05 Dec 1997 09:13:00 -0000 Subject: Compiling for Win32 API Message-ID: <01bd01a0$b0b096c0$2010a396@carvalho1.dss.inpe.br> Hello, I have tried to compile a Windows version of "Hello World", called winnie.cpp written by Bartosz Milewski, 1995, with the command: c++ -DSTRICT -o winnie.exe winnie.cpp Compilation results in many errors about undefined references, like the following: C:\WINDOWS\TEMP\cc0010001.o(.text+0x8f):winnie.cc: undefined reference to `GetMessageA@16' Since I have never programmed for Windows (but only for Unix and XWindows), and I'm not used to specific include files or libraries in this case, could someone give me some hint about what I should do? Thanks, Himilcon - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ggp@informix.com Fri Dec 5 11:30:00 1997 From: ggp@informix.com (Guy Gascoigne - Piggford) Date: Fri, 05 Dec 1997 11:30:00 -0000 Subject: case sensitive directory names References: <01BD0098.CFFDAEE0@rodney.wg.icl.co.uk> <3486ECB1.5C616684@opennt.com> Message-ID: <3.0.3.32.19971205095859.0132bce0@pop.pdx.informix.com> At 10:47 AM 12/4/97 -0700, you wrote: >It's not gnuwin that "remembers" the original case you typed - it's the >filesystem. Whilst this may be true, what I see happening is that under bash (and not under cmd.exe) if I cd to a directory with the name typed in lowercase, certain other non-cygwin programs (clearcase) inherit this wrong pathname. To me this means that he problem is separate from bash' intrenal concept of what the CWD is. >NTFS and FAT16 (under NT, anyway) are case-storing filesystems; Win32 is >case-insensitive when looking at the stored filenames. > >Instead of using the bash built-in pwd, use /bin/pwd to get the "real" >working directory in a case-consistent way. /bin/pwd walks the >filesystem to find out where you are, while the bash built-in tracks it >by assuming a starting point and watching the cd commands fly by. (If >you cd through a symlink, I think you'll get wildly different answers >from the builtin pwd and /bin/pwd; I don't know which is more useful to >your scripts.) Well, actually /bin/pwd prints out the same 'wrong case' version of the CWD as bash does. Guy - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From banders@ECD.Rockwell.COM Fri Dec 5 11:30:00 1997 From: banders@ECD.Rockwell.COM (Bartlee A. Anderson) Date: Fri, 05 Dec 1997 11:30:00 -0000 Subject: Did _you_ ever try to rebuild coolview sources? Message-ID: <34884194.6765@switch.rockwell.com> I would like to find out who has tried to get Sergey's coolview sources rebuilt(libcygwin.a cygwin.dll)? What Platform? Were you successful? After getting the results, I'll compile a list and post. I'd really like to be able to rebuild on 95, but am afraid there are some tricks to get around 95's limitations. And I'll need to know from those who have gone before, what are the tight spots yet. -- Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) Rockwell International Electronic Commerce Division 300 Bauman Ct. banders@ecd.rockwell.com MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 Wood Dale, IL 60191 FAX (630) 227-9771 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fshevlin@maths.tcd.ie Fri Dec 5 11:30:00 1997 From: fshevlin@maths.tcd.ie (Fergal Shevlin) Date: Fri, 05 Dec 1997 11:30:00 -0000 Subject: -fallow-single-precision: can't get to work Message-ID: <9712051845.aa25341@salmon.maths.tcd.ie> Hi, I have some code (generated by f2c) which needs to be compiled such that operations with floats are not promoted to double. The gnu-win32/b18 gcc 2.7.2 -fallow-single precision flag does not seem to have any effect, i.e. the numerical results produced after having used it are the same as those produced without it. This is not the case when the same code is compiled with the OSF and Solaris cc compilers with and without the -float and -fsingle options respectively. Does anybody have any idea how I can get this option to work? It has taken me a long time to diagnose this as the problem so I would really appreciate some help. Fergal ==================================================================== Fergal Shevlin Phone: +353-1-6081209 Fax: 6772204 Dept. of Computer Science http://www.cs.tcd.ie/ Trinity College, Dublin 2, Ireland. Fergal.Shevlin@cs.tcd.ie - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 5 11:30:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 05 Dec 1997 11:30:00 -0000 Subject: gcc (NBY) Message-ID: <19971205173814.28302.qmail@hotmail.com> check the GCC_EXEC_PREFIX environment variable. It must end with a \ or / depending on DOS COMMAND SHELL or BASH. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: "Mark Young (in6)" >To: GNU Mailing List >Subject: gcc >Date: Fri, 5 Dec 1997 14:13:00 -0000 > >Hi >I just installed the gnu-win32 on my win95 system but when I try to run >gcc I get the error that cpp cannot be found. >I checked through the mail archives and have installed it according >to the FAQ (i think) and can't find any answers. >Any suggestions? >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ggp@informix.com Fri Dec 5 11:30:00 1997 From: ggp@informix.com (Guy Gascoigne - Piggford) Date: Fri, 05 Dec 1997 11:30:00 -0000 Subject: coolview error with bashrc References: <01BD00C3.128264C0.mike_lottridge@mentorg.com> Message-ID: <3.0.3.32.19971205102305.00a3c280@pop.pdx.informix.com> At 02:44 PM 12/4/97 -0800, Mike Lottridge wrote: >I just installed the latest coolview patch, and find the following error message when I invoke bash: > >bash.exe: C:/usr/.bashrc: error 0 > >I did not get this error prior to the coolview patch, and haven't changed anything in the bashrc file. This version requires that the .bashrc file (and shell scripts in general) be a unix file and not a dos one, i.e. th line terminators need to be \n and not \r\n. For some reason that I don't understand this one file doesn't appear to be affected by the text!=binary setting on the mount, whereas other shell scripts are OK. Guy - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hasmik@maker.com Fri Dec 5 11:30:00 1997 From: hasmik@maker.com (Hasmik Sarkezians) Date: Fri, 05 Dec 1997 11:30:00 -0000 Subject: TCL8.0 Message-ID: <3.0.32.19971205143842.009045b0@nucleus.maker.com> Does anybody know if TCL8.0 has been incorporated into cygwin32 yet?? Or if there is a plan to do that. thanks, hasmik - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 5 12:44:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 05 Dec 1997 12:44:00 -0000 Subject: install on e: HELP NEEDED (NBY) Message-ID: <19971205204423.661.qmail@hotmail.com> To remount your root directory pointer: cd e:\gnuwin32\b18\H-i386-cygwin32\bin .\umount / .\mount -b e:\ / This assumes you are in the dos command shell. Note the -b mounts the file system as text=binary which sets the default mode of file operation with the cygwin32 compiled modules to BINARY. After doing this you will need to remove the \r (^M) from the script files. You can do this with: cp scriptfile \tmp\scriptfile tr -d "\r" < \tmp\scriptfile > scriptfile don't do this with anything but script files. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: ply >To: "'gnu-win32@cygnus.com'" >Subject: install on e: HELP NEEDED >Date: Wed, 3 Dec 1997 09:22:55 +0100 > >Goal : I want to compile kaffe (java virtual machine) on windows Nt. I follow directives and download cygwin b18. >Having no more space on C: i install Cygwin on E: >configure script won't work, bash don't want to launch it. The only way to do is to launch it with a point before : >. ./configure instead of a standard ./configure >Inside the script execution another script is called, the call fails. In fact every shell call fails. >It's not possible to unmount /, then it is not possible to mount root as e: >I mount root / as e: directly in registery. >I create a /bin directory in e: where i put a link of sh.exe. it didn't work better >I put a copy of sh.exe, it didn't work more >I put a link of all exe in /bin. > >I lauch bash in a command box after calling this .bat : > >SET GCC_EXEC_PREFIX=E:\gnuwin32\b18\H-i386-cygwin32\lib\gcc-lib\ >PATH=%PATH%;\bin;E:\gnuwin32\b18\H-i386-cygwin32\bin;E:\gnuwin32\b18\tcl\bin >SET TCL_LIBRARY=E:/gnuwin32/b18/tcl/lib/tcl7.6 >SET GDBTK_LIBRARY=E:/gnuwin32/b18/share/gdbtcl > >CD /D E:\gnuwin32\b18\H-i386-cygwin32\bin > >when i call cygnwin -V, answer is CYGWIN version ??? > >Where am i wrong ? > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 5 12:50:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 05 Dec 1997 12:50:00 -0000 Subject: Problems with ncurses Message-ID: <19971205204810.17519.qmail@hotmail.com> >From: "Buckingham,Mark,SOLON,NUSA" >To: "'gnu-win32@cygnus.com'" >Subject: Problems with ncurses >Date: Wed, 3 Dec 1997 07:41:40 -0800 > >I'm trying to get ncurses compiled & running. (I downloaded ncurses >1.9.9e from the Slackware linux distribution.) > [snip] >TCL_LIBRARY=C:/gnuwin32/b18/tcl/lib/tcl7.6 >TERM=dumb You need to: export TERM=linux >_=/gnuwin32/b18/H-i386-cygwin32/bin/env > >I know people have gotten ncurses to work ok. Any ideas? > >- - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jazz@opennt.com Fri Dec 5 13:22:00 1997 From: jazz@opennt.com (Jason Zions) Date: Fri, 05 Dec 1997 13:22:00 -0000 Subject: case sensitive directory names References: <01BD0098.CFFDAEE0@rodney.wg.icl.co.uk> <3.0.3.32.19971205095859.0132bce0@pop.pdx.informix.com> Message-ID: <348870C6.CA421D88@opennt.com> Sorry one and all - I thought gnuwin had a /bin/pwd that was distinct from the shell built-in. Might actually be a useful tool to build - one that uses Win32 calls to get the precise as-stored pathname of the current directory. Jason - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kunglao@prairienet.org Fri Dec 5 18:45:00 1997 From: kunglao@prairienet.org (Norm Peterson) Date: Fri, 05 Dec 1997 18:45:00 -0000 Subject: Newsgroup it! Message-ID: <3488BABF.2BCF915C@prairienet.org> Well, I started to investigate writing scripts to do the mail <--> news transferring. My first step was to upgrade my version of INN on my Linux box. Then, lo and behold, there on ISC's ftp site is a little file called newsgate.tar.Z. The second paragraph in the README from this mysterious file is this: With these programs and the right set of mail aliases and news sys and active file entries, it is possible to build any set of moderated, unmoderated, one-way, or bi-directional gateways between any set of news and mail groups and lists that your little heart desires. I'll set it up and further see how well it works. I would offer to do it, but I don't think my little modem would handle too much traffic. So, is there anyone willing to donate a little bandwidth/cpuage to start a private newsgroup mirror? For any interested, the url is ftp://ftp.isc.org/isc/inn/contrib/newsgate.tar.Z - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Fri Dec 5 18:45:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Fri, 05 Dec 1997 18:45:00 -0000 Subject: egcs-1.00 binaries for cygwin32 and mingw32 Message-ID: <9712060207.AA25229@modi.xraylith.wisc.edu> Check out http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (or ftp://ftp.xraylith.wisc.edu/pub/khan/gnu-win32/ ) for binary releases of egcs-1.00 for cygwin32 and mingw32 platforms. For those who don't know what these are, check out Cygnus Gnu-Win32 project page at http://www.cygnus.com/misc/gnu-win32/ . Both the distributions are built with egcs-1.00 source + following: - binutils 971023 - libg++-2.8.0b6 (slightly modified) And mingw32 comes with Colin Peter's 971205 mingw snapshot. Cheers, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ ps: I haven't had a chance to put up the source patches for mingw32 (my CVS tree is a mess), but hopefully within the next few days. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ggp@informix.com Fri Dec 5 18:45:00 1997 From: ggp@informix.com (Guy Gascoigne - Piggford) Date: Fri, 05 Dec 1997 18:45:00 -0000 Subject: case sensitive directory names References: <01BD0098.CFFDAEE0@rodney.wg.icl.co.uk> <3.0.3.32.19971205095859.0132bce0@pop.pdx.informix.com> <348870C6.CA421D88@opennt.com> Message-ID: <3.0.3.32.19971205133833.0096d160@pop.pdx.informix.com> At 02:23 PM 12/5/97 -0700, you wrote: >Sorry one and all - I thought gnuwin had a /bin/pwd that was distinct >from the shell built-in. Might actually be a useful tool to build - one >that uses Win32 calls to get the precise as-stored pathname of the >current directory. Oh there is a separate executable, though it's in C:\gnuwin32\b18\H-i386-cygwin32\bin rather than \bin. Either way, it returns the same thing as the builtin. Guy - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From craft@alacritech.com Fri Dec 5 18:45:00 1997 From: craft@alacritech.com (Peter Craft) Date: Fri, 05 Dec 1997 18:45:00 -0000 Subject: printf/scanf: %Lx format broken? Message-ID: <348883B8.5D4F@alacritech.com> Hi, It seems that the long long size modifier (L) is broken in printf and scanf. If I compile the following program: #include #include main(argc, argv) int argc; char **argv; { long long foo; sscanf(argv[1], "%Lx", &foo); printf("foo %Lx\n", foo); } and run it on my NT system as follows I get: ==> ./foo 0x123456789 foo ffffffff Whereas the same program on a Solaris system it gives me the proper: ==> ./a 0x123456789 foo 123456789 Am I missing something? I have also tried %llx instead of %Lx with the same results. Thanks. Pete - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ggp@informix.com Fri Dec 5 18:45:00 1997 From: ggp@informix.com (Guy Gascoigne - Piggford) Date: Fri, 05 Dec 1997 18:45:00 -0000 Subject: More on getting coolview-bash to work with emacs Message-ID: <3.0.3.32.19971205131402.00bac570@pop.pdx.informix.com> I said a few days ago that I'm managed to get bash working correctly under emacs on one machine and yet have it fail with ^M command not found error messages on another system. Well I found out what the problem was, though it's less than ideal - I was using different versions of the coolview cygwin.dll on the two systems. Regretable the old one (dated 9/32/97) works correctly and the newer one (dated 11/2/97) fails. Is the change of behaviour in the new version configurable in any way? Guy - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sreuss@asc.on.ca Fri Dec 5 18:45:00 1997 From: sreuss@asc.on.ca (Stephen Reuss) Date: Fri, 05 Dec 1997 18:45:00 -0000 Subject: problems w. symbol redeclarations Message-ID: <3488864C.3E1@asc.on.ca> I have a c++ program with which I am using the meschach vector/matrix library, and I am having troubles with the redeclaration of the pair structure. The compile command and error recieved are: g++ -c -D__GNU_COMPILER -DWIN32 -D__WIN32__ -DWINDOWS -O2 -funroll-loops -fexpensive-optimizations -fomit-frame-pointer -ffast-math -malign-loops=2 -malign-jumps=2 -malign-functions=2 -I/usr/Programs/Mesa-2.5/include -I/LIBS/mv++/include -I/usr/Programs/Meschach/include coef.c In file included from /usr/Programs/Meschach/include/sparse2.h:36, from coef.c:109: /usr/Programs/Meschach/include/sparse.h:66: `struct pair' redeclared as different kind of symbol E:\\gnuwin32\\b18\\H-i386-cygwin32\\lib\\gcc-lib\\i386-cygwin32\\cygnus-2.7.2-970404\\../../../../../include/g++/pair.h:35: previous declaration of `template struct pair' /usr/Programs/Meschach/include/sparse.h:66: `typedef struct pair pair' redeclared as different kind of symbol E:\\gnuwin32\\b18\\H-i386-cygwin32\\lib\\gcc-lib\\i386-cygwin32\\cygnus-2.7.2-970404\\../../../../../include/g++/pair.h:35: previous declaration of `template struct pair' g++: Internal compiler error: program cc1plus got fatal signal 33 make: *** [coef.o] Error 1 Is there any way of forcing the compiler to use the desired (meschach) version of the pair structure? Perhaps there is a way to omit the system pair.h header - is this a smart thing? Thanks in advance... -- --------------------------------------------------------------------- Stephen Reuss ===== ====== ===== M.A.Sc. student, Mechanical Engineering == == == == University of Waterloo, Canada = ===== == = E-mail: sreuss@sunwise.uwaterloo.ca == == == == Phone : (519)888-4567 X3859 ===== == ===== Fax : (519)888-6197 *** G R O U P *** --------------------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From justins@iName.com Fri Dec 5 23:13:00 1997 From: justins@iName.com (Justin Smith) Date: Fri, 05 Dec 1997 23:13:00 -0000 Subject: Problems with ncurses References: Message-ID: <3488FB22.66CF9A22@iname.com> Buckingham,Mark,SOLON,NUSA wrote: > > I'm trying to get ncurses compiled & running. (I downloaded ncurses > 1.9.9e from the Slackware linux distribution.) > > I remounted by C: drive as text = binary. No problems. > I ran configure. No problems. > Ran make. No problems. > > When I try to run any of the tests, I get something like the following: > > (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) In cygwin_except_handler exc > C0000005 at 10042894 sp 242E5B0 > (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) Exception trapped! > (c:\gnu\ncurses-1.9.9e\test\view.exe 3445) exception C0000005 at ... > > I get this on both my NT 4.0 machine with NTFS, and my windows 95 > machine. > > Here is the environment on my NT 4.0 machine: > ... > TERM=dumb > _=/gnuwin32/b18/H-i386-cygwin32/bin/env > > I know people have gotten ncurses to work ok. Any ideas? > Change TERM to something other than "dumb" ("linux" is a good start). I am still messing around with this, too, but dumb is, literally, "dumb", and cannot handle most of the functions ncurses can provide. -- Justin Smith, Systems Engineer with the law office of Dewey, Cheetam and Howe. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sat Dec 6 19:07:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sat, 06 Dec 1997 19:07:00 -0000 Subject: Latest mingw32 release has some problems Message-ID: <199712070307.DAA101194@out5.ibm.net> This is what happens while compiling the mingw32 version of libslang: D:\Src\slang\src>gcc -o worm worm.o -L./ -lslang -ltermcap -lm d:\mingw32\i386-mingw32\lib\crt1.o: In function `mingw32_init_mainargs': /homes/src/tmp/mingw32-cdk/MINGW32/mingw/../../mingw/init.c:65: undefined reference to `__GetMainArgs' d:\mingw32\i386-mingw32\lib\crt1.o: In function `mingw32_init_fmode': /homes/src/tmp/mingw32-cdk/MINGW32/mingw/../../mingw/crt1.c:108: undefined reference to `__imp__fmode_dll' collect2: ld returned 1 exit status - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jfmiller@polymail.cpunix.calpoly.edu Sat Dec 6 23:00:00 1997 From: jfmiller@polymail.cpunix.calpoly.edu (John Miller) Date: Sat, 06 Dec 1997 23:00:00 -0000 Subject: _beginthread Problems with win95 and mingw32 using C++ References: <01bd0141$5f8510a0$fa173185@gbird0.fu.is.saga-u.ac.jp> Message-ID: <348A499E.E58B8A57@polymail.calpoly.edu> Colin Peters wrote: > The reason the function doesn't work without __cdecl__ is that it is > getting name mangled like a C++ function (I think, see below). The header > file should feature a pair of sequences like this around the _beginthread > and other prototypes: > > #ifdef __cplusplus > extern "C" { > #endif > > ... function prototypes ... > > #ifdef __cplusplus > } > #endif This is indeed included in the header file. > The version of Mingw32 you are using might have a bug in that header. You > could try downloading version 0.1.4 from my web page, or the latest > snapshot if you feel brave (on second thought, if you download the latest > snapshot only pull out the process.h header file, the new specs won't work > with b18). > Now that it occurs to me, what does that prototype look like? It should be > > unsigned long _beginthread (void (*pfuncStart)(void *), unsigned > unStackSize, void* pArgList); > > It occurs to me that older versions of Mingw32 might have had an attempt > to cdecl pfuncStart, which is unnecessary and may cause problems in some > cases. Grab process.h out of 0.1.4 (I think) or the latest snapshot. Or > just replace the prototype with the one given above. Did so and this did indeed fix the problem (Yes!) but... Now I can compile an object file correctly but I now am not able to link correctly to _beginthread. ld returns error : test.o(.text+0x3d):test.cc: undefined reference to `_beginthread(void (*)(void *), unsigned int, void *)' GCC.EXE: Internal compiler error: program ld got fatal signal 1 according to the source release of mingw32 this finction is provided by crtdll.dll and is there for linked agenst libcrtdll.a. Howerev even after explisitly linking agenst this library I was unable to make ld happy. My only thought on the matter is that my test case was a console program and not a "Windowed" program as of this letter I have yet to test it as such. It is also interesting to note that the exact same program including _beginthread compiles and runs nicley as a "c" program. which makes me suspicios that it has something to do with the c++ compiler it's self and not mingw32. > If none of this fixes things it might be a bug in the C++ parser (I doubt > it though). It works fine for me using Mumit Khan's Mingw32-native version > of EGCS. I can't test against b18 right now (since it isn't on my disk > drive). > > Good luck, > Colin. I am in the prosses of downloading this as I type in hope that it will solve my problems. Thank you for all your help and hard work on this process (Pun Intended) Thanks John Miller - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Sun Dec 7 01:09:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Sun, 07 Dec 1997 01:09:00 -0000 Subject: _beginthread Problems with win95 and mingw32 using C++ References: <01bd0141$5f8510a0$fa173185@gbird0.fu.is.saga-u.ac.jp> <348A499E.E58B8A57@polymail.calpoly.edu> Message-ID: <199712070909.UAA06549@mundook.cs.mu.OZ.AU> John Miller writes: >Colin Peters wrote: > >> The reason the function doesn't work without __cdecl__ is that it is >> getting name mangled like a C++ function (I think, see below). The header >> file should feature a pair of sequences like this around the _beginthread >> and other prototypes: >> >> #ifdef __cplusplus >> extern "C" { >> #endif >> >> ... function prototypes ... >> >> #ifdef __cplusplus >> } >> #endif > >This is indeed included in the header file. Hmm... >> The version of Mingw32 you are using might have a bug in that header. You >> could try downloading version 0.1.4 from my web page, or the latest >> snapshot if you feel brave (on second thought, if you download the latest >> snapshot only pull out the process.h header file, the new specs won't work >> with b18). > >> Now that it occurs to me, what does that prototype look like? It should be >> >> unsigned long _beginthread (void (*pfuncStart)(void *), unsigned >> unStackSize, void* pArgList); >> >> It occurs to me that older versions of Mingw32 might have had an attempt >> to cdecl pfuncStart, which is unnecessary and may cause problems in some >> cases. Grab process.h out of 0.1.4 (I think) or the latest snapshot. Or >> just replace the prototype with the one given above. > >Did so and this did indeed fix the problem (Yes!) but... > >Now I can compile an object file correctly but I now am not able to link >correctly to _beginthread. ld returns error : > >test.o(.text+0x3d):test.cc: undefined reference to `_beginthread(void (*)(void >*), unsigned int, void *)' The linker error message includes the function argument types, which indicates that the name has been mangled (the linker is nice enough to demangle it when printing the error). Are you *sure* the header file has an `extern "C" { ... }' around the declaration of _beginthread()? You could try replacing the prototype with one that explicitly includes `extern "C"', e.g. extern "C" unsigned long _beginthread (void (*pfuncStart)(void *), unsigned unStackSize, void* pArgList); -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sreuss@asc.on.ca Sun Dec 7 07:09:00 1997 From: sreuss@asc.on.ca (Stephen Reuss) Date: Sun, 07 Dec 1997 07:09:00 -0000 Subject: how to avoid mulitple definitions of STL free_list, start_free, end_free,heap_size Message-ID: <348ABB53.62AF@asc.on.ca> First up, I dealt with the pair structure redeclaration issue by renaiming the pair structure in the third party code to mespair. If anyone has ideas of how I can force compiler to choose one declaration or another, please let me know. The new problem regards the: multiple definition of `__default_alloc_template::start_free ::end_free ::heap_size ::free_list in linking a C++ program. The link command is as follows: g++ -o main -D__GNU_COMPILER -DWIN32 -D__CYGWIN32__ -O2 main.o flowsolver.o gui.o graphics.o io.o parms.o mesh.o vert.o cell.o face.o coef.o mgrid.o gridadapt.o -L../lib -lglut -L../lib -lMesaGLU -lMesaGL /usr/Programs/Meschach/lib/meschach.a ../lib/wing32.a -lkernel32 -luser32 -lgdi32 Interrestingly, the above error is given while trying to link all but the io.o and parms.o files. I checked the included headers for these files and they arent any different than headers included in other source files. And each of the object files was created using the command (eg.): g++ -c -D__GNU_COMPILER -DWIN32 -D__CYGWIN32__ -O2 -I/usr/Programs/Mesa-2.5/include -I. -I/LIBS/mv++/include -I/usr/Programs/Meschach/include main.c >From what I understand, the compilation phase has created object files each with the above memory definitions. How can I get it not to do that, if that is the case? If not, can anyone shed some light? Is there any way I can tell the linker its OK, and just use the most recent definition? They are presumably all identical. Thanks in advance -- --------------------------------------------------------------------- Stephen Reuss ===== ====== ===== M.A.Sc. student, Mechanical Engineering == == == == University of Waterloo, Canada = ===== == = E-mail: sreuss@sunwise.uwaterloo.ca == == == == Phone : (519)888-4567 X3859 ===== == ===== Fax : (519)888-6197 *** G R O U P *** --------------------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Sun Dec 7 11:11:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Sun, 07 Dec 1997 11:11:00 -0000 Subject: Latest mingw32 release has some problems References: <199712070307.DAA101194@out5.ibm.net> Message-ID: <9712071911.AA26935@modi.xraylith.wisc.edu> vischne@ibm.net writes: > This is what happens while compiling the mingw32 version of libslang: > > D:\Src\slang\src>gcc -o worm worm.o -L./ -lslang -ltermcap -lm > d:\mingw32\i386-mingw32\lib\crt1.o: In function `mingw32_init_mainargs': > /homes/src/tmp/mingw32-cdk/MINGW32/mingw/../../mingw/init.c:65: undefined > reference to `__GetMainArgs' > d:\mingw32\i386-mingw32\lib\crt1.o: In function `mingw32_init_fmode': > /homes/src/tmp/mingw32-cdk/MINGW32/mingw/../../mingw/crt1.c:108: undefined > reference to `__imp__fmode_dll' > collect2: ld returned 1 exit status > Did you change the default specs (one the uses CRTDLL) to one that uses MSVCRT or MSVCRT20? If so, please copy the specs.crt top specs and retry. Also, you might want to copy me on these type of messages since this mailing list has a very long (upto 48 hours) delivery latency. Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From raf@comdyn.com.au Sun Dec 7 14:53:00 1997 From: raf@comdyn.com.au (raf) Date: Sun, 07 Dec 1997 14:53:00 -0000 Subject: BUG: stdin not a constant Message-ID: <199712072252.JAA10987@mali.comdyn.com.au> >The following code doesn't compile. >------ >#include >static FILE *out = stdout; >void main(void) >{ > /* out = stdout; */ > fprintf(out, "hello world\n"); >} >------ >It gets the following error message: >test.c:3: initializer element is not constant >The fix is to remove the initialiser and add the line which is commented >out. >However the code I am trying to port has these sort of variables spread >all over the place, and I would like to know if there is someway I can >get the compiler to treat it as a constant. stdout is probably defined as "(&__iob[1])" which is just an expression because of the parentheses. #undef it a #define it as "&__iob[1]" which is a constant. raf - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sun Dec 7 15:38:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sun, 07 Dec 1997 15:38:00 -0000 Subject: Merging ming and cygwin, first cut Message-ID: <199712072338.XAA90190@out4.ibm.net> If you put the ming specs into the specs directory of your cygwin system, and use cygwin bash to configure, you may get the following error messages, after which the configure script will continue at the next test: (At the bash prompt:) checking for location of config-file... /usr/local/lib/lynx.cfg checking for ANSI C header files... wait_for_any: WaitForMultipleObjects failed, win32 error 6 wait_for_any: Trying again. yes (From the Windows 95 error message box:) CONFTEST caused an invalid page fault in module CYGWIN.DLL at 0137:1000c198. Registers: EAX=00000000 CS=0137 EIP=1000c198 EFLGS=00010246 EBX=00000000 SS=013f ESP=0253fe14 EBP=0253fe14 ECX=00402010 DS=013f ESI=81574e18 FS=2437 EDX=00510060 ES=013f EDI=81575200 GS=0000 Bytes at CS:EIP: 83 78 58 00 75 0c c7 40 58 01 00 00 00 e8 b2 ff Stack dump: 0253fe20 00401091 00000000 0253fe38 0040106a 00000001 005105a4 00510060 00000000 0253ff78 bff88f75 81575200 81574e18 00000000 666e6f43 74736574  - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Sun Dec 7 16:08:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Sun, 07 Dec 1997 16:08:00 -0000 Subject: egcs-1.00: "Release goofs" for cygwin32/mingw32 Message-ID: <9712080008.AA27127@modi.xraylith.wisc.edu> I've updated the Web page for my gnu-win32 stuff and got rid most of the old cruft there. Now there is a section "Release Goofs" for both cygwin32 and ming3w2 distributions, so if you have any problems, plesae check that out first. http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kellyj@ricochet.net Sun Dec 7 17:22:00 1997 From: kellyj@ricochet.net (J. Kelly Johnson) Date: Sun, 07 Dec 1997 17:22:00 -0000 Subject: Undefined reference to cygwin32_inet_aton in b18. Message-ID: <348B4D2A.4C561932@ricochet.net> I have a small program which needs to use the inet_aton function. When compiled with gcc under b18 (the code includes the arpa/inet.h header), I get the message: undefined reference to 'cygwin32_inet_aton' gcc: Internal compiler error: program ld got fatal signal 1 I use the opposite function in my code (inet_ntoa) with no difficulties. Am I missing something here? Or is the header file incorrect (or the libraries missing a function)? I have attached a short test program which exhibits the problem. Any help would be appreciated, Kelly Johnson From fjh@cs.mu.OZ.AU Sun Dec 7 20:44:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Sun, 07 Dec 1997 20:44:00 -0000 Subject: BUG: stdin not a constant References: <199712072252.JAA10987@mali.comdyn.com.au> Message-ID: <199712080444.PAA04160@mundook.cs.mu.OZ.AU> raf@comdyn.com.au writes: >stdout is probably defined as "(&__iob[1])" which is just an expression >because of the parentheses. #undef it a #define it as "&__iob[1]" which >is a constant. Not a bad guess... but quite wrong, I'm afraid: - stdout is not defined as `(&__iob[1])' - constants are allowed to contain parentheses -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From anazawa@hic-soft.co.jp Sun Dec 7 21:06:00 1997 From: anazawa@hic-soft.co.jp (ANAZAWA Akio) Date: Sun, 07 Dec 1997 21:06:00 -0000 Subject: BUG: stdin not a constant References: <199712072252.JAA10987@mali.comdyn.com.au> Message-ID: <199712080506.OAA10318@mail.hic-soft.co.jp> Hi, raf. In message "Re: BUG: stdin not a constant" on 97/12/08, raf writes: raf> stdout is probably defined as "(&__iob[1])" which is just an raf> expression because of the parentheses. #undef it a #define it as raf> "&__iob[1]" which is a constant. This solution seems meaningless with three reasons. (1) When expression a was constant, "(a)" should be still remaining as constant (expression). If your idea was correct, we may couldn't write following statement as global variable definition/initilizer. int n = (0); (2) Actual definition of stdout came from Cygnus (b18) is; #define stdout (_impure_ptr->_stdout) (3) Your solution (remove parentheses) still causes same error message. Why don't you see actual stdio.h file? And, why don't you try -E option of gcc? # Forgive me if my expression brings unpleasant feelings to someone. # Still I couldn't realize tone of English expressions. -- @___ anazawa@hic-soft.co.jp c>//_ --- Human Interface Communications Co.,Ltd.(H.I.C.) (_)\(_) --- Phone +81 3 5467 2401 / Fax +81 3 5467 2402 ------------------------------------------------------------------------ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sun Dec 7 21:14:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sun, 07 Dec 1997 21:14:00 -0000 Subject: Problem with libwsock32.a Message-ID: <199712080514.FAA12052@out4.ibm.net> I set up a dual cygwin/ming system and compiled the slrn news reader using a ming-compiled libslang and libtermcap and libwsock32.a that is common to ming and cygwin. When I started to run slrn.exe, I got the following: ====================================================== slrn 0.9.4.3 (Dec 7 1997 23:59:53) Reading startup file ./slrn.rc. Connecting to host news3.ibm.net ...news3.ibm.net: Unknown host. slrn fatal error: Failed to initialize server. ======================================================= Now, I know that slrn.exe compiles and runs successfully using other Windows compilers and wsock32.dll. So, this experience makes me distrust libwsock32.a, especially since early link tests indicate that libwsock32.a points several different socket calls to the same entry points. Any constructive (non-denial, non-blaming) comments?  - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From guiboud@netmansys.fr Mon Dec 8 05:15:00 1997 From: guiboud@netmansys.fr (Stéphane Guiboud-Ribaud) Date: Mon, 08 Dec 1997 05:15:00 -0000 Subject: select function on NT Message-ID: <344B4ADF.15438948@netmansys.fr> I'm currently trying to port an UNIX middleware on Windows NT platform, and I'm met the problem of the 'select' function which is not the same on Windows NT and UNIX. On Windows NT file descriptor passed to the select function is mainly restricted to SOCKET object. Do the select function in Cygnus support file descriptor (concerning file, pipe or socket) ? If not, regarding the Win32 API, how can I know if a file handle (or a handle object) is ready to read/write ? -- -------------------------------------------------------------- Stephane Guiboud-Ribaud NETMANSYS S.A. http://www.netmansys.fr/ mailto:guiboud@netmansys.fr http://www.netmansys.com/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kh@wg.icl.co.uk Mon Dec 8 05:15:00 1997 From: kh@wg.icl.co.uk (Kevin Hughes) Date: Mon, 08 Dec 1997 05:15:00 -0000 Subject: Broken .exe files? Message-ID: <01BD03D9.AF3D5220@rodney.wg.icl.co.uk> Some time ago I sent a request to the group regarding broken .exe files being produced by the gnu compiler. I am still stuck and would appreciate help in tracking the problem down. The system I have produces a large .exe file - around 6 meg. The major part of this is a library which it links with As a standard libxxx.a file. When I try to run it I get an error indicating something wrong with the executable. It is not specific - error 0. The file is not a valid executable. If I strip the executable it shrinks and becomes perfectly well behaved. I am really confused. Is there any way of checking the executable? Has anyone else experienced this? Would building a dll for the library and dynamically loading it help? All suggestions would be much appreciated. Kevin Kevin - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From skinner@drea.dnd.ca Mon Dec 8 06:18:00 1997 From: skinner@drea.dnd.ca (Bruce S. Skinner) Date: Mon, 08 Dec 1997 06:18:00 -0000 Subject: Trouble building ncurses 1.9.9g. Message-ID: <3.0.32.19971208101844.009229f0@firm.drea.dnd.ca> I have been trying to build ncurses_1.9.9g under cygwin32 b18.5 & NT4.0. It appears to build O.K. but when I try and run any of the test programs I get: Error opening terminal: linux. If at this point I do a make install everything appears to go smoothly until: linking captoinfo to tic ln: cannot create hard link `/usr/local/bin/captoinfo' to `/usr/local/bin/tic': No such file or directory make: [install.progs] Error 1 (ignored) and later sh ./run_tic.sh /usr/local/bin . /usr/local/share/terminfo ** Building terminfo database, please wait... ** adjusting tabset paths "/tmp/7863", line 268, terminal 'unknown': resolution of use=dumb failed "/tmp/7863", line 772, terminal 'iris-ansi-ap': resolution of use=iris-ansi failed . . . "/tmp/7863", line 12655, terminal 'env230': resolution of use=vt100 failed "/tmp/7863", line 12664, terminal 'ep40': resolution of use=ep4080 failed ? tic could not build /usr/local/share/terminfo make: *** [install.data] Error 1 make: *** [install] Error 2 Please forgive me if these errors should blatantly tell me what is amiss. I'm not a unix person or Winnt person. I have searched the mailing list archives and haven't gotten any clues from there. Any comments? regards :-) Bruce S. Skinner Bruce S. Skinner | skinner@drea.dnd.ca DEFENCE RESEARCH ESTABLISHMENT ATLANTIC | norstead@ns.sympatico.ca Dartmouth, Nova Scotia, CANADA, B2Y 3Z7 | (902) 426-3100 (205) "New opinions are always suspected, and usually opposed without any other reason but because they are not already common." John Locke ------------------------------------------------------------------------------ bash$ pwd /skinner bash$ cd //d/Gnu bash$ bash$ pwd /d/Gnu bash$ ls groff-1.10/ bash$ ls ../ncurses ncurses-1.9.9g.tar bash$ tar xf ../ncurses/ncurses-1.9.9g.tar tar: Could not change access and modification times of ncurses_1.9.9g/test: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/progs: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/panel: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/ncurses: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/misc/tabset: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/misc: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/menu: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/man: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/include: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/form: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/c++: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/Ada95/ada_include: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/Ada95/gen: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/Ada95/html/images: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/Ada95/html: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/Ada95/samples: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/Ada95/ada_objects: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/Ada95/objects: Bad address tar: Could not change access and modification times of ncurses_1.9.9g/Ada95: Bad address tar: Could not change access and modification times of ncurses_1.9.9g: Bad address bash$ ls groff-1.10/ ncurses_1.9.9g/ bash$ cd ncurses_1.9.9g bash$ ./configure --help Usage: configure [options] [host] Options: [defaults in brackets after descriptions] Configuration: --cache-file=FILE cache test results in FILE --help print this message --no-create do not create output files --quiet, --silent do not print `checking...' messages --version print the version of autoconf that created configure Directory and file names: --prefix=PREFIX install architecture-independent files in PREFIX [/usr/local] --exec-prefix=EPREFIX install architecture-dependent files in EPREFIX [same as prefix] --bindir=DIR user executables in DIR [EPREFIX/bin] --sbindir=DIR system admin executables in DIR [EPREFIX/sbin] --libexecdir=DIR program executables in DIR [EPREFIX/libexec] --datadir=DIR read-only architecture-independent data in DIR [PREFIX/share] --sysconfdir=DIR read-only single-machine data in DIR [PREFIX/etc] --sharedstatedir=DIR modifiable architecture-independent data in DIR [PREFIX/com] --localstatedir=DIR modifiable single-machine data in DIR [PREFIX/var] --libdir=DIR object code libraries in DIR [EPREFIX/lib] --includedir=DIR C header files in DIR [PREFIX/include] --oldincludedir=DIR C header files for non-gcc in DIR [/usr/include] --infodir=DIR info documentation in DIR [PREFIX/info] --mandir=DIR man documentation in DIR [PREFIX/man] --srcdir=DIR find the sources in DIR [configure dir or ..] --program-prefix=PREFIX prepend PREFIX to installed program names --program-suffix=SUFFIX append SUFFIX to installed program names --program-transform-name=PROGRAM run sed PROGRAM on installed program names Host type: --build=BUILD configure for building on BUILD [BUILD=HOST] --host=HOST configure for HOST [guessed] --target=TARGET configure for TARGET [TARGET=HOST] Features and packages: --disable-FEATURE do not include FEATURE (same as --enable-FEATURE=no) --enable-FEATURE[=ARG] include FEATURE [ARG=yes] --with-PACKAGE[=ARG] use PACKAGE [ARG=yes] --without-PACKAGE do not use PACKAGE (same as --with-PACKAGE=no) --x-includes=DIR X include files are in DIR --x-libraries=DIR X library files are in DIR --enable and --with options recognized: --with-install-prefix prefixes actual install-location --with-shared generate shared-libraries --with-normal generate normal-libraries (default) --with-debug generate debug-libraries (default) --with-profile generate profile-libraries --disable-overwrite leave out the link to -lcurses --enable-termcap compile in termcap fallback support --enable-getcap fast termcap load, no xrefs to terminfo --enable-symlinks make tic use symbolic links not hard links --enable-bsdpad recognize BSD-style prefix padding --enable-echo display "compiling" commands (default) --enable-warnings turn on GCC compiler warnings --enable-assertions turn on generation of assertion code bash$ configure --program-suffix=.exe --enable-termcap --enable-symlinks=no --with-shared=no --with-profile=no -v creating cache ./config.cache Configuring NCURSES 1.9.9g ABI 3.4 (Mon Dec 8 06:52:54 GMT-4:00 1997) System type is CYGWIN32_NT Configuring for CYGWIN32/NT 4.0 checking for prefix... /usr/local checking for default include-directory... ${prefix}/include checking for gcc... gcc checking whether we are using GNU C... yes checking whether gcc accepts -g... yes checking version of gcc... cygnus-2.7.2-970404 checking how to run the C preprocessor... gcc -E checking whether gcc needs -traditional... no checking for POSIXized ISC... no checking for c++... c++ checking whether we are using GNU C++... yes checking whether c++ accepts -g... yes checking for mawk... no checking for gawk... gawk checking whether make sets ${MAKE}... yes checking for a BSD compatible install... /bin/ginstall -c checking whether ln -s works... yes checking for ranlib... ranlib checking for tdlint... no checking for lint... no checking for alint... no checking for man... no checking for man_db... no checking for makeflags variable... $(MFLAGS) checking format of man-pages... unknown checking for specified models... normal debug checking include directory... ${prefix}/include checking for gcc __attribute__ directives ... scanf ... printf ... unused ... noreturn checking for -lbsd... no checking whether cross-compiling... no checking for ANSI C header files... yes checking for dirent.h that defines DIR... yes checking for -ldir... no checking for regular-expression headers... checking for fcntl.h... yes checking for getopt.h... no checking for libc.h... no checking for limits.h... yes checking for locale.h... yes checking for sys/bsdtypes.h... no checking for sys/ioctl.h... yes checking for sys/param.h... yes checking for poll.h... no checking for sys/select.h... no checking for sys/stropts.h... no checking for sys/time.h... yes checking for sys/times.h... yes checking for termio.h... yes checking for termios.h... yes checking for ttyent.h... no checking for unistd.h... yes checking for values.h... no checking for SCO by checking on /usr/bin/scosh checking for scosh... maybenot checking for working const... yes checking for inline... inline checking for errno external decl... yes checking declaration of sys_errlist... no checking if data-only library module links... yes checking for speed_t... yes checking for gettimeofday... yes checking for getttynam... no checking for memccpy... yes checking for poll... no checking for select... yes checking for setbuffer... no checking for setvbuf... yes checking for sigaction... yes checking for sigvec... no checking for strdup... yes checking for tcgetattr... yes checking for times... yes checking for usleep... yes checking for vfscanf... no checking for vsscanf... no checking whether sigaction needs _POSIX_SOURCE... no checking whether termios.h needs _POSIX_SOURCE... no checking for restartable system calls... yes checking whether setvbuf arguments are reversed... no checking return type of signal handlers... void checking for type sigaction_t... no checking for working TIOCGWINSZ... yes checking for c++... yes checking for libg++... yes checking for builtin c++ bool type... yes checking for size of c++ bool... char checking for gnatmake... no checking for src modules... ncurses progs panel menu form not updating unwritable cache ./config.cache creating ./config.status creating include/config_h creating include/Makefile creating man/Makefile creating ncurses/Makefile creating progs/Makefile creating panel/Makefile creating menu/Makefile creating form/Makefile creating misc/Makefile creating test/Makefile creating c++/Makefile creating Makefile Appending rules for normal model (ncurses) Appending rules for debug model (ncurses) Appending rules for normal model (progs) Appending rules for debug model (progs) Appending rules for normal model (panel) Appending rules for debug model (panel) Appending rules for normal model (menu) Appending rules for debug model (menu) Appending rules for normal model (form) Appending rules for debug model (form) Appending rules for normal model (c++) Appending rules for debug model (c++) creating headers.sh creating include/config.h removing include/config_h bash$ make cd include; make INSTALL_PREFIX="" all sed <./MKterm.h.awk.in >MKterm.h.awk -e "/@VERSION@/s//1.9.9g/" sh ./MKcurses.sh 1.9.9g 1 char . updating include/curses.h sed <./termcap.h.in >termcap.h -e "/@VERSION@/s//1.9.9g/" sh ./MKhashsize.sh ./Caps >hashsize.h sh ./MKparametrized.sh ./Caps >parametrized.h sed <./unctrl.h.in >unctrl.h -e "/@VERSION@/s//1.9.9g/" gawk -f MKterm.h.awk ./Caps > term.h sh ./edit_cfg.sh ../include/config.h term.h ** edit: HAVE_TCGETATTR ** edit: HAVE_TERMIOS_H ** edit: HAVE_TERMIO_H ** edit: BROKEN_LINKER cd man; make INSTALL_PREFIX="" all sh ./MKterminfo.sh ./terminfo.head ./../include/Caps ./terminfo.tail >terminfo.5 cd ncurses; make INSTALL_PREFIX="" all gcc -o make_hash -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -DMAIN_PROGRAM ./comp_hash.c sh ./MKcaptab.awk gawk ./../include/Caps > comp_captab.c sh ./MKfallback.sh >fallback.c gawk -f ./MKkeys.awk ./keys.list > keys.tries gawk -f ./MKkeyname.awk ./keys.list > lib_keyname.c sh ./MKlib_gen.sh "gcc -E" "gawk" <../include/curses.h >lib_gen.c gawk -f ./MKnames.awk ./../include/Caps cat namehdr boolnames boolfnames numnames numfnames strnames strfnames nameftr >names.c cat namehdr boolcodes numcodes strcodes codeftr >codes.c rm -f namehdr nameftr codeftr boolnames boolfnames boolcodes numnames numfnames numcodes strnames strfnames strcodes cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/fallback.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/hashmap.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/hardscroll.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_acs.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_addch.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_addstr.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_beep.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_bkgd.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_baudrate.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_box.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_clear.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_clrbot.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_clreol.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_color.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_data.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_delch.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_delwin.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_doupdate.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_endwin.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_erase.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_gen.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_getch.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_getstr.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_inchstr.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_initscr.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_insch.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_insdel.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_insstr.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_instr.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_isendwin.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_kernel.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_keyname.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_longname.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_mouse.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_move.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_mvcur.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_mvwin.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_newterm.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_newwin.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_options.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_overlay.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_pad.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_print.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_printw.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_raw.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_refresh.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_restart.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_scanw.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_screen.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_scroll.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_scrreg.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_set_term.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_setup.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_slk.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_termcap.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_ti.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_touch.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_tparm.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_tputs.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_trace.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_traceatr.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_tracechr.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_tracedmp.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_tracemse.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_tstp.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_twait.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_unctrl.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_vidattr.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_window.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/names.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/codes.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/read_entry.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/read_termcap.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/write_entry.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/sigaction.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/vsscanf.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/alloc_entry.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/captoinfo.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/comp_captab.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/comp_error.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/comp_hash.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/comp_parse.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/comp_scan.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/parse_entry.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/resizeterm.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/wresize.c cd ../objects; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../ncurses/lib_adabind.c ar rv ../lib/libncurses.a ../objects/fallback.o ../objects/hashmap.o ../objects/hardscroll.o ../objects/lib_acs.o ../objects/lib_addch.o ../objects/lib_addstr.o ../objects/lib_beep.o ../objects/lib_bkgd.o ../objects/lib_baudrate.o ../objects/lib_box.o ../objects/lib_clear.o ../objects/lib_clrbot.o ../objects/lib_clreol.o ../objects/lib_color.o ../objects/lib_data.o ../objects/lib_delch.o ../objects/lib_delwin.o ../objects/lib_doupdate.o ../objects/lib_endwin.o ../objects/lib_erase.o ../objects/lib_gen.o ../objects/lib_getch.o ../objects/lib_getstr.o ../objects/lib_inchstr.o ../objects/lib_initscr.o ../objects/lib_insch.o ../objects/lib_insdel.o ../objects/lib_insstr.o ../objects/lib_instr.o ../objects/lib_isendwin.o ../objects/lib_kernel.o ../objects/lib_keyname.o ../objects/lib_longname.o ../objects/lib_mouse.o ../objects/lib_move.o ../objects/lib_mvcur.o ../objects/lib_mvwin.o ../objects/lib_newterm.o ../objects/lib_newwin.o ../objects/lib_options.o ../objects/lib_overlay.o ../objects/lib_pad.o ../objects/lib_print.o ../objects/lib_printw.o ../objects/lib_raw.o ../objects/lib_refresh.o ../objects/lib_restart.o ../objects/lib_scanw.o ../objects/lib_screen.o ../objects/lib_scroll.o ../objects/lib_scrreg.o ../objects/lib_set_term.o ../objects/lib_setup.o ../objects/lib_slk.o ../objects/lib_termcap.o ../objects/lib_ti.o ../objects/lib_touch.o ../objects/lib_tparm.o ../objects/lib_tputs.o ../objects/lib_trace.o ../objects/lib_traceatr.o ../objects/lib_tracechr.o ../objects/lib_tracedmp.o ../objects/lib_tracemse.o ../objects/lib_tstp.o ../objects/lib_twait.o ../objects/lib_unctrl.o ../objects/lib_vidattr.o ../objects/lib_window.o ../objects/names.o ../objects/codes.o ../objects/read_entry.o ../objects/read_termcap.o ../objects/write_entry.o ../objects/sigaction.o ../objects/vsscanf.o ../objects/alloc_entry.o ../objects/captoinfo.o ../objects/comp_captab.o ../objects/comp_error.o ../objects/comp_hash.o ../objects/comp_parse.o ../objects/comp_scan.o ../objects/parse_entry.o ../objects/resizeterm.o ../objects/wresize.o ../objects/lib_adabind.o a - ../objects/fallback.o a - ../objects/hashmap.o a - ../objects/hardscroll.o a - ../objects/lib_acs.o a - ../objects/lib_addch.o a - ../objects/lib_addstr.o a - ../objects/lib_beep.o a - ../objects/lib_bkgd.o a - ../objects/lib_baudrate.o a - ../objects/lib_box.o a - ../objects/lib_clear.o a - ../objects/lib_clrbot.o a - ../objects/lib_clreol.o a - ../objects/lib_color.o a - ../objects/lib_data.o a - ../objects/lib_delch.o a - ../objects/lib_delwin.o a - ../objects/lib_doupdate.o a - ../objects/lib_endwin.o a - ../objects/lib_erase.o a - ../objects/lib_gen.o a - ../objects/lib_getch.o a - ../objects/lib_getstr.o a - ../objects/lib_inchstr.o a - ../objects/lib_initscr.o a - ../objects/lib_insch.o a - ../objects/lib_insdel.o a - ../objects/lib_insstr.o a - ../objects/lib_instr.o a - ../objects/lib_isendwin.o a - ../objects/lib_kernel.o a - ../objects/lib_keyname.o a - ../objects/lib_longname.o a - ../objects/lib_mouse.o a - ../objects/lib_move.o a - ../objects/lib_mvcur.o a - ../objects/lib_mvwin.o a - ../objects/lib_newterm.o a - ../objects/lib_newwin.o a - ../objects/lib_options.o a - ../objects/lib_overlay.o a - ../objects/lib_pad.o a - ../objects/lib_print.o a - ../objects/lib_printw.o a - ../objects/lib_raw.o a - ../objects/lib_refresh.o a - ../objects/lib_restart.o a - ../objects/lib_scanw.o a - ../objects/lib_screen.o a - ../objects/lib_scroll.o a - ../objects/lib_scrreg.o a - ../objects/lib_set_term.o a - ../objects/lib_setup.o a - ../objects/lib_slk.o a - ../objects/lib_termcap.o a - ../objects/lib_ti.o a - ../objects/lib_touch.o a - ../objects/lib_tparm.o a - ../objects/lib_tputs.o a - ../objects/lib_trace.o a - ../objects/lib_traceatr.o a - ../objects/lib_tracechr.o a - ../objects/lib_tracedmp.o a - ../objects/lib_tracemse.o a - ../objects/lib_tstp.o a - ../objects/lib_twait.o a - ../objects/lib_unctrl.o a - ../objects/lib_vidattr.o a - ../objects/lib_window.o a - ../objects/names.o a - ../objects/codes.o a - ../objects/read_entry.o a - ../objects/read_termcap.o a - ../objects/write_entry.o a - ../objects/sigaction.o a - ../objects/vsscanf.o a - ../objects/alloc_entry.o a - ../objects/captoinfo.o a - ../objects/comp_captab.o a - ../objects/comp_error.o a - ../objects/comp_hash.o a - ../objects/comp_parse.o a - ../objects/comp_scan.o a - ../objects/parse_entry.o a - ../objects/resizeterm.o a - ../objects/wresize.o a - ../objects/lib_adabind.o ranlib ../lib/libncurses.a cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/fallback.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/hashmap.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/hardscroll.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_acs.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_addch.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_addstr.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_beep.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_bkgd.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_baudrate.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_box.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_clear.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_clrbot.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_clreol.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_color.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_data.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_delch.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_delwin.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_doupdate.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_endwin.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_erase.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_gen.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_getch.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_getstr.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_inchstr.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_initscr.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_insch.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_insdel.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_insstr.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_instr.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_isendwin.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_kernel.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_keyname.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_longname.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_mouse.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_move.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_mvcur.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_mvwin.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_newterm.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_newwin.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_options.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_overlay.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_pad.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_print.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_printw.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_raw.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_refresh.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_restart.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_scanw.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_screen.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_scroll.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_scrreg.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_set_term.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_setup.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_slk.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_termcap.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_ti.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_touch.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_tparm.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_tputs.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_trace.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_traceatr.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_tracechr.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_tracedmp.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_tracemse.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_tstp.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_twait.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_unctrl.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_vidattr.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_window.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/names.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/codes.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/read_entry.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/read_termcap.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/write_entry.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/sigaction.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/vsscanf.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/alloc_entry.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/captoinfo.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/comp_captab.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/comp_error.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/comp_hash.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/comp_parse.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/comp_scan.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/parse_entry.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/resizeterm.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/wresize.c cd ../obj_g; gcc -I../ncurses -I./../ncurses -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../ncurses/lib_adabind.c ar rv ../lib/libncurses_g.a ../obj_g/fallback.o ../obj_g/hashmap.o ../obj_g/hardscroll.o ../obj_g/lib_acs.o ../obj_g/lib_addch.o ../obj_g/lib_addstr.o ../obj_g/lib_beep.o ../obj_g/lib_bkgd.o ../obj_g/lib_baudrate.o ../obj_g/lib_box.o ../obj_g/lib_clear.o ../obj_g/lib_clrbot.o ../obj_g/lib_clreol.o ../obj_g/lib_color.o ../obj_g/lib_data.o ../obj_g/lib_delch.o ../obj_g/lib_delwin.o ../obj_g/lib_doupdate.o ../obj_g/lib_endwin.o ../obj_g/lib_erase.o ../obj_g/lib_gen.o ../obj_g/lib_getch.o ../obj_g/lib_getstr.o ../obj_g/lib_inchstr.o ../obj_g/lib_initscr.o ../obj_g/lib_insch.o ../obj_g/lib_insdel.o ../obj_g/lib_insstr.o ../obj_g/lib_instr.o ../obj_g/lib_isendwin.o ../obj_g/lib_kernel.o ../obj_g/lib_keyname.o ../obj_g/lib_longname.o ../obj_g/lib_mouse.o ../obj_g/lib_move.o ../obj_g/lib_mvcur.o ../obj_g/lib_mvwin.o ../obj_g/lib_newterm.o ../obj_g/lib_newwin.o ../obj_g/lib_options.o ../obj_g/lib_overlay.o ../obj_g/lib_pad.o ../obj_g/lib_print.o ../obj_g/lib_printw.o ../obj_g/lib_raw.o ../obj_g/lib_refresh.o ../obj_g/lib_restart.o ../obj_g/lib_scanw.o ../obj_g/lib_screen.o ../obj_g/lib_scroll.o ../obj_g/lib_scrreg.o ../obj_g/lib_set_term.o ../obj_g/lib_setup.o ../obj_g/lib_slk.o ../obj_g/lib_termcap.o ../obj_g/lib_ti.o ../obj_g/lib_touch.o ../obj_g/lib_tparm.o ../obj_g/lib_tputs.o ../obj_g/lib_trace.o ../obj_g/lib_traceatr.o ../obj_g/lib_tracechr.o ../obj_g/lib_tracedmp.o ../obj_g/lib_tracemse.o ../obj_g/lib_tstp.o ../obj_g/lib_twait.o ../obj_g/lib_unctrl.o ../obj_g/lib_vidattr.o ../obj_g/lib_window.o ../obj_g/names.o ../obj_g/codes.o ../obj_g/read_entry.o ../obj_g/read_termcap.o ../obj_g/write_entry.o ../obj_g/sigaction.o ../obj_g/vsscanf.o ../obj_g/alloc_entry.o ../obj_g/captoinfo.o ../obj_g/comp_captab.o ../obj_g/comp_error.o ../obj_g/comp_hash.o ../obj_g/comp_parse.o ../obj_g/comp_scan.o ../obj_g/parse_entry.o ../obj_g/resizeterm.o ../obj_g/wresize.o ../obj_g/lib_adabind.o a - ../obj_g/fallback.o a - ../obj_g/hashmap.o a - ../obj_g/hardscroll.o a - ../obj_g/lib_acs.o a - ../obj_g/lib_addch.o a - ../obj_g/lib_addstr.o a - ../obj_g/lib_beep.o a - ../obj_g/lib_bkgd.o a - ../obj_g/lib_baudrate.o a - ../obj_g/lib_box.o a - ../obj_g/lib_clear.o a - ../obj_g/lib_clrbot.o a - ../obj_g/lib_clreol.o a - ../obj_g/lib_color.o a - ../obj_g/lib_data.o a - ../obj_g/lib_delch.o a - ../obj_g/lib_delwin.o a - ../obj_g/lib_doupdate.o a - ../obj_g/lib_endwin.o a - ../obj_g/lib_erase.o a - ../obj_g/lib_gen.o a - ../obj_g/lib_getch.o a - ../obj_g/lib_getstr.o a - ../obj_g/lib_inchstr.o a - ../obj_g/lib_initscr.o a - ../obj_g/lib_insch.o a - ../obj_g/lib_insdel.o a - ../obj_g/lib_insstr.o a - ../obj_g/lib_instr.o a - ../obj_g/lib_isendwin.o a - ../obj_g/lib_kernel.o a - ../obj_g/lib_keyname.o a - ../obj_g/lib_longname.o a - ../obj_g/lib_mouse.o a - ../obj_g/lib_move.o a - ../obj_g/lib_mvcur.o a - ../obj_g/lib_mvwin.o a - ../obj_g/lib_newterm.o a - ../obj_g/lib_newwin.o a - ../obj_g/lib_options.o a - ../obj_g/lib_overlay.o a - ../obj_g/lib_pad.o a - ../obj_g/lib_print.o a - ../obj_g/lib_printw.o a - ../obj_g/lib_raw.o a - ../obj_g/lib_refresh.o a - ../obj_g/lib_restart.o a - ../obj_g/lib_scanw.o a - ../obj_g/lib_screen.o a - ../obj_g/lib_scroll.o a - ../obj_g/lib_scrreg.o a - ../obj_g/lib_set_term.o a - ../obj_g/lib_setup.o a - ../obj_g/lib_slk.o a - ../obj_g/lib_termcap.o a - ../obj_g/lib_ti.o a - ../obj_g/lib_touch.o a - ../obj_g/lib_tparm.o a - ../obj_g/lib_tputs.o a - ../obj_g/lib_trace.o a - ../obj_g/lib_traceatr.o a - ../obj_g/lib_tracechr.o a - ../obj_g/lib_tracedmp.o a - ../obj_g/lib_tracemse.o a - ../obj_g/lib_tstp.o a - ../obj_g/lib_twait.o a - ../obj_g/lib_unctrl.o a - ../obj_g/lib_vidattr.o a - ../obj_g/lib_window.o a - ../obj_g/names.o a - ../obj_g/codes.o a - ../obj_g/read_entry.o a - ../obj_g/read_termcap.o a - ../obj_g/write_entry.o a - ../obj_g/sigaction.o a - ../obj_g/vsscanf.o a - ../obj_g/alloc_entry.o a - ../obj_g/captoinfo.o a - ../obj_g/comp_captab.o a - ../obj_g/comp_error.o a - ../obj_g/comp_hash.o a - ../obj_g/comp_parse.o a - ../obj_g/comp_scan.o a - ../obj_g/parse_entry.o a - ../obj_g/resizeterm.o a - ../obj_g/wresize.o a - ../obj_g/lib_adabind.o ranlib ../lib/libncurses_g.a cd progs; make INSTALL_PREFIX="" all sh -c "./MKtermsort.sh gawk ./../include/Caps" >termsort.c cd ../objects; gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../progs/tic.c cd ../objects; gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../progs/dump_entry.c gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O ../objects/tic.o ../objects/dump_entry.o ../lib/libncurses.a -o tic cd ../objects; gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../progs/toe.c gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O ../objects/toe.o ../objects/dump_entry.o ../lib/libncurses.a -o toe cd ../objects; gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../progs/infocmp.c gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O ../objects/infocmp.o ../objects/dump_entry.o ../lib/libncurses.a -o infocmp cd ../objects; gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../progs/clear.c gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O ../objects/clear.o ../lib/libncurses.a -o clear cd ../objects; gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../progs/tput.c gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O ../objects/tput.o ../lib/libncurses.a -o tput cd ../objects; gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../progs/tset.c gcc -I../progs -I. -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O ../objects/tset.o ../objects/dump_entry.o ../lib/libncurses.a -o tset cd panel; make INSTALL_PREFIX="" all rm -f ../include/panel.h cp ./panel.h ../include/panel.h cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -c ../panel/panel.c ar rv ../lib/libpanel.a ../objects/panel.o a - ../objects/panel.o ranlib ../lib/libpanel.a cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -DTERMINFO=\"/usr/local/share/terminfo\" -O -g -Dinline= -DTRACE -c ../panel/panel.c ar rv ../lib/libpanel_g.a ../obj_g/panel.o a - ../obj_g/panel.o ranlib ../lib/libpanel_g.a cd menu; make INSTALL_PREFIX="" all rm -f ../include/menu.h cp ./menu.h ../include/menu.h rm -f ../include/eti.h cp ./eti.h ../include/eti.h rm -f ../include/mf_common.h cp ./mf_common.h ../include/mf_common.h cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_attribs.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_cursor.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_driver.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_format.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_global.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_hook.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_item_cur.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_item_nam.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_item_new.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_item_opt.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_item_use.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_item_val.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_item_vis.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_items.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_new.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_opts.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_post.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_req_name.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_spacing.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_userptr.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_win.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../menu/m_adabind.c ar rv ../lib/libmenu.a ../objects/m_attribs.o ../objects/m_cursor.o ../objects/m_driver.o ../objects/m_format.o ../objects/m_global.o ../objects/m_hook.o ../objects/m_item_cur.o ../objects/m_item_nam.o ../objects/m_item_new.o ../objects/m_item_opt.o ../objects/m_item_use.o ../objects/m_item_val.o ../objects/m_item_vis.o ../objects/m_items.o ../objects/m_new.o ../objects/m_opts.o ../objects/m_post.o ../objects/m_req_name.o ../objects/m_spacing.o ../objects/m_userptr.o ../objects/m_win.o ../objects/m_adabind.o a - ../objects/m_attribs.o a - ../objects/m_cursor.o a - ../objects/m_driver.o a - ../objects/m_format.o a - ../objects/m_global.o a - ../objects/m_hook.o a - ../objects/m_item_cur.o a - ../objects/m_item_nam.o a - ../objects/m_item_new.o a - ../objects/m_item_opt.o a - ../objects/m_item_use.o a - ../objects/m_item_val.o a - ../objects/m_item_vis.o a - ../objects/m_items.o a - ../objects/m_new.o a - ../objects/m_opts.o a - ../objects/m_post.o a - ../objects/m_req_name.o a - ../objects/m_spacing.o a - ../objects/m_userptr.o a - ../objects/m_win.o a - ../objects/m_adabind.o ranlib ../lib/libmenu.a cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_attribs.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_cursor.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_driver.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_format.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_global.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_hook.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_item_cur.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_item_nam.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_item_new.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_item_opt.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_item_use.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_item_val.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_item_vis.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_items.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_new.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_opts.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_post.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_req_name.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_spacing.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_userptr.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_win.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../menu/m_adabind.c ar rv ../lib/libmenu_g.a ../obj_g/m_attribs.o ../obj_g/m_cursor.o ../obj_g/m_driver.o ../obj_g/m_format.o ../obj_g/m_global.o ../obj_g/m_hook.o ../obj_g/m_item_cur.o ../obj_g/m_item_nam.o ../obj_g/m_item_new.o ../obj_g/m_item_opt.o ../obj_g/m_item_use.o ../obj_g/m_item_val.o ../obj_g/m_item_vis.o ../obj_g/m_items.o ../obj_g/m_new.o ../obj_g/m_opts.o ../obj_g/m_post.o ../obj_g/m_req_name.o ../obj_g/m_spacing.o ../obj_g/m_userptr.o ../obj_g/m_win.o ../obj_g/m_adabind.o a - ../obj_g/m_attribs.o a - ../obj_g/m_cursor.o a - ../obj_g/m_driver.o a - ../obj_g/m_format.o a - ../obj_g/m_global.o a - ../obj_g/m_hook.o a - ../obj_g/m_item_cur.o a - ../obj_g/m_item_nam.o a - ../obj_g/m_item_new.o a - ../obj_g/m_item_opt.o a - ../obj_g/m_item_use.o a - ../obj_g/m_item_val.o a - ../obj_g/m_item_vis.o a - ../obj_g/m_items.o a - ../obj_g/m_new.o a - ../obj_g/m_opts.o a - ../obj_g/m_post.o a - ../obj_g/m_req_name.o a - ../obj_g/m_spacing.o a - ../obj_g/m_userptr.o a - ../obj_g/m_win.o a - ../obj_g/m_adabind.o ranlib ../lib/libmenu_g.a cd form; make INSTALL_PREFIX="" all rm -f ../include/form.h cp ./form.h ../include/form.h cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fty_regex.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fty_num.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fty_int.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fty_enum.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fty_alpha.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fty_alnum.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fty_ipv4.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/frm_data.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/frm_win.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/frm_user.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/frm_opts.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/frm_hook.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/frm_req_name.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fld_user.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fld_type.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fld_stat.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/fld_def.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/frm_def.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/frm_driver.c cd ../objects; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -c ../form/frm_adabind.c ar rv ../lib/libform.a ../objects/fty_regex.o ../objects/fty_num.o ../objects/fty_int.o ../objects/fty_enum.o ../objects/fty_alpha.o ../objects/fty_alnum.o ../objects/fty_ipv4.o ../objects/frm_data.o ../objects/frm_win.o ../objects/frm_user.o ../objects/frm_opts.o ../objects/frm_hook.o ../objects/frm_req_name.o ../objects/fld_user.o ../objects/fld_type.o ../objects/fld_stat.o ../objects/fld_def.o ../objects/frm_def.o ../objects/frm_driver.o ../objects/frm_adabind.o a - ../objects/fty_regex.o a - ../objects/fty_num.o a - ../objects/fty_int.o a - ../objects/fty_enum.o a - ../objects/fty_alpha.o a - ../objects/fty_alnum.o a - ../objects/fty_ipv4.o a - ../objects/frm_data.o a - ../objects/frm_win.o a - ../objects/frm_user.o a - ../objects/frm_opts.o a - ../objects/frm_hook.o a - ../objects/frm_req_name.o a - ../objects/fld_user.o a - ../objects/fld_type.o a - ../objects/fld_stat.o a - ../objects/fld_def.o a - ../objects/frm_def.o a - ../objects/frm_driver.o a - ../objects/frm_adabind.o ranlib ../lib/libform.a cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fty_regex.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fty_num.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fty_int.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fty_enum.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fty_alpha.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fty_alnum.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fty_ipv4.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/frm_data.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/frm_win.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/frm_user.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/frm_opts.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/frm_hook.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/frm_req_name.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fld_user.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fld_type.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fld_stat.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/fld_def.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/frm_def.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/frm_driver.c cd ../obj_g; gcc -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -O -g -Dinline= -DTRACE -c ../form/frm_adabind.c ar rv ../lib/libform_g.a ../obj_g/fty_regex.o ../obj_g/fty_num.o ../obj_g/fty_int.o ../obj_g/fty_enum.o ../obj_g/fty_alpha.o ../obj_g/fty_alnum.o ../obj_g/fty_ipv4.o ../obj_g/frm_data.o ../obj_g/frm_win.o ../obj_g/frm_user.o ../obj_g/frm_opts.o ../obj_g/frm_hook.o ../obj_g/frm_req_name.o ../obj_g/fld_user.o ../obj_g/fld_type.o ../obj_g/fld_stat.o ../obj_g/fld_def.o ../obj_g/frm_def.o ../obj_g/frm_driver.o ../obj_g/frm_adabind.o a - ../obj_g/fty_regex.o a - ../obj_g/fty_num.o a - ../obj_g/fty_int.o a - ../obj_g/fty_enum.o a - ../obj_g/fty_alpha.o a - ../obj_g/fty_alnum.o a - ../obj_g/fty_ipv4.o a - ../obj_g/frm_data.o a - ../obj_g/frm_win.o a - ../obj_g/frm_user.o a - ../obj_g/frm_opts.o a - ../obj_g/frm_hook.o a - ../obj_g/frm_req_name.o a - ../obj_g/fld_user.o a - ../obj_g/fld_type.o a - ../obj_g/fld_stat.o a - ../obj_g/fld_def.o a - ../obj_g/frm_def.o a - ../obj_g/frm_driver.o a - ../obj_g/frm_adabind.o ranlib ../lib/libform_g.a cd misc; make INSTALL_PREFIX="" all make: Nothing to be done for `all'. cd test; make INSTALL_PREFIX="" all gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./blue.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o blue gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./bs.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o bs gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./firework.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o firework gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./gdc.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o gdc gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./hanoi.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o hanoi gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./hashtest.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o hashtest gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./knight.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o knight gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./lrtest.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o lrtest gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./ncurses.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o ncurses gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./newdemo.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o newdemo gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./rain.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o rain gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./tclock.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -lm -o tclock gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./testcurs.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o testcurs gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./view.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o view gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./worm.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o worm gcc -O -DNDEBUG -I. -I../include -DHAVE_CONFIG_H -I. ./xmas.c ../lib/libform.a ../lib/libmenu.a ../lib/libpanel.a ../lib/libncurses.a -o xmas cd c++; make INSTALL_PREFIX="" all cd ../objects; c++ -DNDEBUG -I. -I../include -I../include -DHAVE_CONFIG_H -O -c ../c++/cursesw.cc cd ../objects; c++ -DNDEBUG -I. -I../include -I../include -DHAVE_CONFIG_H -O -c ../c++/cursesp.cc cd ../objects; c++ -DNDEBUG -I. -I../include -I../include -DHAVE_CONFIG_H -O -c ../c++/demo.cc c++ -o demo ../objects/cursesw.o ../objects/cursesp.o ../objects/demo.o ../lib/libpanel.a ../lib/libncurses.a -lg++ -lm bash$ pwd /d/Gnu/ncurses_1.9.9g bash$ cd progs bash$ ls *.exe clear.exe* infocmp.exe* tic.exe* toe.exe* tput.exe* tset.exe* bash$ cd ../test bash$ ls *.exe blue.exe* hanoi.exe* ncurses.exe* testcurs.exe* bs.exe* hashtest.exe* newdemo.exe* view.exe* firework.exe* knight.exe* rain.exe* worm.exe* gdc.exe* lrtest.exe* tclock.exe* xmas.exe* bash$ printenv TERM linux bash$ testcurs Error opening terminal: linux. bash$ cd ../ bash$ pwd /d/Gnu/ncurses_1.9.9g bash$ make install cd include; make INSTALL_PREFIX="" install installing term.h in /usr/local/include /bin/gginstall -c -m 644 term.h /usr/local/include installing curses.h in /usr/local/include /bin/gginstall -c -m 644 curses.h /usr/local/include installing unctrl.h in /usr/local/include /bin/gginstall -c -m 644 unctrl.h /usr/local/include installing termcap.h in /usr/local/include /bin/gginstall -c -m 644 termcap.h /usr/local/include cd man; make INSTALL_PREFIX="" install sh ./edit_man.sh /usr/local/man terminfo.5 ./*.[0-9]* installing /usr/local/man/man5/terminfo.5 /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man5/terminfo.5 installing /usr/local/man/man1/captoinfo.1m /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man1/captoinfo.1m installing /usr/local/man/man1/clear.1 /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man1/clear.1 installing /usr/local/man/man3/curs_addch.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_addch.3x installing /usr/local/man/man3/curs_addchstr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_addchstr.3x installing /usr/local/man/man3/curs_addstr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_addstr.3x installing /usr/local/man/man3/curs_attr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_attr.3x installing /usr/local/man/man3/curs_beep.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_beep.3x installing /usr/local/man/man3/curs_bkgd.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_bkgd.3x installing /usr/local/man/man3/curs_border.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_border.3x installing /usr/local/man/man3/curs_clear.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_clear.3x installing /usr/local/man/man3/curs_color.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_color.3x installing /usr/local/man/man3/curs_delch.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_delch.3x installing /usr/local/man/man3/curs_deleteln.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_deleteln.3x installing /usr/local/man/man3/curs_getch.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_getch.3x installing /usr/local/man/man3/curs_getstr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_getstr.3x installing /usr/local/man/man3/curs_getyx.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_getyx.3x installing /usr/local/man/man3/curs_inch.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_inch.3x installing /usr/local/man/man3/curs_inchstr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_inchstr.3x installing /usr/local/man/man3/curs_initscr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_initscr.3x installing /usr/local/man/man3/curs_inopts.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_inopts.3x installing /usr/local/man/man3/curs_insch.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_insch.3x installing /usr/local/man/man3/curs_insstr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_insstr.3x installing /usr/local/man/man3/curs_instr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_instr.3x installing /usr/local/man/man3/curs_kernel.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_kernel.3x installing /usr/local/man/man3/curs_mouse.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_mouse.3x installing /usr/local/man/man3/curs_move.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_move.3x installing /usr/local/man/man3/curs_outopts.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_outopts.3x installing /usr/local/man/man3/curs_overlay.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_overlay.3x installing /usr/local/man/man3/curs_pad.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_pad.3x installing /usr/local/man/man3/curs_print.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_print.3x installing /usr/local/man/man3/curs_printw.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_printw.3x installing /usr/local/man/man3/curs_refresh.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_refresh.3x installing /usr/local/man/man3/curs_scanw.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_scanw.3x installing /usr/local/man/man3/curs_scr_dmp.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_scr_dmp.3x installing /usr/local/man/man3/curs_scroll.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_scroll.3x installing /usr/local/man/man3/curs_slk.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_slk.3x installing /usr/local/man/man3/curs_termattrs.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_termattrs.3x installing /usr/local/man/man3/curs_termcap.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_termcap.3x installing /usr/local/man/man3/curs_terminfo.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_terminfo.3x installing /usr/local/man/man3/curs_touch.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_touch.3x installing /usr/local/man/man3/curs_util.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_util.3x installing /usr/local/man/man3/curs_window.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/curs_window.3x installing /usr/local/man/man3/form.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form.3x installing /usr/local/man/man3/form_cursor.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_cursor.3x installing /usr/local/man/man3/form_data.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_data.3x installing /usr/local/man/man3/form_driver.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_driver.3x installing /usr/local/man/man3/form_field.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_field.3x installing /usr/local/man/man3/form_field_attributes.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_field_attributes.3x installing /usr/local/man/man3/form_field_buffer.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_field_buffer.3x installing /usr/local/man/man3/form_field_info.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_field_info.3x installing /usr/local/man/man3/form_field_just.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_field_just.3x installing /usr/local/man/man3/form_field_new.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_field_new.3x installing /usr/local/man/man3/form_field_opts.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_field_opts.3x installing /usr/local/man/man3/form_field_userptr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_field_userptr.3x installing /usr/local/man/man3/form_field_validation.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_field_validation.3x installing /usr/local/man/man3/form_fieldtype.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_fieldtype.3x installing /usr/local/man/man3/form_hook.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_hook.3x installing /usr/local/man/man3/form_new.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_new.3x installing /usr/local/man/man3/form_new_page.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_new_page.3x installing /usr/local/man/man3/form_opts.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_opts.3x installing /usr/local/man/man3/form_page.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_page.3x installing /usr/local/man/man3/form_post.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_post.3x installing /usr/local/man/man3/form_requestname.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_requestname.3x installing /usr/local/man/man3/form_userptr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_userptr.3x installing /usr/local/man/man3/form_win.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/form_win.3x installing /usr/local/man/man1/infocmp.1m /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man1/infocmp.1m installing /usr/local/man/man3/menu.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu.3x installing /usr/local/man/man3/menu_attribs.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_attribs.3x installing /usr/local/man/man3/menu_cursor.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_cursor.3x installing /usr/local/man/man3/menu_driver.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_driver.3x installing /usr/local/man/man3/menu_format.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_format.3x installing /usr/local/man/man3/menu_hook.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_hook.3x installing /usr/local/man/man3/menu_items.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_items.3x installing /usr/local/man/man3/menu_mark.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_mark.3x installing /usr/local/man/man3/menu_new.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_new.3x installing /usr/local/man/man3/menu_opts.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_opts.3x installing /usr/local/man/man3/menu_pattern.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_pattern.3x installing /usr/local/man/man3/menu_post.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_post.3x installing /usr/local/man/man3/menu_requestname.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_requestname.3x installing /usr/local/man/man3/menu_spacing.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_spacing.3x installing /usr/local/man/man3/menu_userptr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_userptr.3x installing /usr/local/man/man3/menu_win.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/menu_win.3x installing /usr/local/man/man3/mitem_current.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/mitem_current.3x installing /usr/local/man/man3/mitem_name.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/mitem_name.3x installing /usr/local/man/man3/mitem_new.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/mitem_new.3x installing /usr/local/man/man3/mitem_opts.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/mitem_opts.3x installing /usr/local/man/man3/mitem_userptr.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/mitem_userptr.3x installing /usr/local/man/man3/mitem_value.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/mitem_value.3x installing /usr/local/man/man3/mitem_visible.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/mitem_visible.3x installing /usr/local/man/man3/ncurses.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/ncurses.3x installing /usr/local/man/man3/panel.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/panel.3x installing /usr/local/man/man3/resizeterm.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/resizeterm.3x installing /usr/local/man/man5/term.5 /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man5/term.5 installing /usr/local/man/man7/term.7 /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man7/term.7 installing /usr/local/man/man5/terminfo.5 /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man5/terminfo.5 installing /usr/local/man/man1/tic.1m /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man1/tic.1m installing /usr/local/man/man1/toe.1m /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man1/toe.1m installing /usr/local/man/man1/tput.1 /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man1/tput.1 installing /usr/local/man/man1/tset.1 /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man1/tset.1 installing /usr/local/man/man3/wresize.3x /bin/gginstall -c -m 644 /tmp/man7077 /usr/local/man/man3/wresize.3x cd ncurses; make INSTALL_PREFIX="" install installing ../lib/libncurses.a as /usr/local/lib/libncurses.a /bin/ginstall -c -m 644 ../lib/libncurses.a /usr/local/lib/libncurses.a /bin/gginstall -c -m 644 ../lib/libncurses.a /usr/local/lib/libncurses.a linking libcurses.a to libncurses.a rm -f /usr/local/lib/libcurses.a ln -s /usr/local/lib/libncurses.a /usr/local/lib/libcurses.a ranlib /usr/local/lib/libncurses.a installing ../lib/libncurses_g.a as /usr/local/lib/libncurses_g.a /bin/ginstall -c -m 644 ../lib/libncurses_g.a /usr/local/lib/libncurses_g.a /bin/gginstall -c -m 644 ../lib/libncurses_g.a /usr/local/lib/libncurses_g.a ranlib /usr/local/lib/libncurses_g.a cd progs; make INSTALL_PREFIX="" install /bin/ginstall -c tic /usr/local/bin/tic /bin/gginstall -c tic.exe /usr/local/bin/tic.exe /bin/ginstall -c toe /usr/local/bin/toe /bin/gginstall -c toe.exe /usr/local/bin/toe.exe /bin/ginstall -c infocmp /usr/local/bin/infocmp /bin/gginstall -c infocmp.exe /usr/local/bin/infocmp.exe /bin/ginstall -c clear /usr/local/bin/clear /bin/gginstall -c clear.exe /usr/local/bin/clear.exe /bin/ginstall -c tput /usr/local/bin/tput /bin/gginstall -c tput.exe /usr/local/bin/tput.exe /bin/ginstall -c tset /usr/local/bin/tset /bin/gginstall -c tset.exe /usr/local/bin/tset.exe linking captoinfo to tic ln: cannot create hard link `/usr/local/bin/captoinfo' to `/usr/local/bin/tic': No such file or directory make: [install.progs] Error 1 (ignored) linking reset to tset cd panel; make INSTALL_PREFIX="" install installing ../lib/libpanel.a as /usr/local/lib/libpanel.a /bin/ginstall -c -m 644 ../lib/libpanel.a /usr/local/lib/libpanel.a /bin/gginstall -c -m 644 ../lib/libpanel.a /usr/local/lib/libpanel.a ranlib /usr/local/lib/libpanel.a installing ../lib/libpanel_g.a as /usr/local/lib/libpanel_g.a /bin/ginstall -c -m 644 ../lib/libpanel_g.a /usr/local/lib/libpanel_g.a /bin/gginstall -c -m 644 ../lib/libpanel_g.a /usr/local/lib/libpanel_g.a ranlib /usr/local/lib/libpanel_g.a installing ./panel.h in /usr/local/include /bin/gginstall -c -m 644 ./panel.h /usr/local/include cd menu; make INSTALL_PREFIX="" install installing ../lib/libmenu.a as /usr/local/lib/libmenu.a /bin/ginstall -c -m 644 ../lib/libmenu.a /usr/local/lib/libmenu.a /bin/gginstall -c -m 644 ../lib/libmenu.a /usr/local/lib/libmenu.a ranlib /usr/local/lib/libmenu.a installing ../lib/libmenu_g.a as /usr/local/lib/libmenu_g.a /bin/ginstall -c -m 644 ../lib/libmenu_g.a /usr/local/lib/libmenu_g.a /bin/gginstall -c -m 644 ../lib/libmenu_g.a /usr/local/lib/libmenu_g.a ranlib /usr/local/lib/libmenu_g.a installing ./eti.h in /usr/local/include /bin/gginstall -c -m 644 ./eti.h /usr/local/include installing ./menu.h in /usr/local/include /bin/gginstall -c -m 644 ./menu.h /usr/local/include cd form; make INSTALL_PREFIX="" install installing ../lib/libform.a as /usr/local/lib/libform.a /bin/ginstall -c -m 644 ../lib/libform.a /usr/local/lib/libform.a /bin/gginstall -c -m 644 ../lib/libform.a /usr/local/lib/libform.a ranlib /usr/local/lib/libform.a installing ../lib/libform_g.a as /usr/local/lib/libform_g.a /bin/ginstall -c -m 644 ../lib/libform_g.a /usr/local/lib/libform_g.a /bin/gginstall -c -m 644 ../lib/libform_g.a /usr/local/lib/libform_g.a ranlib /usr/local/lib/libform_g.a installing ./form.h in /usr/local/include /bin/gginstall -c -m 644 ./form.h /usr/local/include cd misc; make INSTALL_PREFIX="" install sh ./run_tic.sh /usr/local/bin . /usr/local/share/terminfo ** Building terminfo database, please wait... ** adjusting tabset paths "/tmp/7863", line 268, terminal 'unknown': resolution of use=dumb failed "/tmp/7863", line 772, terminal 'iris-ansi-ap': resolution of use=iris-ansi failed "/tmp/7863", line 917, terminal 'pcvt25': resolution of use=pcvtXX failed "/tmp/7863", line 920, terminal 'pcvt28': resolution of use=pcvtXX failed "/tmp/7863", line 923, terminal 'pcvt35': resolution of use=pcvtXX failed "/tmp/7863", line 926, terminal 'pcvt40': resolution of use=pcvtXX failed "/tmp/7863", line 929, terminal 'pcvt43': resolution of use=pcvtXX failed "/tmp/7863", line 932, terminal 'pcvt50': resolution of use=pcvtXX failed "/tmp/7863", line 939, terminal 'pcvt25w': resolution of use=pcvtXX failed "/tmp/7863", line 942, terminal 'pcvt28w': resolution of use=pcvtXX failed "/tmp/7863", line 945, terminal 'pcvt35w': resolution of use=pcvtXX failed "/tmp/7863", line 948, terminal 'pcvt40w': resolution of use=pcvtXX failed "/tmp/7863", line 951, terminal 'pcvt43w': resolution of use=pcvtXX failed "/tmp/7863", line 954, terminal 'pcvt50w': resolution of use=pcvtXX failed "/tmp/7863", line 961, terminal 'x68k': resolution of use=vt220 failed "/tmp/7863", line 1001, terminal 'cons25': resolution of use=cons25w failed "/tmp/7863", line 1023, terminal 'cons25r': resolution of use=cons25w failed "/tmp/7863", line 1038, terminal 'cons25l1': resolution of use=cons25w failed "/tmp/7863", line 1100, terminal 'bsdos-bold': resolution of use=bsdos failed "/tmp/7863", line 1105, terminal 'pc3': resolution of use=bsdos failed "/tmp/7863", line 1244, terminal 'vt100nam': resolution of use=vt100-am failed "/tmp/7863", line 1249, terminal 'vt100-w': resolution of use=vt100-am failed "/tmp/7863", line 1259, terminal 'vt100-nav': resolution of use=vt100 failed "/tmp/7863", line 1271, terminal 'vt100-s': resolution of use=vt100-am failed "/tmp/7863", line 1280, terminal 'vt100-s-bot': resolution of use=vt100-am failed "/tmp/7863", line 1287, terminal 'vt102': resolution of use=vt100 failed "/tmp/7863", line 1306, terminal 'vt125': resolution of use=vt100 failed "/tmp/7863", line 1337, terminal 'vt132': resolution of use=vt100 failed "/tmp/7863", line 1369, terminal 'vt220-w': resolution of use=vt220 failed "/tmp/7863", line 1384, terminal 'vt220d': resolution of use=vt220 failed "/tmp/7863", line 1388, terminal 'vt220-nam': resolution of use=vt220 failed "/tmp/7863", line 1425, terminal 'vt320nam': resolution of use=vt220 failed "/tmp/7863", line 1474, terminal 'vt320-nam': resolution of use=vt320 failed "/tmp/7863", line 1480, terminal 'vt320-w': resolution of use=vt320 failed "/tmp/7863", line 1734, terminal 'dec-vt220': resolution of use=vt220 failed "/tmp/7863", line 1874, terminal 'xterm': resolution of use=xterm-r6 failed "/tmp/7863", line 1987, terminal 'xterm-pcolor': resolution of use=xterm+sl failed "/tmp/7863", line 2079, terminal 'vremote': resolution of use=cbunix failed "/tmp/7863", line 2083, terminal 'pty': resolution of use=cbunix failed "/tmp/7863", line 2200, terminal 'sun-ss5': resolution of use=sun-il failed "/tmp/7863", line 2203, terminal 'sun': resolution of use=sun-il failed "/tmp/7863", line 2402, terminal 'altos3': resolution of use=altos2 failed "/tmp/7863", line 2404, terminal 'altos4': resolution of use=wy50 failed "/tmp/7863", line 2458, terminal 'hp110': resolution of use=hpgeneric failed "/tmp/7863", line 2525, terminal 'hp2621-fl': resolution of use=hpgeneric failed "/tmp/7863", line 2699, terminal 'hp2645': resolution of use=hpgeneric failed "/tmp/7863", line 2845, terminal 'hp2397a': resolution of use=hp+color failed "/tmp/7863", line 2887, terminal 'hp2392': resolution of use=hpsub failed "/tmp/7863", line 2913, terminal 'hpex': resolution of use=hpsub failed "/tmp/7863", line 3046, terminal 'gator-t': resolution of use=gator failed "/tmp/7863", line 3059, terminal 'gator-52': resolution of use=vt52 failed "/tmp/7863", line 3083, terminal 'dku7003': resolution of use=dku7003-dumb failed "/tmp/7863", line 3138, terminal 'adm3a+': resolution of use=adm3a failed "/tmp/7863", line 3200, terminal 'adm21': resolution of use=adm3a failed "/tmp/7863", line 3369, terminal 'adm31-old': resolution of use=adm31 failed "/tmp/7863", line 3373, terminal 'adm36': resolution of use=vt100 failed "/tmp/7863", line 3432, terminal 'pt100w': resolution of use=pt100 failed "/tmp/7863", line 3434, terminal 'pt250': resolution of use=pt100 failed "/tmp/7863", line 3514, terminal 'qvt103-w': resolution of use=qvt103 failed "/tmp/7863", line 3541, terminal 'qvt203': resolution of use=qvt103 failed "/tmp/7863", line 3726, terminal 'tvi912c': resolution of use=tvi912 failed "/tmp/7863", line 3730, terminal 'tvi912-2p': resolution of use=tvi912 failed "/tmp/7863", line 3793, terminal 'tvi920b': resolution of use=tvi912 failed "/tmp/7863", line 4236, terminal 'tvi970-vb': resolution of use=tvi970 failed "/tmp/7863", line 4239, terminal 'tvi970-2p': resolution of use=tvi970 failed "/tmp/7863", line 4383, terminal 'vi200-f': resolution of use=vi200 failed "/tmp/7863", line 4385, terminal 'vi200-rv': resolution of use=vi200 failed "/tmp/7863", line 4410, terminal 'vi300-old': resolution of use=vi300 failed "/tmp/7863", line 4440, terminal 'vi550': resolution of use=vi300 failed "/tmp/7863", line 4451, terminal 'vi603': resolution of use=vt100 failed "/tmp/7863", line 4525, terminal 'wy30-mc': resolution of use=wy30 failed "/tmp/7863", line 4531, terminal 'wy30-vb': resolution of use=wy30 failed "/tmp/7863", line 4581, terminal 'wy50-mc': resolution of use=wy50 failed "/tmp/7863", line 4584, terminal 'wy50-vb': resolution of use=wy50 failed "/tmp/7863", line 4587, terminal 'wy50-w': resolution of use=wy50 failed "/tmp/7863", line 5012, terminal 'wy75-mc': resolution of use=wy75 failed "/tmp/7863", line 5015, terminal 'wy75-vb': resolution of use=wy75 failed "/tmp/7863", line 5018, terminal 'wy75-w': resolution of use=wy75 failed "/tmp/7863", line 5072, terminal 'wy85-vb': resolution of use=wy85 failed "/tmp/7863", line 5077, terminal 'wy85-w': resolution of use=wy85 failed "/tmp/7863", line 5135, terminal 'wy185-24': resolution of use=wy185 failed "/tmp/7863", line 5139, terminal 'wy185-vb': resolution of use=wy185 failed "/tmp/7863", line 5146, terminal 'wy185-w': resolution of use=wy185 failed "/tmp/7863", line 5321, terminal 'wy370': resolution of use=wy370-nk failed "/tmp/7863", line 5335, terminal 'wy370-105k': resolution of use=wy370-nk failed "/tmp/7863", line 5345, terminal 'wy370-EPC': resolution of use=wy370-nk failed "/tmp/7863", line 5381, terminal 'wy160-tek': resolution of use=wy99gt-tek failed "/tmp/7863", line 5466, terminal 'wy520-24': resolution of use=wy520 failed "/tmp/7863", line 5470, terminal 'wy520-vb': resolution of use=wy520 failed "/tmp/7863", line 5477, terminal 'wy520-w': resolution of use=wy520 failed "/tmp/7863", line 5497, terminal 'wy520-epc': resolution of use=wy520 failed "/tmp/7863", line 5526, terminal 'wy520-36': resolution of use=wy520 failed "/tmp/7863", line 5532, terminal 'wy520-48': resolution of use=wy520 failed "/tmp/7863", line 5589, terminal 'wy75ap': resolution of use=wy75 failed "/tmp/7863", line 5619, terminal 'kermit-am': resolution of use=kermit failed "/tmp/7863", line 5629, terminal 'pckermit': resolution of use=kermit failed "/tmp/7863", line 5642, terminal 'pckermit120': resolution of use=kermit failed "/tmp/7863", line 5666, terminal 'msk227am': resolution of use=msk227 failed "/tmp/7863", line 5679, terminal 'msk22714': resolution of use=mskermit227 failed "/tmp/7863", line 5812, terminal 'att2350': resolution of use=att2300 failed "/tmp/7863", line 5849, terminal 'att4410v1-w': resolution of use=att5410v1 failed "/tmp/7863", line 5853, terminal 'att4410': resolution of use=att5410v1 failed "/tmp/7863", line 6018, terminal 'att5420_2-w': resolution of use=att5420_2 failed "/tmp/7863", line 6040, terminal 'att4418-w': resolution of use=att5418 failed "/tmp/7863", line 6090, terminal 'att4424-1': resolution of use=att4424 failed "/tmp/7863", line 6155, terminal 'att5425-nl': resolution of use=att4425 failed "/tmp/7863", line 6159, terminal 'att5425-w': resolution of use=tty5425 failed "/tmp/7863", line 6372, terminal 'att5620-24': resolution of use=att5620 failed "/tmp/7863", line 6374, terminal 'att5620-34': resolution of use=att5620 failed "/tmp/7863", line 6430, terminal 'att605-pc': resolution of use=att605 failed "/tmp/7863", line 6433, terminal 'att605-w': resolution of use=att605 failed "/tmp/7863", line 6469, terminal 'att610-w': resolution of use=att610 failed "/tmp/7863", line 6485, terminal 'att610-103k': resolution of use=att610 failed "/tmp/7863", line 6497, terminal 'att615': resolution of use=att610 failed "/tmp/7863", line 6554, terminal 'att620-w': resolution of use=att620 failed "/tmp/7863", line 6572, terminal 'att620-103k': resolution of use=att620 failed "/tmp/7863", line 6610, terminal 'att630-24': resolution of use=att630 failed "/tmp/7863", line 6864, terminal 'att505-24': resolution of use=att505 failed "/tmp/7863", line 6866, terminal 'tt505-22': resolution of use=att505 failed "/tmp/7863", line 6909, terminal 'ampex175-b': resolution of use=ampex175 failed "/tmp/7863", line 6947, terminal 'ampex219w': resolution of use=ampex219 failed "/tmp/7863", line 7163, terminal 'aaa-18': resolution of use=aaa+unk failed "/tmp/7863", line 7169, terminal 'aaa-20': resolution of use=aaa+unk failed "/tmp/7863", line 7173, terminal 'aaa-22': resolution of use=aaa+unk failed "/tmp/7863", line 7177, terminal 'aaa-24': resolution of use=aaa+unk failed "/tmp/7863", line 7184, terminal 'aaa-26': resolution of use=aaa+unk failed "/tmp/7863", line 7189, terminal 'aaa-28': resolution of use=aaa+unk failed "/tmp/7863", line 7198, terminal 'aaa-30-s': resolution of use=aaa+unk failed "/tmp/7863", line 7211, terminal 'aaa': resolution of use=aaa+unk failed "/tmp/7863", line 7224, terminal 'aaa-36': resolution of use=aaa+unk failed "/tmp/7863", line 7231, terminal 'aaa-40': resolution of use=aaa+unk failed "/tmp/7863", line 7238, terminal 'aaa-48': resolution of use=aaa+unk failed "/tmp/7863", line 7247, terminal 'aaa-60-s': resolution of use=aaa+unk failed "/tmp/7863", line 7254, terminal 'aaa-60': resolution of use=aaa+unk failed "/tmp/7863", line 7266, terminal 'guru': resolution of use=aaa+unk failed "/tmp/7863", line 7282, terminal 'guru-s': resolution of use=guru+s failed "/tmp/7863", line 7293, terminal 'guru-44-s': resolution of use=guru+s failed "/tmp/7863", line 7301, terminal 'guru-76-s': resolution of use=guru+s failed "/tmp/7863", line 7312, terminal 'guru-76-w-s': resolution of use=guru+s failed "/tmp/7863", line 7358, terminal 'regent100': resolution of use=regent failed "/tmp/7863", line 7360, terminal 'regent20': resolution of use=regent failed "/tmp/7863", line 7390, terminal 'screwpoint': resolution of use=viewpoint failed "/tmp/7863", line 7496, terminal 'cit101e-n': resolution of use=cit101e failed "/tmp/7863", line 7499, terminal 'cit101e-132': resolution of use=cit101e failed "/tmp/7863", line 7503, terminal 'cit101e-n132': resolution of use=cit101e failed "/tmp/7863", line 7544, terminal 'citoh': resolution of use=lpr failed "/tmp/7863", line 7750, terminal 'c108-4p': resolution of use=c100 failed "/tmp/7863", line 7826, terminal 'c100-rv': resolution of use=c100 failed "/tmp/7863", line 7829, terminal 'oc100': resolution of use=c100 failed "/tmp/7863", line 7861, terminal 'avt-rv-ns': resolution of use=avt-ns failed "/tmp/7863", line 7864, terminal 'avt-w-ns': resolution of use=avt-ns failed "/tmp/7863", line 7868, terminal 'avt-w-rv-ns': resolution of use=avt-ns failed "/tmp/7863", line 7886, terminal 'avt': resolution of use=avt+s failed "/tmp/7863", line 7886, terminal 'avt': resolution of use=avt-ns failed "/tmp/7863", line 7888, terminal 'avt-rv': resolution of use=avt+s failed "/tmp/7863", line 7889, terminal 'avt-rv': resolution of use=avt-ns failed "/tmp/7863", line 7891, terminal 'avt-w': resolution of use=avt+s failed "/tmp/7863", line 7892, terminal 'avt-w': resolution of use=avt-ns failed "/tmp/7863", line 7896, terminal 'avt-w-rv': resolution of use=avt+s failed "/tmp/7863", line 7896, terminal 'avt-w-rv': resolution of use=avt-ns failed "/tmp/7863", line 7916, terminal 'contel301': resolution of use=contel300 failed "/tmp/7863", line 7954, terminal 'dg211': resolution of use=dg200 failed "/tmp/7863", line 7957, terminal 'dg450': resolution of use=dg200 failed "/tmp/7863", line 8030, terminal 'cs10-w': resolution of use=cs10 failed "/tmp/7863", line 8057, terminal 'dmchat': resolution of use=dm2500 failed "/tmp/7863", line 8074, terminal 'dm3045': resolution of use=dm3025 failed "/tmp/7863", line 8111, terminal 'dm80': resolution of use=vt100 failed "/tmp/7863", line 8217, terminal 'ts100-ctxt': resolution of use=ts100 failed "/tmp/7863", line 8474, terminal 'sbi': resolution of use=sb1 failed "/tmp/7863", line 8498, terminal 'superbeeic': resolution of use=superbee-xsb failed "/tmp/7863", line 8646, terminal 'hz1552': resolution of use=vt52 failed "/tmp/7863", line 8676, terminal 'esprit-am': resolution of use=esprit failed "/tmp/7863", line 8732, terminal 'ibm3151': resolution of use=ibm3163 failed "/tmp/7863", line 8769, terminal 'ibm3164': resolution of use=ibm3163 failed "/tmp/7863", line 8780, terminal 'ibm-apl': resolution of use=dm1520 failed "/tmp/7863", line 8790, terminal 'ibmmono': resolution of use=ibm3101 failed "/tmp/7863", line 9021, terminal 'abm85h': resolution of use=abm85 failed "/tmp/7863", line 9027, terminal 'abm85e': resolution of use=abm85 failed "/tmp/7863", line 9033, terminal 'abm85h-old': resolution of use=abm85 failed "/tmp/7863", line 9180, terminal 'prism5': resolution of use=p4 failed "/tmp/7863", line 9191, terminal 'prism7': resolution of use=p4 failed "/tmp/7863", line 9207, terminal 'prism8': resolution of use=p4 failed "/tmp/7863", line 9272, terminal 'prism9-w': resolution of use=p9 failed "/tmp/7863", line 9300, terminal 'prism12': resolution of use=p9 failed "/tmp/7863", line 9334, terminal 'prism14': resolution of use=p9 failed "/tmp/7863", line 9405, terminal 'act5': resolution of use=act4 failed "/tmp/7863", line 9409, terminal 'mime-fb': resolution of use=mime failed "/tmp/7863", line 9411, terminal 'mime-hb': resolution of use=mime failed "/tmp/7863", line 9446, terminal 'mime3a': resolution of use=adm3a failed "/tmp/7863", line 9652, terminal 'pe7000c': resolution of use=pe7000m failed "/tmp/7863", line 9690, terminal 'tandem6510': resolution of use=adm3a failed "/tmp/7863", line 9739, terminal 'dt100w': resolution of use=dt100 failed "/tmp/7863", line 9776, terminal 'tek4013': resolution of use=tek4012 failed "/tmp/7863", line 9779, terminal 'tek4014': resolution of use=tek4012 failed "/tmp/7863", line 9834, terminal 'tek4025-17': resolution of use=tek4025 failed "/tmp/7863", line 9843, terminal 'tek4025-ex': resolution of use=tek4025 failed "/tmp/7863", line 9894, terminal 'tek4025ex': resolution of use=tek4025 failed "/tmp/7863", line 10040, terminal 'tek4207-s': resolution of use=tek4107 failed "/tmp/7863", line 10087, terminal 'tek4113-34': resolution of use=tek4113 failed "/tmp/7863", line 10145, terminal 'tek4125': resolution of use=vt100 failed "/tmp/7863", line 10310, terminal 'vc303a': resolution of use=vc303 failed "/tmp/7863", line 10320, terminal 'vc404-s': resolution of use=vc404 failed "/tmp/7863", line 10335, terminal 'vc415': resolution of use=vc404 failed "/tmp/7863", line 10351, terminal 'pcplot': resolution of use=vt100 failed "/tmp/7863", line 10414, terminal 'apple2e-p': resolution of use=apple2e failed "/tmp/7863", line 10552, terminal 'liswb': resolution of use=lisa failed "/tmp/7863", line 10582, terminal 'lisaterm-w': resolution of use=lisaterm failed "/tmp/7863", line 10591, terminal 'mac': resolution of use=lisa failed "/tmp/7863", line 10652, terminal 'uniterm': resolution of use=vt220 failed "/tmp/7863", line 10856, terminal 'pc-minix': resolution of use=minix failed "/tmp/7863", line 10896, terminal 'basis': resolution of use=adm3a failed "/tmp/7863", line 10987, terminal 'cbblit': resolution of use=blit failed "/tmp/7863", line 11018, terminal 'bitgraph': resolution of use=bg2.0 failed "/tmp/7863", line 11021, terminal 'bg2.0rv': resolution of use=bg2.0 failed "/tmp/7863", line 11035, terminal 'bg1.25rv': resolution of use=bg1.25 failed "/tmp/7863", line 11037, terminal 'bg1.25nv': resolution of use=bg1.25 failed "/tmp/7863", line 11231, terminal 'ln03-w': resolution of use=ln03 failed "/tmp/7863", line 11284, terminal 'ps300': resolution of use=vt100 failed "/tmp/7863", line 11346, terminal 'h19-bs': resolution of use=h19-b failed "/tmp/7863", line 11365, terminal 'h19-u': resolution of use=h19-b failed "/tmp/7863", line 11370, terminal 'h19-g': resolution of use=h19-b failed "/tmp/7863", line 11373, terminal 'alto-h19': resolution of use=h19 failed "/tmp/7863", line 11458, terminal 'z29a-kc-uc': resolution of use=z29a failed "/tmp/7863", line 11461, terminal 'z29a-nkc-bc': resolution of use=z29a failed "/tmp/7863", line 11464, terminal 'z29a-nkc-uc': resolution of use=z29a failed "/tmp/7863", line 11506, terminal 'p19': resolution of use=h19-b failed "/tmp/7863", line 11584, terminal 'intertube2': resolution of use=intertube failed "/tmp/7863", line 11622, terminal 'graphos-30': resolution of use=graphos failed "/tmp/7863", line 11643, terminal 'modgraph': resolution of use=vt100 failed "/tmp/7863", line 11776, terminal 'rt6221-w': resolution of use=rt6221 failed "/tmp/7863", line 11821, terminal 'hirez100-w': resolution of use=hirez100 failed "/tmp/7863", line 11860, terminal 'soroc120': resolution of use=adm3a failed "/tmp/7863", line 11957, terminal 'tab132': resolution of use=vt100 failed "/tmp/7863", line 12006, terminal 't1061f': resolution of use=t1061 failed "/tmp/7863", line 12071, terminal 'ti916': resolution of use=vt220 failed "/tmp/7863", line 12123, terminal 'ti924w': resolution of use=ti924 failed "/tmp/7863", line 12125, terminal 'ti924-8w': resolution of use=ti924-8 failed "/tmp/7863", line 12140, terminal 'ti926': resolution of use=ti924 failed "/tmp/7863", line 12144, terminal 'ti926-8': resolution of use=ti924-8 failed "/tmp/7863", line 12511, terminal 'diablo1620-m8': resolution of use=diablo1620 failed "/tmp/7863", line 12515, terminal 'diablo1640': resolution of use=diablo1620 failed "/tmp/7863", line 12520, terminal 'diablo1640-lm': resolution of use=diablo1620 failed "/tmp/7863", line 12655, terminal 'env230': resolution of use=vt100 failed "/tmp/7863", line 12664, terminal 'ep40': resolution of use=ep4080 failed ? tic could not build /usr/local/share/terminfo make: *** [install.data] Error 1 make: *** [install] Error 2 bash$ Bruce S. Skinner | skinner@drea.dnd.ca DEFENCE RESEARCH ESTABLISHMENT ATLANTIC | norstead@ns.sympatico.ca Dartmouth, Nova Scotia, CANADA, B2Y 3Z7 | (902) 426-3100 x205 or 538-1765 "New opinions are always suspected, and usually opposed without any other reason but because they are not already common." John Locke - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 8 08:11:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 08 Dec 1997 08:11:00 -0000 Subject: More on getting coolview-bash to work with emacs Message-ID: <19971208161112.16720.qmail@hotmail.com> I don't use emacs so I don't know if this can be done; but, if you can configure emacs to send bash a \n instead of a \r\n then it should work. I believe that Sergey's newest versions of bash forces binary mode always. Is this correct Sergey? Or is it controlled from cygwin.dll? Depending on Sergey's answer you could retrieve the sources for bash and make them yourself checking the line received for the \r\n combination. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Fri, 05 Dec 1997 13:14:02 -0800 >To: gnu-win32@cygnus.com >From: Guy Gascoigne - Piggford >Subject: More on getting coolview-bash to work with emacs > >I said a few days ago that I'm managed to get bash working correctly under >emacs on one machine and yet have it fail with ^M command not found error >messages on another system. Well I found out what the problem was, though >it's less than ideal - I was using different versions of the coolview >cygwin.dll on the two systems. > >Regretable the old one (dated 9/32/97) works correctly and the newer one >(dated 11/2/97) fails. Is the change of behaviour in the new version >configurable in any way? > >Guy > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hiro@lynx.com Mon Dec 8 11:50:00 1997 From: hiro@lynx.com (Hiro Sugawara) Date: Mon, 08 Dec 1997 11:50:00 -0000 Subject: bash ^C problem Message-ID: <348C4FD8.7DE14518@lynx.com> I have bash and other stuff out of b18 and have the following problem: When I hit ^C at bash with child processes, a signal seems to be broadcast to all of its children including background processes. Is this true? This hurts my usual complation habit on large programs like: bash$ make >make.log& [1] 12345 bash$ tail -f make.log .... .... ^C (to terminate only tail) Regards, hiro -- Hiro Sugawara, a ticketed driver on the Information Super-Highway Lynx Real-Time Systems, Inc. mailto:hiro@lynx.com 2239 Samaritan Dr. San Jose, CA 95124 http://www.lynx.com Voice:+1(408)879-3900 Fax:+1(408)879-3920 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Mon Dec 8 11:50:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Mon, 08 Dec 1997 11:50:00 -0000 Subject: Problem with libwsock32.a Message-ID: <199712081638.QAA109330@out1.ibm.net> >> vischne@ibm.net I set up a dual cygwin/ming system and compiled the slrn news reader using a ming-compiled libslang and libtermcap and libwsock32.a that is common to ming and cygwin. When I started to run slrn.exe, I got the following: ====================================================== slrn 0.9.4.3 (Dec 7 1997 23:59:53) Reading startup file ./slrn.rc. Connecting to host news3.ibm.net ...news3.ibm.net: Unknown host. slrn fatal error: << I should point out that using ./configure to set up and compile slrn under cygwin 18 works quite well, and slrn.exe works also. But, it is working with cygwin_gethostbyname, not gethostbyname. The only reason to do a ming version is to test out the ming terminal: a dos window with ansi.sys. This is because the cygwin terminal window is just not good enough. Even without using a libslang application, you can see the problems with it. To prove this to yourself, get a bash$ prompt going in a dos window, and type bash$ env | less What you will see is the `env' screen with text from the window overlaid. This is just one demonstration of how inadequate the present cygwin console window is. Any fast backspacing or back arrow typing in the bash$ window will provide the other demonstration.  - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hiro@lynx.com Mon Dec 8 11:50:00 1997 From: hiro@lynx.com (Hiro Sugawara) Date: Mon, 08 Dec 1997 11:50:00 -0000 Subject: GDB problems Message-ID: <348C4EAB.2F1CF0FB@lynx.com> I'm building Win32-LynxOS cross debuggers out of b18 and having the following problems: 1. Missing function gdb/utils.c complains a missing function of "win32pollquit" which it makes a call to. I replaced the function with a stub and succeeded in building... 2. GDB crashes in ScreedRows() This reports a segmentation violation error. This seems to be a DLL function. I relaced the call with a default constant value of 24 and ScreenCols() with 80 and it no longer crashed. 3. GDB doesn't honor ^C Now I can debug a LynxOS process over TCP/IP, but if the process enters a blocking system call (like sleep()), hitting ^C at GDB is no use. (This may be related to the first problem.) I'm using a Pentium NT box with 32MB. Does anyone have similar symptoms (and solutions!)? Thanks. hiro -- Hiro Sugawara, a ticketed driver on the Information Super-Highway Lynx Real-Time Systems, Inc. mailto:hiro@lynx.com 2239 Samaritan Dr. San Jose, CA 95124 http://www.lynx.com Voice:+1(408)879-3900 Fax:+1(408)879-3920 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ggp@informix.com Mon Dec 8 11:50:00 1997 From: ggp@informix.com (Guy Gascoigne - Piggford) Date: Mon, 08 Dec 1997 11:50:00 -0000 Subject: More on getting coolview-bash to work with emacs References: <19971208161112.16720.qmail@hotmail.com> Message-ID: <3.0.3.32.19971208101531.01a704a0@pop.pdx.informix.com> At 08:11 AM 12/8/97 PST, you wrote: >I don't use emacs so I don't know if this can be done; but, if you can >configure emacs to send bash a \n instead of a \r\n then it should work. > >I believe that Sergey's newest versions of bash forces binary mode >always. Is this correct Sergey? Or is it controlled from cygwin.dll? >Depending on Sergey's answer you could retrieve the sources for bash and >make them yourself checking the line received for the \r\n combination. I was planning to do this, but rather wanted to limit the number of changes in the bash source itself if there was some other 'configuration' way of fixing it. Thanks - Guy - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From diegom@pts.mot.com Mon Dec 8 14:08:00 1997 From: diegom@pts.mot.com (Diego Montanez) Date: Mon, 08 Dec 1997 14:08:00 -0000 Subject: Compiling nvi Message-ID: <9712081733.AA04166@motprux.pr.paging.mot.com> People, I downloaded nvi 1.71 to compile with cygnus gcc b18. The configure phase works OK and the compile phase also. But when I try to run the resulting nvi.exe y get this: nvi: (null) Argh! What could be this??? Thanks in advance. Diego --- D I E G O A M O N T A N E Z - D I E G O M @ PTS . M O T . C O M --------------------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kmoon@www.geocities.com Mon Dec 8 15:08:00 1997 From: kmoon@www.geocities.com (kmoon) Date: Mon, 08 Dec 1997 15:08:00 -0000 Subject: Trying to compile TinyMUSH 2.2.3 Message-ID: <199712082308.PAA09780@norway.it.earthlink.net> I know this has already been ported, but anyone who runs a MUSH knows that you have to be able to compile the program yourself. Anyhow, After running the 'configure' shell script, I checked the config.log file it generated, and I noticed something about the slide.h (include/sys copy), around lines 31-37 (told me errors before '('). It told me there was a problem with this two times. I'm not sure if this is normal or not. However, after the configure, and modifying relevant Makefile and .h's entries, I tried to compile it. However, it failed with a problem with 'getpagesize' not having the same value ('getpagesize' is a symbolic constant which I believed is defined as a function - difficult to tell from code). Has anyone gotten this to work? If so, what kind of environment are you running? Thanks in advance. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From bennett@btinternet.com Mon Dec 8 16:16:00 1997 From: bennett@btinternet.com (Matthew Bennett) Date: Mon, 08 Dec 1997 16:16:00 -0000 Subject: directx example - problem Message-ID: <199712090016.QAA16596@cygint.cygnus.com> Hi I'm using mingw32, and I'm having some problems with the directx example from: http://dragon.acadiau.ca/~025668b/develop/directx/ It compiles without any errors and the program runs fine until it comes to create the surface. The "faulty" line is: ddrval = lpDD->lpVtbl->CreateSurface(lpDD, &ddsd, &lpDDSPrimary, NULL ); ddrval is returned as *not* DD_OK and so the program stops. Looking over the program, I can't find any obvious mistakes... so what is causing this? Thanks a lot for any help - I really want to get directx going! :) Matt - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jfmiller@polymail.cpunix.calpoly.edu Mon Dec 8 16:26:00 1997 From: jfmiller@polymail.cpunix.calpoly.edu (John Miller) Date: Mon, 08 Dec 1997 16:26:00 -0000 Subject: _beginthread Problems with win95 and mingw32 using C++ References: <01bd0141$5f8510a0$fa173185@gbird0.fu.is.saga-u.ac.jp> <348A499E.E58B8A57@polymail.calpoly.edu> Message-ID: <348C906C.D72544DB@polymail.calpoly.edu> Dear Frends, Thank you to all of you who were kind enough to lend a helping hand (esp. Colin Peters). For those of you intrested in the final solution here it is 1.) I change the Process.h file declairation of _beginthread by removing the __cdecl__ this satisfied the cpp error about 'name type' declaired as function returning function. (I think this problem is also fixed in the lates snapshot?) 2.) I am still puzzled by the link error. the function does seem to be within the extern "c" block so it shouldn't be mangled but I can't seem to get it to link. Insetad I Downloaded the mingw32 version of egcs 1.00 and I was able to get the code to compile correctly therefore I am not going to worry about this for now. Once again thank you to all of you for you comments and sugestions. John Miller - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 8 18:48:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 08 Dec 1997 18:48:00 -0000 Subject: Trouble building ncurses 1.9.9g. Message-ID: <19971209024826.5634.qmail@hotmail.com> I've tried to answer this about three times today. Trouble with the mail server. I'm looking for a different mail server as this one seems to be too busy. You need to: configure --enable-symlinks - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From shorti@apci.net Mon Dec 8 19:54:00 1997 From: shorti@apci.net (shorti) Date: Mon, 08 Dec 1997 19:54:00 -0000 Subject: (newbie) please be kind... Message-ID: <3.0.3.32.19971208215231.007ee900@apci.net> hello all: ok - here's the deal: i have very limited experience with c, although i'm figuring out the text-driven stuff fairly well. i would like to write windows programs using my new found resources, but i know next to nothing about API, integrating dll's, etc. does anyone know of any good tutorials/references on win32 (or c in general) on the web? maybe if you have nothing better to do over christmas holiday, you could write one? or is it just easier for me to go out and buy a book? If you have a list of links related to (c, win32) programming, it would be greatly appreciated.. i'll compile it to a webpage somewhere... Also - could someone point me to the things that make bash useful? i'm not sure totally what to do.. right now i'm questioning the wisdom of the person that said "ignorance is bliss"... thanx for your time and patience, - josh - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hgfernan@usp.br Mon Dec 8 23:07:00 1997 From: hgfernan@usp.br (hgfernan@usp.br) Date: Mon, 08 Dec 1997 23:07:00 -0000 Subject: Too Much Mail Volume Message-ID: <9712090709.AA50572@spider.uspnet.usp.br> On 3 Dec 97 at 11:04, Robert.V.Baron wrote: > I think the traffic is more like 20 to 30 pieces a day. For me, the > issue of volume is less important than quality. I'd like to see > several groups (news or mail), for: > newbies > programming questions > interactions with microsoft DLL's, tools > technical discussions > perhaps more. This would allow us to sort and categorize the volume > better. Hi, all! I'd like to give my vote for Robert's proposal. I'm not very sure about the mail categories, but the idea is excellent. That way, the signal to noise in the mailing list would be much higher. For instance, people that read the newbie mailing list would be sure they would find newbie questions: newbies would be free to post newbie questions, anyone else would answer those questions only if they really wanted to. Cum bona fide, some would say. That way, probably people would team up to create the long awaited (more complete) FAQ. I guess that subdivision interests Cygnus too, because there would be more focused light on their product. Regards, ++Hilton ------------------- Hilton Fernandes email: hgfernan@usp.br hilton@lsi www: http://www.lsi.usp.br/~hilton.html M. Sc. Student of Parallel Distributed Applications at LSI (Laboratory of Integrated Systems) University of S. Paulo - Brazil - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Tue Dec 9 00:15:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Tue, 09 Dec 1997 00:15:00 -0000 Subject: how to avoid mulitple definitions of STL free_list, start_free, end_free,heap_size References: <348ABB53.62AF@asc.on.ca> Message-ID: <9712090815.AA29183@modi.xraylith.wisc.edu> Stephen Reuss writes: > First up, I dealt with the pair structure redeclaration issue by > renaiming the pair structure in the third party code to mespair. If > anyone has ideas of how I can force compiler to choose one declaration > or another, please let me know. You can't. No C++ code written in the last 3, especially the last 2, years should've used 'struct pair', unless the author(s) was simply hoping that STL will never make it into the standard ;-) > > The new problem regards the: > multiple definition of `__default_alloc_template::start_free > ::end_free > ::heap_size > ::free_list > > in linking a C++ program. The link command is as follows: > Are you using the b18 C++ compiler? If so, you have other bugs to worry about as well (miscompiled code when using multiple virtual bases and such is just one example). If you are, then I strongly suggest you move to a newer C++ compiler, such as egcs-1.00 or even the newer FSF snapshots. For a binary distribution of egcs-1.00, check the URL below. Now that my obligatory plug is out the way, the problem is that the older versions of C++ front-end simply couldn't handle static template data members properly, and hence the problem. I can suggest a few workarounds: 1. Use -fno-implicit and instantiate everything yourself. I cover a bit about explicit instantiation in my (now grossly outdated) STL Newbie guide (URL below). This is the only portable (ie., something that'll work on *all* platforms) way to do templates with gcc, even with the newest and greatest egcs release. 2. Comment out the default_alloc_template definition in alloc.h (or whatever file it's in in $prefix/include/g++) and supply the definition yourself in one of your object files. 3. Did I mention that you should upgrade your compiler ;-) EGCS : http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ STL : http://www.xraylith.wisc.edu/~khan/software/stl/STL.newbie.html Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Tue Dec 9 01:59:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Tue, 09 Dec 1997 01:59:00 -0000 Subject: More on getting coolview-bash to work with emacs Message-ID: <01BD04A2.3B85F610@gater.krystalbank.msk.ru> Earnie Boyd wrote: > I believe that Sergey's newest versions of bash forces binary mode > always. Is this correct Sergey? Or is it controlled from cygwin.dll? > Depending on Sergey's answer you could retrieve the sources for bash and > make them yourself checking the line received for the \r\n combination. Cygwin.dll forces stdin/stdout/stderr to be opened in a binary mode when connected to a non-tty devices. NTemacs appends \r to every line sent to the shell process. We have to found a way to force emacs to not append \r. -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From skinner@drea.dnd.ca Tue Dec 9 02:45:00 1997 From: skinner@drea.dnd.ca (Bruce S. Skinner) Date: Tue, 09 Dec 1997 02:45:00 -0000 Subject: Trouble building ncurses 1.9.9g. Message-ID: <3.0.32.19971209064824.0094f930@firm.drea.dnd.ca> Hi, At 06:48 PM 12/8/97 PST, you wrote: >configure --enable-symlinks I don't think so but I'll try it again. I built it once before with --enable-symlinks and it still didn't work. I'll try it again though. BTW Thank You for the install script you offered up earlier on the group to get around the lack of a working --program-suffix. regards :-) Bruce S. Bruce S. Skinner | skinner@drea.dnd.ca DEFENCE RESEARCH ESTABLISHMENT ATLANTIC | norstead@ns.sympatico.ca Dartmouth, Nova Scotia, CANADA, B2Y 3Z7 | (902) 426-3100 x205 or 538-1765 "New opinions are always suspected, and usually opposed without any other reason but because they are not already common." John Locke - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From batanov@ascinc.com Tue Dec 9 05:36:00 1997 From: batanov@ascinc.com (Alexander Batanov) Date: Tue, 09 Dec 1997 05:36:00 -0000 Subject: problem with export in GNU make v3.75 Message-ID: <01BD047D.690B8E30@banach.ascinc.com> Dear all, I have a problem with exporting variables in makefile using GNU make v3.75 distributed in Cygnus's gnu-win32-b18 package. It simply does not work. Here is an example with the top-level makefile calling make on subordinate makefile: file: Makefile: ------------------- MYVAR=alex export MYVAR trg1: echo MAKELEVEL is $(MAKELEVEL) echo MYVAR is $(MYVAR) $(MAKE) -f Makefile.sub echo MAKELEVEL is $(MAKELEVEL) echo MYVAR is $(MYVAR) file: Makefile.sub: ------------------------- trg1: echo MAKELEVEL is $(MAKELEVEL) echo MYVAR is $(MYVAR) output of make -f Makefile: -------------------------------------- bash$ make echo MAKELEVEL is 0 MAKELEVEL is 0 echo MYVAR is alex MYVAR is alex make -f Makefile.sub echo MAKELEVEL is 0 MAKELEVEL is 0 echo MYVAR is MYVAR is echo MAKELEVEL is 0 MAKELEVEL is 0 echo MYVAR is alex MYVAR is alex Note that MAKELEVEL environment variable does not work as advertised either. I tried running GNU make v3.73 obtained from elsewhere on the net, and it worked fine! Any ideas how to overcome this problem? Thanks, Alex Batanov Software Engineer Alternative System Concepts, Inc. 22 Haverhill Rd Windham, NH 03087-0128 (603) 437-2234 (603) 437-2722 fax batanov@ascinc.com P.S. Please cc your responses to my email address. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From arlindo@niteroi.gsfc.nasa.gov Tue Dec 9 06:12:00 1997 From: arlindo@niteroi.gsfc.nasa.gov (Arlindo da Silva) Date: Tue, 09 Dec 1997 06:12:00 -0000 Subject: More on getting coolview-bash to work with emacs References: <01BD04A2.3B85F610@gater.krystalbank.msk.ru> Message-ID: <199712091414.JAA06709@woodmore.gsfc.nasa.gov> Sergey, > > Cygwin.dll forces stdin/stdout/stderr to be opened in a binary mode when > connected to a non-tty devices. Is there a deeper reason why you had to enforce binary mode in such cases? Would it make sense (is it possible) to make this feature user configurable? This binary mode thing and bash has caused many people a lot of grief (and e-mail traffic). Just a thought. -- Arlindo da Silva dasilva@alum.mit.edu - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Tue Dec 9 06:41:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Tue, 09 Dec 1997 06:41:00 -0000 Subject: More on getting coolview-bash to work with emacs Message-ID: <01BD04C9.889A67C0@gater.krystalbank.msk.ru> Arlindo da Silva wrote: > Sergey, > > > > > Cygwin.dll forces stdin/stdout/stderr to be opened in a binary mode when > > connected to a non-tty devices. > > Is there a deeper reason why you had to enforce binary mode in such cases? In order to things like cat file | tar xvf - or something similar works at least when started from os prompt. > Would it make sense (is it possible) to make this feature user configurable? > This binary mode thing and bash has caused many people a lot of grief (and > e-mail traffic). Just a thought. I do not understand the needs of text mode at all. Neither NT posix subsystem, nor OpenNT, nor UWIN doesn't have text mode as default. -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ron@mossbayeng.com Tue Dec 9 07:00:00 1997 From: ron@mossbayeng.com (Ron Aaron) Date: Tue, 09 Dec 1997 07:00:00 -0000 Subject: Has anyone got id-utils (mkid, lid) to compile and run? References: <348C4EAB.2F1CF0FB@lynx.com> Message-ID: <199712091441.GAA04225@linus.mbe> Hi all, I recently installed b18 so I could use Win32 versions of my favorite *nix tools. Among them, id-utils. I have been using the DGJPP port, which works - but the DOS extender causes me some problems, plus I am sure it is slower than a native port. Anyway, I untarred id-utils, went to the src directory, did make - everything went fine (I was amazed!). When I try to run the cygwin version of mkid, I get the following messages: mkid: can't lstat `D' from `' mkid: can't chdir to `D' from `/': No such file or directory mkid: can't chdir to `D/vim' from `/': No such file or directory mkid: can't chdir to `D/vim/src' from `/': No such file or directory mkid: can't read entire language map file `/usr/local/share/id-lang.map' Ignoring the 'map' message, which I know how to fix - it looks like somehow the file name is getting parsed incorrectly (the dir is: d:/vim/src) Any ideas, pointers or help would be very much appreciated! Thanks, Ron - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Nick.Ing-Simmons@tiuk.ti.com Tue Dec 9 07:00:00 1997 From: Nick.Ing-Simmons@tiuk.ti.com (Nick.Ing-Simmons@tiuk.ti.com) Date: Tue, 09 Dec 1997 07:00:00 -0000 Subject: ld / dlltool Message-ID: <199712091434.OAA16836@pluto.tiuk.ti.com> I have successfully built perl5.004_55+ using the 'native' Mingw32 gcc/binutils port from: http://agnes.dida.physik.uni-essen.de/~janjaap/mingw32/ It was fairly easy except for the business with dlltool. To build a loadable dll for a perl extension takes two runs of dlltool and two of ld. To build the main perl.dll also uses two passes by generating the import library on one of the other two. This is clumsy at best, the import library generation in particular takes a very long time for the large number of symbols perl exports. (Just about doubles the build time) - and they will nearly always be used together so separate .o file for each symbol seems excessive. Why cannot the linker do the rellocation stuff itself, why does it need this bag-on-the-side? If linker had to be modified to accept and output --dll and --base-file why could it not be taught to write the .reloc section? The other stuff (building the source of export module from .def file, and invoking the assembler) could easily be done with a perl script. (Indeed you could also do the --base-file -> reloc source in perl too if it really must be done that way.) For some reason dlltool did not like the .def file used for VC++/Borland but it was not clear from the 'syntax error line 1' type message what was wrong. I would be willing to assist in/perform or make the effort to clean this up, but would like to know why it was done this way rather than as part of the bfd port for the win32 object format. -- Nick Ing-Simmons - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From JVAPNE.FR.ORACLE.COM@fr.oracle.com Tue Dec 9 09:37:00 1997 From: JVAPNE.FR.ORACLE.COM@fr.oracle.com (Julian VAPNE) Date: Tue, 09 Dec 1997 09:37:00 -0000 Subject: Q: Install gnu-win32/b18 with AZERTY keyboard , pipe lost Message-ID: <199712091738.RAA05368@frseq8.fr.oracle.com> Bonjour, Hello, Could you please help me to configure gnuwin32/b18 with AZERTY keyboard. I am only using bash (no development). I Have a Dell PC with NT4 US, Pack 3 applied and standard French AZERTY keyboard. My French keyboard works fine in cmd.exe window ( pipe symbol etc.) I installed win32/b18 without any problem. If I configure (in Control Panel) US keyboard the QWERTY mapping is OK, but my keyboard is AZERTY! When I switch to AZERTY (in Control Panel) I can not obtain "|" (pipe) from first keyboard row but mapping of other rows is OK. Could you please help me. Does it work with other non US locales ? Thank you in advance. Julian Vapne France - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Tue Dec 9 10:20:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Tue, 09 Dec 1997 10:20:00 -0000 Subject: ld / dlltool Message-ID: <199712091820.SAA109402@out2.ibm.net> >> This is clumsy at best, the import library generation in particular takes a very long time for the large number of symbols perl exports. (Just about doubles the build time) - and they will nearly always be used together so separate .o file for each symbol seems excessive. Why cannot the linker do the rellocation stuff itself, why does it need this bag-on-the-side? If linker had to be modified to accept and output --dll and --base-file why could it not be taught to write the .reloc section? << The ming version of cygwin lets you get a dll directly from the make, using `-dll' as the final gcc option. There still is the problem that import libraries generated by dlltool don't always work, as, for example, those built with wsock32.dll. If you use the impdef.exe program on one of the ming pages for wsock32.dll, you lose all the ordinals associated with each exported function, and the link process lists the wsock32.dll functions you are linking to with _incorrect_ import ordinals associated to them. If you use the cygwin libwsock32.a dll, the link goes to completion without the error messages, but the resulting program doesn't work. Somewhere, dlltool and/or ld are getting the import ordinals mixed up. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From banders@ECD.Rockwell.COM Tue Dec 9 10:20:00 1997 From: banders@ECD.Rockwell.COM (Bartlee A. Anderson) Date: Tue, 09 Dec 1997 10:20:00 -0000 Subject: Did _you_ ever try to rebuild coolview sources? References: <34884194.6765@switch.rockwell.com> Message-ID: <348D894E.4A1E@switch.rockwell.com> Bartlee A. Anderson wrote: > > I would like to find out who has tried to get Sergey's coolview sources > rebuilt(libcygwin.a cygwin.dll)? > > What Platform? Were you successful? > > After getting the results, I'll compile a list and post. I'd really > like to be able to rebuild on 95, but am afraid there are some tricks > to get around 95's limitations. And I'll need to know from those who > have gone before, what are the tight spots yet. 88 Well I've received 1 reply so far. Results for building coolview sources (libcygwin.a and cygwin.dll) ------------------------------------------------------------------- Tester NT Cross/Linux Cross/Other W95 Dean Koester N Y N Bart Anderson N Sergey Okhapkin Y* Y? Y? ------------------------------------------------------------------- N = attempted with no success Y = succeeded * = assumed (yeah, yeah... I know ;) ? = guessed Anyone like to add to the list? - Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) Rockwell International Electronic Commerce Division 300 Bauman Ct. banders@ecd.rockwell.com MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 Wood Dale, IL 60191 FAX (630) 227-9771 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From JONCKHEERE@d0tng.fnal.gov Tue Dec 9 10:20:00 1997 From: JONCKHEERE@d0tng.fnal.gov (JONCKHEERE@d0tng.fnal.gov) Date: Tue, 09 Dec 1997 10:20:00 -0000 Subject: problem with export in GNU make v3.75 Message-ID: <971209114338.268013db@d0tng.fnal.gov> Get Sergey's latest coolview binaries from http://www.lexa.ru/sos/ . unzip and untar. Copy cygwin.dll to *both* places that it exists in your cygwin directory tree. That has fixed this problem for us. Alan ========================================================================== Dear all, I have a problem with exporting variables in makefile using GNU make v3.75 distributed in Cygnus's gnu-win32-b18 package. It simply does not work. Here is an example with the top-level makefile calling make on subordinate makefile: file: Makefile: ------------------- MYVAR=alex export MYVAR trg1: echo MAKELEVEL is $(MAKELEVEL) echo MYVAR is $(MYVAR) $(MAKE) -f Makefile.sub echo MAKELEVEL is $(MAKELEVEL) echo MYVAR is $(MYVAR) file: Makefile.sub: ------------------------- trg1: echo MAKELEVEL is $(MAKELEVEL) echo MYVAR is $(MYVAR) output of make -f Makefile: -------------------------------------- bash$ make echo MAKELEVEL is 0 MAKELEVEL is 0 echo MYVAR is alex MYVAR is alex make -f Makefile.sub echo MAKELEVEL is 0 MAKELEVEL is 0 echo MYVAR is MYVAR is echo MAKELEVEL is 0 MAKELEVEL is 0 echo MYVAR is alex MYVAR is alex Note that MAKELEVEL environment variable does not work as advertised either. I tried running GNU make v3.73 obtained from elsewhere on the net, and it worked fine! Any ideas how to overcome this problem? Thanks, Alex Batanov Software Engineer Alternative System Concepts, Inc. 22 Haverhill Rd Windham, NH 03087-0128 (603) 437-2234 (603) 437-2722 fax batanov@ascinc.com P.S. Please cc your responses to my email address. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From pcanal@fnal.gov Tue Dec 9 12:11:00 1997 From: pcanal@fnal.gov (Canal, Philippe) Date: Tue, 09 Dec 1997 12:11:00 -0000 Subject: Bizarre failure. Message-ID: <01BD04AB.D63D6B30.pcanal@popgtw.fnal.gov> Hi, The following command issued from a bash prompt (cygwin b18): echo statusrel -p | grep "^stuff " | cut -f2 -d ' ' fails in average one of out 10 times with the following messages: in a window: procedure entry point MSWSock.rcmd could not be located in the dynamic library wsock32.dll in the bash window: fork_helper: child died before initialization with win32 error 128 Any ideas ? Thanks, Philippe Canal, Fermilab, pcanal@fnal.gov - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Tue Dec 9 13:14:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Tue, 09 Dec 1997 13:14:00 -0000 Subject: date in a backgrounf task ? Message-ID: <6jWduzp$pfB@mike.franken.de> Hi *, I'm trying to use "date" in a batch file, which is called by an NT service (Legato Networker save). Whenever the date command gets started, it sits there doing nothing, leaving the batch file waiting forever. Only way to get the batch running further is to kill date.exe from taskmanager. Any idea, what might cause this behaviour ? If I start this batch from the command line, everything work as expected. This is NT4 Server SP3 German with Cygnus' cdk B18 and latest coolview. Thx and bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Tue Dec 9 13:14:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Tue, 09 Dec 1997 13:14:00 -0000 Subject: Compiling nvi References: <9712081733.AA04166@motprux.pr.paging.mot.com> Message-ID: <6jWdubs$pfB@mike.franken.de> Hi Diego, [...] > nvi: (null) > > Argh! What could be this??? Thanks in advance. Yes, delete the source tree for nvi, remount (at least) the partition you are compiling on as binary, unpack the source package for nvi again and recompile it. You should think about switching to binary mounts at all. > >Diego Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jazz@opennt.com Tue Dec 9 13:14:00 1997 From: jazz@opennt.com (Jason Zions) Date: Tue, 09 Dec 1997 13:14:00 -0000 Subject: More on getting coolview-bash to work with emacs References: <01BD04C9.889A67C0@gater.krystalbank.msk.ru> Message-ID: <348DB4BE.69EC48C8@opennt.com> > I do not understand the needs of text mode at all. Neither NT posix > subsystem, nor OpenNT, nor UWIN doesn't have text mode as default. If I understand the arguments, Sergey, the basic difference is that the goals of the NT Posix subsystem, OpenNT, and UWIN are different from the goals of cygwin. The first three are aimed are providing (forgive the simplification and trademark abuse) UNIX on top of NT. Users live in a Unix world, where there is only one line-ending and \n is its name. Cygwin has multiple goals; while "Unix on NT" is one of them, it appears to be of less importance than the others: a free development environment for Win32, a way of getting more tools more rapidly into the Win32 world, etc. In meeting those goals, supporting the Win32 line-ending of \r\n is relatively important. Folks living in the Win32 world don't want to give up Win32 tools that operate on text files; because of that, they need cygwin to deal properly with \r\n-style text files. I'd like to contribute to a solution for living in a mixed line-ending world, but I haven't found a coherent way of solving the problem short of living in one world or the other and abandoning attempts to handle both line-endings simultaneously. That's not a helpful position for those who have a goal other than "Unix on NT", so I say nothing. Jason Zions Softway Systems Inc. Makers of OpenNT, POSIX and more for Windows NT - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From banders@ECD.Rockwell.COM Tue Dec 9 13:36:00 1997 From: banders@ECD.Rockwell.COM (Bartlee A. Anderson) Date: Tue, 09 Dec 1997 13:36:00 -0000 Subject: Did _you_ ever try to rebuild coolview sources? References: <34884194.6765@switch.rockwell.com> <348D894E.4A1E@switch.rockwell.com> Message-ID: <348DB1F9.117C@switch.rockwell.com> Bartlee A. Anderson wrote: > > Bartlee A. Anderson wrote: > > > > I would like to find out who has tried to get Sergey's coolview sources > > rebuilt(libcygwin.a cygwin.dll)? > > > > What Platform? Were you successful? > > > > After getting the results, I'll compile a list and post. I'd really > > like to be able to rebuild on 95, but am afraid there are some tricks > > to get around 95's limitations. And I'll need to know from those who > > have gone before, what are the tight spots yet. > > 88 > > Well I've received 1 reply so far. > > Results for building coolview sources (libcygwin.a and cygwin.dll) > > ------------------------------------------------------------------- > Tester NT Cross/Linux Cross/Other W95 > > Dean Koester N Y N My apologies to Ron and Dean (my fault for reading mail while I write it) The above line should read Ron Parker N Y N > Bart Anderson N > Sergey Okhapkin Y* Y? Y? > ------------------------------------------------------------------- > > N = attempted with no success > Y = succeeded > * = assumed (yeah, yeah... I know ;) > ? = guessed > > Anyone like to add to the list? > - > Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) > Rockwell International Electronic Commerce Division > 300 Bauman Ct. banders@ecd.rockwell.com > MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 > Wood Dale, IL 60191 FAX (630) 227-9771 -- Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) Rockwell International Electronic Commerce Division 300 Bauman Ct. banders@ecd.rockwell.com MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 Wood Dale, IL 60191 FAX (630) 227-9771 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ian@cygnus.com Tue Dec 9 13:36:00 1997 From: ian@cygnus.com (ian@cygnus.com) Date: Tue, 09 Dec 1997 13:36:00 -0000 Subject: ld / dlltool References: <199712091434.OAA16836@pluto.tiuk.ti.com> Message-ID: <199712092128.QAA23107@tweedledumb.cygnus.com> In gnu-win32 Nick.Ing-Simmons@tiuk.ti.com writes: >Why cannot the linker do the rellocation stuff itself, why does it need >this bag-on-the-side? If linker had to be modified to accept and output >--dll and --base-file why could it not be taught to write the .reloc section? It can. The ELF shared library support essentially builds all the relocation information inside the linker, and the Win32 support could do the same. The AIX support can even read import and export files when building a shared library, much the way dlltool does. >I would be willing to assist in/perform or make the effort to clean this up, >but would like to know why it was done this way rather than as part of the >bfd port for the win32 object format. I would guess because Steve Chamberlain found it easier to write a separate program than to figure out how the other shared library support worked. No deep reason. It would be great if you, or anybody, volunteered to clean this up. It's not a simple job, though. I think the way to start is to understand how the ELF shared library support works. Ian - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Tue Dec 9 13:36:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Tue, 09 Dec 1997 13:36:00 -0000 Subject: Trouble building ncurses 1.9.9g. References: <3.0.32.19971208101844.009229f0@firm.drea.dnd.ca> Message-ID: <6jWduUz4pfB@mike.franken.de> Hi Bruce, >I have been trying to build ncurses_1.9.9g under cygwin32 b18.5 & NT4.0. >It appears to build O.K. but when I try and run any of the test programs >I get: Error opening terminal: linux. This happens, because the termlib database hasn't been built correctly, as your messages below indicate. [...] > sh ./run_tic.sh /usr/local/bin . /usr/local/share/terminfo > ** Building terminfo database, please wait... > ** adjusting tabset paths > "/tmp/7863", line 268, terminal 'unknown': resolution of use=dumb failed [...] > >Please forgive me if these errors should blatantly tell me what is amiss. >I'm not a unix person or Winnt person. I have searched the mailing list >archives and haven't gotten any clues from there. Any comments? Yes, delete the source tree for curses, remount (at least) the partition you are compiling on as binary, unpack the source package for curses again and recompile it. > >regards :-) >Bruce S. Skinner Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From diegom@pts.mot.com Tue Dec 9 14:10:00 1997 From: diegom@pts.mot.com (Diego Montanez) Date: Tue, 09 Dec 1997 14:10:00 -0000 Subject: nvi again Message-ID: <9712091759.AA19524@motprux.pr.paging.mot.com> Today I tried to compile nvi but this time with the latest coolview patches. Configure worked ok, compiling and installation were ok and alas, when I run vi I get vi: error 0 I'm using binary mounts... Argh! Diego --- D I E G O A M O N T A N E Z - D I E G O M @ PTS . M O T . C O M --------------------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tceraj@public.srce.hr Tue Dec 9 14:51:00 1997 From: tceraj@public.srce.hr (Ivica Ceraj) Date: Tue, 09 Dec 1997 14:51:00 -0000 Subject: how to import dll ? Message-ID: <348DCB42.FFD35861@public.srce.hr> I have one question I have not found in FAQs, and may be interesting and for other people, is there any way to import dll, and build library wrapping it ? [like implib.exe in Borland C] Thanks... Ivan -- @ @ iceraj@geocities.com I may know the word @ iceraj@student.phy.hr but not say it @ iceraj@hotmail.com I may know the truth @ but not face it @ http://student.phy.hr/~iceraj N.Merchant @ http://www.geocities.com/SiliconValley/Vista/8042 @ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dahms@ifk20.mach.uni-karlsruhe.de Tue Dec 9 16:05:00 1997 From: dahms@ifk20.mach.uni-karlsruhe.de (dahms@ifk20.mach.uni-karlsruhe.de) Date: Tue, 09 Dec 1997 16:05:00 -0000 Subject: date in a backgrounf task ? Message-ID: <009BE892.85F755C0.29929@ifk20.mach.uni-karlsruhe.de> Hi Michael, you wrote: : I'm trying to use "date" in a batch file, which is called by an NT service : (Legato Networker save). Whenever the date command gets started, it sits : there doing nothing, leaving the batch file waiting forever. Only way to : get the batch running further is to kill date.exe from taskmanager. Maybe you have different PATH settings and got the "DOS" version first? For that, you need something like: date Hi, Josh, Welcome to the world of "C". Below you will find a list of links I send to newbies. I can suggest you take a look at the O'reilly & Associates publisher for a book on Bash. There are files in the info directory for help on bash but you really need an info reader to view them correctly. I'm not sure but you may find help at the last link below. Goodluck. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Mon, 08 Dec 1997 21:52:31 -0600 >To: gnu-win32@cygnus.com >From: shorti >Subject: (newbie) please be kind... > >hello all: > >ok - here's the deal: i have very limited experience with c, although i'm >figuring out the text-driven stuff fairly well. i would like to write >windows programs using my new found resources, but i know next to nothing >about API, integrating dll's, etc. does anyone know of any good >tutorials/references on win32 (or c in general) on the web? maybe if you >have nothing better to do over christmas holiday, you could write one? or >is it just easier for me to go out and buy a book? If you have a list of >links related to (c, win32) programming, it would be greatly appreciated.. >i'll compile it to a webpage somewhere... > >Also - could someone point me to the things that make bash useful? i'm not >sure totally what to do.. > >right now i'm questioning the wisdom of the person that said "ignorance is >bliss"... > >thanx for your time and patience, >- josh >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From colin@bird.fu.is.saga-u.ac.jp Tue Dec 9 17:24:00 1997 From: colin@bird.fu.is.saga-u.ac.jp (Colin Peters) Date: Tue, 09 Dec 1997 17:24:00 -0000 Subject: how to import dll ? Message-ID: <01bd0509$d08eb5a0$fa173185@gbird0.fu.is.saga-u.ac.jp> From: Ivica Ceraj Date: Wednesday, December 10, 1997 8:44 AM >I have one question I have not found in FAQs, and may be interesting and >for other people, is there any way to import dll, and build library >wrapping it ? >[like implib.exe in Borland C] The program impdef.exe from my homepage (add /gcc-extra.html to either URL in my signature and scroll down near the bottom of the page) can be used to generate a .def file from a DLL. However: It does not work with PASCAL or STDCALL functions, because there is no way to determine the correct @xx value to attach to the end of a function name; and it doesn't seem to work quite right on some DLLs produced with newer versions of MSVC. Do this: impdef foo.dll >foo.def dlltool --def foo.def --output-lib libfoo.a Then link your program which uses functions in foo.dll with the library libfoo.a, e.g.: gcc -o bar.exe bar.o -lfoo Hope this helps, Colin. P.S. Before you (or anyone else) asks, trying to use C++ DLLs from Borland or MSVC with GCC compiled code is unlikely to work for lots of good reasons. Stick to C DLLs or C++ done with the same compiler (and even that can be pretty tricky). P.P.S. Functions with STDCALL or PASCAL calling convention need the number of bytes in their argument list added as @xx (where xx is the number of bytes) to the function name in the .def file (unless the DLL exports the name with @xx attached, which does happen sometimes), and you need to add the -k option to the dlltool command line above. -- Colin Peters -- colin at fu.is.saga-u.ac.jp -- Saga University Dept. of Information Science -- http://www.fu.is.saga-u.ac.jp/~colin -- http://www.geocities.com/Tokyo/Towers/6162 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Tue Dec 9 17:24:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Tue, 09 Dec 1997 17:24:00 -0000 Subject: More on getting coolview-bash to work with emacs References: <01BD04C9.889A67C0@gater.krystalbank.msk.ru> <348DB4BE.69EC48C8@opennt.com> Message-ID: <19971210113649.60115@mundook.cs.mu.OZ.AU> On 09-Dec-1997, Jason Zions wrote: > > I do not understand the needs of text mode at all. Neither NT posix > > subsystem, nor OpenNT, nor UWIN doesn't have text mode as default. > > If I understand the arguments, Sergey, the basic difference is that the > goals of the NT Posix subsystem, OpenNT, and UWIN are different from the > goals of cygwin. > > The first three are aimed are providing (forgive the simplification and > trademark abuse) UNIX on top of NT. Users live in a Unix world, where > there is only one line-ending and \n is its name. > > Cygwin has multiple goals; while "Unix on NT" is one of them, it appears > to be of less importance than the others: a free development environment > for Win32, a way of getting more tools more rapidly into the Win32 > world, etc. In meeting those goals, supporting the Win32 line-ending of > \r\n is relatively important. Folks living in the Win32 world don't want > to give up Win32 tools that operate on text files; because of that, they > need cygwin to deal properly with \r\n-style text files. Yes, absolutely. To give a concrete example: my goal is to provide a port of the Mercury compiler and development tools to Win32. GNU-win32 is just a means to that end. Ideally the fact that the Mercury system was originally written on Unix (making use of Bourne shell scripts, etc.) should be entirely transparent to the Win32 users of Mercury. They should be able to use their favourite Win32 text editor, which no doubt puts \r\n into files, without needing to do anything special. I don't mind having to use "rb" or "wb" when opening binary files. Actually our code already does that, but the point is that I'm willing to invest some additional developer effort solving text-vs-binary hassles in order to spare our users from win32-vs-unix hassles. To achieve my goal, I need to use text-mode mounts. Thus for my purposes, if bash stops working on text-mode mounts, then this is a big step backwards... -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From webhead@web-feats.com Tue Dec 9 22:08:00 1997 From: webhead@web-feats.com (Rich Dorfman) Date: Tue, 09 Dec 1997 22:08:00 -0000 Subject: Gcc compiler b18 newbie probs Message-ID: <3.0.32.19971210000704.0069ca10@globaldialog.com> Just installed Gnu-Win32 b18 on Win95 P100 platform. Paths and environment variables successfully added to autoexec.bat. Tried to compile the following: #include int main(void) { cout << "Come up and C++ me some time."; cout << "\n"; return 0; } Got the following errors: undefined reference to `cout' undefined reference to `ostream::operator<<(char const *)' I presume that iostream.h is not being found. The file resides at: C:\gnuwin32\b18\include\g++\iostream.h I'm using the dos command line. I've tried the -I switch after my compile statement. What is the proper switch? What is the proper syntax? Back slashes or forward slashes in the path? Should I be using BASH shell prompt? Advantages/disadvantages? Thanks list for your help. _____________________________________________________________ \ | / |== Rich Dorfman \|/ eb |== (414) 569-1228 | eats webhead@web-feats.com * Web page design, Internet marketing & instruction * Total connectivity solutions * Firewalls, content filtering & network security --->> www.web-feats.com <<--- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From colin@bird.fu.is.saga-u.ac.jp Tue Dec 9 23:12:00 1997 From: colin@bird.fu.is.saga-u.ac.jp (Colin Peters) Date: Tue, 09 Dec 1997 23:12:00 -0000 Subject: Gcc compiler b18 newbie probs Message-ID: <01bd0539$fbf8c020$fa173185@gbird0.fu.is.saga-u.ac.jp> From: Rich Dorfman To: gnu-win32@cygnus.com Date: Wednesday, December 10, 1997 3:43 PM >Just installed Gnu-Win32 b18 on Win95 P100 platform. Paths and environment >variables successfully added to autoexec.bat. > >Tried to compile the following: > >#include >int main(void) >{ > cout << "Come up and C++ me some time."; > cout << "\n"; > return 0; >} > >Got the following errors: > >undefined reference to `cout' >undefined reference to `ostream::operator<<(char const *)' > >I presume that iostream.h is not being found. The file resides at: > C:\gnuwin32\b18\include\g++\iostream.h The problem is not iostream.h not being found. Note that the linker (not the compiler) is complaining, and it knows about ostream, which it wouldn't if it wasn't reading the headers. The problem is that the iostream library is not being linked. Try adding -lstdc++ on your link line (or using g++ instead of gcc to do your linking). Good luck, Colin. -- Colin Peters -- colin at fu.is.saga-u.ac.jp -- Saga University Dept. of Information Science -- http://www.fu.is.saga-u.ac.jp/~colin -- http://www.geocities.com/Tokyo/Towers/6162 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Tue Dec 9 23:12:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Tue, 09 Dec 1997 23:12:00 -0000 Subject: More on getting coolview-bash to work with emacs Message-ID: <01BD0554.11F03AB0@gater.krystalbank.msk.ru> Jason Zions wrote: > Cygwin has multiple goals; while "Unix on NT" is one of them, it appears > to be of less importance than the others: a free development environment > for Win32, a way of getting more tools more rapidly into the Win32 > world, etc. In meeting those goals, supporting the Win32 line-ending of > \r\n is relatively important. Folks living in the Win32 world don't want > to give up Win32 tools that operate on text files; because of that, they > need cygwin to deal properly with \r\n-style text files. I live in Win32 world too, but why I never needed \r in line separators?-) The file management/editing tools I use works with bot types of line separators, moreover, the editor preserves original line separators on file saving. The only file I have to have with CRLF is NT's boot.ini - ntldr just traps if the file contains single \n as line separators :-) -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Tue Dec 9 23:49:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Tue, 09 Dec 1997 23:49:00 -0000 Subject: Did _you_ ever try to rebuild coolview sources? Message-ID: <01BD0559.68A2A640@gater.krystalbank.msk.ru> Sergey Okhapkin wrote: > Y Y N (configure runs but make fails, > probably environment size problem?) > BTW, any dll compiled with cygnus tools on W95 will not work in most cases because of well-known "garbage filling on seek past end of file" problem. -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Tue Dec 9 23:49:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Tue, 09 Dec 1997 23:49:00 -0000 Subject: Did _you_ ever try to rebuild coolview sources? Message-ID: <01BD0558.9427A910@gater.krystalbank.msk.ru> Bartlee A. Anderson wrote: > Results for building coolview sources (libcygwin.a and cygwin.dll) > > ------------------------------------------------------------------- > Tester NT Cross/Linux Cross/Other W95 > > Dean Koester N Y N > Bart Anderson N > Sergey Okhapkin Y* Y? Y? > ------------------------------------------------------------------- > Y Y N (configure runs but make fails, probably environment size problem?) > N = attempted with no success > Y = succeeded > * = assumed (yeah, yeah... I know ;) > ? = guessed -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From guiboud@netmansys.fr Wed Dec 10 00:19:00 1997 From: guiboud@netmansys.fr (Stéphane Guiboud-Ribaud) Date: Wed, 10 Dec 1997 00:19:00 -0000 Subject: select function Message-ID: <348E504C.A47C9A5@netmansys.fr> Regarding the Win32 API of the select function, only SOCKET objects could be used within fd_set structure. On UNIX, you can use file descriptor and socket descriptor. * Is the select function in cygnus package provide the same functionality ? * If not, is there function(s) of Win32 API or inside cygnus package to check whether a file handle (as a HANDLE object) is ready to read or to write ? * Do a mailing list exist for the Microsoft Visual C++ compiler ? Stephane Guiboud-Ribaud???????????????????????????????????????????????????????????????????????????????????????????????? NETMANSYS S.A. ???????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????? http://www.netmansys.fr/ mailto:guiboud@netmansys.fr ???????????????????????????????????? http://www.netmansys.com/ ???? From JoopvandeWege@mail.mococo.nl Wed Dec 10 00:22:00 1997 From: JoopvandeWege@mail.mococo.nl (Joop van de Wege) Date: Wed, 10 Dec 1997 00:22:00 -0000 Subject: date in a backgrounf task ? References: <6jWduzp$pfB@mike.franken.de> Message-ID: On 09 Dec 1997 20:02:00 +0100 mh@mike.franken.de (Michael Hirmke) wrote: > Hi *, > > I'm trying to use "date" in a batch file, which is called by an NT service > (Legato Networker save). Whenever the date command gets started, it sits > there doing nothing, leaving the batch file waiting forever. Only way to > get the batch running further is to kill date.exe from taskmanager. > > Any idea, what might cause this behaviour ? > If I start this batch from the command line, everything work as > expected. Looks like a rights problem to me. Programs started from services use the 'SYSTEM' account which doesn't have much rights. Looks like you need to create a special account and tell the service to use it. Open control-panel/services, find your service, select it, choose 'Startup' and change 'Logon as'. Another problem you could have is that 'date' is waiting for you to press enter, meaning your redirection is somehow wrong. Joop ---- Joop van de Wege Mobile Computing Consultants +31 (0) 318 553292 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Axel.Riese@risc.uni-linz.ac.at Wed Dec 10 02:20:00 1997 From: Axel.Riese@risc.uni-linz.ac.at (Axel Riese) Date: Wed, 10 Dec 1997 02:20:00 -0000 Subject: More on getting coolview-bash to work with emacs References: <19971210113649.60115@mundook.cs.mu.OZ.AU> Message-ID: <199712101020.AA26190@melmac.risc.uni-linz.ac.at> Hi all ! I strongly support Jason's and Fergus' remarks concerning binary/text mounts: > > Cygwin has multiple goals; while "Unix on NT" is one of them, it appears > > to be of less importance than the others: a free development environment > > for Win32, a way of getting more tools more rapidly into the Win32 > > world, etc. In meeting those goals, supporting the Win32 line-ending of > > \r\n is relatively important. Folks living in the Win32 world don't want > > to give up Win32 tools that operate on text files; because of that, they > > need cygwin to deal properly with \r\n-style text files. I'm one of these guys who uses cygwin mainly for writing Win32 applications. Using UNIX \r line-endings is rather unnatural for this purpose. > To achieve my goal, I need to use text-mode mounts. > Thus for my purposes, if bash stops working on text-mode mounts, > then this is a big step backwards... Actually, Sergey's work is a big step *forwards*, but I would appreciate support for both kinds of line-endings very much. BW, Axel =========================================================================== Axel Riese Research Institute for Symbolic Computation J. Kepler University Linz A-4040 Linz e-Mail: Axel.Riese@risc.uni-linz.ac.at Austria URL: http://www.risc.uni-linz.ac.at/people/ariese/home/ =========================================================================== - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 10 04:49:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 10 Dec 1997 04:49:00 -0000 Subject: how to import dll ? (NBY) Message-ID: <19971210124842.3974.qmail@hotmail.com> Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Tue, 09 Dec 1997 23:50:43 +0100 >From: Ivica Ceraj >Reply-To: iceraj@student.phy.hr >To: gnu-win32@cygnus.com >Subject: how to import dll ? > >I have one question I have not found in FAQs, and may be interesting and >for other people, is there any way to import dll, and build library >wrapping it ? >[like implib.exe in Borland C] > >Thanks... > >Ivan > >-- >@ >@ iceraj@geocities.com I may know the word >@ iceraj@student.phy.hr but not say it >@ iceraj@hotmail.com I may know the truth >@ but not face it >@ http://student.phy.hr/~iceraj N.Merchant >@ http://www.geocities.com/SiliconValley/Vista/8042 >@ >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 10 05:25:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 10 Dec 1997 05:25:00 -0000 Subject: More on getting coolview-bash to work with emacs Message-ID: <19971210132503.2207.qmail@hotmail.com> Here is my 2c worth. Modify bash to always open stdin in binary. Modify bash to default open stdout, stderr in binary and provide settings to modify this. Modify the readline function to replace the \r\n combination with \n. The above modifications will allow bash to operate on both line endings without problems and will provide the means to have a personal preference for the output from bash. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From lane@netti.fi Wed Dec 10 06:20:00 1997 From: lane@netti.fi (Sami Laine) Date: Wed, 10 Dec 1997 06:20:00 -0000 Subject: accept() and other networking functions? Message-ID: <199712101423.QAA27619@sokrates.netti.fi> Hello. I'm having some trouble with accept() under Win95. I use gnu-win32 b18 release from ftp.cygnus.com. Problem I'm having is the fact that accept() fails to function and always returns -1. Same source is working nicely under Debian/GNU Linux and NetBSD-1.1. Are there any special tricks to get accept() working normally? Usual bind(), socket() stuff done before this. --- snip --- while(1) { i = sizeof(sin); cli_sock = accept(serv_sock, (struct sockaddr *) &sin, (int *)&i); if(cli_sock == -1 && errno == EINTR) continue; if(cli_sock == -1) return EXIT_FAILURE; --- snip --- -- Sami Laine - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tomas.fasth@twinspot.net Wed Dec 10 06:20:00 1997 From: tomas.fasth@twinspot.net (Tomas Fasth) Date: Wed, 10 Dec 1997 06:20:00 -0000 Subject: More on getting coolview-bash to work with emacs References: <199712091414.JAA06709@woodmore.gsfc.nasa.gov> Message-ID: <348E9F7D.FDD25961@twinspot.net> Arlindo da Silva wrote: > This binary mode thing and bash has caused many people a lot of grief (and > e-mail traffic). Just a thought. Oh well, talking about grief. Isn't it amazing that three major platforms all use different protocols for text line termination? on MacOS, on Unix, and on DOS. Typical product differentiation in the age of platform warfare? Another classic cause for grief (at least for the non-ascii text communities) was (is?) the 7-bit (ascii) character masking found well into this decade in many (US originated?) software engineerings. Yet another cause for much grief was (is?) 16-bit offset pointers. We programmers are all a bunch of loosers, aren't we? ;-) Tomas - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From duran@vdl2.ca Wed Dec 10 06:20:00 1997 From: duran@vdl2.ca (Duran) Date: Wed, 10 Dec 1997 06:20:00 -0000 Subject: fork() problem with sockets Message-ID: <199712101431.JAA08175@cobra.vdl2.ca> The Fork()???? ???? Problem: The fork child does not inherate the file descriptors when working whith sockets. How to reproduce the problem: Just create a socket server, accept a connection and fork, you will notice that the child will not be able to use the sockets declared before the forking.???? ???? From khan@xraylith.wisc.edu Wed Dec 10 07:40:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Wed, 10 Dec 1997 07:40:00 -0000 Subject: Gcc compiler b18 newbie probs References: <3.0.32.19971210000704.0069ca10@globaldialog.com> Message-ID: <9712101540.AA01573@modi.xraylith.wisc.edu> Rule-of-thumb, especially for new users: - use the right compiler driver to compile/link programs. For C programs, use gcc, for C++ use c++ and for F77 use g77. There are cases where you need a different driver, but then you'd have to know what libraries need to be linked in explicitly and so on. Not worth the hassle for usual projects, but sometimes needed for multi-language projects, especially when mixing F77 and C/C++. Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kh@wg.icl.co.uk Wed Dec 10 08:37:00 1997 From: kh@wg.icl.co.uk (Kevin Hughes) Date: Wed, 10 Dec 1997 08:37:00 -0000 Subject: wind river and cygnus - what is the relationship? Message-ID: <01BD0589.DAD45990@rodney.wg.icl.co.uk> I have been looking at the gnu port to NT for some time and have installed the s/w and am trying hard to use it. I am a user of the Wind River VxWorks hosted on a SPARC/Solaris box. I recently obtained an NT hosted version of the VxWorks and think that there may be some potential incompatibility with the gnu95 system. It appears that there is a cygwin.dll shipped with the VxWorks system - I presume for running the gnu tool chain. It is dated June 96 so I guess it is old BUT it appears to have extra functionality - mmap appears to be there. Any ideas most welcome Kevin - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From banders@ECD.Rockwell.COM Wed Dec 10 10:08:00 1997 From: banders@ECD.Rockwell.COM (Bartlee A. Anderson) Date: Wed, 10 Dec 1997 10:08:00 -0000 Subject: Did _you_ ever try to rebuild coolview sources? References: <01BD0559.68A2A640@gater.krystalbank.msk.ru> Message-ID: <348EDAE3.429E@switch.rockwell.com> Sergey Okhapkin wrote: 8< snip >8 > BTW, any dll compiled with cygnus tools on W95 will not work in most cases > because of well-known "garbage filling on seek past end of file" problem. I thought I saw that there was a couple of proposed fixes for this problem. Is there a possible solution? Will it be in b19? > > -- > Sergey Okhapkin, http://www.lexa.ru/sos > Moscow, Russia > Looking for a job. -- Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) Rockwell International Electronic Commerce Division 300 Bauman Ct. banders@ecd.rockwell.com MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 Wood Dale, IL 60191 FAX (630) 227-9771 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ggp@informix.com Wed Dec 10 11:49:00 1997 From: ggp@informix.com (Guy Gascoigne - Piggford) Date: Wed, 10 Dec 1997 11:49:00 -0000 Subject: More on getting coolview-bash to work with emacs Message-ID: <3.0.3.32.19971210114728.0140c240@pop.pdx.informix.com> At 11:20 AM 12/10/97 +0100, you wrote: >> To achieve my goal, I need to use text-mode mounts. >> Thus for my purposes, if bash stops working on text-mode mounts, >> then this is a big step backwards... > >Actually, Sergey's work is a big step *forwards*, but I would appreciate >support for both kinds of line-endings very much. I have to agree here, I'm not completely convinced by the arguments of 'just use binary mounts', but aside from that, Sergy's work has added a lot of improvements and added functionality. I'd much rather use it than not. Guy - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From banders@ECD.Rockwell.COM Wed Dec 10 12:04:00 1997 From: banders@ECD.Rockwell.COM (Bartlee A. Anderson) Date: Wed, 10 Dec 1997 12:04:00 -0000 Subject: Has anyone got id-utils (mkid, lid) to compile and run? References: <199712101554.HAA07172@linus.mbe> Message-ID: <348EE028.54D5@switch.rockwell.com> I don't see anything obviously wrong. I would suggest re-mounting as binary all filesystems ;) It looks like you can't cd to d from / Try an experiment like I did. cd / cd d doesn't work cd /d that works. may have something to do with separate devices but needing full path to access. I'll forward to the list for further comment. Bart Ron Aaron wrote: > > Bartlee- > > Here is the output from mount: > Device Directory Type Flags > s: /s native no-mixed,text!=binary > \\.\tape1: /dev/st1 native no-mixed,text!=binary > \\.\tape0: /dev/st0 native no-mixed,text!=binary > \\.\b: /dev/fd1 native no-mixed,text!=binary > \\.\a: /dev/fd0 native no-mixed,text!=binary > d: /d native no-mixed,text!=binary > c: / native no-mixed,text!=binary > > > -----Original Message----- > > From: Bartlee A. Anderson [SMTP:banders@ECD.Rockwell.COM] > > Sent: Tuesday, December 09, 1997 12:59 PM > > To: Ron Aaron > > Subject: Re: Has anyone got id-utils (mkid, lid) to compile and run? > > > > > > Ron Aaron wrote: > > > > > > Hi all, > > > > > > I recently installed b18 so I could use Win32 versions of my > > favorite *nix tools. Among them, id-utils. > > > > > > I have been using the DGJPP port, which works - but the DOS > > extender causes me some problems, plus I am sure it is slower than a > > native port. > > > > > > Anyway, I untarred id-utils, went to the src directory, did make - > > everything went fine (I was amazed!). > > > > > > When I try to run the cygwin version of mkid, I get the following messages: > > > mkid: can't lstat `D' from `' > > > mkid: can't chdir to `D' from `/': No such file or directory > > > mkid: can't chdir to `D/vim' from `/': No such file or directory > > > mkid: can't chdir to `D/vim/src' from `/': No such file or directory > > > mkid: can't read entire language map file `/usr/local/share/id-lang.map' > > > > What does your mount output look like? > > > > > > Ignoring the 'map' message, which I know how to fix - it looks like > > somehow the file name is getting parsed incorrectly (the dir is: d:/vim/src) > > > > > > Any ideas, pointers or help would be very much appreciated! > > > > > > Thanks, > > > Ron > > > - > > > For help on using this list (especially unsubscribing), send a message to > > > "gnu-win32-request@cygnus.com" with one line of text: "help". > > > > -- > > Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) > > Rockwell International Electronic Commerce Division > > 300 Bauman Ct. banders@ecd.rockwell.com > > MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 > > Wood Dale, IL 60191 FAX (630) 227-9771 > > > > -- Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) Rockwell International Electronic Commerce Division 300 Bauman Ct. banders@ecd.rockwell.com MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 Wood Dale, IL 60191 FAX (630) 227-9771 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Wed Dec 10 12:04:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Wed, 10 Dec 1997 12:04:00 -0000 Subject: date in a backgrounf task ? References: Message-ID: <6j$mIkBKpfB@mike.franken.de> Hi Joop, [...] >> I'm trying to use "date" in a batch file, which is called by an NT service >> (Legato Networker save). Whenever the date command gets started, it sits >> there doing nothing, leaving the batch file waiting forever. Only way to >> get the batch running further is to kill date.exe from taskmanager. [...] >Looks like a rights problem to me. Programs started from services use >the 'SYSTEM' account which doesn't have much rights. Looks like you need >to create a special account and tell the service to use it. Open >control-panel/services, find your service, select it, choose 'Startup' >and change 'Logon as'. The Networker service, which runs that batch, has its own userid with all necessary permissions. >Another problem you could have is that 'date' is waiting for you to >press enter, meaning your redirection is somehow wrong. It is not the DOS or NT date, I am calling - it is that one from the Cygnus cdk, which does not need any input. And I specified the complete path when calling date.exe - so I don't think this is the problem either. Thx anyway for your answer. > >Joop Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cliff@isis.com Wed Dec 10 12:04:00 1997 From: cliff@isis.com (Cliff Krumvieda) Date: Wed, 10 Dec 1997 12:04:00 -0000 Subject: b18: "Exception trapped" from cat Message-ID: <199712101949.OAA13685@polyphony.sw.stratus.com> Hello: I've received the attached "exception trapped" report and trace from cat.exe several times. I am running beta 18 under NT 4.0. I wish I could produce a repeatable test case, but the error is not deterministic. I'm running a perl script that invokes cat.exe maybe 25 times; I usually get the trace at least once per run, but from different sub-invocations of cat.exe. --Cliff Krumvieda (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) In cygwin_except_handler (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) Exception trapped! (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) exception C0000005 at 10011E58 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) exception: ax 1D bx 1 cx 240F31C dx 0 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) exception: si 0 di 240F280 bp 240F234 sp 240F22C (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) exception is: STATUS_ACCESS_VIOLATION (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) Stack trace: (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 0: sp = 0x240F060, pc = 0x1000CEC2 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 1: sp = 0x240F07C, pc = 0x77F94072 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 2: sp = 0x240F0A0, pc = 0x77F88A53 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 3: sp = 0x240F12C, pc = 0x77F75E82 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 4: sp = 0x240F234, pc = 0x100219C4 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 5: sp = 0x240F250, pc = 0x10031648 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 6: sp = 0x240F260, pc = 0x10031582 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 7: sp = 0x240F2C0, pc = 0x1002E637 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 8: sp = 0x240F2D0, pc = 0x100305B4 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 9: sp = 0x240F2F8, pc = 0x10027891 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 10: sp = 0x240F328, pc = 0x4036CD (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 11: sp = 0x240F3A0, pc = 0x1000C102 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 12: sp = 0x240FF94, pc = 0x1000C113 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 13: sp = 0x240FFA0, pc = 0x4045D5 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 14: sp = 0x240FFB0, pc = 0x40103B (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) frame 15: sp = 0x240FFC0, pc = 0x77F1AFC1 (c:\gnuwin32\b18\h-i386-cygwin32\bin\cat.exe 1000) End of stack trace (more stack frames may be present) TMS: Compare error: code = 0 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fine@mail.cern.ch Wed Dec 10 13:11:00 1997 From: fine@mail.cern.ch (Valery Fine) Date: Wed, 10 Dec 1997 13:11:00 -0000 Subject: date in a backgrounf task ? References: Message-ID: <199712102029.VAA116804@sp049.cern.ch> On 10 Dec 97 at 9:08, Joop van de Wege wrote: > > > On 09 Dec 1997 20:02:00 +0100 > mh@mike.franken.de (Michael Hirmke) wrote: > > > Hi *, > > > > I'm trying to use "date" in a batch file, which is called by an NT service > > (Legato Networker save). Whenever the date command gets started, it sits > > there doing nothing, leaving the batch file waiting forever. Did you try "date /t" ? Cheers, Valery == Dr. Valeri Faine (Valery Fine) ------------ ------------- Phone: +41 22 767 4921 CERN FAX : +41 22 767 7155 CH-1211 Geneva, 23 mailto:fine@mail.cern.ch Switzerland http://nicewww.cern.ch/~fine - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jer@x10sion.com Wed Dec 10 13:11:00 1997 From: jer@x10sion.com (Jeremy Garff) Date: Wed, 10 Dec 1997 13:11:00 -0000 Subject: sigalrm? References: <9710291211.AA07005@ss38.wg.icl.co.uk> Message-ID: Is the SIGALRM signal available under NT? I am trying to use the setitimer function to set the timer for the alarm, however it isn't present in any header I can find, and I can't seem to get it to link either. Thanks in advance, Jeremy Garff jer@x10sion.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 10 15:31:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 10 Dec 1997 15:31:00 -0000 Subject: Gcc compiler b18 newbie probs Message-ID: <19971210220223.5992.qmail@hotmail.com> >To: Rich Dorfman >Cc: gnu-win32@cygnus.com >Subject: Re: Gcc compiler b18 newbie probs >Date: Wed, 10 Dec 1997 09:40:38 -0600 >From: Mumit Khan > >Rule-of-thumb, especially for new users: > > - use the right compiler driver to compile/link programs. For C programs, > use gcc, for C++ use c++ and for F77 use g77. ^^^ shouldn't this be g++ gcc will (or so it is documented to do so) determine if the code is C++. Also file extensions are used to determine C vs C++. The cygwin gcc will compile a file named hello.c as C if you type gcc -c hello.c It will compile it as C++ if you type gcc -c hello.C hello.c and hello.C are the same file as the win32 filesystem is case insensitive when in comes to finding the file on disk. However the command line case is kept intact so that gcc will operate differently on the file based on the lowercase .c or the UPPERCASE .C. > >There are cases where you need a different driver, but then you'd have to >know what libraries need to be linked in explicitly and so on. Not worth >the hassle for usual projects, but sometimes needed for multi-language >projects, especially when mixing F77 and C/C++. > >Regards, >Mumit -- khan@xraylith.wisc.edu > http://www.xraylith.wisc.edu/~khan/ >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jer@x10sion.com Wed Dec 10 15:31:00 1997 From: jer@x10sion.com (Jeremy Garff) Date: Wed, 10 Dec 1997 15:31:00 -0000 Subject: More on getting coolview-bash to work with emacs References: <01BD0554.11F03AB0@gater.krystalbank.msk.ru> Message-ID: Where can I download the lastest cygwin32.dll and the latest libraries that contain all of the new patches which are present on Sergeys site? I have had little luck trying to recompile them on my own. (I think I need more memory) Thanks, Jeremy Garff jer@x10sion.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From skinner@drea.dnd.ca Wed Dec 10 15:31:00 1997 From: skinner@drea.dnd.ca (Bruce S. Skinner) Date: Wed, 10 Dec 1997 15:31:00 -0000 Subject: mount -b C: / breaks /etc/profile &friends Message-ID: <3.0.32.19971210193348.0098d100@firm.drea.dnd.ca> I have been having troubling building the terminfo database required by ncurses and have been told I need to have disks mounted text=binary for it to work. After I do "mount -b c: /" bash starts really having trouble. All the alias' defined in .bash_profile break and when I set PS1 the cursor goes back to the beginning of line after writing the bash prompt. It appears that carriage returns are being read from the .bash_profile as data instead of indicating end of line. Is there a fix for this? Environment is gnuwin32 b18.5 & WinNt4.0. regards :-) Bruce S. Bruce S. Skinner | skinner@drea.dnd.ca DEFENCE RESEARCH ESTABLISHMENT ATLANTIC | norstead@ns.sympatico.ca Dartmouth, Nova Scotia, CANADA, B2Y 3Z7 | (902) 426-3100 x205 or 538-1765 "New opinions are always suspected, and usually opposed without any other reason but because they are not already common." John Locke - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From newsham@lava.net Wed Dec 10 17:42:00 1997 From: newsham@lava.net (Tim Newsham) Date: Wed, 10 Dec 1997 17:42:00 -0000 Subject: 'nohup' in nt? Message-ID: This is more of a general NT question than a cygwin question. How can I start a program (a cygwin program in this case) while logged in to an NT machine that will not get killed off when I log out of the machine? Is it possible to do as a normal user? How about as the administrator? Tim N. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Wed Dec 10 18:19:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Wed, 10 Dec 1997 18:19:00 -0000 Subject: c++ vs g++ [Re: Gcc compiler b18 newbie probs] References: <19971210220223.5992.qmail@hotmail.com> Message-ID: <9712110219.AA02704@modi.xraylith.wisc.edu> "Earnie Boyd" writes: > > > - use the right compiler driver to compile/link programs. For C > programs, > > use gcc, for C++ use c++ and for F77 use g77. > > ^^^ shouldn't this be g++ > Historically these two had a subtle difference: g++ implicitly linked with libg++ while c++ did not. With egcs-1.00, they're equivalent; even before the advent of egcs, I never considereed libg++ as a standard package, and preferred not to use g++ as the C++ driver. I guess I simply don't like unneeded libraried implicitly linked in that standard conforming code(*) does not need. I haven't checked the last FSF gcc-2.8.0 snapshot to check if the distinction is still there or not. Mumit * Now we can finally (almost) use the term "standard conforming" when talking about C++. What a relief. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 10 18:34:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 10 Dec 1997 18:34:00 -0000 Subject: Has anyone got id-utils (mkid, lid) to compile and run? Message-ID: <19971211023127.17144.qmail@hotmail.com> >Date: Wed, 10 Dec 1997 12:32:08 -0600 >From: "Bartlee A. Anderson" >To: Ron Aaron >CC: gnu-win32@cygnus.com >Subject: Re: Has anyone got id-utils (mkid, lid) to compile and run? > >I don't see anything obviously wrong. I would suggest re-mounting as >binary all filesystems ;) > >It looks like you can't cd to d from / >Try an experiment like I did. >cd / >cd d >doesn't work it worked for me on my NT3.51 workstation with the original B18 cygwin.dll. My mounts are binary. >cd /d >that works. >may have something to do with separate devices but needing full path to >access. I'll forward to the list for further comment. > > Bart > [snip] - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ian@cygnus.com Wed Dec 10 20:16:00 1997 From: ian@cygnus.com (ian@cygnus.com) Date: Wed, 10 Dec 1997 20:16:00 -0000 Subject: wind river and cygnus - what is the relationship? References: <01BD0589.DAD45990@rodney.wg.icl.co.uk> Message-ID: <199712110416.XAA08105@tweedledumb.cygnus.com> In gnu-win32 kh@wg.icl.co.uk (Kevin Hughes) writes: >I recently obtained an NT hosted version of the VxWorks and think >that there may be some potential incompatibility with the gnu95 >system. It appears that there is a cygwin.dll shipped with the >VxWorks system - I presume for running the gnu tool chain. It is >dated June 96 so I guess it is old BUT it appears to have extra >functionality - mmap appears to be there. My impression is that Wind River Systems picked up an earlier version of cygwin32, made some local changes, and shipped it with their product. As far as I know, WRS has never contributed anything back to the mainline sources at Cygnus. It is quite likely that you will have problems trying to run two different versions of cygwin.dll simultaneously; I believe this will work somewhat better in the upcoming b19 release. I expect that if you want more information about what WRS ships you will have to contact WRS. Ian - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Wed Dec 10 23:20:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Wed, 10 Dec 1997 23:20:00 -0000 Subject: fork() problem with sockets Message-ID: <01BD061E.701A01A0@gater.krystalbank.msk.ru> Duran wrote: > Problem: The fork child does not inherate the file descriptors when working whith sockets. > > How to reproduce the problem: Just create a socket server, accept a connection and fork, you will notice that the child will not be able to use the sockets declared before the forking. The problem is Win95 specific. Winsock's socket() call returns non-inherittable socket handle on W95. Dowload patched cygwin.dll from my web page. -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Thu Dec 11 02:04:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Thu, 11 Dec 1997 02:04:00 -0000 Subject: One way to get around the cygwin terminal problem Message-ID: <199712111004.KAA278668@out1.ibm.net> One way to get around the limitations of the cygwin terminal is to initiate a second buffer for applications and use the Windows console I/O to set the color attributes, read from the console buffer and move the curser around. I don't have a completely working version yet, but an initial reworking of the libslang.a library to make everything in it work from the console buffer generated an slrn news reader that is lightning fast by comparison to the cygwin version and has no text gaps or problems with overwritten text. In other words, with some work, it is possible to get a professional looking and acting terminal application using the cygwin Unix environment but bypassing the cygwin terminal window. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Thu Dec 11 02:23:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Thu, 11 Dec 1997 02:23:00 -0000 Subject: Gcc compiler b18 newbie probs References: <19971210220223.5992.qmail@hotmail.com> Message-ID: <19971211212021.46772@murlibobo.cs.mu.OZ.AU> Earnie Boyd wrote: > > > >To: Rich Dorfman > >Cc: gnu-win32@cygnus.com > >Subject: Re: Gcc compiler b18 newbie probs > >Date: Wed, 10 Dec 1997 09:40:38 -0600 > >From: Mumit Khan > > > >Rule-of-thumb, especially for new users: > > > > - use the right compiler driver to compile/link programs. For C > programs, > > use gcc, for C++ use c++ and for F77 use g77. > > ^^^ shouldn't this be g++ Depends on what you want to link in. `c++' just links in the standard C++ libraries, I think. `g++' also links in libg++. > gcc will (or so it is documented to do so) determine if the code is C++. > Also file extensions are used to determine C vs C++. The cygwin gcc > will compile a file named hello.c as C if you type > > gcc -c hello.c > > It will compile it as C++ if you type > > gcc -c hello.C Yes, it will _compile_ it as C++, but if you do `gcc hello.C' it will still _link_ it as if it were C (that is, it won't link in the C++ libraries). -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 11 04:24:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 11 Dec 1997 04:24:00 -0000 Subject: mount -b C: / breaks /etc/profile &friends Message-ID: <19971211122348.4620.qmail@hotmail.com> After mounting to text=binary you have to get rid of all \r in all scripts. Use the following with caution (ie: only on scripts): cp scriptfile /tmp/scriptfile tr -d "\r" < /tmp/scriptfile > scriptfile - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Wed, 10 Dec 1997 19:34:47 -0400 >To: gnu-win32@cygnus.com >From: "Bruce S. Skinner" >Subject: mount -b C: / breaks /etc/profile &friends > >I have been having troubling building the terminfo database required by >ncurses and have been told I need to have disks mounted text=binary for it >to work. After I do "mount -b c: /" bash starts really having trouble. >All the alias' defined in .bash_profile break and when I set PS1 the cursor >goes back to the beginning of line after writing the bash prompt. It >appears that carriage returns are being read from the .bash_profile as data >instead of indicating end of line. Is there a fix for this? Environment is >gnuwin32 b18.5 & WinNt4.0. > >regards :-) >Bruce S. > > Bruce S. Skinner | skinner@drea.dnd.ca >DEFENCE RESEARCH ESTABLISHMENT ATLANTIC | norstead@ns.sympatico.ca >Dartmouth, Nova Scotia, CANADA, B2Y 3Z7 | (902) 426-3100 x205 or 538-1765 >"New opinions are always suspected, and usually opposed without any >other reason but because they are not already common." John Locke >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 11 04:37:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 11 Dec 1997 04:37:00 -0000 Subject: c++ vs g++ [Re: Gcc compiler b18 newbie probs] Message-ID: <19971211123624.27195.qmail@hotmail.com> Thanks Mumit, I didn't realize that the c++ option was available. I believe however that the c++.exe and g++.exe supplied from cygnus are the same file with different names. I base this on exec "cmp -c /bin/g++.exe /bin/c++.exe" not returning anything. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >To: gnu-win32@cygnus.com >Subject: c++ vs g++ [Re: Gcc compiler b18 newbie probs] >Date: Wed, 10 Dec 1997 20:19:15 -0600 >From: Mumit Khan > >"Earnie Boyd" writes: >> >> > - use the right compiler driver to compile/link programs. For C >> programs, >> > use gcc, for C++ use c++ and for F77 use g77. >> >> ^^^ shouldn't this be g++ >> > >Historically these two had a subtle difference: g++ implicitly linked >with libg++ while c++ did not. With egcs-1.00, they're equivalent; even >before the advent of egcs, I never considereed libg++ as a standard >package, and preferred not to use g++ as the C++ driver. I guess I >simply don't like unneeded libraried implicitly linked in that standard >conforming code(*) does not need. > >I haven't checked the last FSF gcc-2.8.0 snapshot to check if the >distinction is still there or not. > >Mumit > >* Now we can finally (almost) use the term "standard conforming" when >talking about C++. What a relief. >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Thu Dec 11 05:30:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Thu, 11 Dec 1997 05:30:00 -0000 Subject: One way to get around the cygwin terminal window problem Message-ID: <199712111330.NAA25648@out2.ibm.net> >> Gnu-win32 mailing list archives: One way to get around the cygwin te (p2 of 2) In other words, with some work, it is possible to get a professional looking and acting terminal application using the cygwin Unix environment but bypassing the cygwin terminal window. << Actually, it looks like the problem is the cygwin fread() function which apparently converts cursor-movement and control keys to vtxxx escape sequences Unix-style. If you use the slang library with the Windows console option, then you need to write a GetKey() function that translates from Windows to vtxxx so that programs using libslang.a can find the function keys and things like pageup and pagedown. ???? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dsyed1@umbc.edu Thu Dec 11 05:43:00 1997 From: dsyed1@umbc.edu (daniyal syed) Date: Thu, 11 Dec 1997 05:43:00 -0000 Subject: 'nohup' in nt? References: Message-ID: start /b program.exe On Wed, 10 Dec 1997, Tim Newsham wrote: > > This is more of a general NT question than a cygwin question. > How can I start a program (a cygwin program in this case) while > logged in to an NT machine that will not get killed off when > I log out of the machine? Is it possible to do as a normal user? > How about as the administrator? > > Tim N. > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". > Daniyal Syed. dsyed1@umbc.edu dsyed1@right.ml.org http://www.gl.umbc.edu/~dsyed1 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kh@wg.icl.co.uk Thu Dec 11 05:52:00 1997 From: kh@wg.icl.co.uk (Kevin Hughes) Date: Thu, 11 Dec 1997 05:52:00 -0000 Subject: Shell script performance Message-ID: <01BD063B.E880E380@rodney.wg.icl.co.uk> I have noticed that the performance of my shell scripts leaves a great deal to be desired. On my Sparc boxes they trip along quite nicely - making the users happy but on much more powerful Intel PII machines running NT - wow is it slow. Its certainly not the processors as the sparcs are little hyperSparcs with 64M, the Intel is dual PII at 300Mhz with 256M. I guess this is a bash implementation problem to do with process creation which has been mentioned on the group before. I have noticed that the uwin system runs quicker as does the OpenNT but I would prefer not to go that way if I can avoid it. Does anyone have any ideas if anything can/is being done? Kevin - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kh@wg.icl.co.uk Thu Dec 11 06:07:00 1997 From: kh@wg.icl.co.uk (Kevin Hughes) Date: Thu, 11 Dec 1997 06:07:00 -0000 Subject: converting between UNC and NT files names? Message-ID: <01BD063C.D420B680@rodney.wg.icl.co.uk> Is there any easy way of converting between the unc used by the gnu tools for file names and the standard NT file names? I would like to call native NT tools directly from shell scripts but all the file names I have in variables etc. are in UNC format and the tools only understand the standard names. Any help much appreciated Kevin - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kh@wg.icl.co.uk Thu Dec 11 06:07:00 1997 From: kh@wg.icl.co.uk (Kevin Hughes) Date: Thu, 11 Dec 1997 06:07:00 -0000 Subject: sigalrm? Message-ID: <01BD063E.03F663E0@rodney.wg.icl.co.uk> -----Original Message----- From: Jeremy Garff [SMTP:jer@x10sion.com] Sent: 10 December 1997 21:13 To: gnu-win32@cygnus.com Subject: sigalrm? Is the SIGALRM signal available under NT? I am trying to use the setitimer function to set the timer for the alarm, however it isn't present in any header I can find, and I can't seem to get it to link either. Thanks in advance, Jeremy Garff jer@x10sion.com If you download the latest coolview from sergey`s site ( http://miracle.geol.msu.ru/sos/ ) and install you will get the Functionality you want. You will also need the itimer patches in order to get time.h for the definition of setitimer etc. (this replaces sys/time.h) It works for me - but I only ever tried getting one event, I assume multiple events will work Good luck Kevin - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From erniec@ix.netcom.com Thu Dec 11 06:41:00 1997 From: erniec@ix.netcom.com (Ernest Clayton Cordell, Jr.) Date: Thu, 11 Dec 1997 06:41:00 -0000 Subject: More on getting coolview-bash to work with emacs Message-ID: <01bd063c$90b000e0$4fb4dfcf@magic> Folks, Sergey has done a lot of wonderful work: None of us have put that up for debate. It is my hope to add a thought that might clarify what I see being expressed frequently in the list. My motive for doing so is to help define a convention for (cygwin32-style) binary file mounts that will be acceptable to a broad audience. I wrote a larger, clearer message earlier but I deleted it in favor of the more succinct rubbish that follows. In the few months that I've occasionally played with the cygwin32-toolset (not "in anger"), I have accepted the processing schism of a "text mount" as a temporary challenge. There are three components to this problem: data introduced to processing elements ; the processing elements themselves ; and the environment used for execution of processing elements . These can be vectored as "native" or "foreign," but the order of business is always -->-->. With -->--> (cygwin32/"unix"): no problem, right? Both modes accomodated. With -->-->: unix/binary seems accomodated, not win/dos. With -->-->: "outside" programs have own behavior; no standard possible. With -->-->: "outside" programs only port to "pre-configured" environments. With -->-->: _filters_ required to process foreign data in native environment. With -->-->: policy must be established for cygwin32 target environments. With -->-->: can make accomodations but no guarantees for "imported subsystems." With -->-->: collision possible if cygwin32 and foreign environment don't coordinate. I have focused here on porting and execution because "use" and "compatibility" are vague terms where anything other than the "precise" behavior can be expected in both environments: Such considerations are too complex for an online analysis. In most of these cases everything possible has already been done: In the case of _filters_, it could be argued that if such filters exist, they should be part of the front-end processing of the cygwin32 toolset. In this scenario, shell scripts would be because they are input that is read by cygwin32 utilities: While the shell could be modified to accept either or -alone as possible line terminators, "foreign" programs that refuse to do so cannot be governed by the cygwin32 toolset, team or sponsors. In compilers and utilities, comments are stripped and spaces/blanks are also routinely skipped: Should a utility halt because it finds an intervening blank between the line-continuation marker and the marker? Then should it fail because of "printer-control" characters either? As far as operating cygwin32 tools outside of a cygwin32 environment (not relevant to this particular issue, but frequently mentioned), the goals of the development team must define the policy for "where" (in what environments) cygwin32 tools are designed to operate. I'm sorry that it took me so many words to relate this concept: I'm not gifted with abbreviated expression, Ernie ernie.cordell@computer.org Web pages at http://www.netcom.com/~erniec/ http://www.geocities.com/SouthBeach/Lagoon/7778 -----Original Message----- From: Guy Gascoigne - Piggford To: GNU-Win32 Date: Wednesday, December 10, 1997 8:14 PM Subject: Re: More on getting coolview-bash to work with emacs >At 11:20 AM 12/10/97 +0100, you wrote: >>> To achieve my goal, I need to use text-mode mounts. >>> Thus for my purposes, if bash stops working on text-mode mounts, >>> then this is a big step backwards... >> >>Actually, Sergey's work is a big step *forwards*, but I would appreciate >>support for both kinds of line-endings very much. > >I have to agree here, I'm not completely convinced by the arguments of >'just use binary mounts', but aside from that, Sergy's work has added a lot >of improvements and added functionality. I'd much rather use it than not. > >Guy > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > From cgf@bbc.com Thu Dec 11 06:41:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Thu, 11 Dec 1997 06:41:00 -0000 Subject: More on getting coolview-bash to work with emacs References: <01BD0554.11F03AB0@gater.krystalbank.msk.ru> Message-ID: In article < Pine.LNX.3.94.971210144707.27713C-100000@fire.x10sion.com >, Jeremy Garff wrote: >Where can I download the lastest cygwin32.dll and the latest libraries >that contain all of the new patches which are present on Sergeys site? I >have had little luck trying to recompile them on my own. (I think I need >more memory) Just download the "precompiled binaries" that are mentioned at the top of the web page: http://www.lexa.ru/sos -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Thu Dec 11 07:48:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Thu, 11 Dec 1997 07:48:00 -0000 Subject: c++ vs g++ [Re: Gcc compiler b18 newbie probs] References: <19971211123624.27195.qmail@hotmail.com> Message-ID: <9712111548.AA03306@modi.xraylith.wisc.edu> "Earnie Boyd" writes: > I didn't realize that the c++ option was available. I believe however > that the c++.exe and g++.exe supplied from cygnus are the same file with > different names. I base this on exec "cmp -c /bin/g++.exe /bin/c++.exe" > not returning anything. On systems that support linking, they're the same file. On Win32, they're copies. Until recently, the program would include the different libraries depending on what name it was called with (g++ or c++), much like what gzip and gunzip do. with egcs, there is *no* difference between c++ and g++. Those who want to use libg++, should now use: % c++ -o blah.exe blah.o -lg++ Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From joe_banta@perceptics.com Thu Dec 11 07:48:00 1997 From: joe_banta@perceptics.com (Joseph E. Banta, Esquire) Date: Thu, 11 Dec 1997 07:48:00 -0000 Subject: Is it practical to install cygwin32 using cygwin32? Message-ID: <01BD0622.7C76D820@HERBY> Howdy. I am trying to 'make all install' from the source distribution (I have been using the binary distribution for a while now), and I get a couple of errors. I am compiling with the cygwin gcc compiler, with a directory configured with 'configure --host=i386-cygwin32 --target=i386-cygwin32 --exec_prefix=/usr/local'. The errors I am getting are all minor; just things where I need to go into a Makefile.in and change a line or two (and similar hacks). This question is for experienced installers: Is it practical to install cygwin32 using cygwin32? Did I get the configuration right? Joseph Banta - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Gerard.Milhaud@esil.univ-mrs.fr Thu Dec 11 09:05:00 1997 From: Gerard.Milhaud@esil.univ-mrs.fr (Gerard Milhaud) Date: Thu, 11 Dec 1997 09:05:00 -0000 Subject: Basic question : I don't see man.exe Message-ID: <34901A4C.9066A223@esil.univ-mrs.fr> Hi, I've just installed b18, and i'm very happy to have my gnu environment on NT. But the commands (i.e. ls) are not exactly the same so i've tried "man ls" bash answers me that it can't find man. I'm wondering that man.exe is not in the bin directory where bash is. I've seen all the man stuff in three directories in the distribution, so i think there is a possibility to use it (and the info stuff so, because there is not texinfo.exe or info.exe neither). Could anyone help me ? Thanks in advance. -- Gerard MILHAUD Ecole Superieure d'Ingenieurs de Luminy (ESIL) Luminy - Case 925 - 13288 Marseille Cedex 09 Tel.: 04 91 82 85 31 - Fax.: 04 91 82 85 11 mailto:g@planete-marseille.esil.univ-mrs.fr - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Thu Dec 11 09:14:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Thu, 11 Dec 1997 09:14:00 -0000 Subject: date in a backgrounf task ? References: <199712102029.VAA116804@sp049.cern.ch> Message-ID: <6jdqZ35$pfB@mike.franken.de> Hi Valery, [...] >> > I'm trying to use "date" in a batch file, which is called by an NT >> > service (Legato Networker save). Whenever the date command gets started, >> > it sits there doing nothing, leaving the batch file waiting forever. > > Did you try "date /t" ? I meant the date command from the Cygnus cdk ! > > Cheers, > Valery Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Thu Dec 11 09:25:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Thu, 11 Dec 1997 09:25:00 -0000 Subject: converting between UNC and NT files names? References: <01BD063C.D420B680@rodney.wg.icl.co.uk> Message-ID: On Thu, 11 Dec 1997, Kevin Hughes wrote: > Is there any easy way of converting between the unc used by the gnu tools for file names and the standard > NT file names? I would like to call native NT tools directly from shell scripts but all the file names I have > in variables etc. are in UNC format and the tools only understand the standard names. > The attached piece of code may help you get started. Simply run it with the posix filename(s) on the command line and it'll spit out both relative and absolute win32 pathnames. % posix2win32path . //server/src . --> c:\tmp (.) //server/src --> \\server\src (\\server\src) Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ From sos@buggy.prospect.com.ru Thu Dec 11 10:28:00 1997 From: sos@buggy.prospect.com.ru (Sergey Okhapkin) Date: Thu, 11 Dec 1997 10:28:00 -0000 Subject: Dll import/export Message-ID: <01BD0677.B49F1000@sos> Hi! Does anybody works on implementation dllexport/dllimport function and data attributes in gcc/x86 (like in gcc/ppc)? -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Thu Dec 11 10:28:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Thu, 11 Dec 1997 10:28:00 -0000 Subject: mount -b C: / breaks /etc/profile &friends References: <3.0.32.19971210193348.0098d100@firm.drea.dnd.ca> Message-ID: <6jdtxWm$pfB@mike.franken.de> Hi Bruce, >I have been having troubling building the terminfo database required by >ncurses and have been told I need to have disks mounted text=binary for it >to work. After I do "mount -b c: /" bash starts really having trouble. >All the alias' defined in .bash_profile break and when I set PS1 the cursor >goes back to the beginning of line after writing the bash prompt. It >appears that carriage returns are being read from the .bash_profile as data >instead of indicating end of line. Is there a fix for this? Environment is >gnuwin32 b18.5 & WinNt4.0. Sorry not to have mentioned it - you also have to change all your text files from DOS to Unix style line breaks ( -> ) !!! > >regards :-) >Bruce S. Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Thu Dec 11 10:28:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Thu, 11 Dec 1997 10:28:00 -0000 Subject: 'nohup' in nt? References: Message-ID: <6jdtyFb4pfB@mike.franken.de> Hi Tim, > >This is more of a general NT question than a cygwin question. >How can I start a program (a cygwin program in this case) while >logged in to an NT machine that will not get killed off when >I log out of the machine? Is it possible to do as a normal user? >How about as the administrator? Easiest way is to use the NT scheduler and run the job via the "at" command. > > Tim N. Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ulittj00@mcl.ucsb.edu Thu Dec 11 11:37:00 1997 From: ulittj00@mcl.ucsb.edu (Joshua J Little) Date: Thu, 11 Dec 1997 11:37:00 -0000 Subject: lcc-win32 vs. mingw32/gnuwin32 Message-ID: Does anyone out there have any experience with the c compiler lcc for compiling vanilla win32 C code? It seems to make much smaller executables, compiles a lot faster, and, at least for the examples in Petzold's "Programming Windows 95," it compiles the code pretty much right out-of-the-box, unlike mingw32 or gnuwin32. Are there any disadvantages to it that anyone knows of? Any advice anyone can offer? Thanks in advance for the help. Josh Little ulittj00@mcl.ucsb.edu - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From smueller@microsoft.com Thu Dec 11 13:27:00 1997 From: smueller@microsoft.com (Stephan Mueller) Date: Thu, 11 Dec 1997 13:27:00 -0000 Subject: 'nohup' in nt? Message-ID: <60602D8F29B3D011879800805FD4722401749F5A@red-11-msg.dns.microsoft.com> Start /b will start the application without creating a console window for it. This does not mean the process won't go away when you logout. To do what you want, I believe you should be investigating running your process as a service. Services are a little bit special, but the NT Resource Kit includes a program called srvany.exe which can be used to run any process as a service. (For some definition of 'any.') stephan(); > -----Original Message----- > From: daniyal syed [SMTP:dsyed1@umbc.edu] > Sent: Thursday, December 11, 1997 5:43 AM > To: Tim Newsham > Cc: gnu-win32@cygnus.com > Subject: Re: 'nohup' in nt? > > > > start /b program.exe > > On Wed, 10 Dec 1997, Tim Newsham wrote: > > > > > This is more of a general NT question than a cygwin question. > > How can I start a program (a cygwin program in this case) while > > logged in to an NT machine that will not get killed off when > > I log out of the machine? Is it possible to do as a normal user? > > How about as the administrator? > > > > Tim N. > > - > > For help on using this list (especially unsubscribing), send a message > to > > "gnu-win32-request@cygnus.com" with one line of text: "help". > > > > > > Daniyal Syed. > > dsyed1@umbc.edu > dsyed1@right.ml.org > http://www.gl.umbc.edu/~dsyed1 > > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mark.buckingham@us.nestle.com Thu Dec 11 13:27:00 1997 From: mark.buckingham@us.nestle.com (Buckingham, Mark, SOLON, NUSA) Date: Thu, 11 Dec 1997 13:27:00 -0000 Subject: egcs 1.00 Message-ID: Has anyone gotten egcs 1.00 to work with gnuwin32 yet? I downloaded the tar file (the one for gnuwin32), and installed it. I am running on nt 4.0, with the coolview patch. I try to compile a simple hello, world type program, but after typing gcc -v hello.c -o hello.exe The command line never comes back. I checked into it with ps, and it shows two copies of ld.exe running. In addition, here is the output of the -verbose switch from gcc: Reading specs from /usr/local/lib/gcc-lib/i386-cygwin32\egcs-2.90.21\specs gcc version egcs-2.90.21 971202 (egcs-1.00 release) /usr/local/lib/gcc-lib/i386-cygwin32\egcs-2.90.21\cpp.exe -lang-c -v -iprefix / usr/local/lib/gcc-lib/i386-cygwin32\egcs-2.90.21\ -undef -D__GNUC__=2 -D__GNUC_M INOR__=90 -Di386 -D_WIN32 -DPOSIX -D__CYGWIN32__ -DWINNT -D_X86_=1 -D__STDC__=1 -D__stdcall=__attribute__((__stdcall__)) -D__cdecl=__attribute__((__cdecl__)) -D __i386__ -D_WIN32 -D__POSIX__ -D__CYGWIN32__ -D__WINNT__ -D_X86_=1 -D__STDC__=1 -D__stdcall=__attribute__((__stdcall__)) -D__cdecl=__attribute__((__cdecl__)) -D __i386 -D__POSIX -D__WINNT -Asystem(winnt) -Acpu(i386) -Amachine(i386) -Di386 -A system(unix) -Acpu(i386) -Amachine(i386) -D__i386__ -Asystem(unix) -Acpu(i386) - Amachine(i386) hello.c C:\TEMP\cc001135.i GNU CPP version egcs-2.90.21 971202 (egcs-1.00 release) (80386, BSD syntax) #include "..." search starts here: #include <...> search starts here: /usr/local/lib/gcc-lib/i386-cygwin32\egcs-2.90.21\include /usr/local/i386-cygwin32/include /usr/include End of search list. /usr/local/lib/gcc-lib/i386-cygwin32\egcs-2.90.21\cc1.exe C:\TEMP\cc001135.i -q uiet -dumpbase hello.c -version -o C:\TEMP\cc001135.s GNU C version egcs-2.90.21 971202 (egcs-1.00 release) (i386-cygwin32) compiled b y GNU C version egcs-2.90.21 971202 (egcs-1.00 release). as -o C:\TEMP\cc0011351.o C:\TEMP\cc001135.s /usr/local/lib/gcc-lib/i386-cygwin32\egcs-2.90.21\ld.exe -o hello.exe /usr/loca l/i386-cygwin32/lib\crt0.o -L/usr/local/lib/gcc-lib/i386-cygwin32\egcs-2.90.21 - L/usr/local/lib/gcc-lib -L/usr/local/i386-cygwin32/lib C:\TEMP\cc0011351.o -lgcc -lcygwin -lkernel32 -lgcc By the way, /usr/local is a symlink to /usr/H-i386-cygwin32 Thanks! - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fine@mail.cern.ch Thu Dec 11 16:18:00 1997 From: fine@mail.cern.ch (Valery Fine) Date: Thu, 11 Dec 1997 16:18:00 -0000 Subject: 'nohup' in nt? References: Message-ID: <199712112220.XAA34408@sp051.cern.ch> On 11 Dec 97 at 8:42, daniyal syed wrote: > > > start /b program.exe > I think it is wrong advice. > On Wed, 10 Dec 1997, Tim Newsham wrote: > > > > > This is more of a general NT question than a cygwin question. > > How can I start a program (a cygwin program in this case) while > > logged in to an NT machine Administrator should set a Logon script via "User Manager" for you (for all users would be better). This script may look like === if exist %homedrive%%homepath%\logon.cmd %homedrive%%homepath%\logon.cmd === > > that will not get killed off when > > I log out of the machine? Is it possible to do as a normal user? > > How about as the administrator? What you want is called "start batch program". Just type "AT /?" to get further information. Hope this helps, Valery === Dr. Valeri Faine (Valery Fine) ------------ ------------- Phone: +41 22 767 4921 CERN FAX : +41 22 767 7155 CH-1211 Geneva, 23 mailto:fine@mail.cern.ch Switzerland http://nicewww.cern.ch/~fine - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hasmik@maker.com Thu Dec 11 16:18:00 1997 From: hasmik@maker.com (Hasmik Sarkezians) Date: Thu, 11 Dec 1997 16:18:00 -0000 Subject: "read" routine! Message-ID: <3.0.32.19971211183313.009472c0@nucleus.maker.com> Does anyone know what "read" routine does when a Ctrl-C is issued?? My application starts a Tcl shell and if I issue a Ctrl-C from the debugger it exits the program. I have debugged this to the point that I think the "read" routine is not behaving like the "read" on the unix system. The "read" on NT system returns a 0 (zero) and the Tcl application treats that as the end-of-file and exits the program. Shouldn't the "read" routine return some sort of error (EINTR)?? Thanks, hasmik - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From rickmoll@mvp.net Thu Dec 11 16:18:00 1997 From: rickmoll@mvp.net (Rick Moll) Date: Thu, 11 Dec 1997 16:18:00 -0000 Subject: newbie wants vi Message-ID: <34907C11.3D6771CA@mvp.net> Sorry for the newbie question, but I've just got the gnu-win32 tools installed and I was disappointed that there wasn't any kind of editor. I'm totally new to all of this, gnu-win32, my WinNT system, and GNU in general. I'm completely lost without a vi; what's the easiest way to get one to use with the gnu-win32 tools. I understand that there is a "vim". Is this a vi workalike? If so where's the easiest place to get it, and will it easily port to the gnu-win32 environment? Thanks, Rick - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From iverson@cisco.com Thu Dec 11 16:18:00 1997 From: iverson@cisco.com (Tim Iverson) Date: Thu, 11 Dec 1997 16:18:00 -0000 Subject: Shell script performance References: <01BD063B.E880E380@rodney.wg.icl.co.uk> Message-ID: <199712120017.QAA09228@stars.cisco.com> If you're running stock b18, I'd guess stat(), not fork/exec (though fork/exec is really slow, too). Last I checked, stat() actual opens each file and reads a few bytes to check for symbolic links. IMHO, this is an incredible waste of time, especially considering that W95/NT already has a defined symbolic link standard (.lnk files). I strongly suspect this has changed in b19 and probably Sergey's code, since I recall that this whole topic was beaten to death and finally resolved about 6 months ago. - Tim Iverson iverson@lionheart-com +---------------- | From: Kevin Hughes | To: "Gnuwin95 (E-mail)" | Subject: Shell script performance | Date: Thu, 11 Dec 1997 13:51:39 -0000 | [paraphrased: cygwin32 bash is incredibly slow + some examples] - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dahms@ifk20.mach.uni-karlsruhe.de Thu Dec 11 19:03:00 1997 From: dahms@ifk20.mach.uni-karlsruhe.de (dahms@ifk20.mach.uni-karlsruhe.de) Date: Thu, 11 Dec 1997 19:03:00 -0000 Subject: Gcc compiler b18 newbie probs Message-ID: <009BEA34.CE3795E0.30099@ifk20.mach.uni-karlsruhe.de> Hi Earnie, you wrote: : It will compile it as C++ if you type : : gcc -c hello.C There is a third way besides the compiler name and filename case, namely the -x option, to specify the language. Bye, Heribert (dahms@ifk20.mach.uni-karlsruhe.de) - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Thu Dec 11 19:03:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Thu, 11 Dec 1997 19:03:00 -0000 Subject: egcs 1.00 References: Message-ID: <9712120303.AA04364@modi.xraylith.wisc.edu> "Buckingham,Mark,SOLON,NUSA" writes: > > The command line never comes back. I checked into it with ps, and it > shows two copies of ld.exe running. > Ah! I think I might know what the problem is. Looks like you're basically seeing the following problem: 1. gcc calls collect2 (aka ld in gcc library directory) 2. ld looks for the "real" ld and somehow finds itself and execs it (itself in this case). 3. Repeats 2. On the other hand, the above I think can only happen under weird circumstances. I'm assuming that $prefix/lib/gcc-lib/i386-cygwin32/egcs-2.90.21/ is NOT in your path, right? Could you send me the following info: 1. your PATH and othe GCC related environment vars 2. the output of the following: $ type -all ld.exe ### << if you're running bash $ gcc -print-prog-name=ld $ gcc -v -o test.exe test.c -Wl,--verbose Thanks Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From aubin.mahe@wanadoo.fr Thu Dec 11 19:03:00 1997 From: aubin.mahe@wanadoo.fr (Aubin Mahé) Date: Thu, 11 Dec 1997 19:03:00 -0000 Subject: Basic question : I don't see man.exe References: <34901A4C.9066A223@esil.univ-mrs.fr> Message-ID: <3490A75A.EF1A540B@wanadoo.fr> Hi, fr : C'est normal, lisez le README, l'????quipe bosse sur la mise au point du man... Je suis interress???? aussi par cet outil, pr????venez moi d????s que vous avez du nouveau Il existe un site o???? les pages man ont ????t???? transform????e en fichier WinHelp : http://www.itribe.net/virtunix/winhelp-man-pages/ us : man don't come with release b18, be patient... I'm interesting by the man tool, if it exist I want now that ! Look at this site for man on WinHelp files : http://www.itribe.net/virtunix/winhelp-man-pages/ Gerard Milhaud a ????crit: Hi, ???????????? I've just installed b18, and i'm very happy to have my gnu environment on NT. But the commands (i.e. ls) are not exactly the same so i've tried "man ls" bash answers me that it can't find man.???? I'm wondering that man.exe is not in the bin directory where bash is. I've seen all the man stuff in three directories in the distribution, so i think there is a possibility to use it (and the info stuff so, because there is not texinfo.exe or info.exe neither). ???????????? Could anyone help me ? Thanks in advance. -- Gerard MILHAUD Ecole Superieure d'Ingenieurs de Luminy (ESIL) Luminy - Case 925 - 13288 Marseille Cedex 09 Tel.: 04 91 82 85 31 - Fax.: 04 91 82 85 11 mailto:g@planete-marseille.esil.univ-mrs.fr - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". ???? From paag@tid.es Thu Dec 11 23:41:00 1997 From: paag@tid.es (Pedro A. Aranda Gutiirrez) Date: Thu, 11 Dec 1997 23:41:00 -0000 Subject: Basic question : I don't see man.exe References: <34901A4C.9066A223@esil.univ-mrs.fr> <3490A75A.EF1A540B@wanadoo.fr> Message-ID: <3490EAFE.FA1B4172@tid.es> E: Como empecemos a ser multiling????es, los mensajes van a crecer mucho! D: Wenn wir jetzt multi-kulti werden, k????nnen die e-mails gr????????lich wachsen.! UK: If we start multi-linguality, message length will grow damn fast! :) :) :) Pedro Andres Aranda Gutierrez mailto:paag@tid.es - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dbe@wgn.net Thu Dec 11 23:41:00 1997 From: dbe@wgn.net ($Bill Luebkert) Date: Thu, 11 Dec 1997 23:41:00 -0000 Subject: newbie wants vi References: <34907C11.3D6771CA@mvp.net> Message-ID: <3490E8D7.3AA562CD@wgn.net> Rick Moll wrote: > > Sorry for the newbie question, but I've just got the gnu-win32 tools > installed and I was disappointed that there wasn't any kind of editor. > I'm totally new to all of this, gnu-win32, my WinNT system, and GNU in > general. > > I'm completely lost without a vi; what's the easiest way to get one to > use with the gnu-win32 tools. I understand that there is a "vim". Is > this a vi workalike? If so where's the easiest place to get it, and > will it easily port to the gnu-win32 environment? Try one of these: http://www.halcyon.com/gvr/vim.html # win32 ftp://ftp.halcyon.com/local/gvr/vim/ # Win32 http://www.tiac.net/users/cgf/cygvim50h.bin.tar.gz # VIM 5.0h binary for cygwin32 -- ,-/- __ _ _ $Bill Luebkert (_/ / ) // // DBE Collectibles / ) /--< o // // http://www.wgn.net/~dbe/ -/-' /___/_<_ Hi, I am new to gnuwin32, and I would like to port ssh to win95, can any one tell me sites where I can get info about porting unix programs? Thanks Livhu - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From periat@ens.ascom.ch Fri Dec 12 02:00:00 1997 From: periat@ens.ascom.ch (Periat Henri) Date: Fri, 12 Dec 1997 02:00:00 -0000 Subject: Definition of VPATH Message-ID: <199712120952.KAA14993@stva89.ens.ascom.ch> Hi everybody I posted this message a few days ago, but i didn't go much reply on it. I'm very new to cygnu and have to make a UNIX gmake file to run also on MS-NT boxes. Now the problem which stops me in the moment is the ':' problem: - with 'VPATH:=/home/periat/src:D:/foo/src' make can't find the source files if they are on drive 'D:' only. - with 'VPATH:=/home/periat/src;D:/foo/src' or 'VPATH:=/home/periat/src://D/foo/src' '$<' contains '//D/foo/src/foo.c' and non cygnus programs can't deal with such file names. - with 'VPATH:=/home/periat/src:/foo/src' and 'D:/foo' binary mounted as '/foo' 'make' finds files, but again non cygnus programs don't know anything about 'mount' and so they can't find '/fool/file' Can anyone tell me how i can solve this problem? Thanks Henri I have b18 and the 'coolview' patches. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". ------------- End Forwarded Message ------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@buggy.prospect.com.ru Fri Dec 12 02:00:00 1997 From: sos@buggy.prospect.com.ru (Sergey Okhapkin) Date: Fri, 12 Dec 1997 02:00:00 -0000 Subject: One way to get around the cygwin terminal problem Message-ID: <01BD06F3.CBA2B6E0@sos> vischne@ibm.net wrote: > > In other words, with some work, it is possible to get a professional > looking and acting terminal application using the cygwin Unix > environment but bypassing the cygwin terminal window. It's a bad idea to replace stdout with direct console writings. You application, for example, will not work in a telnet session or in xterm window. -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ltshisik@zibbi.mikom.csir.co.za Fri Dec 12 04:59:00 1997 From: ltshisik@zibbi.mikom.csir.co.za (Livhu Tshisikule) Date: Fri, 12 Dec 1997 04:59:00 -0000 Subject: Help, configure error!!! Message-ID: <199712121258.OAA11643@zibbi.mikom.csir.co.za> Hi I am not familiar with gnuwin32, when I run configure I am getting the following error: checking host system type... Invalid configuration 'i386-unknown-cygwin32': system 'cygwin32' not recognized What should I do? Thanks in advance. Livhu - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net-nospam Fri Dec 12 05:03:00 1997 From: vischne@ibm.net-nospam (root) Date: Fri, 12 Dec 1997 05:03:00 -0000 Subject: Slang library slw32tty.c that overlays the cygwin console - slw32tty.c [1/2] Message-ID: <199712121037.KAA13880@out5.ibm.net> Content-Type: Application/octet-stream; name=slw32tty.c Content-Transfer-Encoding: x-uue begin 755 slw32tty.c M+RH@0V]P>7)I9VAT("AC*2`Q.3DR+"`Q.3DU($IO:&X@12X@1&%V:7,*("H@ M06QL(')I9VATF5R;RX*("H@3D]?1DQ/5U]#3TY44D],(&%N9"!/4$]35"!A2!F;W(@8V]M<&%T:6)L:71Y(&%N9"!A&ET("HO"B`@2$%.1$Q%(&-O;G-O;&4@ M/2!#2P@;VX@<')O9W)A;2!E>&ET+"!T:&4@;W)I9VEN86P@ M2`J+PH*+RHM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM M+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM*EP*("H@ M($9U;F-T:6]N.@EV;VED(%-,86YG7W)E2`H=F]I9"D["B`J"B`J M(')E7!E(#T]($M%65]%5D5.5"D*"0D@("`@ M("`F)B!R96-O41O=VXI(`H@("`@("`@ M+RH@;VLL('1H97)E(&ES(&$@:V5Y<')E7-?9V5T M:V5Y("AV;VED*3L*("H*("H@=V%I="!F;W(@86YD(&=E="!T:&4@;F5X="!A M=F%I;&%B;&4@:V5Y3`L(&ME>3$["@IV;VED($=E=$ME>2`H=F]I M9"D*>PH@('5N3`@/2`H:V5Y,2`](#`I.PH@(&EF("A296%D0V]N M41O=VXI"B`@("!["B`@("`@(&ME>3`@ M/2`H=6YS:6=N960@:6YT*7)E8V]R9"Y%=F5N="Y+97E%=F5N="Y!3$@/2!R96-O5!R97-S960Z("HO"B\J(&-H87(@*F5S8V%P95]S=')I;F<@/2`H8VAA7-?9V5T:V5Y("AV;VED*0I[ M"B!I9B`H97-C87!E7W-T3`@/2`H=6YS:6=N960@:6YT*2IE7,G.B`J+PH)8V%S92!62U]%3D0Z"65S8V%P95]S=')I;F<@/2`B6S(U2"([ M(&)R96%K.PH)8V%S92!62U](3TU%.@EE3`["B`@('T["B!R971U Vim is Vi Improved. You can find references to vim on the /misc/gnu-win32 page. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Thu, 11 Dec 1997 17:49:37 -0600 >From: Rick Moll >To: gnu-win32 >CC: Rick Moll >Subject: newbie wants vi > >Sorry for the newbie question, but I've just got the gnu-win32 tools >installed and I was disappointed that there wasn't any kind of editor. >I'm totally new to all of this, gnu-win32, my WinNT system, and GNU in >general. > >I'm completely lost without a vi; what's the easiest way to get one to >use with the gnu-win32 tools. I understand that there is a "vim". Is >this a vi workalike? If so where's the easiest place to get it, and >will it easily port to the gnu-win32 environment? > >Thanks, > > Rick > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 12 05:19:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 12 Dec 1997 05:19:00 -0000 Subject: Basic question : I don't see man.exe (NBY) Message-ID: <19971212131920.25008.qmail@hotmail.com> Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Thu, 11 Dec 1997 17:52:28 +0100 >From: Gerard Milhaud >To: gnu-win32@cygnus.com >Subject: Basic question : I don't see man.exe > >Hi, > > I've just installed b18, and i'm very happy to have my gnu >environment on NT. But the commands (i.e. ls) are not exactly the same >so i've tried "man ls" bash answers me that it can't find man. I'm >wondering that man.exe is not in the bin directory where bash is. I've >seen all the man stuff in three directories in the distribution, so i >think there is a possibility to use it (and the info stuff so, because >there is not texinfo.exe or info.exe neither). > > Could anyone help me ? > >Thanks in advance. > >-- >Gerard MILHAUD >Ecole Superieure d'Ingenieurs de Luminy (ESIL) >Luminy - Case 925 - 13288 Marseille Cedex 09 >Tel.: 04 91 82 85 31 - Fax.: 04 91 82 85 11 > mailto:g@planete-marseille.esil.univ-mrs.fr > > > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From periat@ens.ascom.ch Fri Dec 12 06:06:00 1997 From: periat@ens.ascom.ch (Periat Henri) Date: Fri, 12 Dec 1997 06:06:00 -0000 Subject: newbie wants vi Message-ID: <199712121405.PAA15163@stva89.ens.ascom.ch> > From rickmoll@mvp.net Fri Dec 12 10:58:35 1997 > Date: Thu, 11 Dec 1997 17:49:37 -0600 > From: Rick Moll > Mime-Version: 1.0 > To: gnu-win32 > Cc: Rick Moll > Subject: newbie wants vi > Content-Transfer-Encoding: 7bit > > Sorry for the newbie question, but I've just got the gnu-win32 tools > installed and I was disappointed that there wasn't any kind of editor. > I'm totally new to all of this, gnu-win32, my WinNT system, and GNU in > general. > > I'm completely lost without a vi; what's the easiest way to get one to > use with the gnu-win32 tools. I understand that there is a "vim". Is > this a vi workalike? If so where's the easiest place to get it, and > will it easily port to the gnu-win32 environment? > > Thanks, > > Rick > > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". Hi Rick I have something even better then 'vi' 'vim'. You can get binaries here: http://www.pcisys.net/~kscott/vim/ If you don't know 'vim' yet: It's a improved 'vi' with syntax highliting, GUI and other goodis. The basic is the same as for 'vi'! Cheers Henri - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ywu@internic.net Fri Dec 12 08:57:00 1997 From: ywu@internic.net (Yibing Wu) Date: Fri, 12 Dec 1997 08:57:00 -0000 Subject: Bug Report: b18 path involves drive letter Message-ID: <34916235.C1E0091D@internic.net> I have two disks, named C: and E: respectively under NT. C:\ is the / directory of gnuwin32 in my case and the problem rose when I tried to access E: drive. For example, E:\ has a "WorkArea" directory, going there by "cd //E/WorkArea" was successful. However, it was impossible to go one level up from there. Neither "cd .." or "cd //E/" worked. Sometimes, the directory path did not change at all (reading from `pwd`), sometimes, the `pwd` reading was correct, but you knew you were not in the right place since the contents (reading from `ls`) were still the same as before. The only way around was to go back to / (C:\) then "cd //E/" or "cd -". Bing - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From William.Gacquer@obspm.fr Fri Dec 12 08:57:00 1997 From: William.Gacquer@obspm.fr (root) Date: Fri, 12 Dec 1997 08:57:00 -0000 Subject: problem with EGCS on gnu-win32 b18!! HELP Message-ID: <34917470.2ABCD359@obspm.fr> Hi! When I compile a C++ code with egcs (over gnuwin32 b18) on win95, I get no errors. (the code works fine on my linux box). But when I run the executable, I get an error message telling me that I miss the "abort.dll" file. What the hell could this be due to? Thanks! William Gacquer. PS : I promise I will help the EGCS team after my PhD!! (please wait one more year!) - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 12 09:07:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 12 Dec 1997 09:07:00 -0000 Subject: Newbie: Porting Unix programs to Win95. (NBY) Message-ID: <19971212170705.11650.qmail@hotmail.com> Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >To: gnu-win32@cygnus.com >Subject: Newbie: Porting Unix programs to Win95. >Date: Fri, 12 Dec 1997 11:59:28 +0200 >From: Livhu Tshisikule > >Hi, > >I am new to gnuwin32, and I would like to port ssh to win95, can any one tell >me sites where I can get info about porting unix programs? > >Thanks >Livhu > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 12 09:17:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 12 Dec 1997 09:17:00 -0000 Subject: Definition of VPATH Message-ID: <19971212171622.7205.qmail@hotmail.com> Hi Henri, Don't know if this will work but here is a suggestion: VPATH:=/home/periat/src:D\:\\foo\\src - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Fri, 12 Dec 1997 10:52:28 +0100 >From: periat@ens.ascom.ch (Periat Henri) >Subject: Definition of VPATH >To: gnu-win32@cygnus.com > > >Hi everybody > >I posted this message a few days ago, but i didn't go much reply on it. > >I'm very new to cygnu and have to make a UNIX gmake file to run also >on MS-NT boxes. >Now the problem which stops me in the moment is the ':' problem: > >- with 'VPATH:=/home/periat/src:D:/foo/src' make can't find the > source files if they are on drive 'D:' only. > >- with 'VPATH:=/home/periat/src;D:/foo/src' or > 'VPATH:=/home/periat/src://D/foo/src' > '$<' contains '//D/foo/src/foo.c' and non cygnus programs can't deal > with such file names. > >- with 'VPATH:=/home/periat/src:/foo/src' and 'D:/foo' binary mounted > as '/foo' > 'make' finds files, but again non cygnus programs don't know anything > about 'mount' and so they can't find '/fool/file' > > >Can anyone tell me how i can solve this problem? > > > >Thanks >Henri > > > >I have b18 and the 'coolview' patches. > > > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". >------------- End Forwarded Message ------------- > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mark.n@usa.net Fri Dec 12 13:52:00 1997 From: mark.n@usa.net (Mark Newnham) Date: Fri, 12 Dec 1997 13:52:00 -0000 Subject: libdbm.a Message-ID: <95DE24D1FCFED0118B910060940A334004C0@mailhub.adacorp.com> Does anyone have a libdbm.dll , .h and .a available that I could have please.... Regards Mark ################################################## # Application Design Associates, Inc # # 6021 S.Syracuse Way #302 # # Englewood CO 80111 Tel +1 (303)779 8958 # # USA Fax +1 (303)779 0949 # ################################################## - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From skachman@unl.edu Fri Dec 12 14:40:00 1997 From: skachman@unl.edu (Steve Kachman) Date: Fri, 12 Dec 1997 14:40:00 -0000 Subject: WFSO failure running make in emacs Message-ID: <3.0.5.32.19971212142535.007ffb40@popmail.unl.edu> When I run make in emacs I get the following error gcc -g -O1 -I../../include -I. -fcheck-new -ffor-scope -c lu.cc (F:\USR\H-I386-CYGWIN32\BIN\MAKE.EXE 1117) sig_send: WFSO error, pid 1117, rc 258, signal 0, last signal -1, error 0 gcc -g -O1 -I../../include -I. -fcheck-new -ffor-scope -c ginverse1.cc gcc -g -O1 -I../../include -I. -fcheck-new -ffor-scope -c symm_eigen.cc gcc -g -O1 -I../../include -I. -fcheck-new -ffor-scope -c nonsymm_eigen.cc gcc -g -O1 -I../../include -I. -fcheck-new -ffor-scope -c sweep.cc It happens with a variety of source files and always on the first one. I am using Sergey Okhapkin's coolview cygwin32 and GNU Emacs 19.34.6 (i386-*-nt4.0) of Thu Sep 25 1997 on ESME. Any suggestions? Steve Kachman skachman@unl.edu - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From noer@cygnus.com Fri Dec 12 14:40:00 1997 From: noer@cygnus.com (Geoffrey Noer) Date: Fri, 12 Dec 1997 14:40:00 -0000 Subject: Shell script performance References: <01BD063B.E880E380@rodney.wg.icl.co.uk> Message-ID: <199712122238.OAA11970@cirdan.cygnus.com> Kevin Hughes wrote: [...] > Does anyone have any ideas if anything can/is being done? At this point, performance is an active concern. Everything's a lot faster than it was in beta 18 or earlier but there's still a ways to go. -- Geoffrey Noer noer@cygnus.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From lordgate@mail.geocities.com Fri Dec 12 17:21:00 1997 From: lordgate@mail.geocities.com (LordGate [Ste]) Date: Fri, 12 Dec 1997 17:21:00 -0000 Subject: what is coolview? Message-ID: <199712130109.RAA11891@geocities.com> I keep hearing about it but never see where to get it or what it is.... so what is it, where is it, and is it any good? Thanks Free Compiler List: http://right.ml.org/srobin1/compilers/ Personal Homepage: http://right.ml.srobin1/lordgate/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dahms@ifk20.mach.uni-karlsruhe.de Fri Dec 12 17:21:00 1997 From: dahms@ifk20.mach.uni-karlsruhe.de (dahms@ifk20.mach.uni-karlsruhe.de) Date: Fri, 12 Dec 1997 17:21:00 -0000 Subject: "read" routine! Message-ID: <009BEAF8.A91D2440.30151@ifk20.mach.uni-karlsruhe.de> Hi Hasmik, you wrote: : My application starts a Tcl shell and if I issue a Ctrl-C from : the debugger it exits the program. I have debugged this to the point : that I think the "read" routine is not behaving like the "read" on the unix : system. The "read" on NT system returns a 0 (zero) and the Tcl application : treats that as the end-of-file and exits the program. Shouldn't the "read" : routine return some sort of error (EINTR)?? I guess it would, but first the default signal handler terminates it. You should ignore SIGINT or install a handler, maybe which writes "ouch!" to fd 2 (stderr). After ^c, (my) programs by default die under Unix, too, and so does tclsh. And I'm pretty shure tclsh retries after EINTR. If (debugger == gdb), then under Unix, it doesn't pass SIGINT to the debuggee by default. Check output of "info signals" on cygwin32 and Unix... Bye, Heribert (dahms@ifk20.mach.uni-karlsruhe.de) - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Fri Dec 12 17:31:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Fri, 12 Dec 1997 17:31:00 -0000 Subject: problem with EGCS on gnu-win32 b18!! HELP References: <34917470.2ABCD359@obspm.fr> Message-ID: <9712130130.AA05682@modi.xraylith.wisc.edu> root writes: > Hi! > When I compile a C++ code with egcs (over gnuwin32 b18) on win95, I get > no errors. (the > code works fine on my linux box). But when I run the executable, I get > an error > message telling me that I miss the "abort.dll" file. What the hell could > this be due to? Dumping egcs right over gnu-win32 b18 is probably idea, given that others seem to be having trouble as well. One thing to watch out for is that you CANNOT use import libraries (eg., kernel32.a or libcygwin32.a) that were created with b18 tools. You must use the egcs supplied ones. If you're using Sergey's Coolview, delete the libcygwin.a that came with coolview and replace with egcs supplied one. If you're not using coolview, please get back to me and we can figure out where the problem is. Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hgfernan@usp.br Fri Dec 12 18:22:00 1997 From: hgfernan@usp.br (hgfernan@usp.br) Date: Fri, 12 Dec 1997 18:22:00 -0000 Subject: how to avoid mulitple definitions of STL free_list, start_fr References: <348ABB53.62AF@asc.on.ca> <9712090815.AA29183@modi.xraylith.wisc.edu> Message-ID: <9712130157.AA48056@spider.uspnet.usp.br> On 9 Dec 97 at 2:15, Mumit Khan wrote: > Now that my obligatory plug is out the way, the problem is that the older > versions of C++ front-end simply couldn't handle static template data > members properly, and hence the problem. I can suggest a few workarounds: > > 1. Use -fno-implicit and instantiate everything yourself. I cover a bit > about explicit instantiation in my (now grossly outdated) STL Newbie > guide (URL below). This is the only portable (ie., something that'll > work on *all* platforms) way to do templates with gcc, even with the > newest and greatest egcs release. > 2. Comment out the default_alloc_template definition in alloc.h (or > whatever file it's in in $prefix/include/g++) and supply the > definition yourself in one of your object files. > 3. Did I mention that you should upgrade your compiler ;-) > Another alternative is the page "Portable STL - Note for GCC users", at http://www.metabyte.com/~fbp/stl/READ ------------------- Hilton Fernandes email: hgfernan@usp.br hilton@lsi www: http://www.lsi.usp.br/~hilton.html M. Sc. Student of Parallel Distributed Applications at LSI (Laboratory of Integrated Systems) University of S. Paulo - Brazil - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From bzinger@iName.com Fri Dec 12 18:22:00 1997 From: bzinger@iName.com (Micheal A. Benzinger) Date: Fri, 12 Dec 1997 18:22:00 -0000 Subject: newbie wants vi References: <34907C11.3D6771CA@mvp.net> Message-ID: <3.0.3.32.19971212202247.00914690@pop.flash.net> Rick, You can get a beta version of VIM that compiles right out of the box on GNU-Win32 and it has color syntax highlighting. Here's the URL: ftp://ftp.oce.nl/pub/misc/vim/beta-test/unix/vim-5.0s.tar.gz If you do download and compile it, do the following before installing: export SUFFIX=.exe Mike Benzinger At 05:49 PM 12/11/97 -0600, Rick Moll wrote: >Sorry for the newbie question, but I've just got the gnu-win32 tools >installed and I was disappointed that there wasn't any kind of editor. >I'm totally new to all of this, gnu-win32, my WinNT system, and GNU in >general. > >I'm completely lost without a vi; what's the easiest way to get one to >use with the gnu-win32 tools. I understand that there is a "vim". Is >this a vi workalike? If so where's the easiest place to get it, and >will it easily port to the gnu-win32 environment? > >Thanks, > > Rick > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hgfernan@usp.br Fri Dec 12 20:26:00 1997 From: hgfernan@usp.br (hgfernan@usp.br) Date: Fri, 12 Dec 1997 20:26:00 -0000 Subject: (newbie) please be kind... References: <3.0.3.32.19971208215231.007ee900@apci.net> Message-ID: <9712130157.AA48066@spider.uspnet.usp.br> On 8 Dec 97 at 21:52, shorti wrote: > hello all: > > ok - here's the deal: i have very limited experience with c, although i'm > figuring out the text-driven stuff fairly well. i would like to write > windows programs using my new found resources, but i know next to nothing > about API, integrating dll's, etc. does anyone know of any good > tutorials/references on win32 (or c in general) on the web? maybe if you > have nothing better to do over christmas holiday, you could write one? or > is it just easier for me to go out and buy a book? If you have a list of > links related to (c, win32) programming, it would be greatly appreciated.. > i'll compile it to a webpage somewhere... Hi! Concerning C, you got to read the founding fathers: Kernighan & Ritchie's The C Programming Language. Other interesting book, if a little out of date is Holub's The C Companion, that explains intuitively and in detail things about pointers et al. that we only learn after 10 years of C programming. :-) Concerning the Win32 API, DLLs etc., there's another book almost as good as K & R: it is Programming Windows 95, by Charles Petzold with Paul Yao. It was a classic for the first Windows (2.X & 3.X). Now it shows its age, but it is probably the best introduction to the subject available. Concerning Web sites, please take a look at the page Hello Windows!, at http://www.relisoft.com/win32/winnie.html by Reliable Software. The page http://www.relisoft.com/index.htm gives access to other interesting things at their site. Good links can be found at Webopaedia http://www.pcwebopedia.com/ and whatis.com http://www.whatis.com/ Regards, ++Hilton ------------------- Hilton Fernandes email: hgfernan@usp.br hilton@lsi www: http://www.lsi.usp.br/~hilton.html M. Sc. Student of Parallel Distributed Applications at LSI (Laboratory of Integrated Systems) University of S. Paulo - Brazil - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Fri Dec 12 20:26:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Fri, 12 Dec 1997 20:26:00 -0000 Subject: Re; One way to get around the cygwin terminal problem Message-ID: <199712130426.EAA81212@out2.ibm.net> vischne@ibm.net wrote: > > In other words, with some work, it is possible to get a professional > looking and acting terminal application using the cygwin Unix > environment but bypassing the cygwin terminal window. It's a bad idea to replace stdout with direct console writings. You application, for example, will not work in a telnet session or in xterm window. I don't see the point of an xterm window if you are running a Unix environment under Windows 95. Simply calling up bash.exe ought to generate the equivalent of an xterm window. But, running lynx or slrn or midnight commander in the current version of a cygwin Windows/xterm window just doesn't work right. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From pgarceau@submergedsoftware.com Fri Dec 12 21:42:00 1997 From: pgarceau@submergedsoftware.com (Paul Garceau) Date: Fri, 12 Dec 1997 21:42:00 -0000 Subject: Generating gnuwin32 compatible headers Message-ID: <199712130540.AAA06695@bw.webex.net> Hi everyone, Recently, myself and two others have been playing with the possibility of generating headers for gnuwin32 that can, with little or no effort on the part of the cross-platform developer, generate those headers based on information found within non-gnuwin32 based header collections, specifically those defined as C/C++ development components. After some discussion, it was agreed that this might be something worth pursuing on the mailing list. In my mind, two scenarios arose: 1) Create a tool that automatically patches a set of external header files based on a set of external header files which have not been generated or formatted to accomodate gnu-win32 header format. 2) Create a tool that is smart enough to generate GNU-Win32 compatible header files based on a set of external header files that are not formatted to accomodate a gnu-win32 header format. Can these things be accomplished? If so, where do we go from here? Paul G. Chronicles of Aurum: Book I A Journey Into the Fantastic http://www.submergedsoftware.com/ ICQ#977283 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From pgarceau@submergedsoftware.com Fri Dec 12 21:42:00 1997 From: pgarceau@submergedsoftware.com (Paul Garceau) Date: Fri, 12 Dec 1997 21:42:00 -0000 Subject: Generating gnuwin32 compatible headers Message-ID: <199712130540.AAA06613@bw.webex.net> Hi everyone, Recently, myself and two others have been playing with the possibility of generating headers for gnuwin32 that can, with little or no effort on the part of the cross-platform developer, generate those headers based on information found within non-gnuwin32 based header collections, specifically those defined as C/C++ development components. After some discussion, it was agreed that this might be something worth pursuing on the mailing list. In my mind, two scenarios arose: 1) Create a tool that automatically patches a set of external header files based on a set of external header files which have not been generated or formatted to accomodate gnu-win32 header format. 2) Create a tool that is smart enough to generate GNU-Win32 compatible header files based on a set of external header files that are not formatted to accomodate a gnu-win32 header format. Can these things be accomplished? If so, where do we go from here? Paul G. Chronicles of Aurum: Book I A Journey Into the Fantastic http://www.submergedsoftware.com/ ICQ#977283 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sat Dec 13 06:26:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sat, 13 Dec 1997 06:26:00 -0000 Subject: One way to get around the cygwin terminal window Message-ID: <199712131426.OAA13754@out1.ibm.net> >> I don't see the point of an xterm window if you are running a Unix environment under Windows 95. Simply calling up bash.exe ought to generate the equivalent of an xterm window. But, running lynx or slrn or midnight commander in the current version of a cygwin Windows/xterm window just doesn't work right. << One more point: While a properly implemented xterm window ought to run applications that use console escape sequences to move and color the cursor better than the current b18 cygwin terminal, it is implemented using so many layers of interpreted Windows code, including the code in the server, that the resulting slowdown seems like an excessive time penalty. If you really want complete Unix compatibility for remote communications via X-windows, then you can _still_ configure libslang.a as a pure Unix library under cygwin by replacing slw32tty.c by slutty.c. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From v.canestrale@isnet.it Sat Dec 13 08:38:00 1997 From: v.canestrale@isnet.it (Vincenzo Canestrale) Date: Sat, 13 Dec 1997 08:38:00 -0000 Subject: Newbie Problem with Linker! Message-ID: <16363979629779@isnet.it> Hi to all, I think mine is a very basic question..... I'd like to compile a program that opens a simple window in Windows95. I obtain the object file..... but when i had to link my file Prog.o with the library the linker don't know where to find the library. My file include: #include I tried gcc -c c:\Prog.C [OK] ld c:\Prog.o [Doesn't work :-(] Which is the right command line? Thanks. v.canestrale@isnet.it http://www.ipsnet.it/netpages/room466 (non attiva) http://www.geocities.com/WallStreet/5385 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Sat Dec 13 17:08:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Sat, 13 Dec 1997 17:08:00 -0000 Subject: WFSO failure running make in emacs References: <3.0.5.32.19971212142535.007ffb40@popmail.unl.edu> Message-ID: In article < 3.0.5.32.19971212142535.007ffb40@popmail.unl.edu >, Steve Kachman wrote: >When I run make in emacs I get the following error > > >gcc -g -O1 -I../../include -I. -fcheck-new -ffor-scope -c lu.cc >(F:\USR\H-I386-CYGWIN32\BIN\MAKE.EXE 1117) sig_send: WFSO error, pid 1117, >rc 258, signal 0, last signal -1, error 0 >gcc -g -O1 -I../../include -I. -fcheck-new -ffor-scope -c ginverse1.cc >gcc -g -O1 -I../../include -I. -fcheck-new -ffor-scope -c symm_eigen.cc >gcc -g -O1 -I../../include -I. -fcheck-new -ffor-scope -c nonsymm_eigen.cc >gcc -g -O1 -I../../include -I. -fcheck-new -ffor-scope -c sweep.cc > >It happens with a variety of source files and always on the first one. I am >using Sergey Okhapkin's coolview cygwin32 and GNU Emacs 19.34.6 >(i386-*-nt4.0) of Thu Sep 25 1997 on ESME. > >Any suggestions? Unfortunately, the only suggestion that I have is to wait for B19. The error you are seeing is from code that I contributed to Cygnus. It has an occasional problem with execed processes that are (very hopefully) fixed in B19. I'm sorry that I don't have better news, but there have been substantial changes to the code and back porting things from B19 to B18 would be difficult. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dunkel@tocsy.chem.utah.edu Sat Dec 13 20:07:00 1997 From: dunkel@tocsy.chem.utah.edu (Reinhard Dunkel) Date: Sat, 13 Dec 1997 20:07:00 -0000 Subject: How to set directory time stamps? Message-ID: <199712140407.VAA22002@tocsy.chem.utah.edu> I am trying to port a program which depends on being able to update directory time stamps. Using GNU-Win32 b18 (with various patches applied), updating the time stamps of a plain file works. But I did not succeed to update the time stamps of a directory (Windows NT 4.0 on Pentium II, FAT filesystem): bash-2.01$ mkdir TEST bash-2.01$ touch TEST touch: TEST: Permission denied The same problem appears for a C program using: utime(directoryName, NULL); (perror() returns something like "Is a directory".) NuTCRACKER fails too, Microsoft's "_utime()" doesn't work, but the MKS Toolkit "touch" command works flawlessly (...so it is do-able after all!). Somebody indicated to me that using "backup semantics to open the directory" and using "SetFileTime on that handle" should work. Well, I just entered the Microsoft World of programming. Part two sounds familiar, but what is "backup semantics"? Any help on solving the mystery of NT directory time stamp updates would be greatly appreciated! -- __ _ | ) o | | | \ | | | / _ |_ _ _ _| | | | _ | |/\ /_) | ^ ^ | | _) | / | | | | | ^ ^ |_) /_) | | \_/\___|_| | |_| |_(_|__|___\_| |_/__|_|_| | |_| \_/\___| Department of Chemistry Office: (801) 581-5894 2020 Henry Eyring Building b125 Home: (801) 582-7516 University of Utah FAX: (801) 581-8433 Salt Lake City, UT 84112 dunkel@chemistry.utah.edu - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sat Dec 13 21:16:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sat, 13 Dec 1997 21:16:00 -0000 Subject: Experimental slangwin.tar.gz at deninc.com Message-ID: <199712140516.FAA106000@out4.ibm.net> While it's there, you can get the experimental slang with console buffer support source at the incoming directory of ftp.deninc.com.us. You can also get the very latest development lynx.exe.gz in that directory (supply your own lynx.cfg, or lynx will). P. S. Maybe this is already known, but compiling without the -g option still leaves executables that shrink when stipped. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hgfernan@usp.br Sun Dec 14 02:13:00 1997 From: hgfernan@usp.br (hgfernan@usp.br) Date: Sun, 14 Dec 1997 02:13:00 -0000 Subject: Newbie Problem with Linker! References: <16363979629779@isnet.it> Message-ID: <9712141016.AA42696@spider.uspnet.usp.br> On 13 Dec 97 at 16:49, Vincenzo Canestrale wrote: > From: "Vincenzo Canestrale" > To: > Subject: Newbie Problem with Linker! > Date: Sat, 13 Dec 1997 16:49:12 +0100 > Hi to all, > > I think mine is a very basic question..... > > I'd like to compile a program that opens a > simple window in Windows95. Hi, Vicenzo! In the Cygnus page http://www.cygnus.com/misc/ there's reference to a help page (FAQ) you should consult. Programming MS Windows is a little difficult :-), but you will find some help in the page "Introduction to the Win32 Application Programming Interface", at http://www.microsoft.com/WIN32DEV/APIEXT/INTRO32.HTM Microsoft created this mess, so it should explain it. :-) A complete "Hello Windows" program is at http://www.relisoft.com/win32/winnie.html that contains indications to an almost complete practical Windows programming course. A must read. If we talk about books, Programming Windows 95, by Petzold & Yao, is still the best, as far as i know. Not related to Cygnus Gnu-Win32 -- a. k. a. cygwin --, is the excellent DJGPP FAQ, by Eli Zaretski et al. Maybe you know that DJGPP is -- like cygwin -- another implementation of GCC for the MS world. Actually DJGPP is for MS-DOS, but the FAQ has almost everything you should know to compile a program under GCC -- or a version of it. Please find it somewhere in http://www.delorie.com Follow the (easy to follow) links in this site and you'll get there. :-) Best regards, ++Hilton ------------------- Hilton Fernandes email: hgfernan@usp.br hilton@lsi www: http://www.lsi.usp.br/~hilton.html M. Sc. Student of Parallel Distributed Applications at LSI (Laboratory of Integrated Systems) University of S. Paulo - Brazil - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Sun Dec 14 05:00:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Sun, 14 Dec 1997 05:00:00 -0000 Subject: Newbie Problem with Linker! References: <16363979629779@isnet.it> Message-ID: <9712141300.8535@mulga.cs.mu.OZ.AU> Vincenzo Canestrale, you wrote: > I tried > gcc -c c:\Prog.C [OK] > ld c:\Prog.o [Doesn't work :-(] > > Which is the right command line? You should use `gcc' (or, for C++ programs, g++) to do the linking, rather than `ld'. Instead of `ld Prog.o' you should use `gcc Prog.o'. -- Fergus Henderson | Designing grand concepts is fun; fjh@cs.mu.oz.au | finding nitty little bugs is just work. http://www.cs.mu.oz.au/~fjh | -- Brooks, in "The Mythical Man-Month". PGP key fingerprint: 00 D7 A2 27 65 09 B6 AC 8B 3E 0F 01 E7 5D C4 3F - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Sun Dec 14 10:24:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Sun, 14 Dec 1997 10:24:00 -0000 Subject: Experimental slangwin.tar.gz at deninc.com References: <199712140516.FAA106000@out4.ibm.net> Message-ID: In article < 199712140516.FAA106000@out4.ibm.net >, wrote: >P. S. Maybe this is already known, but compiling without >the -g option still leaves executables that shrink when >stipped. This is the case with every version of cc or gcc that I've ever seen. ld always puts symbols in the file unless you link with '-s'. When you specify -g it only means that more debugging information may be placed in the final executable. This extra information is usually things like line number info and possibly symbol entries for static variables. Without -g you'll probably only see global symbols in your executable. Exactly what is left in the executable is visible by using the `nm' command. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Sun Dec 14 12:02:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Sun, 14 Dec 1997 12:02:00 -0000 Subject: what is coolview? References: <199712130109.RAA11891@geocities.com> Message-ID: <6jqBx4WKpfB@mike.franken.de> Hi, >I keep hearing about it but never see where to get it or what it >is.... so what is it, where is it, and is it any good? Mainly it is an improved cygwin.dll by Sergey Okhapkin. You can get it at http://miracle.geol.msu.ru/sos/ ! >Thanks Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jan.dubois@ibm.net Sun Dec 14 12:06:00 1997 From: jan.dubois@ibm.net (Jan Dubois) Date: Sun, 14 Dec 1997 12:06:00 -0000 Subject: How to set directory time stamps? References: <199712140407.VAA22002@tocsy.chem.utah.edu> Message-ID: <34963b25.33670766@smtp1.ibm.net> On Sat, 13 Dec 1997 21:07:51 -0700 (MST), you wrote: >I am trying to port a program which depends on being able to update >directory time stamps. Using GNU-Win32 b18 (with various patches >applied), updating the time stamps of a plain file works. But I >did not succeed to update the time stamps of a directory (Windows NT >4.0 on Pentium II, FAT filesystem): > > bash-2.01$ mkdir TEST > bash-2.01$ touch TEST > touch: TEST: Permission denied > >The same problem appears for a C program using: > > utime(directoryName, NULL); > >(perror() returns something like "Is a directory".) NuTCRACKER fails >too, Microsoft's "_utime()" doesn't work, but the MKS Toolkit "touch" >command works flawlessly (...so it is do-able after all!). Somebody >indicated to me that using "backup semantics to open the directory" >and using "SetFileTime on that handle" should work. Well, I just >entered the Microsoft World of programming. Part two sounds familiar, >but what is "backup semantics"? Any help on solving the mystery of NT >directory time stamp updates would be greatly appreciated! Windows NT: You can obtain a handle to a directory by calling the CreateFile function with the FILE_FLAG_BACKUP_SEMANTICS flag set, as follows: hDir = CreateFile ( DirName, GENERIC_READ, FILE_SHARE_READ|FILE_SHARE_DELETE, NULL, OPEN_EXISTING, FILE_FLAG_BACKUP_SEMANTICS, NULL ); You can pass a directory handle to the following SetFileTime function. -Jan - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From VZELL@de.oracle.com Sun Dec 14 13:08:00 1997 From: VZELL@de.oracle.com (Volker Zell) Date: Sun, 14 Dec 1997 13:08:00 -0000 Subject: UNIX file command ??? Message-ID: <199712142109.WAA10710@mailm1> Hi folks Does anybody know about the source code location of the UNIX file command. I need it for Midnight Commander. Or can anybody provide me with a cygwin b18 compiled binary ? Ciao Volker - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hollo@freemail.c3.hu Sun Dec 14 14:50:00 1997 From: hollo@freemail.c3.hu (Zoltan Pasztor) Date: Sun, 14 Dec 1997 14:50:00 -0000 Subject: Bugs with gdb and bash ; Message-ID: <00e501bd08e2$a7c9e720$0100007f@localhost> Hi Here are attached two dump's zipped, (in DrWatson-s WLG format). Cygwin b18 on Win98 beta. One is at closing of gdb The another is if a try ./configure or make, the bash caused the fault. I am not on the gnu-win list. Regards Zoltan -------------- next part -------------- A non-text attachment was scrubbed... Name: cygdump.zip Type: application/zip Size: 41269 bytes Desc: not available URL: From earnie_boyd@hotmail.com Sun Dec 14 15:30:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Sun, 14 Dec 1997 15:30:00 -0000 Subject: Bug Report: b18 path involves drive letter (NBY) Message-ID: <19971214233009.23962.qmail@hotmail.com> You need to: mount -b E:\\ /e then you can: cd /e pwd /e cd .. pwd / and be in the correct places. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Fri, 12 Dec 1997 11:11:33 -0500 >From: Yibing Wu >Reply-To: ywu@internic.net >To: gnu-win32@cygnus.com >Subject: Bug Report: b18 path involves drive letter > >I have two disks, named C: and E: respectively under NT. >C:\ is the / directory of gnuwin32 in my case and the problem rose when >I tried to access E: drive. > >For example, E:\ has a "WorkArea" directory, going there by "cd >//E/WorkArea" was successful. However, it was impossible to go one >level up from there. Neither "cd .." or "cd //E/" worked. Sometimes, >the directory path did not change at all (reading from `pwd`), >sometimes, the `pwd` reading was correct, but you knew you were not in >the right place since the contents (reading from `ls`) were still the >same as before. The only way around was to go back to / (C:\) then "cd >//E/" or "cd -". > >Bing > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Sun Dec 14 15:35:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Sun, 14 Dec 1997 15:35:00 -0000 Subject: what is coolview? (NBY) Message-ID: <19971214233508.24301.qmail@hotmail.com> Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: "LordGate [Ste]" >To: gnu-win32@cygnus.com >Date: Fri, 12 Dec 1997 20:17:48 +0000 >Subject: what is coolview? > >I keep hearing about it but never see where to get it or what it >is.... so what is it, where is it, and is it any good? >Thanks > >Free Compiler List: > http://right.ml.org/srobin1/compilers/ > >Personal Homepage: > http://right.ml.srobin1/lordgate/ >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Sun Dec 14 15:45:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Sun, 14 Dec 1997 15:45:00 -0000 Subject: Newbie Problem with Linker! (NBY) Message-ID: <19971214234429.1132.qmail@hotmail.com> >From: "Vincenzo Canestrale" >To: >Subject: Newbie Problem with Linker! >Date: Sat, 13 Dec 1997 16:49:12 +0100 > >Hi to all, > >I think mine is a very basic question..... > >I'd like to compile a program that opens a >simple window in Windows95. > >I obtain the object file..... but >when i had to link my file Prog.o with the library >the linker don't know where to find the library. > >My file include: >#include > >I tried >gcc -c c:\Prog.C [OK] >ld c:\Prog.o [Doesn't work :-(] instead of ld do: gcc -o c:\Prog.exe c:\Prog.o > >Which is the right command line? > >Thanks. > >v.canestrale@isnet.it > http://www.ipsnet.it/netpages/room466 (non attiva) > http://www.geocities.com/WallStreet/5385 >- Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From raf@comdyn.com.au Sun Dec 14 19:11:00 1997 From: raf@comdyn.com.au (raf) Date: Sun, 14 Dec 1997 19:11:00 -0000 Subject: Help, configure error!!! Message-ID: <199712150310.OAA07211@mali.comdyn.com.au> >I am not familiar with gnuwin32, when I run configure I am getting the >following error: >checking host system type... Invalid configuration 'i386-unknown-cygwin32': >system 'cygwin32' not recognized >What should I do? download the latest autoconf or automake from gnu and replace every instance you have of the files 'config.sub' 'config.guess' with the ones in either of these packages. these are the files that 'guess' the platform and the latest version knows about cygwin32. raf - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From raf@comdyn.com.au Sun Dec 14 19:17:00 1997 From: raf@comdyn.com.au (raf) Date: Sun, 14 Dec 1997 19:17:00 -0000 Subject: Newbie: Porting Unix programs to Win95. Message-ID: <199712150317.OAA07273@mali.comdyn.com.au> >I am new to gnuwin32, and I would like to port ssh to win95, can any one tell >me sites where I can get info about porting unix programs? look at http://www.zip.com.au/~raf2/lib/ssh/ which contains patches and instructions for compiling the ssh-1.2.21 client (only tested on NT). look at the http://www.cygnus.com/misc/gnu-win32/ (Related Sites) for links to general resources on porting unix programs. raf - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Sun Dec 14 21:15:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Sun, 14 Dec 1997 21:15:00 -0000 Subject: [ANN] updated windres for cygwin32 and mingw32 Message-ID: <9712150515.AA07257@modi.xraylith.wisc.edu> If you use windres, you might want to pick up the latest version for both cygwin32 and mingw32. This is for egcs-1.00, and I have no idea if it works with b18 cdk (the original development tools from Cygnus). Go to http://www.xraylith.wisc.edu/~khan/software/gnu-win32/egcs.html and follow the fixes/updates link. ftp://ftp.xraylith.wisc.edu/pub/khan/gnu-win32/cygwin32/releases/fixes for direct ftp access. Replace "cygwin32" with "mingw32" if you're using latter distribution. enjoy, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From russellm@iName.com Sun Dec 14 21:21:00 1997 From: russellm@iName.com (Russell K Montgomery) Date: Sun, 14 Dec 1997 21:21:00 -0000 Subject: Mixed mounting doesn't seem to work Message-ID: <3494BE46.7C17759D@iname.com> I have installed CDK.EXE (B18) and Coolview on Windows 95B.???? I would prefer to work in "mixed mode" because of several scripts that I have brought over from a SCO machine that uses the same name but with differently capitalized filenames. I have modified all mount points to be both mixed and binary.???? mount outputs the following information: Device???????????????????????????????????????? Directory???????????????????????????????????????? Type???????????????????????????? Flags G:\emacs-19.34???????? /emacs???????????????????????????????????????????????????? native???????????????????? mixed,text=binary \\.\tape1:???????????????????????? /dev/st1???????????????????????????????????????????? native???????????????????? mixed,text=binary \\.\tape0:???????????????????????? /dev/st0???????????????????????????????????????????? native???????????????????? mixed,text=binary \\.\b:???????????????????????????????????????? /dev/fd1???????????????????????????????????????????? native???????????????????? mixed,text=binary \\.\a:???????????????????????????????????????? /dev/fd0???????????????????????????????????????????? native???????????????????? mixed,text=binary G:\GNU-Win-32???????????? /???????????????????????????????????????????????????????????????????????? native???????????????????? mixed,text=binary The basic method I used to validate if it is working is: bash.exe-2.01$ echo "HELLO" > test bash.exe-2.01$ echo "Hi" > Test bash.exe-2.01$ ls -l -rw-r--r--???????? 1 500???????????????????? everyone???????????????????????????? 3 Dec 14 23:18 test If someone could provide any input or is successfully using "Mixed" mounting I would appreciate any feedback you could offer. ---- Russell Montgomery Austin, TX, USA ???? From vischne@ibm.net Sun Dec 14 21:52:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sun, 14 Dec 1997 21:52:00 -0000 Subject: problem with EGCS on gun-win32 b18!! HELP Message-ID: <199712150552.FAA53122@out5.ibm.net> root writes: > Hi! > When I compile a C++ code with egcs (over gnuwin32 b18) on win95, I get > no errors. (the > code works fine on my linux box). But when I run the executable, I get > an error > message telling me that I miss the "abort.dll" file. What the hell could > this be due to? Dumping egcs right over gnu-win32 b18 is probably idea, given that others seem to be having trouble as well. One thing to watch out for is that you CANNOT use import libraries (eg., kernel32.a or libcygwin32.a) that were created with b18 tools. You must use the egcs supplied ones. That's interesting. The first version of ming were compatible with kernel32. I wonder why it is necessary to have two separate setups, one cygwin and one mingw32, when the cygwin ./configure and make capabilities are so much better at this time. It seems to make more sense to have ming as an auxiliary system, and use the cygwin tools to run ming. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From periat@ens.ascom.ch Mon Dec 15 00:59:00 1997 From: periat@ens.ascom.ch (Periat Henri) Date: Mon, 15 Dec 1997 00:59:00 -0000 Subject: Definition of VPATH Message-ID: <199712150859.JAA15761@stva89.ens.ascom.ch> > From earnie_boyd@hotmail.com Sat Dec 13 09:56:56 1997 > X-Originating-Ip: [199.186.21.253] > From: "Earnie Boyd" > To: periat@ens.ascom.ch > Cc: gnu-win32@cygnus.com > Subject: Re: Definition of VPATH > Date: Fri, 12 Dec 1997 09:16:21 PST > > Hi Henri, > > Don't know if this will work but here is a suggestion: > > VPATH:=/home/periat/src:D\:\\foo\\src > > - \\||// > ---o0O0--Earnie--0O0o---- > -earnie_boyd@hotmail.com- > ------ooo0O--O0ooo------- > > >Date: Fri, 12 Dec 1997 10:52:28 +0100 > >From: periat@ens.ascom.ch (Periat Henri) > >Subject: Definition of VPATH > >To: gnu-win32@cygnus.com > > > > > >Hi everybody > > > >I posted this message a few days ago, but i didn't go much reply on it. > > > >I'm very new to cygnu and have to make a UNIX gmake file to run also > >on MS-NT boxes. > >Now the problem which stops me in the moment is the ':' problem: > > > >- with 'VPATH:=/home/periat/src:D:/foo/src' make can't find the > > source files if they are on drive 'D:' only. > > > >- with 'VPATH:=/home/periat/src;D:/foo/src' or > > 'VPATH:=/home/periat/src://D/foo/src' > > '$<' contains '//D/foo/src/foo.c' and non cygnus programs can't deal > > with such file names. > > > >- with 'VPATH:=/home/periat/src:/foo/src' and 'D:/foo' binary mounted > > as '/foo' > > 'make' finds files, but again non cygnus programs don't know anything > > about 'mount' and so they can't find '/fool/file' > > > > > >Can anyone tell me how i can solve this problem? > > > > > > > >Thanks > >Henri > > > > > > > >I have b18 and the 'coolview' patches. > > Hi Earnie Thanks for your repeated efforts. But your suggestion doesn't work, the file still can't be found. Version 3.76.1 of 'GNU make' solves this problem, but it has big problems with '"' in combination with cygnus 'bash', even '@echo "Hello world"' doesn't work. This problem seems to be known and Rob Tulloh recommended me to load the latest patches (a, b, c and d). But i was not able to install these patches and get a running, patched 'make'. In 'gnu.utils.help' i asked for a binary but got no response yet. Henri - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From neil@riskware.co.uk Mon Dec 15 05:21:00 1997 From: neil@riskware.co.uk (neil@riskware.co.uk) Date: Mon, 15 Dec 1997 05:21:00 -0000 Subject: Definition of VPATH Message-ID: <9712151311.AA19842@> Henri, I too am very new to cygnu but like you I am trying to build Unix source on NT so I recognise the VPATH problems you have. The difference is that we use Imake - we have one file that defines all the paths, that file sets up a "root" variable which is platform dependent, all paths then include this root variable so the paths get changes depending on the platform. Could you do something similar with gmake? More info on imake from http://www.primate.wisc.edu/software/imake-stuff or the O-Reilly book "Software Portability with imake". If you wish to reply, please do so directly - I just unsubscribed from the list to avoid a huge mailbox on return from the Christmas break. Regards, Neil J Mackie. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 15 06:00:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 15 Dec 1997 06:00:00 -0000 Subject: problem with EGCS on gun-win32 b18!! HELP Message-ID: <19971215135959.12392.qmail@hotmail.com> >From: vischne@ibm.net >Date: Mon, 15 Dec 1997 05:52:37 GMT >Subject: Re: problem with EGCS on gun-win32 b18!! HELP >To: gnu-win32@cygnus.com.Mumit.Khan.Fri (khan@xraylith.wisc.edu), > "12.Dec.1997.19":30:36.-0600@out5.ibm.net > >root writes: >> Hi! >> When I compile a C++ code with egcs (over gnuwin32 b18) on win95, I get >> no errors. (the >> code works fine on my linux box). But when I run the executable, I get >> an error >> message telling me that I miss the "abort.dll" file. What the hell could >> this be due to? >Dumping egcs right over gnu-win32 b18 is probably idea, given that others >seem to be having trouble as well. One thing to watch out for is that you >CANNOT use import libraries (eg., kernel32.a or libcygwin32.a) that were >created with b18 tools. You must use the egcs supplied ones. > >That's interesting. The first version of ming were compatible with kernel32. >I wonder why it is necessary to have two separate setups, one cygwin and one >mingw32, when the cygwin ./configure and make capabilities are so much better >at this time. It seems to make more sense to have ming as an auxiliary system, >and use the cygwin tools to run ming. I have read, I believe on the gnu-win32 webpage, that this is planned for a future release; however, it won't be in B19. > > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 15 06:15:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 15 Dec 1997 06:15:00 -0000 Subject: Definition of VPATH Message-ID: <199712151412.GAA08868@f134.hotmail.com> >Date: Mon, 15 Dec 1997 09:59:02 +0100 >From: periat@ens.ascom.ch (Periat Henri) >To: earnie_boyd@hotmail.com >Subject: Re: Definition of VPATH >Cc: gnu-win32@cygnus.com > [snip] >Hi Earnie > >Thanks for your repeated efforts. But your suggestion doesn't work, the file >still can't be found. >Version 3.76.1 of 'GNU make' solves this problem, but it has big problems with >'"' in combination with cygnus 'bash', even '@echo "Hello world"' doesn't work. I find it interesting that you are have a problem with the '"' in make. In the original mingw32 it was a known problem; had to do with the way that win32 was handling a quoted string. To fix the problem I had to escape the '"' with the '\' character and depending on the shell nesting level sometimes the escape would have to be escaped. I haven't given Mumit Kahn's EGCS version a try yet but I know he worked on changes specific to Win32. I suggest: 1st try Mumit's version. 2nd try compiling the source under cygwin. >This problem seems to be known and Rob Tulloh >recommended me to load the latest patches (a, b, c and d). But i was not able >to install these patches and get a running, patched 'make'. In 'gnu.utils.help' >i asked for a binary but got no response yet. > > >Henri > - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 15 06:19:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 15 Dec 1997 06:19:00 -0000 Subject: Mixed mounting doesn't seem to work Message-ID: <199712151418.GAA12577@f140.hotmail.com> You'll find in the mail archives a reply from Geoff Noer concerning this that mixed mode no longer functions. It was removed in an earlier version. I don't know why but I suspect that it caused a lot of headaches between cygwin and win32. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Sun, 14 Dec 1997 23:21:10 -0600 >From: Russell K Montgomery >To: Gnu-Win-32 >Subject: Mixed mounting doesn't seem to work > >I have installed CDK.EXE (B18) and Coolview on Windows 95B. I would >prefer to work in "mixed mode" because of several scripts that I have >brought over from a SCO machine that uses the same name but with >differently capitalized filenames. > >I have modified all mount points to be both mixed and binary. mount >outputs the following information: >Device Directory Type Flags >G:\emacs-19.34 /emacs native mixed,text=binary >\\.\tape1: /dev/st1 native mixed,text=binary >\\.\tape0: /dev/st0 native mixed,text=binary >\\.\b: /dev/fd1 native mixed,text=binary >\\.\a: /dev/fd0 native mixed,text=binary >G:\GNU-Win-32 / native mixed,text=binary > >The basic method I used to validate if it is working is: >bash.exe-2.01$ echo "HELLO" > test >bash.exe-2.01$ echo "Hi" > Test >bash.exe-2.01$ ls -l >-rw-r--r-- 1 500 everyone 3 Dec 14 23:18 test > >If someone could provide any input or is successfully using "Mixed" >mounting I would appreciate any feedback you could offer. > >---- >Russell Montgomery >Austin, TX, USA > > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cvn@interchain.nl Mon Dec 15 08:42:00 1997 From: cvn@interchain.nl (Kees van Veen) Date: Mon, 15 Dec 1997 08:42:00 -0000 Subject: DLL's in console applications Message-ID: <34956C4C.389D2ABB@interchain.nl> Hi, This may be a little off topic, but... I've managed to get my console applications running with b18 and Sergey's patches. I've been able to build a couple of DLL's, linked to them, runs ok (slow in comparison to Linux, but it runs). But now it looks like every time an instance of the application starts, all the DLL's are loaded again as well. The DLL's together amount to about 10Mb, and this will increase in the future. My aim was to get them to work as a sort of shared libraries. Now I heard my collegue mumble something about console applications running in a sort of protected environment, that's why all the DLLs are loaded again. I know little about NT. Does any of you have experience with console applications using large DLLs? Would it help to link the DLLs as 'windows application'. Does any of you know where I could find more on this topic. Any help or comments will be greatly appreciated, Kees - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From svr@cagent.com Mon Dec 15 13:37:00 1997 From: svr@cagent.com (Ramesh Subrahmaniam) Date: Mon, 15 Dec 1997 13:37:00 -0000 Subject: problem with ownership of files under NT4.0 under b18 Message-ID: <3495953D.A66EEFFB@cagent.com> I installed NT4.0 Spervice Pack 3 on a SCSI disk with the FAT file system. I am having a problem compiling (with gnumake) the files installed on this file system. I have tried NTFS also without great success. When I do a ls -l on the source files it always shows the owner as being 0 and group as being everyone. The first source of the bug is when I try to do a make it complains that I do not own the files. I tried to login as administrator and I go into a DOS window and try to run chown. It always complains: I did the following combinations: chown 500 bullit . Error: bullit: Invalid Argument chown svr bullit. Error : Invalid User chown -v 500 bullit. Says changing bullit owner to 500 and then says bullit Invalid argument. I then tried NTFS because the next natural thing might be to suspect the file system. The owner of the file now became 544 which is in range of the UIDs when I type mkpasswd -l. But still chown does not work. The next thing I tried is to modify the permissions. I did a chmod +w . It failed silently meaning it did not change the permissions. I did an ls -l to verify this. Now I wanted to make a tmp directory and etc directory on my /. My / is not a the C: drive instead I want it to be D:. I have installed NT4.0 on a SCSI disk which I added. How do I change my / mount point to be D: instead of C:. I tried the mount command as follows: mount d: / : Error fails with Device busy So I did the following: umount C:. Failed silently; did not do anything. umount /. Failed siliently again. I am trying desperately to get this working for me. Can you please help? I will also be happy to go over any FAQs for these kind of problems. Also how I modify the source to see what the problem is? Where can I find the cygwin.dll source code? Thanks Ramesh - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From stanton@haas.berkeley.edu Mon Dec 15 14:28:00 1997 From: stanton@haas.berkeley.edu (Richard Stanton) Date: Mon, 15 Dec 1997 14:28:00 -0000 Subject: Has anyone created DLL file to link into Excel? Message-ID: <199712152217.OAA15982@grether.haas.berkeley.edu> Has anyone successfully created a DLL file and linked it into an Excel spreadsheet? I've installed the mingw32 package, and can get the DLL test program to work, but can't get Excel to load my function successfully. If anyone has done this, and could share even a tiny example of how to do it, I'd be very grateful. Thanks a lot. Richard Stanton - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From batanov@ascinc.com Mon Dec 15 14:28:00 1997 From: batanov@ascinc.com (Alexander Batanov) Date: Mon, 15 Dec 1997 14:28:00 -0000 Subject: linking with vc-built dlls Message-ID: <01BD097E.5E5D8B40@banach.ascinc.com> Dear all, Are there any special considerations for linking object files built with Cygnus g++ with dlls built with Visual C++? Dll has a C interface, so when I do nm on .lib file I see symbols which are not decorated but prefixed with a single underscore (maybe this is a problem). Quikview on .dll file shows the same symbols without leading underscores, though. I am integrating a gcc project with FlexLM license manager. mylibrary.lib is in the working directory and mylibrary.dll is in /winnt/system32. When I am trying to do this g++ mylibrary.lib myobjectfile.o -o myproject.exe I am getting some "undefined refrence" errors: undefined reference to `lp_checkout'. Thanks, Alex Batanov Software Engineer Alternative System Concepts, Inc. 22 Haverhill Rd Windham, NH 03087-0128 (603) 437-2234 (603) 437-2722 fax batanov@ascinc.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From geier@forwiss.uni-passau.de Mon Dec 15 14:29:00 1997 From: geier@forwiss.uni-passau.de (Roland Geier) Date: Mon, 15 Dec 1997 14:29:00 -0000 Subject: Mixed mounting doesn't seem to work References: <3494BE46.7C17759D@iname.com> Message-ID: <199712152229.XAA13821@bombadill.forwiss.uni-passau.de> Are your sources placed on a samba-share? If so, check 'man smb.conf' for preservation of filename-case on general filenames (option 'preserve case') and dos-style '8.3'-ones (option 'preserve short case' or something like that :). If your mounted drive is a native NT/Win95 one, just forget about the samba-stuff. I can somehow remember a posting of Geoffrey Noer mentioning that the mixed-support is broken in b18. Perhaps check the mail-archives on that topic or correct me if I'm wrong here! hth, --- Roland. --- Roland Geier / University of Passau, GER / geier@forwiss.uni-passau.de - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 15 17:54:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 15 Dec 1997 17:54:00 -0000 Subject: problem with ownership of files under NT4.0 under b18 (NBY) Message-ID: <199712160152.RAA10241@f66.hotmail.com> Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Mon, 15 Dec 1997 12:38:21 -0800 >From: Ramesh Subrahmaniam >To: gnu-win32@cygnus.com >Subject: problem with ownership of files under NT4.0 under b18 > >I installed NT4.0 Spervice Pack 3 on a SCSI disk with the FAT file >system. I am having a problem compiling (with gnumake) the files >installed on this file system. I have tried NTFS also without great >success. > >When I do a ls -l on the source files it always shows the owner as being >0 and group as being everyone. > >The first source of the bug is when I try to do a make it complains >that I do not own the files. I tried to login as administrator and I go >into a DOS window and try to run chown. It always complains: > >I did the following combinations: > > chown 500 bullit . Error: bullit: Invalid >Argument > chown svr bullit. Error : Invalid User > chown -v 500 bullit. Says changing bullit owner to 500 > and then says bullit Invalid >argument. > >I then tried NTFS because the next natural thing might be to suspect the >file system. The owner of the file now became 544 which is in range of >the UIDs when I type mkpasswd -l. But still chown does not work. > >The next thing I tried is to modify the permissions. I did a chmod +w >. It failed silently meaning it did not change the >permissions. I did an ls -l to verify this. > >Now I wanted to make a tmp directory and etc directory on my /. My / is >not a the C: drive instead I want it to be D:. I have installed NT4.0 on >a SCSI disk which I added. How do I change my / mount point to be D: >instead of C:. I tried the mount command as follows: > mount d: / : Error fails with Device busy > >So I did the following: > > umount C:. Failed silently; did not do anything. > umount /. Failed siliently again. > >I am trying desperately to get this working for me. > >Can you please help? I will also be happy to go over any FAQs for these >kind of problems. Also how I modify the source to see what the problem >is? Where can I find the cygwin.dll source code? > >Thanks > >Ramesh > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From redmond.urbino@bigfoot.com Mon Dec 15 17:54:00 1997 From: redmond.urbino@bigfoot.com (Redmond Urbino) Date: Mon, 15 Dec 1997 17:54:00 -0000 Subject: opengl, glut, gnat, and gnu-win32 Message-ID: <3495D53D.6CF0@bigfoot.com> I finally got OpenGL, GLUT, GNAT, NT and gnu-win32 all working together. If anybody needs help they can email me. I'm also using the beta version of GLUT 3.6. So far no problems found. Major problems. 1. Dont use GNAT 3.10, use 3.09, the 3.10 is based on the minimalist version, and 3.09 goes with th gnu-win32 b18 from cygnus 2. The cygwin.dll that comes with GNAT 3.09, works with GLUT. Don't use the one which comes with gnu-win32 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From noer@cygnus.com Mon Dec 15 17:54:00 1997 From: noer@cygnus.com (Geoffrey Noer) Date: Mon, 15 Dec 1997 17:54:00 -0000 Subject: Experimental slangwin.tar.gz at deninc.com References: Message-ID: <199712160106.RAA12754@cirdan.cygnus.com> Christopher Faylor wrote: [...] > This is the case with every version of cc or gcc that I've ever seen. > ld always puts symbols in the file unless you link with '-s'. When you > specify -g it only means that more debugging information may be placed > in the final executable. This extra information is usually things like > line number info and possibly symbol entries for static variables. [...] Also keep in mind that there are other object files that may contain debugging information (like crt0.o) which are linked in when you compile. -- Geoffrey Noer noer@cygnus.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Mon Dec 15 17:54:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Mon, 15 Dec 1997 17:54:00 -0000 Subject: Experimental slangwin.tar.gz at deninc.com Message-ID: <199712160013.AAA23042@out4.ibm.net> >>While it's there, you can get the experimental slang with console buffer support source at the incoming directory of ftp.deninc.com.us. You can also get the very latest development lynx.exe.gz in that directory (supply your own lynx.cfg, or lynx will). << The owner of deninc.com has kindly moved the two files to the following URL: ftp://ftp.deninc.com/pub/ports/Lynx ???? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From colin@bird.fu.is.saga-u.ac.jp Mon Dec 15 17:54:00 1997 From: colin@bird.fu.is.saga-u.ac.jp (Colin Peters) Date: Mon, 15 Dec 1997 17:54:00 -0000 Subject: problem with EGCS on gun-win32 b18!! HELP Message-ID: <01bd09c0$da871aa0$fa173185@gbird0.fu.is.saga-u.ac.jp> From: vischne@ibm.net To: khan@xraylith.wisc.edu ; 12.Dec.1997.19 <:30:36.-0600@out5.ibm.net> Date: Monday, December 15, 1997 3:50 PM Subject: Re: problem with EGCS on gun-win32 b18!! HELP >root writes: >> Hi! >> When I compile a C++ code with egcs (over gnuwin32 b18) on win95, I get >> no errors. (the >> code works fine on my linux box). But when I run the executable, I get >> an error >> message telling me that I miss the "abort.dll" file. What the hell could >> this be due to? [Mumit] >Dumping egcs right over gnu-win32 b18 is probably [not a good] idea, given that others >seem to be having trouble as well. One thing to watch out for is that you >CANNOT use import libraries (eg., kernel32.a or libcygwin32.a) that were >created with b18 tools. You must use the egcs supplied ones. > [vischne] >That's interesting. The first version of ming were compatible with kernel32. >I wonder why it is necessary to have two separate setups, one cygwin and one >mingw32, when the cygwin ./configure and make capabilities are so much better >at this time. It seems to make more sense to have ming as an auxiliary system, >and use the cygwin tools to run ming. I just want to point out that it is not a problem with what Mumit Khan has done with Mingw32 (nothing except use and bundle it as far as I can tell). The same problems will occur if one moves from Cygwin32 b18 to the EGCS version of Cygwin32 (I believe). There is an incompatibility between the format of files used by the EGCS compiler and linker and the format used by the b18 compiler and linker. Probably these very same problems could happen if you are not careful when upgrading to b19 from b18, as I understand it will include the same fix. You can still use the EGCS version of Cygwin32 and install Mingw32 as an "auxiliary system", in fact many changes in EGCS and the latest snapshots of Mingw32 are designed to make this easier (perhaps too easy... but that's another topic). In fact, although I haven't tried it, you could also probably set up to build Cygwin32 programs using the Mingw32 version of EGCS pretty easily (relatively speaking of course). The problem is not related to whether the run time libraries are Mingw32 or Cygwin32, it is related to whether you are using EGCS (either version) or the GCC supplied with Cygwin32 (or Jan-Jaap's native Mingw32 version of GCC). Sorry about the bandwidth, Colin. P.S. As to whether it makes more sense to use Mingw32 under Cygwin32-based tools: I, for one, do not really miss configure, bash, or GNU make with full shell scripting very much, but I would miss the disk space they would take up if I installed them (I am on a tight disk space budget :-/). For people like me a native Mingw32 option is very useful, and, as I pointed out, you can still use Cygwin32 tools to do Mingw32 stuff if you want to, so thank you to Mr. Khan. -- Colin Peters -- colin at fu.is.saga-u.ac.jp -- Saga University Dept. of Information Science -- http://www.fu.is.saga-u.ac.jp/~colin -- http://www.geocities.com/Tokyo/Towers/6162 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From colin@bird.fu.is.saga-u.ac.jp Mon Dec 15 20:16:00 1997 From: colin@bird.fu.is.saga-u.ac.jp (Colin Peters) Date: Mon, 15 Dec 1997 20:16:00 -0000 Subject: DLL's in console applications Message-ID: <01bd09d1$c047fcc0$fa173185@gbird0.fu.is.saga-u.ac.jp> From: Kees van Veen Date: Tuesday, December 16, 1997 2:25 AM >But now it looks like every time an instance of the application starts, >all the DLL's are loaded again as well. The DLL's together amount to >about 10Mb, and this will increase in the future. My aim was to get them >to work as a sort of shared libraries. It is my understanding that this is natural. When any instance of an application starts it should map all of it's DLLs into it's own memory space. I believe that whether the DLLs are relocated or not can influence their "memory" use (that is, whether they are backed by the DLL files or by the swap file) but in either case each instance will be getting it's own pages of real memory. I also cannot conceive of any way of sanely doing static data for DLLs except giving each instance it's own copy of the data pages (probably with "copy on write" set). Some expert please correct me if I'm wrong. They are shared libraries, in the sense that each instance uses the same files to obtain the shared code to build the executable image in memory. However, I don't believe you are going to reduce memory usage of multiple instances of an application by using shared libraries under Win32. You might want to investigate ways to run your application as a single instance, or splitting up the functionality of the DLLs into smaller chunks so that only the DLLs needed can be linked to the application. >Now I heard my collegue mumble something about console applications >running in a sort of protected environment, that's why all the DLLs are >loaded again. Your colleague was probably thinking about DOS applications, which each run in a separate virtual machine. Win32 console applications run in the same virtual machine as all other Win32 applications AFAIK. >Would it help to link the DLLs as 'windows application'. I don't think so, but you could try it and see. Colin. -- Colin Peters -- colin at fu.is.saga-u.ac.jp -- Saga University Dept. of Information Science -- http://www.fu.is.saga-u.ac.jp/~colin -- http://www.geocities.com/Tokyo/Towers/6162 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ronsolo@stomped.com Mon Dec 15 22:47:00 1997 From: ronsolo@stomped.com (Ron Crisco) Date: Mon, 15 Dec 1997 22:47:00 -0000 Subject: Partial Distribution of GNU-Win32 Utilities Message-ID: <001201bd09ee$7a9cba40$d6b6f1cd@ronsolo.ronsolo.net> I need to know what the requirements are for re-distributing only the md5sum.exe and cygwin.dll. This is non-profit of course, just a convenience so that our users don't need to download the full 4M usertools.exe. Is it enough to just copy the GNU license from the FAQ and distribute that along with those 2 files? Or is there a more stringent requirement? Also, must we provide the md5sum source in that zip file? Or may we point to the Cygnus site for source availablity. I would like to do this all on the up and up :) FYI, this will be used as part of the OpenQuake initiative - www.openquake.org to help verify file downloads by providing the MD5 checksum. RonSolo - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Mon Dec 15 23:26:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Mon, 15 Dec 1997 23:26:00 -0000 Subject: DLL's in console applications Message-ID: <01BD0A0B.99021350@gater.krystalbank.msk.ru> Kees van Veen wrote: > But now it looks like every time an instance of the application starts, > all the DLL's are loaded again as well. The DLL's together amount to > about 10Mb, and this will increase in the future. My aim was to get them > to work as a sort of shared libraries. > It just seems to you. The code segment of a dll shares between applications, but data segment is loaded for every app. -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hmwang@csit.edu.cn Mon Dec 15 23:26:00 1997 From: hmwang@csit.edu.cn (S.Q.Yang) Date: Mon, 15 Dec 1997 23:26:00 -0000 Subject: autoconf/automake Message-ID: <9712160725.AA07162@ns> Hi: I don't know where to find autoconf or automake. Would anybody give me some suggestions? Thanks! - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hmwang@csit.edu.cn Mon Dec 15 23:26:00 1997 From: hmwang@csit.edu.cn (S.Q.Yang) Date: Mon, 15 Dec 1997 23:26:00 -0000 Subject: questions on porting postgresql Message-ID: <9712160718.AA07152@ns> I'd like to port postgresql-v6.1 to winNT. I had successfully installed gnuwin32.When I run configure file of postgresql-v6.1,it reports the following errors: -ASSERT CHECKING disabled checking for gcc... (cached) gcc checking whether the C compiler (gcc ) works... no configure: error: installation or configuration problem: C compiler cannot create executables. I don't know how to overcome it. Any suggestion will be very appreciated! - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cvn@interchain.nl Tue Dec 16 00:15:00 1997 From: cvn@interchain.nl (Kees van Veen) Date: Tue, 16 Dec 1997 00:15:00 -0000 Subject: DLL's in console applications References: <01BD0A0B.99021350@gater.krystalbank.msk.ru> Message-ID: <349646F8.3CDAB55F@interchain.nl> Sergey Okhapkin wrote: > > Kees van Veen wrote: > > But now it looks like every time an instance of the application starts, > > all the DLL's are loaded again as well. The DLL's together amount to > > about 10Mb, and this will increase in the future. My aim was to get them > > to work as a sort of shared libraries. > > > > It just seems to you. The code segment of a dll shares between applications, but data segment is loaded for every app. > > Thanks Colin and Sergey for your response. It seemed to me when looking at the performance tab in the task manager. Every time I start a new instance of the application I see the memory usage increase with about the size of the application. Also the startup time of the application doesn't decrease when an instance is already loaded. On Unix I'm used to wait a little longer the first time to get the shared libraries loaded, and the second instance to start up quickly. On NT I cannot say I notice this, altough I may be short on memory here. 64Mb on a Pro 200 with NT server 4 S P3 is a minimum configuration, I'm told. Kees - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Tue Dec 16 03:08:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Tue, 16 Dec 1997 03:08:00 -0000 Subject: One way to get around the cygwin terminal problem Message-ID: <199712161108.LAA89738@out2.ibm.net> vischne@ibm.net wrote: > > In other words, with some work, it is possible to get a professional > looking and acting terminal application using the cygwin Unix > environment but bypassing the cygwin terminal window. >>It's a bad idea to replace stdout with direct console writings. You application, for example, will not work in a telnet session or in xterm window. << Oh, now I know why you wrote this. Your implementation of cygwin.dll and BASH-4.01$ doesn't support generation of a new console buffer to overlay the cygwin console buffer. In fact, your web site claims that such applications won't work. Tsk, tsk, tsk. How can you claim that is an improvement? I stopped using your earlier developmental binaries because they broke so many of the things that I had come to take for granted with cygwin b18. The operative words here are _compatibility_ and _transparency_ of operation. -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job And, there are lots of jobs in America for Russian emigre programmers, especially in the San Francisco peninsula, where there are startup software companies founded by emigres.  - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Tue Dec 16 05:18:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Tue, 16 Dec 1997 05:18:00 -0000 Subject: autoconf/automake (NBY) Message-ID: <19971216130217.12885.qmail@hotmail.com> ftp.gnu.org or any of it's mirrors. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hmwang@csit.edu.cn Tue Dec 16 05:18:00 1997 From: hmwang@csit.edu.cn (S.Q.Yang) Date: Tue, 16 Dec 1997 05:18:00 -0000 Subject: how to get cdk Message-ID: <9712161317.AA07710@ns> I'm in China. I had used Getright to download cdk.exe,but failed. Because the server does not support breakpoint. Is there any other method to download the file? Thanks! - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Tue Dec 16 05:18:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Tue, 16 Dec 1997 05:18:00 -0000 Subject: linking with vc-built dlls (NBY) Message-ID: <19971216131426.23294.qmail@hotmail.com> This has been discussed before, check the mail archives. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ http://www.cygnus.com/pubs/gnupro/ http://www.cygnus.com/misc/gnu-win32/ http://www.cygnus.com/ml/gnu-win32 (this one is the mail archive) http://www.lexa.ru/sos http://www.fu.is.saga-u.ac.jp/~colin/gcc.html http://www.itribe.net/virtunix/winhelp-man-pages/ - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: Alexander Batanov >To: "'GNU-Win32@cygnus.com'" >Cc: "'batanov@ascinc.com'" >Subject: linking with vc-built dlls >Date: Mon, 15 Dec 1997 17:24:49 -0500 > >Dear all, > >Are there any special considerations for linking object files built with Cygnus g++ with dlls built with Visual C++? > >Dll has a C interface, so when I do nm on .lib file I see symbols which are not decorated but prefixed with a single underscore (maybe this is a problem). Quikview on .dll file shows the same symbols without leading underscores, though. > >I am integrating a gcc project with FlexLM license manager. >mylibrary.lib is in the working directory and mylibrary.dll is in /winnt/system32. >When I am trying to do this > >g++ mylibrary.lib myobjectfile.o -o myproject.exe > >I am getting some "undefined refrence" errors: > >undefined reference to `lp_checkout'. > > >Thanks, >Alex Batanov >Software Engineer >Alternative System Concepts, Inc. >22 Haverhill Rd >Windham, NH >03087-0128 >(603) 437-2234 >(603) 437-2722 fax >batanov@ascinc.com >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jclaiseca@fagorautomation.es Tue Dec 16 07:13:00 1997 From: jclaiseca@fagorautomation.es (J.C. LAISECA) Date: Tue, 16 Dec 1997 07:13:00 -0000 Subject: Buiilding Vxds with gnu-win32 Message-ID: Hi all! I've been looking through the archives but I've found no answer to a previous post regarding building VxDs with gnu-win32. Sorry if I bother you with an obvious question (and answer) but, is it possible to build those famous VxDs with gnu-win32? Has anybody done it or tried it? Any other suggestions? Maybe other tools? Thanks a lot in advance. JC Laiseca - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Tue Dec 16 10:20:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Tue, 16 Dec 1997 10:20:00 -0000 Subject: how to get cdk (NBY) Message-ID: <199712161818.KAA16113@f128.hotmail.com> Below is a list of mirrors with cdk.exe. You should also find a directory of cdk.exe split into multiple files. USA: California: ftp.cygnus.com:/pub/gnu-win32 Texas: ftp.deninc.com Worldwide: Austria: gd.tuwien.ac.at France: ftp.fwntug.org or ftp.inria.fr Italy: aaue.matrice.it Japan: ftp.crl.go.jp Slovenia: sunsite.fri.uni-lj.si Spain: ftp.rediris.es UK: sunsite.doc.ic.ac.uk Let me also point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: "S.Q.Yang" >To: >Subject: how to get cdk >Date: Tue, 16 Dec 1997 21:20:53 +0800 > >I'm in China. I had used Getright to download cdk.exe,but failed. Because >the server does not support breakpoint. Is there any other method to >download the file? >Thanks! >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 16 11:49:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Tue, 16 Dec 1997 11:49:00 -0000 Subject: One way to get around the cygwin terminal problem References: <199712161108.LAA89738@out2.ibm.net> Message-ID: In article < 199712161108.LAA89738@out2.ibm.net >, wrote: >vischne@ibm.net wrote: >> >> In other words, with some work, it is possible to get a professional >> looking and acting terminal application using the cygwin Unix >> environment but bypassing the cygwin terminal window. > >>>It's a bad idea to replace stdout with direct console writings. You >application, for example, will not work in a telnet session or in xterm >window. << > >Oh, now I know why you wrote this. Your implementation of cygwin.dll and >BASH-4.01$ doesn't support generation of a new console buffer to overlay >the cygwin console buffer. In fact, your web site claims that such >applications won't work. Tsk, tsk, tsk. How can you claim that is an >improvement? I stopped using your earlier developmental binaries because >they broke so many of the things that I had come to take for granted with >cygwin b18. That's interesting. I'm sure that you understand that Sergey's "developmental binaries" actually represent snapshots of Cygnus's own development tree, right? So, you were undoubtedly really diligent in reporting all of this "broken" behavior so that it would be corrected for B19. Phew. That's a relief. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From noer@cygnus.com Tue Dec 16 15:58:00 1997 From: noer@cygnus.com (Geoffrey Noer) Date: Tue, 16 Dec 1997 15:58:00 -0000 Subject: Partial Distribution of GNU-Win32 Utilities References: <001201bd09ee$7a9cba40$d6b6f1cd@ronsolo.ronsolo.net> Message-ID: <199712162356.PAA13436@cirdan.cygnus.com> Ron Crisco wrote: > > I need to know what the requirements are for re-distributing only the > md5sum.exe and cygwin.dll. Read the GPL for the full details. My understanding as a non-legal person is that you do not need to distribute sources for either automatically. You are required to provide the sources for both md5sum.exe and cygwin.dll to anyone you distribute them to who asks you for them. -- Geoffrey Noer noer@cygnus.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Tue Dec 16 16:28:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Tue, 16 Dec 1997 16:28:00 -0000 Subject: Help needed: FAQ for cygwin32/mingw32 egcs Message-ID: <9712170028.AA10325@modi.xraylith.wisc.edu> I need some help from folks who have *successfully* installed and used egcs on cygwin32 and/or mingw32. My own setup always seems to work, but there are many for whom this is not the case, and right now I just don't have the time to investigate each problem. What I would like to do is to create FAQ for egcs on both cygwin32 and mingw32 and hopefully that will help the new users, especially those who don't necessarily know how gcc finds its includes, libraries, specs file etc. Have any of you kept notes on any of the following cases: 1. egcs-1.00 dumped on top of cygwin32 b18. Anything special that needs to be done (eg., making sure that the env variables now point to the right place)? 2. egcs-1.00 and cygwin32 b18, but under 2 different trees. That's how I keep mine, but others have had problems. I personally like the clean separation, at the expense of disk space. The only thing I had to make sure was that the egcs/bin directory came before gnuwin32/b18/bin in PATH. 3. egcs-1.00 for cygwin32/mingw32 *and* b18. 4. Other general pitfalls. * ld.exe goes to la-la land. One quick workaround is to copy the ld.exe in bin directory to gcc library directory (where cc1.exe lives) under the name collect-ld.exe; I still haven't figured out what the problem really is, so if you know the answer, don't hold back. * Coolview: coolview's libcygwin.a needs to be built with egcs dlltool (basically a newer version of dlltool than the one that came with cygwin32 b18). Has anybody tried rebuilding libcygwin.a from Sergey's coolview cygwin.dll and see if that works? If so, a shell script can automate that. * whatever else you've noticed. I would like to hear from you on what you've done to get things working so we can help the rest of the users having problems. Thanks in advance, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Tue Dec 16 18:40:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Tue, 16 Dec 1997 18:40:00 -0000 Subject: Buiilding Vxds with gnu-win32 Message-ID: <199712170240.CAA342902@out1.ibm.net> >> Hi all! << That's `y'all'. >> I've been looking through the archives but I've found no answer to a previous post regarding building VxDs with gnu-win32. Sorry if I bother you with an obvious question (and answer) but, is it possible to build those famous VxDs with gnu-win32? Has anybody done it or tried it?<< That comes under the heading of those extras that even commercial compilers don't cover. Like adding a library of DPMI calls to the Windows server, so that applications written for another gcc compiler can be compiled under cygwin. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hmwang@csit.edu.cn Tue Dec 16 19:37:00 1997 From: hmwang@csit.edu.cn (S.Q.Yang) Date: Tue, 16 Dec 1997 19:37:00 -0000 Subject: why not find c compiler (gcc) Message-ID: <9712170336.AA08704@ns> Thank you, raf. The configure program first trys to find where gcc exists. But it doesn't find the compiler. How could I do to overcome the question? Or how to set the configure file? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From turtill@ihug.co.nz Tue Dec 16 23:36:00 1997 From: turtill@ihug.co.nz (Kurt Häusler) Date: Tue, 16 Dec 1997 23:36:00 -0000 Subject: first time compiling windows app with gcc Message-ID: <000c01bd0abe$5ce1ae40$4e804dd1@cs116bf.ihug.co.nz> hello, this question has already been asked by carvalho@penelope.dss.inpe.br but it wasnt answered. how can I compile programs when i get error messages like undefined reference etc. this is just a program using the windows api to pop up one window and saw hello world. I have compiled it before under two other compilers and one of them I had to use a option like -lwin32 or something. What is the equivalent option with this compiler? gcc hello.c is all I have been using. Please answer, I would be very greatful. Thanks from Kurt - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From juliano@apptive.co.uk Wed Dec 17 00:47:00 1997 From: juliano@apptive.co.uk (Julian W H Osborne) Date: Wed, 17 Dec 1997 00:47:00 -0000 Subject: Problems with malloc Message-ID: <971217084523206@snoopy.apptive.co.uk> Hi All, I've recently downloaded the gnuwin32 bits and pieces including cdk. I'm trying to compile some source which is falling over when it looks for malloc.h xalloc.c:16: malloc.h: No such file or directory xalloc.c is my own file ;-) I've had a look through the documentation that I have but haven't found anything. Could anyone help. Thanks in advance Julian - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From g.guilliams@cger.be Wed Dec 17 01:27:00 1997 From: g.guilliams@cger.be (Guido Guilliams) Date: Wed, 17 Dec 1997 01:27:00 -0000 Subject: Problem running bash(s) (b18), as different users, simultaneously Message-ID: <01BD0AD5.E7F7A400@W0106652.aslk-cger.be> This is my problem: I use usertools from cygnus. I can execute many bash as one user but the concurrent execution of the same bash as another user (for example in a service as system account) fails. It seems to share ressources (memory,...) with other occurences of bash. I suppose this is caused by security conflicts. Can you help me ? Bien ???? vous, Guido. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net-nospam Wed Dec 17 01:32:00 1997 From: vischne@ibm.net-nospam (root) Date: Wed, 17 Dec 1997 01:32:00 -0000 Subject: Improved slw32tty.c that allows puts and printf to work with the libslang display - slw32tty.c [1/1] Message-ID: <199712170931.JAA197464@out1.ibm.net> Enclosed in response to constructive criticism received on this conference is an improved slw32tty.c for the slangw32.tar.gz package that does _not_ overlay the cygwin.dll terminal window. begin 755 slw32tty.c M+RH@0V]P>7)I9VAT("AC*2`Q.3DR+"`Q.3DU($IO:&X@12X@1&%V:7,*("H@ M06QL(')I9VAT2YC M(&5X<&5C=&5D(&)Y(&%P<&QI8V%T:6]N5]I;G1R7VAO;VLI("AV;VED*3L*:6YT(%-,86YG7U147T)A=61?4F%T M92`](#`["FEN="!33&%N9U]45%]296%D7T9$(#T@+3$["@IV;VED(%-,='1Y M7W-E=%]S=7-P96YD7W-T871E("AI;G0@;6]D92D*>PI]"@II;G0@:6YI=%]3 M3'5N:7@H*0I["B!R971U2!A;F0@87)E(&EG;F]R960N"EPJ+2TM+2TM M+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM M+2TM+2TM+2TM+2TM+2TM+2TM+2HO"@I(04Y$3$4@:%-T9&]U="P@:%-T9&EN M.PI#3TY33TQ%7U-#4D5%3E]"549&15)?24Y&3R!C2`H:6YT(&-A;F-E;%]C:&%R+"!I;G0@;F]?9FQO=U]C M;VYT&ES="!W:71H M(&)O=&@@=&AE("HO"B`@+RH@4E1,1$Q,+D1,3"!A;F0@0UE'5TE.+D1,3"!T M97)M:6YA;"!W:6YD;W=S+B`@($DN(&4N+"!C86QL:6YG("HO"B`@+RH@9G!R M:6YT9B!A;F0@9G!U=',@9&]E3H@*B\*"B`@+RH@&ET:6YG"EPJ+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM M+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2TM+2HO"G9O:60@ M4TQA;F=?7-?:6YP=71?<&5N9&EN9R`H:6YT('1S96-S*3L*("H@('-L965P(&9O M7-?:6YP=71?<&5N9&EN9R`H:6YT('1S96-S*0I["B`@($E. M4%547U)%0T]21"!R96-OF5O9B@I(')E='5R;G,@ M=&AE('-TF5O M9B@B6T(B*3L@8G)E86L["@D)+RH@4W1R:6YG('9A;'5E71EF5O9B@B6S1^(BD[(&)R96%K.PH) M"6-A71EF5O9B@B6UM# M(BD[(&)R96%K.PH)"6-A71EF5O9B@B6S$W?B(I.R!B71EF5O9B@B6S(P?B(I.R!BF5O9B@B6S(Q?B(I.R!BF5O9B@B6S(S?B(I.R!B MF5O9B@B6S(T M?B(I.R!B71E71E71E3`L(&ME>3$["@IV;VED M($=E=$ME>2`H=F]I9"D*>PH@('5N3`@/2`H:V5Y,2`](#`I.PH@ M("\J("%B2V5Y1&]W;B`]/2!B2V5Y57`@/R`J+PH@(&EF("A296%D0V]N45V96YT+D%S8VEI0VAA4-O9&4["B`@("!].PI]"@HO*B!$96-L87)E9"!A8F]V92!F;W(@2V5Y4')E MPH@("`@:V5Y,"`]("AU;G-I9VYE9"!I;G0I*F5S8V%P95]S M=')I;FPH)8V%S92!62U],1494.@EE Sorry to pound on this subject again, but I still have problems to connect with telnet/in.telnetd. I setup all the variables as system variables (including CYGWIN_TTY and CYGWIN_NOTITLE to 1), I created an account called inetd with all the rights specified in Okhapkin's instructions, I removed all tcp/ip services, launched inetd with Autoexnt, and I still get this message from the inetd code in the Event log window: The description for Event ID ( 0 ) in Source ( inetd ) could not be found. It contains the following insertion string(s): inetd : Win32 Process Id = 0x15C : CygWin32 Process Id = 0x42D : select: error 0. I am running on WindowsNT 4.0.1381 Can somebody help? Thanks! nicb ------------------------------------------------------------------------ Nicola Bernardini Via Aldo Manuzio, 16 I - 00153 Roma Tel.: ++396/5743196 E-mail: nicb@axnet.it Re graphics: A picture is worth 10K words -- but only those to describe the picture. Hardly any sets of 10K words can be adequately described with pictures. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From andrewi@harlequin.co.uk Wed Dec 17 03:43:00 1997 From: andrewi@harlequin.co.uk (Andrew Innes) Date: Wed, 17 Dec 1997 03:43:00 -0000 Subject: [Workaround] Re: M-x shell + bash = Infinite error messages? References: Message-ID: <199712171141.LAA29386@shirow.long.harlequin.co.uk> On 16 Dec 1997 13:44:45 -0500, Paul Stodghill said: >I figured out what was causing my problem: [problem was bash crashing when run in an Emacs shell buffer] >I was invoking emacs by using a command similar to, > > bash -c "runemacs ... some options ..." > >Changing the command to simply > > runemacs ... some options ... > >made by M-x shell problems go away. > >Why? runemacs starts Emacs then exits, so possibly an environment variable is being set by bash which is passed on to Emacs, causing confusion later. Another possibility is that Emacs inherits (via runemacs) some handles from bash (eg. file handles, semaphore handles, etc) which are causing problems. AndrewI >>I'm having a real pain of a problem with NTEmacs 19.34.6 and the bash from >>gnu-win32 beta 18. >> >>It appears that I am able to invoke M-x shell successfully the first time >>that I run NTEmacs, but if I restart NTEmacs and M-x shell again, I get the >>following in the *shell* buffer. >> >>(unknown) In cygwin_except_handler >>(unknown) Exception trapped! >>(unknown) exception C0000005 at 10018910 >>(unknown) exception: ax 0 bx 3EA cx A5 dx 64 >>(unknown) exception: si 244FA28 di 0 bp 244FA3C sp 244FA18 >>(unknown) exception is: STATUS_ACCESS_VIOLATION >>(unknown) Stack trace: >>(unknown) frame 0: sp = 0x244F84C, pc = 0x1000CEC2 >>(unknown) frame 1: sp = 0x244F868, pc = 0x77F94512 >>(unknown) frame 2: sp = 0x244F88C, pc = 0x77F88EEB >>(unknown) frame 3: sp = 0x244F918, pc = 0x77F76266 >>(unknown) frame 4: sp = 0x244FA3C, pc = 0x1000BBED >>(unknown) frame 5: sp = 0x244FF94, pc = 0x1000C113 >>(unknown) frame 6: sp = 0x244FFA0, pc = 0x43AE95 >>(unknown) frame 7: sp = 0x244FFB0, pc = 0x40103B >>(unknown) frame 8: sp = 0x244FFC0, pc = 0x77F1B304 >>(unknown) frame 9: sp = 0x244FFF0, pc = 0x0 >>(unknown) End of stack trace >>(unknown) In cygwin_except_handler >>(unknown) Error while dumping state (probably corrupted stack) >>(unknown) In cygwin_except_handler >>(unknown) Error while dumping state (probably corrupted stack) >>(unknown) In cygwin_except_handler >>(unknown) Error while dumping state (probably corrupted stack) >>(unknown) In cygwin_except_handler >>. >>. >>. >>. >> >>and so on. The only fix to the problem that I have found is to reboot the >>machine. :-( >> >>Any clues as to how to fix this? Thanks. >> >>-- >>Paul Stodghill >> http://www.cs.cornell.edu/home/stodghil/home.html - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From weiku@utkux.utcc.utk.edu Wed Dec 17 09:25:00 1997 From: weiku@utkux.utcc.utk.edu (Wei Ku) Date: Wed, 17 Dec 1997 09:25:00 -0000 Subject: why not find c compiler (gcc) Message-ID: <004c01bd0af0$70ec6010$8b14a980@wku1.utk.edu> Hello, Yang: There are messages that describe how to set up the environment ( mounting, symbolic link, initial files etc.) You can search for them to know all the detail. The attached files are the batch file that set up the DOS environment variables and the initial file for bash. Hope this help. Wei Ku -----Original Message----- From: S.Q.Yang To: gnu-win32@cygnus.com Date: Wednesday, December 17, 1997 3:07 AM Subject: why not find c compiler (gcc) >Thank you, raf. >The configure program first trys to find where gcc exists. >But it doesn't find the compiler. >How could I do to overcome the question? Or how to set the configure file? >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > -------------- next part -------------- set TERM=linux set GCC_EXEC_PREFIX=d:\gnuwin32\b18\H-i386-cygwin32\lib\gcc-lib\ set TCL_LIBRARY=d:/gnuwin32/b18/tcl/lib/tcl7.6 set GDBTK_LIBRARY=d:/gnuwin32/b18/share/gdbtcl set PATH=d:\gnuwin32\b18\H-i386-cygwin32\bin;d:\gnuwin32\b18\tcl\bin;d:\usr\local\bin;%HOMEDRIVE%%HOMEPATH%\bin;%PATH% set LIBRARY_PATH=d:\Cygnus\b18\H-i386-cygwin32\lib set MAN_PATH=d:\Cygnus\man set C_INCLUDE_PATH=d:\Cygnus\include set CPLUS_INCLUDE_PATH=d:\Cygnus\include\g++ set DISPLAY=localhost:0.0 d:\gnuwin32\b18\H-i386-cygwin32\bin\bash.exe -login From j.bessels@g-bank.nl Wed Dec 17 09:25:00 1997 From: j.bessels@g-bank.nl (jbessels) Date: Wed, 17 Dec 1997 09:25:00 -0000 Subject: b18: fileaccess LS UNreliable Message-ID: <199712171725.SAA21080@solair1.inter.NL.net> Hi, I've a few years of experience with Unix and now my company will migrate to NT (damned). I'm currently looking at Cygnus-gnu32 and I'm experiecing some VERY weird things. I've listed my results below. Configuration. Two different NT systems (one Pentium166 and one Pentium II 266) both using NT4.0 with SP3. Observation1: For Bash and Command Line (CLI) is true: ls > dummy is OK. ls -l > ../dummy or ls -ptl > ../dummy is OK. ls -l > dummy or ls -ptl > dummy gives following error: dummy Permission denied. (checked the above for a couple of dirs) Conclusion: redirection doesn't work if the file is in the same directory. Have tried it on WIn95 as well same result. Works fine under Linus though, should work !! Observation2: Standing in rootdir / CLI. Dir pagefile.sys is OK. BASH. Dir pagefile.sys is ERROR (using cygnus dir. cmd) BASH. ls pagefile.sys is ERROR. CLI. ls pagefile.sys is ERROR. In all three cases error is --> dir: pagefile.sys No such file or directory. Ohter files can be accessed, pagefile is swapfile. Conclusion: Under Dos using dir all is OK, using cygnus ls (and possibly other ones as well) give error. Observation3: Standing in rootdir / (have installed SMS) dirs SMSdata.dat smslog.dat are allowed to be accessed by the adminstrator and I'm logged in as adminstrator. CLI. dir SMSdata.dat and dir SMSlog.dat is OK. CLI + BASH. ls SMSdata.log and ls SMSlog.dat is ERROR --> permission denied. obvervation4: This one is beautiful. Standingin /Winnt CLI. dir net*.* ==> two files shown, netlogon.chk and network.wri. BASH. dir net*.* (cygnus dir) and ls net*.* only show the file network.wri. NO error is displayed. BASH ls -l net*.*. only shows network.wri and no error. BASH ls -l net*.* | head -5 gives networ.wri and sometimes the error "ls: netlogon.chk. permission denied." I've tried to be as accurate as possible, but maybe a mistake has slipped in. My conclusion is. Cygnus-gnuwin32 seems nice, pipes et al works like as charm just like all commands BUT I can't trust Cygnus-gnuwin32 because sometimes not al files are shown or aren't handled correctly. This makes it totally unusable for me. Am I doing something wrong here or..... I've tried to download coolview (november version) but that site is so unbelievable slow I can't download coolview at all. If someone could send me the LATEST coolview version this would be much appreciated. OK guys, I'm completely lost here. I don't know whats going on. Most examples can be tried by everone. I'm very interested if you can reproduce my errors. Please let me know.... Cheers, (gonna hit the bottle hard this evening) Jan Bessels ======== - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From joe_banta@perceptics.com Wed Dec 17 09:25:00 1997 From: joe_banta@perceptics.com (Joseph E. Banta, Esquire) Date: Wed, 17 Dec 1997 09:25:00 -0000 Subject: patch and dos text Message-ID: <01BD0AE2.A9E844A0@HERBY> Has anyone had the problem that 'patch' works strangely with dos text files, or is it jus me? Joseph Banta - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 17 09:25:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 17 Dec 1997 09:25:00 -0000 Subject: why not find c compiler (gcc) Message-ID: <19971217154522.8785.qmail@hotmail.com> The answer is ENVIRONMENT VARIABLES. You need to read the README and FAQ located at the ftp site in the list of URL's I sent you. Also, take a look at the other pointers as well. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: "S.Q.Yang" >To: >Subject: why not find c compiler (gcc) >Date: Wed, 17 Dec 1997 11:40:00 +0800 > >Thank you, raf. >The configure program first trys to find where gcc exists. >But it doesn't find the compiler. >How could I do to overcome the question? Or how to set the configure file? >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Wed Dec 17 09:25:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Wed, 17 Dec 1997 09:25:00 -0000 Subject: first time compiling windows app with gcc (NBY) Message-ID: <19971217154855.14113.qmail@hotmail.com> Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: "=?iso-8859-1?Q?Kurt_H=E4usler?=" >To: >Subject: first time compiling windows app with gcc >Date: Wed, 17 Dec 1997 20:35:30 +1300 > >hello, this question has already been asked by carvalho@penelope.dss.inpe.br >but it wasnt answered. > >how can I compile programs when i get error messages like undefined >reference etc. this is just a program using the windows api to pop up one >window and saw hello world. > >I have compiled it before under two other compilers and one of them I had to >use a option like -lwin32 or something. What is the equivalent option >with this compiler? > >gcc hello.c is all I have been using. > >Please answer, I would be very greatful. > >Thanks from Kurt > > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From andyp@parallax.co.uk Wed Dec 17 09:25:00 1997 From: andyp@parallax.co.uk (Andy Piper) Date: Wed, 17 Dec 1997 09:25:00 -0000 Subject: PIMAGE_SECTION_HEADER? Message-ID: <3.0.1.32.19971217142958.0099d180@mailhost> The supplied win32 headers do not include things like PIMAGE_SECTION_HEADER which are defined in WINNT.H under VC. What is the correct sequence of hoops to get this sort of info into the distributed headers? How was the original information put in there? andy ___ ____ Dr Andy Piper / _ \___ ________ _/ / Solutions_ (require 'disclaimer) / ___/ _ `/ __/ _ `/ / / _ `/\ \ / andyp@parallax.co.uk /_/ \_,_/_/ \_,_/_/_/\_,_//_\_\ boot /vmemacs - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From banders@ECD.Rockwell.COM Wed Dec 17 09:25:00 1997 From: banders@ECD.Rockwell.COM (Bartlee A. Anderson) Date: Wed, 17 Dec 1997 09:25:00 -0000 Subject: Beta 19 status References: <199709280507.WAA12657@rtl.cygnus.com> Message-ID: <3497EC77.7C28@switch.rockwell.com> Dear Geoffrey, Any chance for a b19 for Christmas? Thanks, Bart Geoffrey Noer wrote: > > Greetings, > > I guess I should know better even hinting at dates before beta 19 is > completely finished. :-( You should know that I am delaying the > release for a variety of reasons -- in a nutshell, some things have not > progressed enough and other things have progressed too much such that it > will take more time to get a new gnu-win32 to a releasable stage. That's > life I guess. > > That said, the Cygwin32 library (cygwin.dll) has made a lot of progress > since the last release as those of you using Sergey's cygwin.dlls have > seen. For now, my current plan is to make a new cygwin.dll and its sources > available soon that can be added to beta 18 to make beta 18.5. > > Finally, just a heads up so you know we haven't been slacking off :-). > Cygnus has just announced a new IDE called Cygnus Foundry (see the link > from the main Cygnus Web page http://www.cygnus.com for more information). > I will let you all know how this will play with future gnu-win32 beta > releases when I can... > > Best regards, > > Geoffrey Noer > noer@cygnus.com > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". -- Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) Rockwell International Electronic Commerce Division 300 Bauman Ct. banders@ecd.rockwell.com MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 Wood Dale, IL 60191 FAX (630) 227-9771 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From BSimon@randata.com.au Wed Dec 17 14:38:00 1997 From: BSimon@randata.com.au (Brendan Simon) Date: Wed, 17 Dec 1997 14:38:00 -0000 Subject: FW: Buiilding Vxds with gnu-win32 Message-ID: >>---------- >>From: vischne@ibm.net[SMTP:vischne@ibm.net] >>Sent: Wednesday, December 17, 1997 1:40 PM >>To: gnu-win32@cygnus.com.J.C.LAISECA >>Subject: Re: Buiilding Vxds with gnu-win32 >> >>>> Hi all! << >> >>> That's `y'all'. > >Only if you are American AND arrogant !! > >Brendan Simon > > > > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From malcolm@eml.co.uk Wed Dec 17 14:44:00 1997 From: malcolm@eml.co.uk (malcolm) Date: Wed, 17 Dec 1997 14:44:00 -0000 Subject: rx.h missing ? Message-ID: <349855E5.4C4F33B@eml.co.uk> I tried to build some code that uses regex.h, but regex.h includes rx.h (and nothing else) which does not seem to be part of the cdk. Do I have to down load the sources to get this, or is there another solution? Why not use the GNU standard regex.h? (Just curious about how regex works internally.) tia malcolm -- --------------------------------------------------------------- Malcolm Melville email: malcolm@eml.co.uk Enterprise Melville Limited Tel: +44 1435 869911 Network and Aviation Consultancy Fax: +44 1435 869912 --------------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dkegel@activision.com Wed Dec 17 14:44:00 1997 From: dkegel@activision.com (Dan Kegel) Date: Wed, 17 Dec 1997 14:44:00 -0000 Subject: du fails under winnt 4.0 after reading Recycled directory Message-ID: <34984160.30FB@activision.com> When I run beta18's du, it works fine until it hits the Recycled directory, after which, it says " No such file or directory" for all following files in the directory being checked. For example, 1 ./Recycled du: ./gtest: No such file or directory du: ./ffastun0.ffx: No such file or directory du: ./VSS: No such file or directory du: ./temp: No such file or directory . . and so on. this problem only happens on the D: drive, not on the C: drive, even though both have a hidden Recycled folder. This problem happens on drive D: even with the command du Recycled some_other_directory It seems dipping into the Recycled folder confuses du from then on, regardless of whether it gets there by recursion or because I specifies it explicitly. Has anybody else seen this? Thanks, Dan Kegel - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From leisner@sdsp.mc.xerox.com Wed Dec 17 15:04:00 1997 From: leisner@sdsp.mc.xerox.com (Marty Leisner) Date: Wed, 17 Dec 1997 15:04:00 -0000 Subject: how to get cdk (NBY) References: <199712161818.KAA16113@f128.hotmail.com> Message-ID: <9712172216.AA03912@gnu.sdsp.mc.xerox.com> Can an uncompressed tree be installed someplace? Or pieces zipped together... I really don't like the "install it my way approach" to system administation (from 15 years of experience). One of the things I like about unix is you should be able to do things how you want to... I also have had too many problems with uninstall on win95... -- marty leisner@sdsp.mc.xerox.com The Feynman problem solving Algorithm 1) Write down the problem 2) Think real hard 3) Write down the answer Murray Gell-mann in the NY Times - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ulittj00@mcl.ucsb.edu Wed Dec 17 15:04:00 1997 From: ulittj00@mcl.ucsb.edu (Joshua J Little) Date: Wed, 17 Dec 1997 15:04:00 -0000 Subject: first time compiling windows app with gcc References: <000c01bd0abe$5ce1ae40$4e804dd1@cs116bf.ihug.co.nz> Message-ID: I think you need to type something like: gcc -o hello.exe hello.c -lwin32 -Wl,-subsystem,windows Hope this helps - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From lizt@synopsys.com Wed Dec 17 15:04:00 1997 From: lizt@synopsys.com (B. Elizabeth Trojan) Date: Wed, 17 Dec 1997 15:04:00 -0000 Subject: looking for cygnus compatible perl Message-ID: <34983C75.1D706F6D@synopsys.com> Is there a version of Perl available for win32 that is Cygnus compatible? By Cygnus compatible I mean accepts UNC paths and knows about cygnus mounts. -- Liz Trojan Synopsys (503)531-2297 Office Beaverton (503)690-6906 FAX - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From stodghil@cs.cornell.edu Wed Dec 17 15:04:00 1997 From: stodghil@cs.cornell.edu (Paul Stodghill) Date: Wed, 17 Dec 1997 15:04:00 -0000 Subject: [Solved!] Re: [Workaround] Re: M-x shell + bash = Infinite error messages? References: <199712171141.LAA29386@shirow.long.harlequin.co.uk> Message-ID: Andrew Innes said: > On 16 Dec 1997 13:44:45 -0500, Paul Stodghill said: > >I figured out what was causing my problem: > > [problem was bash crashing when run in an Emacs shell buffer] > > >I was invoking emacs by using a command similar to, > > > > bash -c "runemacs ... some options ..." > > > >Changing the command to simply > > > > runemacs ... some options ... > > > >made by M-x shell problems go away. > > > >Why? > > runemacs starts Emacs then exits, so possibly an environment variable is > being set by bash which is passed on to Emacs, causing confusion later. Bingo! The environment variable in question is PID. Setting PID (either by using bash to invoke runemacs, or by M-x setenv) causes M-x shell to die a horrible death. Adding (setenv "PID" nil) to my ~/.emacs fixes the problem. Thanks for your help! -- Paul Stodghill http://www.cs.cornell.edu/home/stodghil/home.html - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Wed Dec 17 15:04:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Wed, 17 Dec 1997 15:04:00 -0000 Subject: Problems with malloc References: <971217084523206@snoopy.apptive.co.uk> Message-ID: <9712171942.AA11148@modi.xraylith.wisc.edu> "malloc.h" is not mandated by ANSI C and hence not needed by standard conforming programs. All the "malloc" and friends are declared in , so include that instead. Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From colin@bird.fu.is.saga-u.ac.jp Wed Dec 17 17:43:00 1997 From: colin@bird.fu.is.saga-u.ac.jp (Colin Peters) Date: Wed, 17 Dec 1997 17:43:00 -0000 Subject: first time compiling windows app with gcc Message-ID: <01bd0b56$149453c0$fa173185@gbird0.fu.is.saga-u.ac.jp> Given that your hello world program is *not* the example from Petzold's Programming Windows 95 (which uses the multi-media API to play a sound file) then it will probably compile if you do: gcc -o hello.exe hello.c -mwindows or gcc -o hello.exe hello.c -luser32 -lgdi32 -Wl,--subsystem,windows Which is roughly equivalent. Under Mingw32 0.1.4 you have to use -windows instead of -mwindows (that should change in the next version BTW) for the first type. For multiple module applications use the -mwindows option at the linking step (it's not necessary when just compiling). Colin. PS. If it *is* the example from Petzold then you need to add -lwinmm to either of the above command lines, and you need the extra mmsystem.h file that isn't currently included in the Win32 API headers from Cygnus (which I can give you if necessary). PPS. You might want to look at my (incomplete) tutorial on Win32 programming with GCC. Add gcc-tutor-contents.html to the URL in my signature (either one). -- Colin Peters -- colin at fu.is.saga-u.ac.jp -- Saga University Dept. of Information Science -- http://www.fu.is.saga-u.ac.jp/~colin -- http://www.geocities.com/Tokyo/Towers/6162 -----Original Message----- From: Kurt Hdusler To: gnu-win32@cygnus.com Date: Wednesday, December 17, 1997 5:25 PM Subject: first time compiling windows app with gcc >hello, this question has already been asked by carvalho@penelope.dss.inpe.br >but it wasnt answered. > >how can I compile programs when i get error messages like undefined >reference etc. this is just a program using the windows api to pop up one >window and saw hello world. > >I have compiled it before under two other compilers and one of them I had to >use a option like -lwin32 or something. What is the equivalent option >with this compiler? > >gcc hello.c is all I have been using. > >Please answer, I would be very greatful. > >Thanks from Kurt > > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From nordwick@graft.XCF.Berkeley.EDU Wed Dec 17 17:43:00 1997 From: nordwick@graft.XCF.Berkeley.EDU (Jason Alan Nordwick) Date: Wed, 17 Dec 1997 17:43:00 -0000 Subject: make and /bin/sh Message-ID: <199712180057.QAA11741@graft.XCF.Berkeley.EDU> this is my first time on the list, so this may have been answered before but i didnt see it in any of the searches (although I didnt search the list archive, b/c it was too much work). i have a clean rule in the makefile that does an rm *.o blah blah blah and make complains that it cant find /bin/sh, I tried making a \bin\sh but it didnt work. Anybody know a work around ? thanks, jay -- Join the FreeBSD Revolution. Support the FSF, buy GNU. http://xcf.berkeley.edu/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From turtill@ihug.co.nz Wed Dec 17 17:43:00 1997 From: turtill@ihug.co.nz (Kurt Häusler) Date: Wed, 17 Dec 1997 17:43:00 -0000 Subject: more information on my problem Message-ID: <006801bd0b43$6aaa4520$26804dd1@cs116bf.ihug.co.nz> This is in regard to being able to compile console apps fine but not a program using the win32 api. I have attached the source file. This is my command line: gcc -o hellowin.exe hellowin.c And this is compiler output stuff like: C:\WINDOWS\TEMP\cc0010611.o(.text+0x13f):hellowin.c: undefined reference to `BeginPaint@8' C:\WINDOWS\TEMP\cc0010611.o(.text+0x151):hellowin.c: undefined reference to `GetClientRect@8' C:\WINDOWS\TEMP\cc0010611.o(.text+0x167):hellowin.c: undefined reference to `DrawTextA@20' C:\WINDOWS\TEMP\cc0010611.o(.text+0x174):hellowin.c: undefined reference to `EndPaint@8' C:\WINDOWS\TEMP\cc0010611.o(.text+0x17f):hellowin.c: undefined reference to `PostQuitMessage@4' C:\WINDOWS\TEMP\cc0010611.o(.text+0x199):hellowin.c: undefined reference to `DefWindowProcA@16' gcc: Internal compiler error: program ld got fatal signal 1 Three people tried to help by pointing to various sites etc. that I have already read and understand but don't cover my problem. They seem to be only concerned with writing UNIX style console apps (but how many of them need to be written for a GUI like windows?). One person suggested explicitly linking libraries in this manner : gcc -o hello.exe hello.c -lwin32 -Wl,-subsystem,windows But it says Wl,-subsystem,windows is an undefined flag. (I cant find it anywhere on the docs either) Someone said adding -lkernal32 -luser32 but the compiler says: file -lkernal32 not found which looks as though its not recognizing the -l flag!! I have set up all the paths and environment vars correctly. Please help . I like the idea of the Cygnus project and I want to give it a fair go before I give up and move on to paying for Visual C++ Thanks a lot Kurt -------------- next part -------------- A non-text attachment was scrubbed... Name: Hellowin.c Type: text/x-c Size: 2743 bytes Desc: not available URL: From colin@bird.fu.is.saga-u.ac.jp Wed Dec 17 19:36:00 1997 From: colin@bird.fu.is.saga-u.ac.jp (Colin Peters) Date: Wed, 17 Dec 1997 19:36:00 -0000 Subject: more information on my problem Message-ID: <01bd0b65$d64fd340$fa173185@gbird0.fu.is.saga-u.ac.jp> From: Kurt H?sler To: gnu-win32@cygnus.com Date: Thursday, December 18, 1997 12:06 PM Subject: more information on my problem >One person suggested explicitly linking libraries in this manner : > >gcc -o hello.exe hello.c -lwin32 -Wl,-subsystem,windows > >But it says Wl,-subsystem,windows is an undefined flag. (I cant find it >anywhere on the docs either) > >Someone said adding -lkernal32 -luser32 but the compiler says: > >file -lkernal32 not found which looks as though its not recognizing the -l >flag!! You are the victim of multiple typos, as well as some bad (or misdirected) advice. The first person was missing a dash in the option (should be -Wl,--subsystem,windows) and also telling you to link a library that doesn't exist (-lwin32 will not work). The second person meant -lkernel32, but that's a bad idea too, because kernel32 is linked automatically and linking it twice can cause problems. It is normal, by the way, for the linker to report not being able to find the "file" -lfoo when really it can't find the library libfoo.a... just one of the quirks of the GNU linker. Here is a command line that works (I know it works because I cut and pasted from the window where I just ran it): gcc -o hellowin.exe hellowin.c -Wl,--subsystem,windows -luser32 -lgdi32 The command line above tells the linker to use a (yes, not officially documented yet, though it is mentioned in the FAQ at Cygnus) subsystem option to make a GUI application. It also tells it to link two extra libraries libuser32.a (user interface, window management and such) and libgdi32.a (graphics device interface) (BTW, I believe that is also mentioned in the FAQ at Cygnus). There is also an option -mwindows which sets -Wl,--subsystem,windows and links those libraries, along with a few others. So gcc -o hellowin.exe hellowin.c -mwindows Should also work (and I think that is also mentioned in the FAQ). Feel free to ask if you have more problems. Good luck, Colin. -- Colin Peters -- colin at fu.is.saga-u.ac.jp -- Saga University Dept. of Information Science -- http://www.fu.is.saga-u.ac.jp/~colin -- http://www.geocities.com/Tokyo/Towers/6162 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Wed Dec 17 21:04:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Wed, 17 Dec 1997 21:04:00 -0000 Subject: looking for cygnus compatible perl References: <34983C75.1D706F6D@synopsys.com> Message-ID: <9712180504.AA11666@modi.xraylith.wisc.edu> "B. Elizabeth Trojan" writes: > Is there a version of Perl available for win32 that is Cygnus > compatible? By Cygnus compatible I mean accepts UNC paths and > knows about cygnus mounts. > -- Check out Chris Faylor's port at http://www.tiac.net/users/cgf/ . A good place to check first is the "Ported Software" link on Cygnus' Gnu-Win32 home page ( http://www.cygnus.com/misc/gnu-win32/ ). Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Wed Dec 17 23:00:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Wed, 17 Dec 1997 23:00:00 -0000 Subject: still inetd problems Message-ID: <01BD0B9B.C5770530@gater.krystalbank.msk.ru> Nicola Bernardini wrote: > all the rights specified in Okhapkin's instructions, I removed all > tcp/ip services, launched inetd with Autoexnt, and I still get > this message from the inetd code in the Event log window: > > > The description for Event ID ( 0 ) in Source ( inetd ) could not be found. > It contains the following insertion string(s): > inetd : Win32 Process Id = 0x15C : CygWin32 Process Id = 0x42D : > select: error 0. > Looks like you've removed not only tcp/ip services, but the tcp/ip protocol too... :-) -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From noer@cygnus.com Wed Dec 17 23:51:00 1997 From: noer@cygnus.com (Geoffrey Noer) Date: Wed, 17 Dec 1997 23:51:00 -0000 Subject: Beta 19 status References: <3497EC77.7C28@switch.rockwell.com> Message-ID: <199712180751.XAA28617@rtl.cygnus.com> Bartlee A. Anderson wrote: > > Any chance for a b19 for Christmas? Soon, I hope, but it's vacation time for me now until the end of the year. I really will let people know when I have a hard date to announce. The "Coolview" dll from Sergey (see project web pages for link) has a lot of the improvements to Cygwin32 that will be in b19 so that's worth checking out for now. I can say that the longer it takes for it to come out, the higher quality the release will be but that might not be much consolation. :-) Happy holidays all! Geoffrey Noer noer@cygnus.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From colin@bird.fu.is.saga-u.ac.jp Wed Dec 17 23:51:00 1997 From: colin@bird.fu.is.saga-u.ac.jp (Colin Peters) Date: Wed, 17 Dec 1997 23:51:00 -0000 Subject: more information on my problem Message-ID: <01bd0b70$cbc57500$fa173185@gbird0.fu.is.saga-u.ac.jp> From: Colin Peters To: Kurt H?sler Cc: GNU-Win32 Date: Thursday, December 18, 1997 1:18 PM Subject: Re: more information on my problem [snip] >gcc -o hellowin.exe hellowin.c -mwindows > >Should also work (and I think that is also mentioned in the FAQ). Feel >free to ask if you have more problems. Actually that won't work (sorry). -mwindows can only be used as an option for linking, so you have to do it in two steps if you want to do it that way: gcc -o hellowin.o hellowin.c gcc -o hellowin.exe hellowin.o -mwindows That should work. Sorry about that, Colin. -- Colin Peters -- colin at fu.is.saga-u.ac.jp -- Saga University Dept. of Information Science -- http://www.fu.is.saga-u.ac.jp/~colin -- http://www.geocities.com/Tokyo/Towers/6162 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From noer@cygnus.com Wed Dec 17 23:51:00 1997 From: noer@cygnus.com (Geoffrey Noer) Date: Wed, 17 Dec 1997 23:51:00 -0000 Subject: PIMAGE_SECTION_HEADER? References: <3.0.1.32.19971217142958.0099d180@mailhost> Message-ID: <199712180745.XAA24204@rtl.cygnus.com> Andy Piper wrote: > > The supplied win32 headers do not include things like PIMAGE_SECTION_HEADER > which are defined in WINNT.H under > VC. What is the correct sequence of hoops to get this sort of info into the > distributed headers? How was the original information put in there? You need to find the information that pertains to the missing header file lines and then write entries for them without looking at the actual Microsoft headers. Finally, send me the diffs to the winsup/include/Windows32 files you modified to get things working with Cygwin32. -- Geoffrey Noer noer@cygnus.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From William.Gacquer@obspm.fr Thu Dec 18 00:12:00 1997 From: William.Gacquer@obspm.fr (root) Date: Thu, 18 Dec 1997 00:12:00 -0000 Subject: Why EGCS does not contain G++? ( ->implies problems !!!) Message-ID: <3498F6F9.12A24FEB@obspm.fr> First of all, I would like to thank everyone for the answers to my last posting. Well, I check that EGCS does not provide a g++ compiler but only a c++ one which is not tuned to accept the standard (c++) includes and libraries.( ex: ) Why is it like this? I would like to mention that a lot of makefiles do need g++ instead of c++? Egcs seems to be a good project but it misses g++!!! Thanks! William Gacquer PS: if you offer positions in physics or computer science, please mail me! - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Axel.Riese@risc.uni-linz.ac.at Thu Dec 18 00:41:00 1997 From: Axel.Riese@risc.uni-linz.ac.at (Axel Riese) Date: Thu, 18 Dec 1997 00:41:00 -0000 Subject: How to avoid Warning: Redefinition of ... Message-ID: <199712180840.AA01499@melmac.risc.uni-linz.ac.at> Dear cygwiners, I managed to patch most of the headers from the latest Platform SDK to make them useable for gcc (i.e., the preprocessor doesn't produce errors). However, I came across the following problem: A lot of header files typedef some symbols more than once, thus producing warnings like ... Warning: redefinition of blah_blah ... ... blah_blah previously defined in ... Is there a way to avoid these warnings that is not contained in the following list: - adding for each typedef something like #define blah_blah_defined and checking #ifndef blah_blah_defined ... #endif - forgetting about the Microsoft headers - closing my eyes during preprocessing and ignoring the warnings Any suggestions appreciated, Axel =========================================================================== Axel Riese Research Institute for Symbolic Computation J. Kepler University Linz A-4040 Linz e-Mail: Axel.Riese@risc.uni-linz.ac.at Austria URL: http://www.risc.uni-linz.ac.at/people/ariese/home/ =========================================================================== - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From P.J.Kersten@dia.eur.nl Thu Dec 18 00:48:00 1997 From: P.J.Kersten@dia.eur.nl (P.J. Kersten) Date: Thu, 18 Dec 1997 00:48:00 -0000 Subject: b18: fileaccess LS UNreliable References: <199712171725.SAA21080@solair1.inter.NL.net> Message-ID: <3498E343.3E68C90@dia.eur.nl> jbessels wrote: > [snip] > Observation1: For Bash and Command Line (CLI) is true: > ls > dummy is OK. > ls -l > ../dummy or ls -ptl > ../dummy is OK. > ls -l > dummy or ls -ptl > dummy gives following error: > dummy Permission denied. (checked the above for a couple of dirs) > Conclusion: redirection doesn't work if the file is in the same > directory. > Have tried it on WIn95 as well same result. Works fine under Linus > though, should work !! > Not so fast. NT uses an exclusive locking scheme by deafult. If you pipe the ls-results into a file, bash ownes that file. ls sees it, but can't access it (because it is locked) and gives an error as result. As long as you ommit the -l parm it works ok. The order of access in ls prevents it from being more verbose. Using a native ls it is able to report at least is size and date before it encounters the same problem. > Observation2: Standing in rootdir / > CLI. Dir pagefile.sys is OK. > BASH. Dir pagefile.sys is ERROR (using cygnus dir. cmd) > BASH. ls pagefile.sys is ERROR. > CLI. ls pagefile.sys is ERROR. In all three cases error is --> dir: > pagefile.sys No such file or directory. > Ohter files can be accessed, pagefile is swapfile. > Conclusion: Under Dos using dir all is OK, using cygnus ls (and possibly > other ones as well) give error. Same as before: pagefile.sys is owned by NT itself. It is therefore locked. As long as you ommit the -l flag, it works ok. > Observation3: Standing in rootdir / (have installed SMS) > dirs SMSdata.dat smslog.dat are allowed to be accessed by the > adminstrator > and I'm logged in as adminstrator. > CLI. dir SMSdata.dat and dir SMSlog.dat is OK. > CLI + BASH. ls SMSdata.log and ls SMSlog.dat is ERROR --> permission > denied. > I don't know about SLS internals, but reasoning from the above I say it is a service up and running and owning those files? > obvervation4: This one is beautiful. Standingin /Winnt > CLI. dir net*.* ==> two files shown, netlogon.chk and network.wri. > BASH. dir net*.* (cygnus dir) and ls net*.* only show the file > network.wri. > NO error is displayed. > BASH ls -l net*.*. only shows network.wri and no error. > BASH ls -l net*.* | head -5 gives networ.wri and sometimes the error > "ls: > netlogon.chk. permission denied." I'm still running NT3.51 and I don't have these files, but I suspect an instance of the above mentioned mechanism. > > I've tried to be as accurate as possible, but maybe a mistake has > slipped > in. My conclusion is. Cygnus-gnuwin32 seems nice, pipes et al works like > as > charm just like all commands BUT I can't trust Cygnus-gnuwin32 because > sometimes not al files are shown or aren't handled correctly. This makes > it > totally unusable for me. Am I doing something wrong here or..... I've > tried > to download coolview (november version) but that site is so unbelievable > > slow I can't download coolview at all. If someone could send me the > LATEST > coolview version this would be much appreciated. > If you like, I can sent it to you by email. > OK guys, I'm completely lost here. I don't know whats going on. Most > examples can be tried by everone. I'm very interested if you can > reproduce > my errors. Please let me know.... [snap] HTH, -- Pieter J. Kersten Erasmus Universiteit Rotterdam dir. I&A/I&S, WH2-10 Postbus 1738 3000 DR Rotterdam Nederland Tel +31-10-4082599 Fax +31-10-4527236 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From andyp@parallax.co.uk Thu Dec 18 01:57:00 1997 From: andyp@parallax.co.uk (Andy Piper) Date: Thu, 18 Dec 1997 01:57:00 -0000 Subject: gcc executable file format? Message-ID: <3.0.1.32.19971218095817.00967330@mailhost> Hi, Does gcc from gnu-win32 generate the same executable file format as MSVC? The reason I ask is that I am hoping to build unexnt.c from emacs using gcc but dumping obviously relies on executable file format. andy ___ ____ Dr Andy Piper / _ \___ ________ _/ / Solutions_ (require 'disclaimer) / ___/ _ `/ __/ _ `/ / / _ `/\ \ / andyp@parallax.co.uk /_/ \_,_/_/ \_,_/_/_/\_,_//_\_\ boot /vmemacs - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From andyp@parallax.co.uk Thu Dec 18 02:06:00 1997 From: andyp@parallax.co.uk (Andy Piper) Date: Thu, 18 Dec 1997 02:06:00 -0000 Subject: Pre-built cvs-1.9.20 for NT Message-ID: <3.0.1.32.19971218095629.00963bc0@mailhost> Hi, I have put a pre-built version of cvs-1.9.20 client/server for NT with my pserverd patches on my web-page http://www.parallax.co.uk/~andyp . You should find all of the necessary files to get up and running on that page. 1.9.20 has $Log$ problems which means I personally don't currently use it as a server, but it is easy to get up and running because you don't need patch or rcs, and I don't currently have a patched prebuilt 1.9.18 lying around. andy ___ ____ Dr Andy Piper / _ \___ ________ _/ / Solutions_ (require 'disclaimer) / ___/ _ `/ __/ _ `/ / / _ `/\ \ / andyp@parallax.co.uk /_/ \_,_/_/ \_,_/_/_/\_,_//_\_\ boot /vmemacs - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Thu Dec 18 02:06:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Thu, 18 Dec 1997 02:06:00 -0000 Subject: Bug Report: b18 path involves drive letter (NBY) References: <19971214233009.23962.qmail@hotmail.com> Message-ID: <19971218210519.18631@murlibobo.cs.mu.OZ.AU> On 14-Dec-1997, Earnie Boyd wrote: > You need to: > > mount -b E:\\ /e > > then you can: > > cd /e > pwd > /e > cd .. > pwd > / > > and be in the correct places. That's a reasonable work-around, but it's not a fix. -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From alex@greco2.polytechnique.fr Thu Dec 18 02:26:00 1997 From: alex@greco2.polytechnique.fr (Alessandro Mirone) Date: Thu, 18 Dec 1997 02:26:00 -0000 Subject: questions from an average user Message-ID: <34990805.56E05C26@greco2.polytechnique.fr> Hi people , iI tried sometimes ago to use your gnuwin32 b18 to build some DLLs but after many tries I gave it up because it was impossible for my level of knowledge, although I have a long experienc in c and c++ programming, to get a working DLL, even following your faq and the Henderson suggestions. As I keep looking now and then at your http site, and I don't find any useful information about this issue I would like to ask you: - Has anything changed respect to the b18 of september 97? - Could you provide a better and updated documentation about building DLLs? - Is gnuwin32 made for eperimented hacker only? - If someone who is able to build DLLs could contact me it would be of great help for me Thank you very much a dum average user Alessandro - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From andrewi@harlequin.co.uk Thu Dec 18 04:33:00 1997 From: andrewi@harlequin.co.uk (Andrew Innes) Date: Thu, 18 Dec 1997 04:33:00 -0000 Subject: [Bug] M-x setenv PID + bash -> M-x shell breaks References: Message-ID: <199712181232.MAA24367@shirow.long.harlequin.co.uk> On 17 Dec 1997 17:22:44 -0500, Paul Stodghill said: >Setting the environment variable "PID" causes shell commands to break when >bash is the shell that emacs invokes. > >Using NTEmacs 19.34.6 and Cygnus gnu-win32 tools "b18", do the following. > >1. start emacs > > c:\> runemacs -q > >2. Evaluate the following in the *scratch* buffer > > (setq shell-file-name "bash") > (setq explicit-shell-file-name shell-file-name) > (setenv "PID" 100) > >3. Invoke M-x shell. Boom. > >4. Evalute the following in the *scratch* buffer > > (setenv "PID" nil) > >5. Invoke M-x shell. It works! > >Why this happens, I have no idea. I'm not even sure if its an emacs or bash >bug... It is presumably a bug in bash. Emacs doesn't care about PID. AndrewI - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 18 04:48:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 18 Dec 1997 04:48:00 -0000 Subject: patch and dos text (NBY) Message-ID: <19971218124737.1459.qmail@hotmail.com> Joseph, Most of these tools will not like the \r\n combination as they are ports from the friendly UNIX world. You must have your directories mounted as text=binary. You must use the cygnus supplied gzip and tar programs to unzip the file sets. There has been much discussion about text=binary on the list, please check the mail archives pointer below. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: "Joseph E. Banta, Esquire" >To: "'gnu-win32@cygnus.com'" >Subject: patch and dos text >Date: Wed, 17 Dec 1997 11:55:24 -0500 > >Has anyone had the problem that 'patch' works strangely with dos text files, >or is it jus me? > >Joseph Banta >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 18 04:48:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 18 Dec 1997 04:48:00 -0000 Subject: questions from an average user (NBY) Message-ID: <19971218124211.2694.qmail@hotmail.com> Alessandro, I don't know if you have this list of pointers or not. Possibly you could get the coolview updates from Sergey Okhapkin's page. Most of the changes he has made will be in release B19. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Thu, 18 Dec 1997 12:24:53 +0100 >From: Alessandro Mirone >To: gnu-win32@cygnus.com >Subject: questions from an average user > >Hi people , > > iI tried sometimes ago to use your gnuwin32 b18 >to build some DLLs but after many tries I gave it up >because it was impossible for my level of knowledge, >although I have a long experienc in c and c++ programming, >to get a working DLL, even following your faq and the Henderson >suggestions. > >As I keep looking now and then at your http site, and I don't find any >useful information about this issue I would like to ask you: > > - Has anything changed respect to the b18 of september 97? > - Could you provide a better and updated documentation about >building DLLs? > - Is gnuwin32 made for eperimented hacker only? > - If someone who is able to build DLLs could contact me it would >be of great help for me > > Thank you very much > > a dum average user > > Alessandro > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 18 05:01:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 18 Dec 1997 05:01:00 -0000 Subject: make and /bin/sh Message-ID: <19971218130037.29809.qmail@hotmail.com> Hi Jay, Searching through the archives for information isn't too much work, it is necessary work. Now the Y2K problem, this is an example of necessary work that is TOO MUCH work (more work than can be accomplished in the allotted amount of time). To solve your problem: BASH# mount -b C:\\gnuwin32\\b18\\H-i386-cygwin32\\bin /bin This example assumes: You've installed in the default directories Your executing from BASH BASH# is the command line prompt string from bash. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Wed, 17 Dec 1997 16:57:58 -0800 (PST) >From: Jason Alan Nordwick >To: gnu-win32@cygnus.com >Subject: make and /bin/sh > > >this is my first time on the list, so this may have been answered before >but i didnt see it in any of the searches (although I didnt search the >list archive, b/c it was too much work). > >i have a clean rule in the makefile that does an rm *.o blah blah blah >and make complains that it cant find /bin/sh, I tried making a \bin\sh >but it didnt work. > >Anybody know a work around ? > >thanks, >jay >-- >Join the FreeBSD Revolution. >Support the FSF, buy GNU. > http://xcf.berkeley.edu/ >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 18 05:09:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 18 Dec 1997 05:09:00 -0000 Subject: How to avoid Warning: Redefinition of ... Message-ID: <19971218130854.13277.qmail@hotmail.com> Axel, I found in the gcc.info file(s) in the Overall Option node a section on switches about for warnings. I believe that you want the -Wno-redundant-decls switch. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Subject: How to avoid Warning: Redefinition of ... >To: gnu-win32@cygnus.com (GNU-Win32) >Date: Thu, 18 Dec 1997 09:40:53 +0100 (MET) >From: "Axel Riese" > >Dear cygwiners, > >I managed to patch most of the headers from the latest Platform SDK to >make them useable for gcc (i.e., the preprocessor doesn't produce errors). > >However, I came across the following problem: > >A lot of header files typedef some symbols more than once, thus producing >warnings like > > ... Warning: redefinition of blah_blah ... > ... blah_blah previously defined in ... > > >Is there a way to avoid these warnings that is not contained in the following >list: > >- adding for each typedef something like #define blah_blah_defined and > checking #ifndef blah_blah_defined ... #endif >- forgetting about the Microsoft headers >- closing my eyes during preprocessing and ignoring the warnings > > >Any suggestions appreciated, > > Axel > > >=========================================================================== >Axel Riese >Research Institute for Symbolic Computation >J. Kepler University Linz >A-4040 Linz e-Mail: Axel.Riese@risc.uni-linz.ac.at >Austria URL: http://www.risc.uni-linz.ac.at/people/ariese/home/ >=========================================================================== >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Thu Dec 18 06:31:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Thu, 18 Dec 1997 06:31:00 -0000 Subject: [Bug] M-x setenv PID + bash -> M-x shell breaks References: <199712181232.MAA24367@shirow.long.harlequin.co.uk> Message-ID: <19971219013059.30687@mundook.cs.mu.OZ.AU> On 18-Dec-1997, Andrew Innes wrote: > On 17 Dec 1997 17:22:44 -0500, Paul Stodghill said: > >Setting the environment variable "PID" causes shell commands to break when > >bash is the shell that emacs invokes. ... > >Why this happens, I have no idea. I'm not even sure if its an emacs or bash > >bug... > > It is presumably a bug in bash. Emacs doesn't care about PID. I think it is more likely to be a bug in gnu-win32 (i.e. cygwin.dll and/or libcygwin.a), which does care about PID. -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From v.canestrale@isnet.it Thu Dec 18 08:24:00 1997 From: v.canestrale@isnet.it (Vincenzo Canestrale) Date: Thu, 18 Dec 1997 08:24:00 -0000 Subject: How to compile a .RC file to obtain a .RES file? Message-ID: <16250261745196@isnet.it> Many many thanks to everybody in this list that helped me to compile my _first_ Win95 C++ program (thanks to Earnie, Fergus, Scott and hgfernan) >>>>Now I have a new Win95 program problem... :-) <<<<<<<<< >>>> How can I compile a .RC file to obtain a .RES file??? <<<< To use a menu in a Win95 prog I have to link a .RES file to my main program, but I don't know which is the application that convert a .RC file like this: #include "menu.h" MYMENU MENU { POPUP "&Uno" { MENUITEM "&Alpha", IDM_ALPHA .................. } into a .RES file. I hope You can help me. v.canestrale@isnet.it http://www.geocities.com/WallStreet/5385 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From juolja@utu.fi Thu Dec 18 09:18:00 1997 From: juolja@utu.fi (Juha Jäykkä) Date: Thu, 18 Dec 1997 09:18:00 -0000 Subject: Problem with ln Message-ID: <34995AC9.D23E3FC5@utu.fi> I am trying to compile the CDK on a NT 4.0 machine, but I have a problem. While running configure, it always fails due to the same error: --cut-- ln: /cdk/libiberty/alloca-norm.h: No such file or directory *** /cdk/configure: unable to link "alloca-conf.h" to \ "/cdk/libiberty/alloca-norm.h". --cut-- I have confirmed the file exists and that I have write access to . and even rw access to the /cdk/libiberty/alloca-norm.h. I cannot figure out why this might be. Can someone suggest a solution? -- ----------------------------------------------- | Juha J????ykk????, juolja@utu.fi | | home: http://www.utu.fi/~juolja/ | | pgp-key: http://www.utu.fi/~juolja/pgp.txt | ----------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@buggy.prospect.com.ru Thu Dec 18 11:21:00 1997 From: sos@buggy.prospect.com.ru (Sergey Okhapkin) Date: Thu, 18 Dec 1997 11:21:00 -0000 Subject: PIMAGE_SECTION_HEADER? Message-ID: <01BD0C01.D41CF4D0@sos> Geoffrey Noer wrote: > You need to find the information that pertains to the missing header > file lines and then write entries for them without looking at the > actual Microsoft headers. Finally, send me the diffs to the Hmm... The _only_ source of information (direct or indirect) is MS headers... -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From u-gwb@nta.com Thu Dec 18 11:21:00 1997 From: u-gwb@nta.com (Gene Buckle) Date: Thu, 18 Dec 1997 11:21:00 -0000 Subject: CYGWIN.DLL and VC5 Message-ID: Is it possible to use the cygwin.dll with Visual C 5? I've got a project I'm working on that I need to be able to access some of the non win32 functions (setitimer(), etc). I really can't use gcc to build with as I've got way too many pre-compiled libraries to work with. Thanks. g. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Thu Dec 18 11:21:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Thu, 18 Dec 1997 11:21:00 -0000 Subject: Why EGCS does not contain G++? ( ->implies problems !!!) References: <3498F6F9.12A24FEB@obspm.fr> Message-ID: <9712181628.AA12219@modi.xraylith.wisc.edu> My apologies for not having more time to help with egcs installation problems. Right now I'm a bit pressed for time with other projects. I would like to clarify a few things so there is no misunderstanding: 1 egcs-1.00 provides a C++ compiler. Originally the executable used to called g++, and then there was also c++, which was typically just linked to g++. The difference was (there is no difference in egcs-1.00) that g++ will also link with -lg++ (which is not a standard library) in addition to whatever library was needed for C++ run-time suport. Currently, c++ and g++ both just link to -lstdc++ to profide the C++ run-time support. libg++ is deprecated, or at least not supported, so it's not linked in by default anymore. EGCS provides *both* executables, c++ and g++, which are basically the same thing. 2 egcs-1.00, just like FSF gcc distribution, uses on environment variables to find paths to include files, libraries, compilers etc; when you use a binary distribution like the ones for gnu-win32/mingw32, you have to set these correctly to get things working. Here's a list of the most essential ones (for the shell and egcs): * PATH: Your egcs binary directory must come before cgywin32 b18 binary directory if you have these under different trees. eg., PATH=c:\egcs\bin;c:\gnuwin32\b18\H-i386-cygwin32\bin;%PATH% * GCC_EXEC_PREFIX: gcc uses this one to find the actual compilers such as cc1 and cc1plus and also other files such libgcc.a and crt0.o. eg., GCC_EXEC_PREFIX=c:\egcs\lib\gcc-lib\ The trailing slash is important for a variety of reasons. * LIBRARY_PATH: gcc looks for various libraries using this path. eg., LIBRARY_PATH=//c/egcs/H-i386-cygwin32/i386-cygwin32/lib: //c/egcs/H-i386-cygwin32/lib * C_INCLUDE_PATH: include path list for "C" include files (stdio.h eg) eg., C_INCLUDE_PATH=//c/egcs/H-i386-cygwin32/i386-cygwin32/include: //c/egcs/include * C_PLUS_INCLUDE_PATH: include path list for "C++" includes. Must also specify the paths to C includes as well. eg., CPLUS_INCLUDE_PATH=//c/egcs/include/g++://c/egcs/include: //c/egcs/H-i386-cygwin32/i386-cygwin32/include * OBJC_INCLUDE_PATH: include path list for "objC" includes. Usually the same as C include path: eg., OBJC_INCLUDE_PATH=%C_INCLUDE_PATH% And others such as COMPILER_PATH, but usually you can do without it if you define the ones above. 3 EGCS provides *both* c++.exe and g++.exe, and if you don't have one of the two, something is wrong with the tar program used to unpack the distribution. In the binary distribution, g++.exe is a link to c++.exe, and tar should have done the right thing (either use symbolic link or make a new file). If you have one that is zero-sized, simply copy over the other one on top. Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ulittj00@mcl.ucsb.edu Thu Dec 18 11:21:00 1997 From: ulittj00@mcl.ucsb.edu (Joshua Little) Date: Thu, 18 Dec 1997 11:21:00 -0000 Subject: first time compiling windows app with gcc Message-ID: <01bd0460$e0c4b040$7ffd6f80@Slittj00.ucsb.edu> Sorry for the wrong advice. I was at work and was giving the example from memory. As already stated by someone Colin Peters, it should be: gcc -o hellowin.exe hellowin.c -Wl,--subsystem,windows -luser32 ... Sorry for the mistakes. Josh - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From rdmiller@execpc.com Thu Dec 18 11:21:00 1997 From: rdmiller@execpc.com (Rick Miller) Date: Thu, 18 Dec 1997 11:21:00 -0000 Subject: Compiling "V" 1.16 - Specific Instructions? Message-ID: <01bd0bea$8e0e5a40$0651040a@rmiller.cooperpower.com> Hi all! After taking the bait in the gnu-win32 FAQ, "What is V?" I went out and got the sources and tried to take the given advice for compiling... Problem is, there are at least FOUR sets of such advice that I know of and at least two of them didn't work. Would someone be kind enough to give specific instructions for building V-1.16 under gnuwin32-b18? Some questions that I didn't see addressed at all (they may not be important): o When building V, does it matter whether I've mounted my file system mixed? o Which source code should I use? "v-1.16.tar.gz" contains all the nicely-cased UNIXy filenames but "vwin116.zip" has them mostly in upper case. GNU-make is being picky about case, so the Makefile has to match it as far as I know. o Are these instructions already posted somewhere? WORKING ones? The default Config.mk tells me to use ARCH=intel, there's one in ftp://ftp.cs.unm.edu/pub/wampler/config/gnuwin32.make that tells me to use ARCH=gnu-win32... then there's other stuff in the "v/contrib/gnuwin32.tar" file in the source code... I'm so confused! Does anyone have instructions I can use? I'd hate to have to try all the combinations of the different ones I've already found, it would take months. Rick Miller http://www.execpc.com/~rdmiller/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sysadmin@almus.com Thu Dec 18 12:10:00 1997 From: sysadmin@almus.com (Jon S. Jaques) Date: Thu, 18 Dec 1997 12:10:00 -0000 Subject: looking for cygnus compatible perl Message-ID: <01bd0be6$8b20d410$0200dede@epsilon> >Is there a version of Perl available for win32 that is Cygnus >compatible? By Cygnus compatible I mean accepts UNC paths and >knows about cygnus mounts. That would be "the" Perl; ie that is v5.004_04, the very same that is compiled under Unix/Linux. ( http://www.perl.org/CPAN/src/latest.tar.gz ) Is it easy? Supposedly, although I've not yet gotten it to work, but haven't given up. Assuming that you've installed "cdk.exe" into the default location, there're two updates you need: 1.) The new cygwin.dll, and 2.) Chris Faylor's Perl patch which is specific to Cygwin32 installation/compilation. (The cygwin.dll seems to be distributed from a users' site, along with a new Bash.exe, among other things, and I can't tell if their presence, or lack thereof, affects the perl install/compile.) These are the steps that were advised to me: >>Here is what I would try: >> >> mkdir /usr/src/perl >> mount -b c:\usr\src\perl /perlsrc >> bash >> cd /perlsrc >> tar xzf /wherever/perl5.004_04.tar.gz >> mv perl5.004_04/* . >> tar xzf /whereever/cygperl5.004_04.pat.tar.gz >> /bin/sh cygperlpat5.004_04 >> patch -p1 -N < cygperlpat5.004_04 >> sh Configure And then, when it didn't work, we went here: >It is a prerequisite that you are able to compile programs. Does your >gcc work? Can you successfully build a simple "hello world" program? > >Also, the patched README file *does not* say to copy files anywhere. Note this last, which IS different from the original Perl README. If you've already followed those instructions, then I'd reccomend deleting the files "gcc2" and "ld2" which you theoretically already copied into your path. Personally, I can compile a simple file, program1.cpp ("hello world") with *g++* but NOT gcc (a friend suggested that the file was a "plus-plus" and that gcc is not, but I don't really know much about either... my primary reason for this thing is just to get the new Perl compiled on my system to open up some options in my Perl programming... one language at a time, ple ase! ) Please, if you have no luck, then lets' continue this dialogue-- I'd really like to get this resolved. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Thu Dec 18 12:10:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Thu, 18 Dec 1997 12:10:00 -0000 Subject: libX11 fix [Re: Egcs-1.0] References: <01bd0b3f$2b842e50$0100007f@coent.demon.co.uk> Message-ID: [I'm cc'ing this to gnu-win32 list as well. The problem is that applications linked against the X11 dlls fail when using egcs-1.00 release.] Hi David, The reason that executables linked against libX11.a from Silvio's package fail is because libX11.a built with an older version of dlltool that had a subtle bug which shows up in the newer ld that comes with egcs. Fortunately there is a trivial fix all the export libraries (eg., libX11.a) if you have the corresponding export list (eg., libX11.def). I don't have the export list for libX11.dll, but I just used impdef from BC5.x to make one (attached). Here's the command I used to regenerate libX11.a that fixes the problem: % dlltool --def libX11.def --output-lib libX11.a --dllname libX11.dll Now all you have to do is relink your application and things should be back to normal. Mumit From jeffdbREMOVETHIS@netzone.com Thu Dec 18 12:10:00 1997 From: jeffdbREMOVETHIS@netzone.com (Mikey) Date: Thu, 18 Dec 1997 12:10:00 -0000 Subject: SOLVED!! linking against MS .lib files in B18 (and possibly egcs/B19) References: <199710300353.WAA18385@subrogation.cygnus.com> Message-ID: <349971b2.3318041@smtp.netzone.com> Hi All ;^) I have some good news. Attached are patches for producing a beta 18 compatible (and possibly egcs/B19 compatible) ld.exe that will allow linking with MS .lib files, AND libcygwin.a type files. I just posted copies of fixed ld.exe objdump.exe dlltool.exe the specs file, and a couple of needed .o's to ftp://ftp.deninc.com/incoming as B18-patched-link-compat-ld.tgz for people who don't want to rebuild from source. also posted SDK_4_ming_cygwin32.tar.gz which has the header patches for using the Platform SDK headers with ming & cygwin The patch makes use of a couple of the routines posted to the list by Marcus Hall. marcus@bighorn.dr.lucent.com Thanks Marcus!!! ;^) User visible changes include strip.exe works on images with .reloc sections now. ie strip --strip-unneeded cygwin.dll gives a usable .dll (still can't strip files with .rsrc sections) .reloc sections are now always terminated properly on win95, and NT w/out C2 security (the tradeoff is that you will always have a .reloc section unless you strip it off ;^) new options for ld.exe --subsystem wwindows --subsystem wconsole (entry @ _wWinMainCRTStartup _wmainCRTStartup respectively for UNICODE apps currently only useful when linking against the real msvcrt.lib & using vc++ headers) -lLIBNAME option to ld.exe (small L) searches first for LIBNAME.lib files, then libLIBNAME.a (should make for easier use under command.com/cmd.exe) new options for gcc.exe -windows (gui apps pull in the patched Platform SDK headers if you have them see SDK_4_ming_cygwin32.tar.gz) -console (console ditto ;^) use -mwindows or -mconsole for correct .o's with the FSF headers/libs -dll (links in the appropriate stub, main_dll.o for windows dll's WinMain_dll.o for console dll's) -wm (mingw32+Platform SDK) change the paths in specs as appropriate for your system entry point WinMainCRTStartup added to crt0.c/o in binary dist (gets rid of those pesky warnings ;^) relocatable .dll's & .exe's work with the specs file included in binary dist when linked against libcygwin.a (only tested with coolview) so to produce a simple nonrelocatable gui dll from source with a DllMain entry point dlltool --dllname dllmain.dll --output-exp dllmain.exp --output-lib dllmain.a --def dllmain.def gcc -windows -dll -Wl,--image-base,0x1C000000 -o dllmain.dll dllmain.exp dllmain.c for a relocatble .dll you must still do 2 passes gcc -windows -dll -Wl,--base-file=dllmain.base -o dllmain.dll dllmain.c dlltool --dllname dllmain.dll --base-file dllmain.base --output-exp dllmain.exp --output-lib dllmain.a --def dllmain.def gcc -windows -dll -o dllmain.dll dllmain.exp dllmain.c Also a patch that should do the same thing for egcs/B19, produced against the gas-971123 snap (MAYBE ;-) >From MS's PEI spec >5.5.6. COMDAT Sections (Object Only) >The Selection field of the Section Definition auxiliary format is >applicable if the section is a COMDAT section: a section that >can be defined by more than one object file. (The flag >IMAGE_SCN_LNK_COMDAT is set in the Section Flags field >of the section header.) The way that the linker resolves the multiple >definitions of COMDAT sections is determined by the Selection field. >The first symbol having the section value of the COMDAT section >is the section symbol This symbol has the name of the section, >Value field equal to 0, the section number of the COMDAT section >in question, Type field equal to IMAGE_SYM_TYPE_NULL, Class >field equal to IMAGE_SYM_CLASS_STATIC, and one auxiliary >record. The second symbol is called "the COMDAT symbol" and >is used by the linker in conjunction with the Selection field. <<<<<<<<< the main thing I have added a routine to the bfd backend (in pei-i386.c) to offer the comdat symbol/associated section info in a seperate routine pei_get_comdat_info() then fixed ld and objdump to use it. Also added a struct to asection->pe_comdat_info to offer the info to applications, this seemed to be the lesser of all evils, if it breaks other backends, the #ifdefs and extern funcs wouldn't be hard to set up. also minor changes to dlltool to allow inter operability with vc++ generated import libraries/objects. exported symbols can start with ? (c++) as well as _ (c) this will allow generated .def files with v(c++) symbols to compile also corrects _imp__ symbol names. The changes were made against the gas-971123 snap, as well as the B18 source tree. / * THIS SOFTWARE IS NOT COPYRIGHTED * * This source code is offered for use in the public domain. You may * use, modify or distribute it freely. * * This code is distributed in the hope that it will be useful but * WITHOUT ANY WARRANTY. ALL WARRENTIES, EXPRESS OR IMPLIED ARE HEREBY * DISCLAMED. This includes but is not limited to warranties of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ TODO objcopy/strip fixed for stripping exe's/dll's with .rsrc sections objcopy/strip associated sections removed when COMDAT section stripped (not critical as the new ld should ignore associated sections without a COMDAT) dlltool still needs to be fixed to produce more MS compatible import libraries so link.exe can be used with them On Wed, 29 Oct 1997 22:53:23 -0500, you wrote: > From: jeffdbREMOVETHIS@netzone.com (Mikey) > Date: Thu, 30 Oct 1997 01:11:06 GMT > > There are quite a few binary incompatibilities like this left between cygwin32 > and MS pei that need to be corrected, but it has been so long since the last < snip> > >That particular problem was fixed some months ago. It means that if >you want to use the new linker and/or the new assembler, you must >recompile and reassemble every .o and .a file. > >Ian > (jeffdbREMOVETHIS@netzone.com) delete REMOVETHIS from the above to reply Mikey Windows vs Linux is a no-win situation. Windows 95: -- 32-bit extensions and a GUI shell for a 16-bit patch to an 8-bit OS originally coded for a 4-bit processor written by a 2-bit company that can't produce 1 bit of quality. -------------- next part -------------- A non-text attachment was scrubbed... Name: bfd_ld_comdat_sym_B18.dif.gz Type: application/x-gzip Size: 6945 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: bfd_ld_comdat_sym_gas971123.dif.gz Type: application/x-gzip Size: 6600 bytes Desc: not available URL: From rduke@alphatronix.com Thu Dec 18 12:10:00 1997 From: rduke@alphatronix.com (Bob Duke) Date: Thu, 18 Dec 1997 12:10:00 -0000 Subject: Fw: GNU-Win32 info Message-ID: <01bd0bf0$9b315500$7bca09c0@lynx> -----Original Message----- From: Bob Duke To: info@cygnus.com Date: Thursday, December 18, 1997 1:27 PM Subject: GNU-Win32 info >Can anyone give me information about the status of the GNU-Win32 tools for >the Windows NT/Intel platform? We are porting a Unix app that is built in a >Sun environment with various GNU tools (gmake, flex, g++, etc.) and are >primarily interested in just being able to use reasonably functional and/or >supported tools. I have tried your B18 beta, and had problems with file >system corruption being reported when using tar and rm -rf on an ntfs volume >(I think that rm was just confused, but it made me want something past the >beta). Anyway, is anything past the beta available? > Thanks - Bob Duke > Alphatronix > > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From derek@silver.clti.com Thu Dec 18 13:22:00 1997 From: derek@silver.clti.com (Derek C. Lee) Date: Thu, 18 Dec 1997 13:22:00 -0000 Subject: Disappearing mount points with WinNT 3.51/4.0 Message-ID: <9712182119.AA03729@silver.clti.com.> In win95 I don't need to remount my mount points once they are set. In WinNT 4.0 and WinNT 3.51, my mount points disappear if I close my bash session. How do I make the mount points persistent in WinNT? I know the mount points live in the registry but I'm not sure how to keep them there. Any help would be greatly appreciated. Thanks, Derek C. Lee derek@clti.com Chain Link Technologies, Inc. http://www.clti.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Thu Dec 18 16:04:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Thu, 18 Dec 1997 16:04:00 -0000 Subject: How to compile a .RC file to obtain a .RES file? Message-ID: <19971218235740.26343.qmail@hotmail.com> Go to Mumit Khan's page and get windres. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: "Vincenzo Canestrale" >To: "GnuWin32 Mailing List" >Subject: How to compile a .RC file to obtain a .RES file? >Date: Wed, 17 Dec 1997 14:00:28 +0100 > >Many many thanks to everybody in this list >that helped me to compile my _first_ Win95 C++ program >(thanks to Earnie, Fergus, Scott and hgfernan) > >>>>>Now I have a new Win95 program problem... :-) <<<<<<<<< >>>>> How can I compile a .RC file to obtain a .RES file??? <<<< > >To use a menu in a Win95 prog I have to link a >.RES file to my main program, but I don't know >which is the application that convert a .RC file like this: > >#include "menu.h" >MYMENU MENU >{ > POPUP "&Uno" > { > MENUITEM "&Alpha", IDM_ALPHA > .................. >} > >into a .RES file. >I hope You can help me. >v.canestrale@isnet.it > http://www.geocities.com/WallStreet/5385 >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dahms@ifk20.mach.uni-karlsruhe.de Thu Dec 18 16:04:00 1997 From: dahms@ifk20.mach.uni-karlsruhe.de (dahms@ifk20.mach.uni-karlsruhe.de) Date: Thu, 18 Dec 1997 16:04:00 -0000 Subject: Compiling "V" 1.16 - Specific Instructions? Message-ID: <009BEFA4.E6F2D2A0.30435@ifk20.mach.uni-karlsruhe.de> Hi Rick, you wrote: : o Which source code should I use? "v-1.16.tar.gz" contains all the : nicely-cased UNIXy filenames but "vwin116.zip" has them mostly in upper : case. GNU-make is being picky about case, so the Makefile has to match it : as far as I know. Try using an unzipper which allows to force all lowercase names. Bye, Heribert (dahms@ifk20.mach.uni-karlsruhe.de) - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jeffdbREMOVETHIS@netzone.com Thu Dec 18 17:11:00 1997 From: jeffdbREMOVETHIS@netzone.com (Mikey) Date: Thu, 18 Dec 1997 17:11:00 -0000 Subject: Problem with ln References: <34995AC9.D23E3FC5@utu.fi> Message-ID: <3499c7ce.25365806@smtp.netzone.com> rename .../H-i386-cygwin32/bin/install.exe ginstall.exe rename .../H-i386-cygwin32/bin/ln.exe ln-real.exe put the 2 scripts in .../H-i386-cygwin32/bin add this to your ~/.bashrc ln(){ #hard links for win32 if test x"$*" = x"`echo $*|sed s/-s//`";then /bin/install $* else ln-real.exe $* || ln-real.exe $1.exe $2 fi } export ln or upgrade to 5.0/NTFS for native hard links ;^) On Thu, 18 Dec 1997 19:18:01 +0200, you wrote: >I am trying to compile the CDK on a NT 4.0 machine, but I have a problem. >While running configure, it always fails due to the same error: >--cut-- >ln: /cdk/libiberty/alloca-norm.h: No such file or directory >*** /cdk/configure: unable to link "alloca-conf.h" to \ >"/cdk/libiberty/alloca-norm.h". >--cut-- > I have confirmed the file exists and that I have write access to . and even >rw access to the /cdk/libiberty/alloca-norm.h. I cannot figure out why this >might be. Can someone suggest a solution? > >-- > ----------------------------------------------- > | Juha J????ykk????, juolja@utu.fi | > | home: http://www.utu.fi/~juolja/ | > | pgp-key: http://www.utu.fi/~juolja/pgp.txt | > ----------------------------------------------- >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > (jeffdbREMOVETHIS@netzone.com) delete REMOVETHIS from the above to reply Mikey Windows vs Linux is a no-win situation. Windows 95: -- 32-bit extensions and a GUI shell for a 16-bit patch to an 8-bit OS originally coded for a 4-bit processor written by a 2-bit company that can't produce 1 bit of quality. -------------- next part -------------- A non-text attachment was scrubbed... Name: install Type: text/x-shellscript Size: 6197 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: ln Type: text/x-shellscript Size: 28 bytes Desc: not available URL: From ksr@lp.nm.fujitsu.co.jp Thu Dec 18 18:14:00 1997 From: ksr@lp.nm.fujitsu.co.jp (TANAKA Keishiro) Date: Thu, 18 Dec 1997 18:14:00 -0000 Subject: how to get source of resource compiler and ld that supports .res file? Message-ID: <199712190216.LAA07949@kumando.lp.nm.fujitsu.co.jp> Hi. I could build gnu-win32 cross on FreeBSD2.2.2. (gcc, as, ld etc) But I cannot link my objects with .res file(resource object file). How to get source of resource compiler and ld that supports .res file? It causes error. I'd like to build cross development environment on FreeBSD and get exe(executalbe file) on FreeBSD without using Windows. Best Regards. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From TYann@vet.com.au Thu Dec 18 22:20:00 1997 From: TYann@vet.com.au (Trevor Yann) Date: Thu, 18 Dec 1997 22:20:00 -0000 Subject: PIMAGE_SECTION_HEADER? References: <01BD0C01.D41CF4D0@sos> Message-ID: <349A12B7.66CD76DC@vet.com.au> Sergey Okhapkin wrote: > > Geoffrey Noer wrote: > > You need to find the information that pertains to the missing header > > file lines and then write entries for them without looking at the > > actual Microsoft headers. Finally, send me the diffs to the > > Hmm... The _only_ source of information (direct or indirect) is MS headers... I seem to remember that there were some Microsoft System Journal articles that described parts of the file. I'll have a look. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dhm@paradigm.webvision.com Thu Dec 18 22:27:00 1997 From: dhm@paradigm.webvision.com (dave madden) Date: Thu, 18 Dec 1997 22:27:00 -0000 Subject: CreateProcess() (NT-style) Message-ID: <199712190627.WAA03562@paradigm.webvision.com> I've written a native telnet server for NT (it uses standard Win32 calls rather than the cygwin DLL) but I'm having trouble communicating with processes spawned from the CMD.EXE that I start up after the user logs in. Builtins like VER and SET work fine, as do a few other very simple commands (DIR & TYPE [builtin?], DATE), but when I run a substantial program like bash or CL, I get no output. I end up killing the process with the Task Manager to get back to the shell. However, if I run CL as the process that my daemon starts, I get its output as expected. (Bash starts up if executed as the first process, but apparently can't/won't talk over the anonymous pipes I've set up.) I suspect that children of the shell process telnetd spawns aren't inheriting the appropriate DETACHED_PROCESS and CREATE_NO_WINDOW flags, but I don't know how to fix it. Can anyone in gnuwin32-land help with this? BTW, I'm using native calls because I have another project which must be done natively, and which will need to use pipes to communicate with child processes. I figured telnetd would be both a good learning experience and a useful program to have. (If anybody wants what I have so far, let me know. I intend to GPL it.) d. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From gswart@nc.com Thu Dec 18 22:52:00 1997 From: gswart@nc.com (Garret Swart) Date: Thu, 18 Dec 1997 22:52:00 -0000 Subject: Running Cygnus tools under Visual Studio is Broken in b18 Message-ID: <349A1A76.97C4DA69@nc.com> I've been running GNU make and friends under Microsoft's Visual Studio (aka Visual C++, aka MSDEV) using the Makefile project type. This was working fine until I upgraded to b18 from b17.1. Now when I run processes from MSDEV all I get is the error messages at the end of the message. When I instead try to use the compile command on GNU emacs 19.34.1 i386-*-nt4.0 of Aug 28, 1996, I get: cd ~/cygnus/ make -k /c: /c: No such file or directory Compilation exited abnormally with code 1 at Thu Dec 18 22:41:37 When running GNU tools in a console window, I've noticed that the process name of the currently executing sub-process now appears in the window banner. In b17.1 of the Cygnus tools did not happen. Is there a way of turning this off? Having the banner track the 100's of calls per minute to various GNU tools being run by make is distracting and seems to cause a loss of stability. Could this be related to the problem above? Here is what gets printed out when I run make under MSDEV, note that Win32 error 87 is invalid parameter. (unknown) heap_init: unable to allocate heap, win32 error 87 cygwin: terminating (unknown) In cygwin_except_handler (unknown) Exception trapped! (unknown) exception C0000005 at 1001E4B2 (unknown) exception: ax 0 bx 0 cx 401 dx FFFFFFFF (unknown) exception: si 0 di 0 bp 242FB38 sp 242FB38 (unknown) exception is: STATUS_ACCESS_VIOLATION (unknown) Stack trace: (unknown) frame 0: sp = 0x242F96C, pc = 0x1001282A (unknown) frame 1: sp = 0x242F988, pc = 0x77F94512 (unknown) frame 2: sp = 0x242F9AC, pc = 0x77F88EEB (unknown) frame 3: sp = 0x242FA38, pc = 0x77F76266 (unknown) frame 4: sp = 0x242FB38, pc = 0x10011D4D (unknown) frame 5: sp = 0x242FB4C, pc = 0x10018972 (unknown) frame 6: sp = 0x242FB64, pc = 0x10011615 (unknown) frame 7: sp = 0x242FF9C, pc = 0x10011BFF (unknown) frame 8: sp = 0x242FFA8, pc = 0x41F284 (unknown) frame 9: sp = 0x242FFB8, pc = 0x401012 (unknown) frame 10: sp = 0x242FFC0, pc = 0x77F1B304 (unknown) frame 11: sp = 0x242FFF0, pc = 0x0 (unknown) End of stack trace (unknown) In cygwin_except_handler (unknown) Error while dumping state (probably corrupted stack) (unknown) In cygwin_except_handler ... repeats indefinitely - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Axel.Riese@risc.uni-linz.ac.at Fri Dec 19 00:30:00 1997 From: Axel.Riese@risc.uni-linz.ac.at (Axel Riese) Date: Fri, 19 Dec 1997 00:30:00 -0000 Subject: How to avoid Warning: Redefinition of ... References: <19971218130854.13277.qmail@hotmail.com> Message-ID: <199712190809.AA23370@melmac.risc.uni-linz.ac.at> Hi Earnie ! > I found in the gcc.info file(s) in the Overall Option node a section on > switches about for warnings. I believe that you want the > -Wno-redundant-decls switch. > > - \\||// > ---o0O0--Earnie--0O0o---- > -earnie_boyd@hotmail.com- > ------ooo0O--O0ooo------- Unfortunatley, this does not work (btw, this option is the default). - Axel My original message: > >Dear cygwiners, > > > >I managed to patch most of the headers from the latest Platform SDK to > >make them useable for gcc (i.e., the preprocessor doesn't produce > errors). > > > >However, I came across the following problem: > > > >A lot of header files typedef some symbols more than once, thus > producing > >warnings like > > > > ... Warning: redefinition of blah_blah ... > > ... blah_blah previously defined in ... > > > > > >Is there a way to avoid these warnings that is not contained in the > following > >list: > > > >- adding for each typedef something like #define blah_blah_defined and > > checking #ifndef blah_blah_defined ... #endif > >- forgetting about the Microsoft headers > >- closing my eyes during preprocessing and ignoring the warnings > > > > > >Any suggestions appreciated, > > > > Axel > > > > =========================================================================== Axel Riese Research Institute for Symbolic Computation J. Kepler University Linz A-4040 Linz e-Mail: Axel.Riese@risc.uni-linz.ac.at Austria URL: http://www.risc.uni-linz.ac.at/people/ariese/home/ =========================================================================== - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mag@bunuel.tii.matav.hu Fri Dec 19 02:37:00 1997 From: mag@bunuel.tii.matav.hu (Magossa'nyi A'rpa'd) Date: Fri, 19 Dec 1997 02:37:00 -0000 Subject: question about pipes Message-ID: Hi! I am now porting scp to win32, using the crosscompiler on a Debian GNU/Linux host. The cygwin32 version is 17.1-1 I am using the initial source of therapy@guardian.htu.tuwien.ac.at The ssh runs nearly well, but looks like it is unable to handle filenames properly. My problem at this point with the handling of pipes. scp works like this: It opens two pipes, forks, the child exec()s ssh with the proper arguments, and the child and the parent tries to communicate through the pipes. It seems that the data stuck in the pipe, and it causes a deadlock. If I modify the source to write() into the pipe in the child, and immediately close it, the parent gets the data, and the child complains because it cannot write() into the closed file descriptor. But if I didn't screw it up, the parent seems to hang on the first read(). Anyone have any idea about it? --- GNU GPL: csak tiszta forr????sb????l - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Fri Dec 19 05:14:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Fri, 19 Dec 1997 05:14:00 -0000 Subject: CreateProcess() NT style Message-ID: <199712191314.NAA16012@out2.ibm.net> >> dave madden (dhm@paradigm.webvision.com) Thu, 18 Dec 1997 22:27:05 -0800 I've written a native telnet server for NT (it uses standard Win32 calls rather than the cygwin DLL) but I'm having trouble communicating with processes spawned from the CMD.EXE that I start up after the user logs in. Builtins like VER and SET work fine, as do a few other very simple commands (DIR & TYPE [builtin?], DATE), but when I run a substantial program like bash or CL, I get no output. I end up killing the process with the Task Manager to get back to the shell. << Did you by any chance remember to issue the initialization call to the wsoc32.dll? ???? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 19 05:14:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 19 Dec 1997 05:14:00 -0000 Subject: looking for cygnus compatible perl Message-ID: <19971219125959.13260.qmail@hotmail.com> I compiled the Perl package without much ado. When the compiler complained I either #ifdef'ed or #define'd the problems. Be sure to look at the configure --help and make a determination about switches. I did not compile the utilites just perl. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: "Jon S. Jaques" >To: "B. Elizabeth Trojan" >Cc: >Subject: Re: looking for cygnus compatible perl >Date: Thu, 18 Dec 1997 13:55:43 -0500 > >>Is there a version of Perl available for win32 >that is Cygnus >>compatible? By Cygnus compatible I mean accepts >UNC paths and >>knows about cygnus mounts. > > >That would be "the" Perl; ie that is v5.004_04, >the very same that is compiled under Unix/Linux. >( http://www.perl.org/CPAN/src/latest.tar.gz ) > >Is it easy? Supposedly, although I've not yet >gotten it to work, but haven't given up. > >Assuming that you've installed "cdk.exe" into the >default location, there're two updates you need: >1.) The new cygwin.dll, and 2.) Chris Faylor's >Perl patch which is specific to Cygwin32 >installation/compilation. (The cygwin.dll seems to >be distributed from a users' site, along with a >new Bash.exe, among other things, and I can't tell >if their presence, or lack thereof, affects the >perl install/compile.) > >These are the steps that were advised to me: > >>>Here is what I would try: >>> >>> mkdir /usr/src/perl >>> mount -b c:\usr\src\perl /perlsrc >>> bash >>> cd /perlsrc >>> tar xzf /wherever/perl5.004_04.tar.gz >>> mv perl5.004_04/* . >>> tar xzf /whereever/cygperl5.004_04.pat.tar.gz >>> /bin/sh cygperlpat5.004_04 >>> patch -p1 -N < cygperlpat5.004_04 >>> sh Configure > >And then, when it didn't work, we went here: > >>It is a prerequisite that you are able to compile >programs. Does your >>gcc work? Can you successfully build a simple >"hello world" program? >> >>Also, the patched README file *does not* say to >copy files anywhere. > >Note this last, which IS different from the >original Perl README. If you've already followed >those instructions, then I'd reccomend deleting >the files "gcc2" and "ld2" which you theoretically >already copied into your path. > >Personally, I can compile a simple file, >program1.cpp ("hello world") with *g++* but NOT >gcc (a friend suggested that the file was a >"plus-plus" and that gcc is not, but I don't >really know much about either... my primary reason >for this thing is just to get the new Perl >compiled on my system to open up some options in >my Perl programming... one language at a time, ple >ase! ) > >Please, if you have no luck, then lets' continue >this dialogue-- I'd really like to get this >resolved. > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 19 05:14:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 19 Dec 1997 05:14:00 -0000 Subject: Fw: GNU-Win32 info (NBY) Message-ID: <19971219124926.6883.qmail@hotmail.com> Hi Bob, I suggest you give it another try. The system corruption errors from the rm -rf command is due to the way that inode is mimmicked. I am giving you a list of URL's to check out. Sergey Okhapkin has a patched cygwin.dll and libcygwin.a to take care of B18 errors. Most of this will be incorporated in B19. MAKE CERTAIN THAT YOU MOUNT YOU DIRECTORIES AS text=binary. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: "Bob Duke" >To: >Subject: Fw: GNU-Win32 info >Date: Thu, 18 Dec 1997 15:07:45 -0500 > > >-----Original Message----- >From: Bob Duke >To: info@cygnus.com >Date: Thursday, December 18, 1997 1:27 PM >Subject: GNU-Win32 info > > >>Can anyone give me information about the status of the GNU-Win32 tools for >>the Windows NT/Intel platform? We are porting a Unix app that is built in >a >>Sun environment with various GNU tools (gmake, flex, g++, etc.) and are >>primarily interested in just being able to use reasonably functional and/or >>supported tools. I have tried your B18 beta, and had problems with file >>system corruption being reported when using tar and rm -rf on an ntfs >volume >>(I think that rm was just confused, but it made me want something past the >>beta). Anyway, is anything past the beta available? >> Thanks - Bob Duke >> Alphatronix >> >> > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 19 05:14:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 19 Dec 1997 05:14:00 -0000 Subject: Running Cygnus tools under Visual Studio is Broken in b18 (NBY) Message-ID: <19971219123930.28045.qmail@hotmail.com> Hi Garret, You need to check Sergey Okhapkins page listed below. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Thu, 18 Dec 1997 22:55:51 -0800 >From: Garret Swart >Reply-To: gswart@nc.com >To: gnu-win32@cygnus.com >Subject: Running Cygnus tools under Visual Studio is Broken in b18 > >I've been running GNU make and friends under Microsoft's Visual Studio >(aka Visual C++, aka MSDEV) using the Makefile project type. This was >working fine until I upgraded to b18 from b17.1. Now when I run >processes from MSDEV all I get is the error messages at the end of the >message. > >When I instead try to use the compile command on GNU emacs 19.34.1 >i386-*-nt4.0 of Aug 28, 1996, I get: > > cd ~/cygnus/ > make -k > /c: /c: No such file or directory > > Compilation exited abnormally with code 1 at Thu Dec 18 22:41:37 > >When running GNU tools in a console window, I've noticed that the >process name of the currently executing sub-process now appears in the >window banner. In b17.1 of the Cygnus tools did not happen. Is there a >way of turning this off? Having the banner track the 100's of calls per >minute to various GNU tools being run by make is distracting and seems >to cause a loss of stability. Could this be related to the problem >above? > >Here is what gets printed out when I run make under MSDEV, note that >Win32 error 87 is invalid parameter. > >(unknown) heap_init: unable to allocate heap, win32 error 87 >cygwin: terminating >(unknown) In cygwin_except_handler >(unknown) Exception trapped! >(unknown) exception C0000005 at 1001E4B2 >(unknown) exception: ax 0 bx 0 cx 401 dx FFFFFFFF >(unknown) exception: si 0 di 0 bp 242FB38 sp 242FB38 >(unknown) exception is: STATUS_ACCESS_VIOLATION >(unknown) Stack trace: >(unknown) frame 0: sp = 0x242F96C, pc = 0x1001282A >(unknown) frame 1: sp = 0x242F988, pc = 0x77F94512 >(unknown) frame 2: sp = 0x242F9AC, pc = 0x77F88EEB >(unknown) frame 3: sp = 0x242FA38, pc = 0x77F76266 >(unknown) frame 4: sp = 0x242FB38, pc = 0x10011D4D >(unknown) frame 5: sp = 0x242FB4C, pc = 0x10018972 >(unknown) frame 6: sp = 0x242FB64, pc = 0x10011615 >(unknown) frame 7: sp = 0x242FF9C, pc = 0x10011BFF >(unknown) frame 8: sp = 0x242FFA8, pc = 0x41F284 >(unknown) frame 9: sp = 0x242FFB8, pc = 0x401012 >(unknown) frame 10: sp = 0x242FFC0, pc = 0x77F1B304 >(unknown) frame 11: sp = 0x242FFF0, pc = 0x0 >(unknown) End of stack trace >(unknown) In cygwin_except_handler >(unknown) Error while dumping state (probably corrupted stack) >(unknown) In cygwin_except_handler >... repeats indefinitely > > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Fri Dec 19 05:14:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Fri, 19 Dec 1997 05:14:00 -0000 Subject: question about pipes (NBY) Message-ID: <19971219123735.5154.qmail@hotmail.com> This has been discussed before. Check the mail archives. I also suggest that you upgrade to version B18 and get a patched cygwin.dll from Sergey Okhapkin. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Fri, 19 Dec 1997 11:45:01 +0100 >From: "Magossa'nyi A'rpa'd" >To: gnu-win32@cygnus.com >Subject: question about pipes > >Hi! > >I am now porting scp to win32, using the crosscompiler on a Debian GNU/L= >inux >host. >The cygwin32 version is 17.1-1 >I am using the initial source of therapy@guardian.htu.tuwien.ac.at >The ssh runs nearly well, but looks like it is unable to handle filenames >properly. >My problem at this point with the handling of pipes. >scp works like this: >It opens two pipes, forks, the child exec()s ssh with the proper argument= >s, >and the child and the parent tries to communicate through the pipes. >It seems that the data stuck in the pipe, and it causes a deadlock. >If I modify the source to write() into the pipe in the child, and >immediately close it, the parent gets the data, and the child complains >because it cannot write() into the closed file descriptor. >But if I didn't screw it up, the parent seems to hang on the first read(). > >Anyone have any idea about it? > >--- >GNU GPL: csak tiszta forr=E1sb=F3l > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Fri Dec 19 06:21:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Fri, 19 Dec 1997 06:21:00 -0000 Subject: Fw: GNU-Win32 info References: <01bd0bf0$9b315500$7bca09c0@lynx> Message-ID: In article < 01bd0bf0$9b315500$7bca09c0@lynx >, Bob Duke wrote: >Can anyone give me information about the status of the GNU-Win32 tools >for the Windows NT/Intel platform? We are porting a Unix app that is >built in a Sun environment with various GNU tools (gmake, flex, g++, >etc.) and are primarily interested in just being able to use reasonably >functional and/or supported tools. I have tried your B18 beta, and had >problems with file system corruption being reported when using tar and >rm -rf on an ntfs volume (I think that rm was just confused, but it >made me want something past the beta). Anyway, is anything past the >beta available? The 'rm -rf' message is just a symptom of rm confusion. It is due to the fact that there are no inodes available for directories under cygwin so they are calculated based on the directory name. That means that if you have a directory structure like c:\foo\bar\foo, and attempt an 'rm -r foo', you will receive an error about "file system corruption" because rm has found that two foo's with the same "inode". B18 is the latest version of the source available but you'll want to get the coolview version of B18 available from: http://www.lexa.ru/sos . en -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cvn@interchain.nl Fri Dec 19 06:30:00 1997 From: cvn@interchain.nl (Kees van Veen) Date: Fri, 19 Dec 1997 06:30:00 -0000 Subject: questions from an average user References: <34990805.56E05C26@greco2.polytechnique.fr> Message-ID: <349A7D51.4702197D@interchain.nl> Alessandro Mirone wrote: > > - If someone who is able to build DLLs could contact me it would > be of great help for me > I have had terrible trouble to get DLLs build and running on Windows95 until I got hold of dllfix.exe (a tool to repair bad .reloc section endings). Currently I'm running on W95 and NT 4.0 SP3 B18 with Sergey's coolview with applications calling 5 or 6 DLLs without problems. One problem is still that I seem to have to call _getenv instead of getenv, _execl instead of execv, etc. to get it running when linked to DLLs. I don't know why, see my mail on environment variables and DLLs sometime back. I attached my shell script to build DLLs with, maybe it is of some help to you (you may need to edit it). Kees P.S. if you can't find dllfix.exe, I'll send it to you From joe_banta@perceptics.com Fri Dec 19 06:37:00 1997 From: joe_banta@perceptics.com (Joseph E. Banta, Esquire) Date: Fri, 19 Dec 1997 06:37:00 -0000 Subject: How to make gdb work? Message-ID: <01BD0C61.EA222830@HERBY> Anybody: I am just curious: does gdb come with a graphical interface, or something? Every time I try to invoke it in NT dos, I get: Tcl_Init failed: Can't find a usable init.tcl in the following directories: {} ./lib/tcl7.6 c:/Progra~1/Cygnus/tcl7.6/library c:/Progra~1/Cygnus/man/lib rary This probably means that Tcl wasn't installed properly. or if I do find init.tcl, it wants main.tcl: Unable to find main.tcl in /pizza/mushroom/noer/beta18/install-i386-cygwin32/sha re/gdbtcl You might want to set GDBTK_LIBRARY What is the proper environment configuration to run gdb? Thanx, Joseph Banta - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From joe_banta@perceptics.com Fri Dec 19 06:57:00 1997 From: joe_banta@perceptics.com (Joseph E. Banta, Esquire) Date: Fri, 19 Dec 1997 06:57:00 -0000 Subject: Cygnus info / man reader Message-ID: <01BD0C64.600A69C0@HERBY> Another question: Has anyone put together a binary release of the standalone info browser with a 32-bit compiler? I'm using the djgpp one now, but its a little quirky in NT. Thanks, Joseph Banta - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Fri Dec 19 07:06:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Fri, 19 Dec 1997 07:06:00 -0000 Subject: rx.h missing ? References: <349855E5.4C4F33B@eml.co.uk> Message-ID: <6k5gFkj$pfB@mike.franken.de> Hi malcolm, >I tried to build some code that uses regex.h, but regex.h includes rx.h >(and nothing else) which does not seem to be part of the cdk. Do I have >to down load the sources to get this, or is there another solution? Why >not use the GNU standard regex.h? (Just curious about how regex works >internally.) Indeed you have to get the sources to have rx.h - Cygnus forgot to supply this file with the cdk.exe package. I would send it to the list, but it is about 3500 lines in size - so I won't do it, if there are not at least four or five people, who'd like to get it. > >tia >malcolm Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dhm@paradigm.webvision.com Fri Dec 19 13:37:00 1997 From: dhm@paradigm.webvision.com (dave madden) Date: Fri, 19 Dec 1997 13:37:00 -0000 Subject: CreateProcess() (NT-style) References: Message-ID: <199712192136.NAA07305@paradigm.webvision.com> =>From: newsham@lava.net (Tim Newsham) =>... =>I have a similar program at http://www.lava.net/~newsham . It =>doesn't seem to suffer the problems you mention. Feel free to =>grab it to see how it differs from yours. Bingo! Thanks to you, I now have a full, native-NT telnetd server! (The problem was, I was passing the DETACHED_PROCESS and CREATE_NO_WINDOW flags to CreateProcess...I should have just passed 0.) I have to fiddle with the CR/LF handling a bit (you have to type ctrl-j now to get commands recognized) but that should be finished by this afternoon. I'll put the source code, binary, and a readme in a tarball at: ftp://ftp.webvision.com/pub/nttelnetd.tar.gz Thanks! d. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ads@rmbmain.rmb.co.za Fri Dec 19 13:37:00 1997 From: ads@rmbmain.rmb.co.za (ads@rmbmain.rmb.co.za) Date: Fri, 19 Dec 1997 13:37:00 -0000 Subject: bash builtin "echo" default behaviour wrong? Message-ID: <9711198825.AA882590468@smtpw.rmb.co.za> Hi, How do I change to default behaviour of the bash builtin "echo" command to process the backslash-escape? bash-2.01$ echo "first \n next" first \n next bash-2.01$ echo -e "First \n next" bash-2.01$ echo -e "first \n next" first next bash-2.01$ bash --posix does not help. I know I can use an alias or a function, but why is this different from the Bourne Shell (/bin/sh) and Korn Shell) (/bin/ksh) and /bin/echo Thanks Alberto da Silva ads@rmb.co.za - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From shane@caraveo.com Fri Dec 19 13:37:00 1997 From: shane@caraveo.com (Shane Caraveo) Date: Fri, 19 Dec 1997 13:37:00 -0000 Subject: CYGWIN.DLL and VC5 References: Message-ID: <199712191754.MAA16979@michael.webdevelop.com> > Is it possible to use the cygwin.dll with Visual C 5? I've got a project > I'm working on that I need to be able to access some of the non win32 > functions (setitimer(), etc). I really can't use gcc to build with as > I've got way too many pre-compiled libraries to work with. > > Thanks. > g. In most cases you can easily port to something vc has. Use settimer in place of setitimer. There are very few functions you should have a real problem porting (fork, accept, fcntl, signal, and a few others). Shane - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Fri Dec 19 20:38:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Fri, 19 Dec 1997 20:38:00 -0000 Subject: looking for cygnus compatible perl References: <19971219125959.13260.qmail@hotmail.com> Message-ID: If you're using my patches and Sergey's coolview, there should be no reason to #ifdef or #define anything. The main problems that people seem to be having are: 1) They extract the perl files, read the README.cygwin, install my patch, and do the operations in the README. Unfortunately, my patch changes the README, so they're following old instructions. 2) They do not extract the perl files to a 'binary' mounted directory and the patch does not proceed correctly. Or, B18 patch screws up so they need to grab and build a new version of patch. 3) Most important: Many people do not have all of the environment variables set up correctly to compile and link C programs. I have had a number of success stories so I am pretty sure that my patch is correct. I think it is just a matter of making sure that the above details have been covered. In article < 19971219125959.13260.qmail@hotmail.com >, Earnie Boyd wrote: >I compiled the Perl package without much ado. When the compiler >complained I either #ifdef'ed or #define'd the problems. Be sure to >look at the configure --help and make a determination about switches. I >did not compile the utilites just perl. > >>From: "Jon S. Jaques" > >>Personally, I can compile a simple file, >>program1.cpp ("hello world") with *g++* but NOT >>gcc (a friend suggested that the file was a >>"plus-plus" and that gcc is not, but I don't >>really know much about either... my primary reason >>for this thing is just to get the new Perl >>compiled on my system to open up some options in >>my Perl programming... one language at a time, ple >>ase! ) >> >>Please, if you have no luck, then lets' continue >>this dialogue-- I'd really like to get this >>resolved. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jman@lx.net Fri Dec 19 23:06:00 1997 From: jman@lx.net (jman) Date: Fri, 19 Dec 1997 23:06:00 -0000 Subject: Cygnus info / man reader References: <01BD0C64.600A69C0@HERBY> Message-ID: <3.0.5.32.19971220010640.0084e6c0@208.221.108.11> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I've uploaded tkinfo I use with Gnu-Win32 to ftp://ftp.deninc.com/pub/ports/tkinfo Using the tcl/tk package distributed with Gnuwin32 it works real nice, I've setup the default paths in it but you can add your own. Just type tkinfo from the bash shell.... At 09:55 AM 12/19/97 -0500, Joseph E. Banta, Esquire wrote: >Another question: > >Has anyone put together a binary release of the standalone info >browser with a 32-bit compiler? I'm using the djgpp one now, but >its a little quirky in NT. > >Thanks, >Joseph Banta >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > -----BEGIN PGP SIGNATURE----- Version: PGP for Personal Privacy 5.0 Charset: noconv iQA/AwUBNJtufw6ne3t4b32aEQJ6ygCgs6vRDP5x5bZBwgjhqsXP73A3BkEAn2Ix 4K9YVBgN1rpHObBShDsjX1y8 =sLIK -----END PGP SIGNATURE----- ------------------------------------------------------- Jason L. Esman aka _Jman Owner Den Internet Services System Admin. Network Consulting http://www.deninc.com Email jman@lx.net or root@lx.net Finger jman@lx.net for PGP Public Keys... ------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From gaffer@zip.com.au Fri Dec 19 23:29:00 1997 From: gaffer@zip.com.au (Gaffer) Date: Fri, 19 Dec 1997 23:29:00 -0000 Subject: prometheus truecolor References: <19971219123735.5154.qmail@hotmail.com> Message-ID: <349B72E5.EA0771B6@zip.com.au> G'day :) I've recently just ported my graphics API to Cygnus GNU-Win32. If you are doing graphical output under win32 using cygnus i highly recommend you check it out. Crusty old unix people can safely ignore this post... :) Information =========== PTC is a fast, portable and free truecolor graphics system for C++ programmers. PTC handles all the low level details such as mode setting, memory surfaces and blitting while providing the programmer with a simple platform independant interface. Portability =========== Using PTC you can write graphics code that compiles without change on both DOS and Win32 platforms using VGA, VESA or GDI for display output. Pixel format conversion ======================= PTC does more than any other truecolor graphics system. PTC provides virtual 8, 16 and 32bit modes with pixel format conversion that is so fast there is no overhead. This means that you can work virtually in 32bit ARGB8888 or 16bit RGB565 and not care what pixel format the display is in. PTC supports output to all the standard truecolor and highcolor graphics formats, plus 9 VGA fakemodes, and even greyscale and RGB332 output. Stability ========= PTC is rock solid. Code using PTC will work flawlessly under DOS, Windows 95, Windows 98, and Windows NT. Compilers ========= -Watcom C++ 10.5, 10.6, 11.0 -Visual C++ 4.x, 5.x -Cygnus GNU-Win32 -DJGPP 2.x Changes since PTC053 ==================== -Completed GDI interface for fast windowed output under Win32 -Fully functional BitBlt between surfaces of any dimension -Added support for surface orientations (topdown or bottomup) -New "Clipper" object to implement clipping with surfaces -Fixed PTC to coexist with Midas and Allegro under DJGPP 2.x -Added the ability to share a single palette across any number of surfaces to save memory -Added fast dithered output for 256 color displays under Win32 -Fixed 256 color output under Win32 with windowed palettes -Largely redesigned internal blitting core -Added Watcom C++ Win32 compiler support -HUGE increase in stability thanks to BoundsChecker Download now! ============= http://www.zip.com.au/~gaffer/ptc cheers -- Gaffer [xtatic] gaffer@bigfoot.com 6800A007B013CD104F33C0B13FAA40A7E2FB57B1C0AB47E2FC5F4779ECB81210 99B1FFCD10061E0741BF8001B3638B857F0202C40F92C48A95810203C28A9500 0503C2C1E802740148AA03C2D1E888853F01E2DAB9400103F94B75D2E5400FC1 0600018AE0ABABE2F30733FFBEC002B53CF366A5B401CD1674ABB80300CD10C3 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sat Dec 20 00:28:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sat, 20 Dec 1997 00:28:00 -0000 Subject: Cygnus / info man reader Message-ID: <199712200828.IAA63792@out2.ibm.net> >> Joseph E. Banta, Esquire (joe_banta@perceptics.com) Fri, 19 Dec 1997 09:55:42 -0500 Has anyone put together a binary release of the standalone info browser with a 32-bit compiler? I'm using the djgpp one now, but its a little quirky in NT. << You can get TkInfo to work. Of course that means getting the latest Windows release of tkxx.dll and tclxx.dll, then generating import libraries for cygwin using dlltool and impdef, then compiling and linking TkInfo. And, it's a little fussy to operate, but can work. As for man.exe, you need the gnu archives at prep.ai.mit.edu, where you download the man sources and its helper apps. ???? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Sat Dec 20 08:15:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Sat, 20 Dec 1997 08:15:00 -0000 Subject: bash builtin "echo" default behaviour wrong? References: <9711198825.AA882590468@smtpw.rmb.co.za> Message-ID: <6kDregCKpfB@mike.franken.de> Hi, [...] >How do I change to default behaviour of the bash builtin "echo" command to >process the backslash-escape? Do you have to use the builtin or would the echo.exe command be sufficent for you ? echo.exe resides in the Cygnus bin directory. [...] >Alberto da Silva Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Sat Dec 20 08:15:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Sat, 20 Dec 1997 08:15:00 -0000 Subject: How to make gdb work? References: <01BD0C61.EA222830@HERBY> Message-ID: <6kDreMkppfB@mike.franken.de> Hi Joseph, [...] > >What is the proper environment configuration to run gdb? GDBTK_LIBRARY=C:/gnuwin32/b18/share/gdbtcl TCL_LIBRARY=C:/gnuwin32/b18/tcl/lib/tcl7.6 Don't forget to use the drive in these paths. I had these vars without the drive and got the same error messages you have. [...] > >Thanx, >Joseph Banta Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sat Dec 20 11:40:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sat, 20 Dec 1997 11:40:00 -0000 Subject: Why would clear.exe persist after exiting? Message-ID: <199712201940.TAA93412@out4.ibm.net> I wrote and compiled the following program as a ming32 CRTDLL.DLL program using the slangw32 version of libslang. When I tested it under the Dos prompt, it worked perfectly. When I ran it under bash$, the mouse pointer retained the hourglass after clear.exe ran, and, pushing CTRL-ALT-DEL (the three-finger salute) showed that clear and rnaapp were still present, even though running ps.exe did not show that. Command line for compiling: bash$ gcc -O -o clear clear.c -lslang (libtermcap not needed here) Program code: ========================================================================= void SLtt_cls (void); int SLang_init_tty (int, int, int); int main (int argc, char **argv) { SLang_init_tty (0, 0, 0); SLtt_cls(); return 0; } ========================================================================== - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From aneilson@usa.net Sat Dec 20 12:37:00 1997 From: aneilson@usa.net (Arthur W. Neilson III) Date: Sat, 20 Dec 1997 12:37:00 -0000 Subject: exception problem Message-ID: <3.0.3.32.19971220103709.0082ca80@pop.netaddress.com> I wrote a short program that dumps memory in hex format. It takes a hex address argument off the command line and displays the contents of that memory address on the screen in hex format like so: C:\gnuwin32\src\src\peek f000 0000F000 A5 85 A5 85 A5 85 A5 85 A5 85 B1 8A A5 85 A5 85 * ................ * 0000F010 A5 85 A5 85 A7 85 A7 85 A7 85 A7 85 A7 85 A7 85 * ................ * 0000F020 A7 85 A5 85 BB 89 A5 85 A5 85 5A 8B BB 89 A5 85 * ..........Z..... * 0000F030 A5 85 5F 89 5F 89 FA 88 5F 89 5F 89 DE 85 A7 85 * .._._..._._..... * 0000F040 A7 85 5F 89 DE 85 5F 89 FD 88 DE 85 DE 85 BD 89 * .._..._......... * 0000F050 A5 85 A5 85 A5 85 EC 8B 73 8C A5 85 5F 89 A7 85 * ........s..._... * 0000F060 38 90 55 90 A5 85 A5 85 A7 85 A5 85 E3 89 DE 85 * 8.U............. * 0000F070 A5 85 A7 85 5F 89 5F 89 DE 85 A5 85 A7 85 0F 89 * ...._._......... * 0000F080 12 89 A5 85 55 90 A7 85 A7 85 A7 85 A7 85 A7 85 * ....U........... * 0000F090 A7 85 A7 85 A7 85 A7 85 15 89 1E 2E 8E 1E 02 00 * ................ * 0000F0A0 80 3E 16 14 00 75 03 E9 EB 00 1F 45 55 8B EC 1E * .>...u.....EU... * 0000F0B0 53 2E 8E 1E 02 00 80 FC 12 76 0A C6 06 45 14 00 * S........v...E.. * 0000F0C0 C6 06 46 14 00 80 3E 45 14 00 74 05 C6 06 46 14 * ..F...>E..t...F. * 0000F0D0 01 80 3E 4E 14 00 75 05 C6 06 4C 14 01 F7 06 EC * ..>N..u...L..... * 0000F0E0 14 02 00 75 08 F7 06 0D 16 02 00 74 18 80 3E 4E * ...u.......t..>N * 0000F0F0 14 00 75 11 60 06 0F A0 0F A8 9A 42 63 1F 01 0F * ..u.`......Bc... * et cetera .... anyways, I get an exception in the gnu-win32 handler when I try to display system memory areas, like low memory 00000417 for example (numlock status byte). What do I have to do to be allowed to examine any area of storage? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From chuck@ethereal.ru Sat Dec 20 15:33:00 1997 From: chuck@ethereal.ru (Chuck Bogorad) Date: Sat, 20 Dec 1997 15:33:00 -0000 Subject: ncftp v2.4.2 patch to put current status in console title Message-ID: <01bd0d9f$334a4b30$0100a8c0@hysteria.spb.ru> here's this small patch to see current percentage in console title. -------------- next part -------------- A non-text attachment was scrubbed... Name: dif Type: text/x-diff Size: 3475 bytes Desc: not available URL: From chuck@ethereal.ru Sat Dec 20 15:53:00 1997 From: chuck@ethereal.ru (Chuck Bogorad) Date: Sat, 20 Dec 1997 15:53:00 -0000 Subject: anyone tried to build UUCP (Taylor) and smail ? Message-ID: <01bd0da1$a82a4480$0100a8c0@hysteria.spb.ru> i'm looking into building a semi-complete mail system with uucp/smail/elm. elm is ok, uucp is also ok (more or less), smail is in the works. i just wonder if anyone has already done this and would like to share some experience. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mike@mbsun.mlb.org Sat Dec 20 21:16:00 1997 From: mike@mbsun.mlb.org (Mike Bernson) Date: Sat, 20 Dec 1997 21:16:00 -0000 Subject: anyone tried to build UUCP (Taylor) and smail ? References: <01bd0da1$a82a4480$0100a8c0@hysteria.spb.ru> Message-ID: I have a number of changes to the cygwin.dll. I have added a bsd/posix terminal driver. I have a version of uucp-1.06 that works under win'95 I am about ready to release the changes that I made. I have a number of users testing them and when thing get stable i will put my winsup and newlib up for ftp. This should happen after the first of the year. > i'm looking into building a semi-complete mail system with uucp/smail/elm. > elm is ok, uucp is also ok (more or less), smail is in the works. i just > wonder if anyone has already done this and would like to share some > experience. > > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". > -- Mike Bernson (mike@mbsun.mlb.org) - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From w.derks@nl.cis.philips.com Sun Dec 21 01:00:00 1997 From: w.derks@nl.cis.philips.com (wiljan) Date: Sun, 21 Dec 1997 01:00:00 -0000 Subject: exception problem Message-ID: <01BD0D92.96026E90@modem036.lss.cp.philips.com> The peek program that you wrote, looks in you own virtual memory. This means that the addresses you specify are virtual address of you program You expect to look into physical memory of the machine. NT takes care (as most virtual memory systems) that you can not read or write those areas under normal circumstances. If you really want to access the physical memory you need a special device driver to have this done. Wiljan -----Original Message----- From: Arthur W. Neilson III [SMTP:aneilson@usa.net] Sent: Saturday, December 20, 1997 9:37 PM To: gnu-win32@cygnus.com Subject: exception problem I wrote a short program that dumps memory in hex format. It takes a hex address argument off the command line and displays the contents of that memory address on the screen in hex format like so: C:\gnuwin32\src\src\peek f000 0000F000 A5 85 A5 85 A5 85 A5 85 A5 85 B1 8A A5 85 A5 85 * ............... * 0000F010 A5 85 A5 85 A7 85 A7 85 A7 85 A7 85 A7 85 A7 85 * ............... * 0000F020 A7 85 A5 85 BB 89 A5 85 A5 85 5A 8B BB 89 A5 85 * .........Z..... * 0000F030 A5 85 5F 89 5F 89 FA 88 5F 89 5F 89 DE 85 A7 85 * ._._..._._..... * 0000F040 A7 85 5F 89 DE 85 5F 89 FD 88 DE 85 DE 85 BD 89 * ._..._......... * 0000F050 A5 85 A5 85 A5 85 EC 8B 73 8C A5 85 5F 89 A7 85 * .......s..._... * 0000F060 38 90 55 90 A5 85 A5 85 A7 85 A5 85 E3 89 DE 85 * 8.U............. * 0000F070 A5 85 A7 85 5F 89 5F 89 DE 85 A5 85 A7 85 0F 89 * ..._._......... * 0000F080 12 89 A5 85 55 90 A7 85 A7 85 A7 85 A7 85 A7 85 * ...U........... * 0000F090 A7 85 A7 85 A7 85 A7 85 15 89 1E 2E 8E 1E 02 00 * ............... * 0000F0A0 80 3E 16 14 00 75 03 E9 EB 00 1F 45 55 8B EC 1E * .>...u.....EU... * 0000F0B0 53 2E 8E 1E 02 00 80 FC 12 76 0A C6 06 45 14 00 * S........v...E.. * 0000F0C0 C6 06 46 14 00 80 3E 45 14 00 74 05 C6 06 46 14 * .F...>E..t...F. * 0000F0D0 01 80 3E 4E 14 00 75 05 C6 06 4C 14 01 F7 06 EC * .>N..u...L..... * 0000F0E0 14 02 00 75 08 F7 06 0D 16 02 00 74 18 80 3E 4E * ..u.......t..>N * 0000F0F0 14 00 75 11 60 06 0F A0 0F A8 9A 42 63 1F 01 0F * .u.`......Bc... * et cetera .... anyways, I get an exception in the gnu-win32 handler when I try to display system memory areas, like low memory 00000417 for example (numlock status byte). What do I have to do to be allowed to examine any area of storage? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From duran@cobra.vdl2.ca Sun Dec 21 03:12:00 1997 From: duran@cobra.vdl2.ca (Alfredo Duran) Date: Sun, 21 Dec 1997 03:12:00 -0000 Subject: c++ vs g++ [Re: Gcc compiler b18 newbie probs] Message-ID: <199712211127.GAA03670@cobra.vdl2.ca> - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sun Dec 21 04:57:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sun, 21 Dec 1997 04:57:00 -0000 Subject: Fwd: Patch for slrn to enable calling Windows sendmail Message-ID: <199712211257.MAA26420@out4.ibm.net> From: root@tailor.aleim.net (root) Newsgroups: news.software.readers Subject: Patch for slrn to enable calling Windows sendmail Organization: The Light Reply-To: vischne@ibm.net-nospam X-Newsreader: slrn (0.9.4.3 UNIX) NNTP-Posting-Host: 32.100.34.253 Message-ID: <349cd0e2.0@news3.ibm.net> Date: 21 Dec 97 08:18:42 GMT Lines: 65 Path: news3.ibm.net!root This allows slrn to call the Windows version of the sendmail command, which has the following sort of command line: Welcome to sendmail/NT, a mailer stub for Windows/NT and Windows/95. (c) Nigel R. Ellis 1994 - 1995. This version built on Jun 5 1995 at 14:56:11 Usage: SENDMAIL.EXE [-h localhostname] [-r smtphost] [-m transport] [-U MAPIUser] [-P MAPIPasswd] -t {ToList} -f From [-s Subject] -F File {ToList} is a space, comma or semi-colon delimited list of recipients Valid transports are XMAPI (allow edit before send), MAPI and SMTP Default transport is SMTP The fork() command still causes problems in the b18 cygwin gcc release, to it's better to un-define HAS_POSIX_SIGNALS in config.h. --- misc.c.orig Sat Jul 12 16:29:01 1997 +++ misc.c Sun Dec 21 02:59:27 1997 @@ -690,7 +690,7 @@ int slrn_mail_file (char *file, int edit, unsigned int editline, har *to, char *subject) /*{{{*/ { char buf [256 + 2 * SLRN_MAX_PATH_LEN]; -#ifdef __os2__ +#if defined(__os2__) || defined(_WIN32) char outfile [SLRN_MAX_PATH_LEN]; #endif @@ -735,7 +735,7 @@ if (fp == NULL) return (-1); slrn_mime_scan_file (fp); -# ifdef __os2__ +# if defined(__os2__) || defined(_WIN32) pp = slrn_open_tmpfile (outfile, "w"); # else pp = slrn_popen (Slrn_SendMail_Command, "w"); @@ -770,7 +770,11 @@ putc('\n', pp); } slrn_fclose (fp); -#ifdef __os2__ +#ifdef _WIN32 + slrn_fclose (pp); + sprintf(buf, "%s -t %s -F %s", Slrn_SendMail_Command, to, outfile); + slrn_posix_system (buf, 0); +#elif defined(__os2__) slrn_fclose (pp); sprintf (buf, "%s %s", Slrn_SendMail_Command, outfile); slrn_posix_system (buf, 0); @@ -780,7 +784,9 @@ } else { -#ifdef __os2__ +#ifdef _WIN32 + sprintf(buf, "%s -t %s -F %s", Slrn_SendMail_Command, to, outfile); +#elif defined(__os2__) sprintf (buf, "%s %s", Slrn_SendMail_Command, file); #else sprintf (buf, "%s < %s", Slrn_SendMail_Command, file); - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tursiop@micronet.fr Sun Dec 21 06:40:00 1997 From: tursiop@micronet.fr (Christophe Ivorra) Date: Sun, 21 Dec 1997 06:40:00 -0000 Subject: rx.h missing ? References: <6k5gFkj$pfB@mike.franken.de> Message-ID: <349D29FB.1E68@micronet.fr> Michael Hirmke wrote: > > Hi malcolm, > > >I tried to build some code that uses regex.h, but regex.h includes rx.h > >(and nothing else) which does not seem to be part of the cdk. Do I have > >to down load the sources to get this, or is there another solution? Why > >not use the GNU standard regex.h? (Just curious about how regex works > >internally.) > > Indeed you have to get the sources to have rx.h - Cygnus forgot to > supply this file with the cdk.exe package. > > I would send it to the list, but it is about 3500 lines in size - so I > won't do it, if there are not at least four or five people, who'd like > to get it. > > > > >tia > >malcolm > > Bye. > Michael. > -- > Michael Hirmke | Telefon +49 (911) 557999 > Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 > 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de > | WWW http://minimike.franken.de/ > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". I am interested too. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jman@lx.net Sun Dec 21 14:03:00 1997 From: jman@lx.net (jman) Date: Sun, 21 Dec 1997 14:03:00 -0000 Subject: Gnu-Win32 Central Website... Message-ID: <3.0.5.32.19971221160341.007af600@208.221.108.11> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, Due to well lack of time, I am removing the Gnu-Win32 Website it is just to much for me to maintain at this time. I am NOT going to remove the ftp site, it will stay and I will continue to update it. This is just to let you know to remove links to http://www.deninc.com/~gnu-win32 Thanks, -----BEGIN PGP SIGNATURE----- Version: PGP for Personal Privacy 5.0 Charset: noconv iQA/AwUBNJ2SPA6ne3t4b32aEQKrUQCggmPJoyiP1L5ujwcBfVi5/8seAqQAoJgf IEhy5KLjWDetwlWKDPEMw0WC =YMUF -----END PGP SIGNATURE----- ------------------------------------------------------- Jason L. Esman aka _Jman Owner Den Internet Services System Admin. Network Consulting http://www.deninc.com Email jman@lx.net or root@lx.net Finger jman@lx.net for PGP Public Keys... ------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From p.dalgaard@biostat.ku.dk Sun Dec 21 15:33:00 1997 From: p.dalgaard@biostat.ku.dk (Peter Dalgaard BSA) Date: Sun, 21 Dec 1997 15:33:00 -0000 Subject: c++ vs g++ [Re: Gcc compiler b18 newbie probs] References: <199712211127.GAA03670@cobra.vdl2.ca> Message-ID: Alfredo Duran writes: > > > Well, for a man with nothing to say, you express yourself very succinctly... ;) -- O__ ---- Peter Dalgaard Blegdamsvej 3 c/ /'_ --- Dept. of Biostatistics 2200 Cph. N (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk) FAX: (+45) 35327907 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From raf@comdyn.com.au Sun Dec 21 19:22:00 1997 From: raf@comdyn.com.au (raf) Date: Sun, 21 Dec 1997 19:22:00 -0000 Subject: Disappearing mount points with WinNT 3.51/4.0 Message-ID: <199712220322.OAA11405@mali.comdyn.com.au> >In win95 I don't need to remount my mount points >once they are set. In WinNT 4.0 and WinNT 3.51, >my mount points disappear if I close my bash >session. How do I make the mount points >persistent in WinNT? I know the mount points >live in the registry but I'm not sure how >to keep them there. Any help would be greatly >appreciated. you can place mount commands in /etc/bashrc or ~/.bashrc this is tacky and slow (to start bash) but it works. i hope you find a better method. raf - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Mon Dec 22 02:37:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Mon, 22 Dec 1997 02:37:00 -0000 Subject: Question about pipes Message-ID: <199712221037.KAA26892@out5.ibm.net> The slrn news reader, that uses pipes (popen, etc.) to drive the Unix sendmail program, causes interesting disk clatter noises when the Windows 95 sendmail.exe program (which doesn't expect its email to be piped to it) is used instead. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Mon Dec 22 02:40:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Mon, 22 Dec 1997 02:40:00 -0000 Subject: libXaw and the latest X11 port Message-ID: <199712221040.KAA15818@out5.ibm.net> The writeup on using the ported libX points out that libXaw (the Athena widgets toolkit) doesn't work well with the port. Athena buttons don't register on the X-window. Has anyone tried this out with libXaw3d or libXaw95, which are workalikes for libXaw with, respectively, Motif-like and Windows95-like buttons? Without libXaw, applications like xfm, xfilemanager, xfontsel, etc., don't work. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jjaques@grovehillsys.com Mon Dec 22 05:17:00 1997 From: jjaques@grovehillsys.com (Jon S. Jaques) Date: Mon, 22 Dec 1997 05:17:00 -0000 Subject: looking for cygnus compatible perl Message-ID: <3.0.32.19971222081836.0095f250@mail.pair.com> Have you recently tried this from a "ground up" install? I still have not gotten it to work, and so I still wonder about what else is required. --jsj At 04:38 AM 12/20/97 GMT, Christopher Faylor wrote: >If you're using my patches and Sergey's coolview, there should be no >reason to #ifdef or #define anything. > >The main problems that people seem to be having are: > >1) They extract the perl files, read the README.cygwin, install my patch, > and do the operations in the README. Unfortunately, my patch changes > the README, so they're following old instructions. > >2) They do not extract the perl files to a 'binary' mounted directory and > the patch does not proceed correctly. Or, B18 patch screws up so > they need to grab and build a new version of patch. > >3) Most important: Many people do not have all of the environment variables > set up correctly to compile and link C programs. > >I have had a number of success stories so I am pretty sure that my patch >is correct. I think it is just a matter of making sure that the above >details have been covered. > >In article < 19971219125959.13260.qmail@hotmail.com >, >Earnie Boyd wrote: >>I compiled the Perl package without much ado. When the compiler >>complained I either #ifdef'ed or #define'd the problems. Be sure to >>look at the configure --help and make a determination about switches. I >>did not compile the utilites just perl. >> >>>From: "Jon S. Jaques" >> >>>Personally, I can compile a simple file, >>>program1.cpp ("hello world") with *g++* but NOT >>>gcc (a friend suggested that the file was a >>>"plus-plus" and that gcc is not, but I don't >>>really know much about either... my primary reason >>>for this thing is just to get the new Perl >>>compiled on my system to open up some options in >>>my Perl programming... one language at a time, ple >>>ase! ) >>> >>>Please, if you have no luck, then lets' continue >>>this dialogue-- I'd really like to get this >>>resolved. >-- > http://www.bbc.com/ cgf@bbc.com "Strange how unreal >VMS=>UNIX Solutions Boston Business Computing the real can be." >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > > -- Jon S. Jaques -- The Grove Hill Pages -- -- Now on the PerlRing! -- -- http://www.grovehillsys.com/scripting -- -- jjaques@grovehillsys.com -- -- Linux: Salvation from Bill, Steve and Marc! -- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mason@primenet.com.au Mon Dec 22 05:52:00 1997 From: mason@primenet.com.au (Geoff Wing) Date: Mon, 22 Dec 1997 05:52:00 -0000 Subject: libtermcap around? Message-ID: Heyla, has anyone made a standalone termcap (not terminfo) lib? If not, I'll write one since I need one and it'll be about as quick as porting one. -- Geoff Wing [gcw@pobox.com] Phone : +61-3-9818 2977 Technical Manager: PrimeNet Computer Consultants Facsimile: +61-3-9818 5155 Mobile : 0412 162 441 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From arlindo@niteroi.gsfc.nasa.gov Mon Dec 22 08:52:00 1997 From: arlindo@niteroi.gsfc.nasa.gov (Arlindo da Silva) Date: Mon, 22 Dec 1997 08:52:00 -0000 Subject: libXaw and the latest X11 port References: <199712221040.KAA15818@out5.ibm.net> Message-ID: <199712221654.LAA28205@woodmore.gsfc.nasa.gov> Hello, > > The writeup on using the ported libX points out that libXaw (the Athena > widgets toolkit) doesn't work well with the port. Athena buttons don't > register on the X-window. > > Has anyone tried this out with libXaw3d or libXaw95, which are workalikes > for libXaw with, respectively, Motif-like and Windows95-like buttons? I am not sure which writeup you are refering to. The X11R6.3 client libs in ftp://ftp.deninc.com/pub/ports/X11R6.3 or ftp://niteroi.gsfc.nasa.gov/pub/win32/X11R6.3 include both libXaw and libXaw3d which work on both win95 and winNT (try the bezier example in the examples/ directory). The only problem that I am aware of is that Xaw does not work well with the MI/X server due to a bug in this X server (you can verify that this is a bug in the X server by running xfig on a Unix box and displaying the results on the PC). I contacted MicroImages and their answer was: "we don't support the Athena widgets". > Without libXaw, applications like xfm, xfilemanager, xfontsel, etc., don't > work. I wish the MicroImages folks would agree with you. Arlindo. PS: BTW, could you upload your ntxlib sources to ftp.deninc.com? -- Arlindo da Silva dasilva@alum.mit.edu - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From arlindo@niteroi.gsfc.nasa.gov Mon Dec 22 11:13:00 1997 From: arlindo@niteroi.gsfc.nasa.gov (Arlindo da Silva) Date: Mon, 22 Dec 1997 11:13:00 -0000 Subject: libXaw and the latest X11 port References: <349EAE6D.7964DBB6@microimages.com> Message-ID: <199712221916.OAA28440@woodmore.gsfc.nasa.gov> Jeff, > It isn't a bug. We don't need to support Athena widgets for > our software, so we don't. Please keep in mind that MI/X was > designed to provide a cross platform interface for our > professional GIS software, TNTmips. We have provided it free > to those who may also find it useful. Development of MI/X > will be driven by the needs of TNTmips and our clients who use > TNTmips. As I mentioned, we don't need to support the Athena > widgets for our software, so we don't. Sorry, I've mispoken here and I can certainly see your reason for priotizing your development. However, you guys should consider releasing the sources for MI/X server under GPL (GNU Public License). This way people in the community could help out implementing additional features, including those features that may be useful for your product. Just a thought. -- Arlindo da Silva dasilva@alum.mit.edu - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 22 12:35:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 22 Dec 1997 12:35:00 -0000 Subject: (egcs-mingw32/cygwin32) The collect2/ld problem workaround Message-ID: <9712222035.AA16731@modi.xraylith.wisc.edu> For those of who've found the mysterious "collect2 exited with status 1" error after seemingly being hung for a while, here's what happens: - gcc invokes ld.exe in gcc's library directory (you can see what that is by running `gcc -print-prog-name=ld'). This is the program otherwise known as collect2 and needed for C++, especially for -frepo and exception handling. - collect2 (installed as ld.exe in library directory) does its thing and then looks for the *real* ld.exe (GNU ld) according to a set of rules (LIBRARY_PATH+COMPILER_PATH env variables, PATH, etc) and execs that to do the final link editing. Due to a bug in collect2, it's finding itself first (since it's called ld.exe) and execing itself causing a recursion and eventually chokes. Solution? Install the real ld.exe (GNU ld that's installed as ld.exe in the binary directory) as collect-ld.exe in the directory where cc1.exe lives (that's called gcc library/compiler directory). This is the directory you see when you use the command `gcc -print-prog-name=cc1'. Now ld.exe will exec collect-ld.exe instead and all will be well. I know the problem exists for cygwin32, but not sure about mingw32. The workaround above is safe for either case, so I recommend you do it. I'll hopefully have a fix out soon. Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From stanton@haas.berkeley.edu Mon Dec 22 13:23:00 1997 From: stanton@haas.berkeley.edu (Richard Stanton) Date: Mon, 22 Dec 1997 13:23:00 -0000 Subject: DLL help needed... Message-ID: <199712222111.NAA04315@grether.haas.berkeley.edu> I'm getting close to creating working DLLs. With a lot of help from Giandomenico De Sanctis, I've actually created a DLL file that passes/returns short ints, and can be used in Excel 97, but I'm having trouble when the functions pass/return double arguments. In fact, I'm having trouble even in plain C. Here's a simple example, which produces the following (odd looking) output [c:\projects\excel]exe i 10.000000, j 11.000000, k -1.#IND00 When I pass everything as short instead of double, it all seems to work OK. Any idea what I'm doing wrong? I attach all the necessary files to get this to compile. By the way, I'm compiling this using mingw32. Thanks a lot. Richard Stanton ------------------------------------------------------------------------ exe.c: ------------------------------------------------------------------------ #include double RHS (double x, double y) __attribute__((stdcall)); int main() { double i, j, k; i = 10; j = 11; k = RHS(i,j); printf ("i %lf, j %lf, k %lf\n", i, j, k); return 0; } ------------------------------------------------------------------------ dll.c: ------------------------------------------------------------------------ double RHS (double x, double y) { return (x + y); } ------------------------------------------------------------------------ dll.def: ------------------------------------------------------------------------ EXPORTS RHS RHS@16=RHS ------------------------------------------------------------------------ makefile: ------------------------------------------------------------------------ dll.dll: dll.c dll.def dlltool --def dll.def --output-exp dll.exp --output-lib dll.a --dllname dll.dll -k gcc -g -o dll.dll dll.c dll.exp -Xlinker --image-base=0x40000000 -dll -mrtd -ggdb3 exe.exe: exe.c dll.dll gcc exe.c dll.a -g -o exe.exe ------------------------------------------------------------------------ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From swhatley@blkbox.com Mon Dec 22 13:43:00 1997 From: swhatley@blkbox.com (Steven Whatley) Date: Mon, 22 Dec 1997 13:43:00 -0000 Subject: Gnu-Win32 Central Website... References: <3.0.5.32.19971221160341.007af600@208.221.108.11> Message-ID: On Sun, 21 Dec 1997, jman wrote: > Due to well lack of time, I am removing the Gnu-Win32 Website it is > just to much for me to maintain at this time. I am NOT going to > remove the ftp site, it will stay and I will continue to update it. > This is just to let you know to remove links to > http://www.deninc.com/~gnu-win32 What is the FTP site? I wasn't aware of the web site. Any info will be appreciated. Thanks, Steven http://www.blkbox.com/~swhatley/ _|_ | _|_ "I am the way and the truth and Steven Whatley | --|-- | the life. No one comes to the swhatley@blkbox.com | | | Father except through me." Houston, Texas | -- Jesus Christ (John 14:6 NIV) | - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jborjess@mail.on.rogers.wave.ca Mon Dec 22 14:55:00 1997 From: jborjess@mail.on.rogers.wave.ca (Johnny Boerjesson) Date: Mon, 22 Dec 1997 14:55:00 -0000 Subject: DLLFIX.EXE Message-ID: <349EEF04.2888FA49@rogers.wave.ca> Could somebody please repost dllfix.exe, as the one in the mailing list archive seems mangled. Thanks ! JB - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jman@lx.net Mon Dec 22 17:25:00 1997 From: jman@lx.net (jman) Date: Mon, 22 Dec 1997 17:25:00 -0000 Subject: Gnu-Win32 Central Website... References: <3.0.5.32.19971221160341.007af600@208.221.108.11> Message-ID: <3.0.5.32.19971222192608.00860100@208.221.108.11> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 03:43 PM 12/22/97 -0600, Steven Whatley wrote: > >What is the FTP site? I wasn't aware of the web site. Any info will be >appreciated. > >Thanks, >Steven > http://www.blkbox.com/~swhatley/ > _|_ | _|_ "I am the way and the truth and >Steven Whatley | --|-- | the life. No one comes to the >swhatley@blkbox.com | | | Father except through me." >Houston, Texas | -- Jesus Christ (John 14:6 NIV) it is ftp://ftp.deninc.com (Totally dedicated to Gnu-Win32) -----BEGIN PGP SIGNATURE----- Version: PGP for Personal Privacy 5.0 Charset: noconv iQA/AwUBNJ8TLg6ne3t4b32aEQIEigCbBR7EyEV2aiZTmh9+1rq/+x74jPwAmwc6 xneL44vwxt8GS7frB7XOT5xZ =DeYb -----END PGP SIGNATURE----- ------------------------------------------------------- Jason L. Esman aka _Jman Owner Den Internet Services System Admin. Network Consulting http://www.deninc.com Email jman@lx.net or root@lx.net Finger jman@lx.net for PGP Public Keys... ------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hgfernan@usp.br Mon Dec 22 18:59:00 1997 From: hgfernan@usp.br (hgfernan@usp.br) Date: Mon, 22 Dec 1997 18:59:00 -0000 Subject: Compiling "V" 1.16 - Specific Instructions? References: <01bd0bea$8e0e5a40$0651040a@rmiller.cooperpower.com> Message-ID: <9712230302.AA28544@spider.uspnet.usp.br> On 18 Dec 97 at 13:24, Rick Miller wrote: > From: "Rick Miller" > To: > Subject: Compiling "V" 1.16 - Specific Instructions? > Date: Thu, 18 Dec 1997 13:24:26 -0600 > Hi all! > > After taking the bait in the gnu-win32 FAQ, "What is V?" I went out and got > the sources and tried to take the given advice for compiling... Problem > is, there are at least FOUR sets of such advice that I know of and at least > two of them didn't work. > The default Config.mk tells me to use ARCH=intel, there's one in > ftp://ftp.cs.unm.edu/pub/wampler/config/gnuwin32.make that tells me to use > ARCH=gnu-win32... then there's other stuff in the "v/contrib/gnuwin32.tar" > file in the source code... Hi! You can get V specific information writing to vgui-discuss@ringtail.hna.com.au which is obviously a V discussion list. Regards, ++Hilton ------------------- Hilton Fernandes email: hgfernan@usp.br www: http://www.lsi.usp.br/~hilton.html (inactive) M. Sc. Student of Parallel Distributed Applications at Escola Politecnica (Polytechic School) University of S. Paulo - Brazil - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ksr@lp.nm.fujitsu.co.jp Mon Dec 22 21:32:00 1997 From: ksr@lp.nm.fujitsu.co.jp (TANAKA Keishiro) Date: Mon, 22 Dec 1997 21:32:00 -0000 Subject: multithread safe program Message-ID: <199712230535.OAA14498@kumando.lp.nm.fujitsu.co.jp> Hi. Can I create multithread program with gnu-win32 gcc ? The library doesn't seem to be multithread safe. I tried that I linked my objects with libcmt.a imported from Microsoft libcmt.lib. # like ld -e _mainCRTStartup mine.o libcmt.a libkernel32.a Several warnings are issued but executable is successfuly generated. But the executable is not able to run (DLL not found etc.). Best Regards. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From P.J.Kersten@dia.eur.nl Tue Dec 23 00:57:00 1997 From: P.J.Kersten@dia.eur.nl (P.J. Kersten) Date: Tue, 23 Dec 1997 00:57:00 -0000 Subject: libtermcap around? References: Message-ID: <349F7CCD.41553AB8@dia.eur.nl> Geoff Wing wrote: > > Heyla, > has anyone made a standalone termcap (not terminfo) lib? If not, I'll write > one since I need one and it'll be about as quick as porting one. > -- > Geoff Wing [gcw@pobox.com] Phone : +61-3-9818 2977 > Technical Manager: PrimeNet Computer Consultants Facsimile: +61-3-9818 5155 > Mobile : 0412 162 441 > - Look on one of the GNU mirrors for termcap-1.3.tar.gz. It compiles out-of-the-box under GNU-Win32. Leaves you some time for X-Mas. HTH, -- Pieter J. Kersten Erasmus Universiteit Rotterdam dir. I&A/I&S, WH2-10 Postbus 1738 3000 DR Rotterdam Nederland Tel +31-10-4082599 Fax +31-10-4527236 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From matts@tibco.com Tue Dec 23 01:11:00 1997 From: matts@tibco.com (Matt Stupple) Date: Tue, 23 Dec 1997 01:11:00 -0000 Subject: gdb questions Message-ID: <349F7FEE.58638D18@tibco.com> Two questions about gdb - hope these aren't FAQs as I couldn't find answers to either on line.... 1) How do I disable the Tk GUI for gdb? In general I find the command line interface more useful and more forgiving. 2) Is it possible to use gdb to debug code built using MSVC? I'm using the M$ compiler, but building from the command line with gmake etc and would like to be able to debug in similar fashion; the alternative (using the MSVC IDE) really doesn't appeal! Thanks, Matt. -- /\/\att Stupple TIBCO, Gracechurch House, London EC3V OJA matts@tibco.com http://www.tibco.com/ +44(0)171 2956053 "How can you lie there and think of England, When you don't even know who's in the team?" -- Billy Bragg - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cvn@interchain.nl Tue Dec 23 01:38:00 1997 From: cvn@interchain.nl (Kees van Veen) Date: Tue, 23 Dec 1997 01:38:00 -0000 Subject: strncat Message-ID: <349F86DE.3377054C@interchain.nl> Hi, I have an application using strncat() which relies on NULL termination. I've looked in the coolview sources at strncat() and saw it is implemented with strncpy(). I'm not sure about this, but shouldn't strncpy() only return a NULL terminated string when its argument is NULL terminated before reaching the length, whereas strncat() always returns a NULL terminated string, regardless of the input string ? Kees - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From map@graph-tech.ch Tue Dec 23 02:14:00 1997 From: map@graph-tech.ch (Martin Portmann) Date: Tue, 23 Dec 1997 02:14:00 -0000 Subject: Diff of executable files Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I want to do a 'diff' of binary files on the Win32 platform (*.exe and *.dll). The problem is that the binary files seems to contain timestamps so that if I compile the same source twice I get different binaries. Where can I get information about the binary format used on Win32 platforms? Thanks a lot ----------------------------------------------------- Martin Portmann Mobile +41 79 330 60 12 Software Department Phone +41 62 896 42 40 Graph-Tech AG, Switzerland map@graph-tech.ch http://www.graph-tech.ch/english/people/map/ -----BEGIN PGP SIGNATURE----- Version: PGPfreeware 5.0i for non-commercial use Charset: noconv iQA/AwUBNJ+A7v/DIHAOmbwcEQJ46ACePIFpOrUvHa9SWQXKm71QhVfgqhAAn3/G FuCnUKGG/iRXX8SIW2ya2OtH =66xu -----END PGP SIGNATURE----- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From acherman@inf.ethz.ch Tue Dec 23 02:36:00 1997 From: acherman@inf.ethz.ch (Erwin Achermann) Date: Tue, 23 Dec 1997 02:36:00 -0000 Subject: looking for cygnus compatible perl References: <19971219125959.13260.qmail@hotmail.com> Message-ID: <349F9413.B3635699@inf.ethz.ch> Hi maybe we elaborate a HowTo altogether.... I have Sergeys patches: bash.exe 04.07.97 in:C:\bin C:\gnuwin32\b18\H-i386-cygwin32\bin C:\gnuwin32\b18\H-i386-cygwin32\i386-cygwin32\bin sh.exe == bash.exe in:C:\bin C:\gnuwin32\b18\H-i386-cygwin32\bin cygwin.dll 03.11.97 in:C:\gnuwin32\b18\H-i386-cygwin32\bin C:\gnuwin32\b18\H-i386-cygwin32\i386-cygwin32\lib libcygwin.a 03.11.97 in:C:\gnuwin32\b18\H-i386-cygwin32\i386-cygwin32\lib I extract the perl5.004.04 into the directory c:\perl I make sure, that /perl is mounted binary, i.e $ mount -b c:\\perl /perl Now I extract Chris's patch and save it. (mighty emacs makes sure, there are no ^M [but that's no problem anyhow, cause we're on a binary mounted disk area] {to be clear: I save it in emacs' text-mode}) I save it at /perl/src/perl5.004_04-mpe/cygerlpat5.004_04 The patch says i should do a (it removes a file apparently not needed...) $ sh cygperlpat5.004_04 Next we do as in the header of tha patch said: $ patch -p1 -N < cygperlpat5.004_04 this produces a lot of output. Saying amongst other: Hunk #1 succeeded at .. etc (Just to test: in util.c there is now an #ifdef __CYGWIN32__ at line 1337, it is the last patch applied by the patch command, so let's hope all the others went well too) Now I load the 'new' README.cygwin32 and proceed as it says. I run $ sh Configure I get tons of : command not found : command not found : command not found Second try: I browse trough Configure and save it this time in emacs' binary-mode (it tells me "no changes need to be saved"... hmmm) I run $ sh Configure Again I get tons of : command not found : command not found : command not found Third try: I change a comment line in Configure change it back, (hehe, now emacs saves it) I run $ sh Configure sh Configure -ds -e sh Configure -ds -e -ds -e sh Configure -ds -e -ds -e -ds -e sh Configure -ds -e -ds -e -ds -e -ds -e sh Configure -ds -e -ds -e -ds -e -ds -e -ds -e sh Configure -ds -e -ds -e -ds -e -ds -e -ds -e -ds -e sh Configure -ds -e -ds -e -ds -e -ds -e -ds -e -ds -e -ds -e sh Configure -ds -e -ds -e -ds -e -ds -e -ds -e -ds -e -ds -e -ds -e sh Configure -ds -e -ds -e -ds -e -ds -e -ds -e -ds -e -ds -e -ds -e -ds -e I dont't have to mount this directory as binary as well, do I?? (did that, no change) I don't have to set environmetn Vars like CC?? (done that: /perl/src/perl5.004_04-mpe 29>sh Configure sh Configure -ds -e -Dcc='gcc' sh Configure -ds -e -ds -e -Dcc='gcc' -Dcc='gcc' sh Configure -ds -e -ds -e -ds -e -Dcc='gcc' -Dcc='gcc' -Dcc='gcc' sh Configure -ds -e -ds -e -ds -e -ds -e -Dcc='gcc' -Dcc='gcc' -Dcc='gcc' -Dcc=' gcc' Please help, what's wrong here??? Christopher Faylor wrote: > > If you're using my patches and Sergey's coolview, there should be no > reason to #ifdef or #define anything. > > The main problems that people seem to be having are: > > 1) They extract the perl files, read the README.cygwin, install my patch, > and do the operations in the README. Unfortunately, my patch changes > the README, so they're following old instructions. > > 2) They do not extract the perl files to a 'binary' mounted directory and > the patch does not proceed correctly. Or, B18 patch screws up so > they need to grab and build a new version of patch. > > 3) Most important: Many people do not have all of the environment variables > set up correctly to compile and link C programs. > > I have had a number of success stories so I am pretty sure that my patch > is correct. I think it is just a matter of making sure that the above > details have been covered. > > In article < 19971219125959.13260.qmail@hotmail.com >, > Earnie Boyd wrote: > >I compiled the Perl package without much ado. When the compiler > >complained I either #ifdef'ed or #define'd the problems. Be sure to > >look at the configure --help and make a determination about switches. I > >did not compile the utilites just perl. > > > >>From: "Jon S. Jaques" > > > >>Personally, I can compile a simple file, > >>program1.cpp ("hello world") with *g++* but NOT > >>gcc (a friend suggested that the file was a > >>"plus-plus" and that gcc is not, but I don't > >>really know much about either... my primary reason > >>for this thing is just to get the new Perl > >>compiled on my system to open up some options in > >>my Perl programming... one language at a time, ple > >>ase! ) > >> > >>Please, if you have no luck, then lets' continue > >>this dialogue-- I'd really like to get this > >>resolved. > -- > http://www.bbc.com/ cgf@bbc.com "Strange how unreal > VMS=>UNIX Solutions Boston Business Computing the real can be." > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". -- Erwin Achermann Tel: ++41 1 632 74 40 Institut fuer Wissenschaftliches Rechnen Fax: ++41 1 632 11 72 ETH Zentrum, IFW C29.2 email: achermann@inf.ethz.ch CH-8092 Zuerich url: http://www.inf.ethz.ch/personal/acherman/ > Perfection is reached, not when there is no longer anything to add, < > but when there is no longer anything to take away. < > -- Antoine de Saint-Exupery < - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tdryden@iil.intel.com Tue Dec 23 02:38:00 1997 From: tdryden@iil.intel.com (tdryden) Date: Tue, 23 Dec 1997 02:38:00 -0000 Subject: df anyone? Message-ID: <01BD0F9F.938E11A0.tdryden@iil.intel.com> Does anyone have 'df' working? I successfully compiled gnu/fileutils-3.16 but df doesn't quite work yet. It is looking for the mount table (/etc/mtab, presumably) but can't seem to find it. I hand generated this file and stored it under /etc but still get the same error: bash$ df df: cannot read table of mounted filesystems df does work if invoked with a target disk and from a directory located on that disk: bash$ cd //c/bin bash$ df . Filesystem 1024-blocks Used Available Capacity Mounted on - 1542208 986929 555279 64% Any solutions for having this table maintained by a daemon process? Thanks, Thomas Dryden Intel Corporation, Haifa Israel tdryden@iil.intel.com Phone: 972-4-865-5871 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mason@primenet.com.au Tue Dec 23 05:10:00 1997 From: mason@primenet.com.au (Geoff Wing) Date: Tue, 23 Dec 1997 05:10:00 -0000 Subject: alpha test ports of rxvt & zsh Message-ID: Heyla, I've got versions of rxvt (2.4.4) & zsh (3.0.5) in alpha test stage. http://www.primenet.com.au/GNU-Win32/ zsh currently has some process control problems at the moment which I'll be working on. And also something weird in the command hashing - so you might have to run unhash. Comments welcome. -- Geoff Wing [gcw@pobox.com] Phone : +61-3-9818 2977 Technical Manager: PrimeNet Computer Consultants Facsimile: +61-3-9818 5155 Mobile : 0412 162 441 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From DHammond@fujisec.com Tue Dec 23 07:03:00 1997 From: DHammond@fujisec.com (Don Hammond) Date: Tue, 23 Dec 1997 07:03:00 -0000 Subject: looking for cygnus compatible perl Message-ID: <01017FB1F1C6D011852400805FC18B256070@NTNY-EXCH> Jon - I managed to do this yesterday using GNU sources and Christopher Faylor's patch. I am a user, *not* a developer, so am out of my element when packages don't build easily. I'm not sure how well I did -- the only thing I've tried so far is the Perl TEST which comes with the distribution, and it only succeeded on 79% of the tests. I haven't had time to try anything else. This was my second attempt, having made the mistake mentioned by Chris Faylor of reading the README.cygwin before applying the patch. The only significant change I made during Configure was to add the cc flag: -DDEBUGGING. I don't know if that was a good choice or not, but it appeared to me that this was necessary for full functionality of the Perl debugger. I kept fairly comprehensive notes of the whole Configure process after the first attempt failed. I would be happy to forward the notes on to you if you think it will help, although all they will show is that I accepted most of the default choices, except for things like host and domain name. I also have the redirected std out and std err from running make and the Perl TEST, if anybody wants to see those for any reason. The only non-standard thing I did was edit the config.sh (in a separate bash window) before running make depend. This was to correct a mistake I had made on man page location. (I had a problem using vi in a shell escape from Configure due to another problem.) The moral of the story: if I can do it, most anyone can. [BTW - Thanks to all who have contributed to cygwin32. It is a Godsend to this user forced to NT. And thanks to Christopher Faylor for the Perl patch, and others making their work available. I sure wish I could contribute, but I'm barely smart enough to use this stuff. Your talents are admired and appreciated.] - Don On Monday, December 22, 1997 8:19 AM, Jon S. Jaques writes: > Have you recently tried this from a "ground up" install? I still > have not gotten it to work, and so I still wonder about what else > is required. > > --jsj > > At 04:38 AM 12/20/97 GMT, Christopher Faylor wrote: > >If you're using my patches and Sergey's coolview, there should be no > >reason to #ifdef or #define anything. > > > >The main problems that people seem to be having are: > > > > [...] - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From woodwrk@cyberramp.net Tue Dec 23 07:16:00 1997 From: woodwrk@cyberramp.net (Richard Henry Wood) Date: Tue, 23 Dec 1997 07:16:00 -0000 Subject: Disappearing mount points with WinNT 3.51/4.0 References: <199712220322.OAA11405@mali.comdyn.com.au> Message-ID: <349FD5B7.7C282413@cyberramp.net> This is strange. I don't see my mount points disappear when I close my bash session. I'm running WinNT Server 4.0 Service Pack 3. What variant of NT are you running, exactly? -dick raf wrote: > > >In win95 I don't need to remount my mount points > >once they are set. In WinNT 4.0 and WinNT 3.51, > >my mount points disappear if I close my bash > >session. How do I make the mount points > >persistent in WinNT? I know the mount points > >live in the registry but I'm not sure how > >to keep them there. Any help would be greatly > >appreciated. > > you can place mount commands in /etc/bashrc or ~/.bashrc > this is tacky and slow (to start bash) but it works. > i hope you find a better method. > > raf > > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 23 09:32:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Tue, 23 Dec 1997 09:32:00 -0000 Subject: looking for cygnus compatible perl Message-ID: <199712231732.MAA01201@hardy.bbc.com> >This was my second attempt, having made the mistake >mentioned by Chris Faylor of reading the README.cygwin >before applying the patch. The only significant change >I made during Configure was to add the cc flag: >-DDEBUGGING. I don't know if that was a good choice or >not, but it appeared to me that this was necessary for >full functionality of the Perl debugger. This is not necessary for the debugger. It has nothing to do with the debugger. -DDEBUGGING is for debugging the perl program itself. >I kept fairly comprehensive notes of the whole Configure >process after the first attempt failed. I would be >happy to forward the notes on to you if you think it will >help, although all they will show is that I accepted most >of the default choices, except for things like host and >domain name. I also have the redirected std out and std >err from running make and the Perl TEST, if anybody wants >to see those for any reason. > >The only non-standard thing I did was edit the config.sh >(in a separate bash window) before running make depend. >This was to correct a mistake I had made on man page >location. (I had a problem using vi in a shell escape >from Configure due to another problem.) It *is* annoying when you've gone most of the way through the Configure process and enter a typo. >The moral of the story: if I can do it, most anyone can. Thanks for verifying that it can be done. cgf - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 23 09:32:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Tue, 23 Dec 1997 09:32:00 -0000 Subject: looking for cygnus compatible perl Message-ID: <199712231729.MAA01171@hardy.bbc.com> There are two configure's in the perl release. One begins with a 'c' and one begins with a 'C'. Make sure that you only extract the one that begins with a 'C'. cgf - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From stanton@haas.berkeley.edu Tue Dec 23 09:32:00 1997 From: stanton@haas.berkeley.edu (Richard Stanton) Date: Tue, 23 Dec 1997 09:32:00 -0000 Subject: DLL help needed... References: <349f7828.48009513@smtp.netzone.com> Message-ID: <9712231654.AA0664@volans75.wco.com> >>>>> ">>" == Mikey writes: > [c:\projects\excel]exe > i 10.000000, j 11.000000, k -1.#IND00 >> Could this help? >> __attribute__((__stdcall__)) double RHS (double x, double y) > { > return (x + y); > } Thanks for the suggestion. It makes some difference, but unfortunately not to the output! With this change, I no longer need to compile with the -mrtd option, and I need to change my dll.def file - using -mrtd I need RHS RHS@16=RHS Using the explicit stdcall attribute in the .C file, and compiling without -mrtd, the DEF file needs to be the other way round: RHS@16 RHS=RHS@16 However, the program output is the same: C:\PROJECTS\excel>exe i 10.000000, j 11.000000, k -1.#IND00 Again, if I use exactly the same files, but replacing every "double" with a "short", (and RHS@4 instead of RHS@16 where appropriate) the program runs perfectly. I just tried again, this time removing all stdcall references. The program still fails, but this time I get C:\PROJECTS\excel>exe i 10.000000, j 11.000000, k 0.000000 (the last number is supposed to be the sum of the other two) I'm still confused... Richard Stanton - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 23 12:55:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Tue, 23 Dec 1997 12:55:00 -0000 Subject: looking for cygnus compatible perl References: <19971219125959.13260.qmail@hotmail.com> <349F9413.B3635699@inf.ethz.ch> Message-ID: In article < 349F9413.B3635699@inf.ethz.ch >, Erwin Achermann wrote: >I run >$ sh Configure > >I get tons of >: command not found >: command not found >: command not found This is a sign that you have a file with ^Ms in it. If you send the output to a file and look at it with the appropriate utility, you should see the ^Ms. Instead of using emacs, try using something like 'od -c'. If you see '\r's then that is a problem. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From epeschko@den-mdev1.tci.com Tue Dec 23 12:55:00 1997 From: epeschko@den-mdev1.tci.com (Ed Peschko) Date: Tue, 23 Dec 1997 12:55:00 -0000 Subject: telnet, ftp for cygwin32 Message-ID: <199712232055.NAA05408@den-mdev1.tci.com> hey all: I was looking for a Unix compatible telnet and ftp (standalone). Is this possible through cygwin32? Ed - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 23 12:55:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Tue, 23 Dec 1997 12:55:00 -0000 Subject: gdb questions References: <349F7FEE.58638D18@tibco.com> Message-ID: In article < 349F7FEE.58638D18@tibco.com >, Matt Stupple wrote: >1) How do I disable the Tk GUI for gdb? In general I find the command >line interface more useful and more forgiving. gdb -nw whatever.exe >2) Is it possible to use gdb to debug code built using MSVC? I'm using >the M$ compiler, but building from the command line with gmake etc and >would like to be able to debug in similar fashion; the alternative >(using the MSVC IDE) really doesn't appeal! I don't think so. GNU C and MSVC use different formats to store debugging information. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jman@lx.net Tue Dec 23 13:38:00 1997 From: jman@lx.net (jman) Date: Tue, 23 Dec 1997 13:38:00 -0000 Subject: Legal aspects of using the GNU-Win32 tools? Message-ID: <3.0.5.32.19971223153918.007a9630@208.221.108.11> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > >Did you get a chance to have your lawyer look over the Cygwin >GNU-Win32 copyrights? Can commercial software be dynamically linked >against cygwin.dll if the software is shipped without the library? As of now all the Gnu-Win32 included in the sdk, cdk are under the GNU License. The way she reads it is that you can include it. She's going to do something even better then that tho. I thougth I sent this mail but I must not have. I can't find the original she sent me but that is the jist of it. If you have a specific question or questions about the use of it or the use of the gnuwin32 tools or Gnu License you can contact her legal@deninc.com and she will get back with you asap. Relise she is a practicing lawyer an may not get back with you super fast but she is offering this to the GNU-Win32 community. Hope this helps those with legal questions.. that emailed me... -----BEGIN PGP SIGNATURE----- Version: PGP for Personal Privacy 5.0 Charset: noconv iQA/AwUBNKAvDA6ne3t4b32aEQIoggCg87WGIcH7peA/f5Q5EBBwu4lFyeMAoLqK OwclvOhmS9JdlkdQWAFZhCUo =b1zh -----END PGP SIGNATURE----- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From rnoer@cygnus.com Tue Dec 23 14:03:00 1997 From: rnoer@cygnus.com (Geoffrey Noer) Date: Tue, 23 Dec 1997 14:03:00 -0000 Subject: strncat References: <349F86DE.3377054C@interchain.nl> Message-ID: <199712232203.OAA26514@rtl.cygnus.com> This has been fixed in more recent sources -gjn Kees van Veen wrote: > > I have an application using strncat() which relies on NULL termination. [...] -- Geoffrey Noer noer@cygnus.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From rnoer@cygnus.com Tue Dec 23 14:03:00 1997 From: rnoer@cygnus.com (Geoffrey Noer) Date: Tue, 23 Dec 1997 14:03:00 -0000 Subject: gdb questions References: <349F7FEE.58638D18@tibco.com> Message-ID: <199712232201.OAA26490@rtl.cygnus.com> Matt Stupple wrote: > > 1) How do I disable the Tk GUI for gdb? In general I find the command > line interface more useful and more forgiving. Invoke gdb with the -nw argument ("no windows"). > 2) Is it possible to use gdb to debug code built using MSVC? No, not really. MSVC uses a different debug format which gdb does not understand. -- Geoffrey Noer noer@cygnus.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hgfernan@usp.br Tue Dec 23 15:07:00 1997 From: hgfernan@usp.br (hgfernan@usp.br) Date: Tue, 23 Dec 1997 15:07:00 -0000 Subject: Compiling "V" 1.16 - Specific Instructions? References: <349FE59A.2098DF17@cyberramp.net> Message-ID: <9712232309.AA40350@spider.uspnet.usp.br> On 23 Dec 97 at 10:23, Richard Henry Wood wrote: > Date: Tue, 23 Dec 1997 10:23:54 -0600 > From: Richard Henry Wood > To: hgfernan@usp.br > Subject: Re: Compiling "V" 1.16 - Specific Instructions? > Hilton, > > If the address you give is a discussion group, could you please tell me > how one subscribes to it? > > Thanks, > -dick > Hi! Sorry for the incomplete information. I'm quite busy lately. The home of V is http://www.cs.unm.edu/~wampler/vgui/vgui.html Here is more info about V discussion lists got from http://www.cs.unm.edu/~wampler/vgui/vmaillist.html Please be warned that V is mainly a Unix GUI library; its support to Windows 95/NT is still to be done. People wanting to program for Win95/NT should take a look at wxWindows, at http://web.ukonline.co.uk/julian.smart/wxwin/ Maybe wxWindows should be added to Cygnus page. HTH ++Hilton --------------------------------------------------------------------- On V mailing lists --------------------------------------------------------------------- There are two V mailing lists. The first is a low frequency, announce only list used for announcements of new V releases. To get on that mailing list, mail to vgui@cs.unm.edu with ADD as the Subject and your e-mail address in the body. To delete yourself from that mailing list, send to that address with DELETE as the subject, and your e-mail address in the body. There is also a general mailing list served by a standard list server. This list is used by the V community for help, questions, etc. To add yourself to the general V mail list, send mail to vgui-discuss-request@ringtail.hna.com.au with subscribe vgui-discuss as the body of the message. After you subscribe, you can then post contributions to the list by e-mailing to vgui-discuss@ringtail.hna.com.au. ------------------- Hilton Fernandes email: hgfernan@usp.br www: http://www.lsi.usp.br/~hilton.html (inactive) M. Sc. Student of Parallel Distributed Applications at Escola Politecnica (Polytechic School) University of S. Paulo - Brazil - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ksr@lp.nm.fujitsu.co.jp Tue Dec 23 16:09:00 1997 From: ksr@lp.nm.fujitsu.co.jp (TANAKA Keishiro) Date: Tue, 23 Dec 1997 16:09:00 -0000 Subject: multithread safe program References: <199712230535.OAA14498@kumando.lp.nm.fujitsu.co.jp> Message-ID: <199712240013.JAA15407@kumando.lp.nm.fujitsu.co.jp> Hi. Thanks replying to my mail > Pat, Geoffrey. I knew that the gnu-win32 cygnus library has not supported multithread-safe yet. And I have tried also with mingw32 and libcrtdll.a, but it resulted in failure. I wonder that maybe crtdll.dll in $WINDOWS directory is not multithread-safe. ># like ld -e _mainCRTStartup mine.o libcmt.a libkernel32.a >Several warnings are issued but executable is successfuly generated. >But the executable is not able to run (DLL not found etc.). Using objects Microsoft VC++ generated with gld, gld often issues that the object has duplicated .text section. nm (in gnuwin32) reports that two .text sections exists in the object. In gld (ldlang.c), if section is duplicated and multiple same section exists, the later section is ignored. I think that the gld cannot link with objects created by Microsoft C++ compiler in a few case. And I patched for gld that _imp__ is used for prefix of indirection of symbols instead of __imp_. Best Regards. --- From: TANAKA Keishiro Subject: multithread safe program Date: Tue, 23 Dec 1997 14:35:39 +0900 Message-ID: < 199712230535.OAA14498@kumando.lp.nm.fujitsu.co.jp > ksr> Hi. ksr> ksr> Can I create multithread program with gnu-win32 gcc ? ksr> The library doesn't seem to be multithread safe. ksr> ksr> I tried that I linked my objects with libcmt.a imported from Microsoft ksr> libcmt.lib. ksr> ksr> # like ld -e _mainCRTStartup mine.o libcmt.a libkernel32.a ksr> Several warnings are issued but executable is successfuly generated. ksr> But the executable is not able to run (DLL not found etc.). ksr> ksr> Best Regards. ksr> - ksr> For help on using this list (especially unsubscribing), send a message to ksr> "gnu-win32-request@cygnus.com" with one line of text: "help". ksr> - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From derek@silver.clti.com Tue Dec 23 18:15:00 1997 From: derek@silver.clti.com (Derek C. Lee) Date: Tue, 23 Dec 1997 18:15:00 -0000 Subject: Disappearing mount points with WinNT 3.51/4.0 References: <349FD5B7.7C282413@cyberramp.net> Message-ID: <9712240214.AA00877@silver.clti.com.> > > This is strange. I don't see my mount points disappear when I > close my bash session. I'm running WinNT Server 4.0 Service > Pack 3. > > What variant of NT are you running, exactly? I'm running NT 4.0 Service Pack 3, NT 3.51 Service Pack 5. After reading your email, I did some more investigation. I forgot to mention that I have been logging into the NT servers using a telnet server and then setting the mount points. When I set the mount points this way, I do not have permission to permanently alter the system registry which stores the mount points. If I login from the machine console, my mount points remain. When bash is run, it changes the registry entries in HKEY_CURRENT_USER->Software->Cygnus Solutions->CYGWIN.DLL setup ->b15.0->mounts (says b15 even though I'm running b18) If I login remotely, the registry settings for the current user are used instead of the user being logged in. I'm not sure how to get around this problem other than adding mount commands in the .profile or .bashrc. A problem arises if another user is logged in that has setup different mount points or has removed some. The remote user will see those mounts. I don't suppose there is a way to setup global mounts for all users... Can I make a similar registry setting in HKEY_LOCAL_MACHINE? Thanks, Derek > -dick > > raf wrote: > > > > >In win95 I don't need to remount my mount points > > >once they are set. In WinNT 4.0 and WinNT 3.51, > > >my mount points disappear if I close my bash > > >session. How do I make the mount points > > >persistent in WinNT? I know the mount points > > >live in the registry but I'm not sure how > > >to keep them there. Any help would be greatly > > >appreciated. > > > > you can place mount commands in /etc/bashrc or ~/.bashrc > > this is tacky and slow (to start bash) but it works. > > i hope you find a better method. > > > > raf > > > > - > > For help on using this list (especially unsubscribing), send a message to > > "gnu-win32-request@cygnus.com" with one line of text: "help". > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". > > Derek C. Lee derek@clti.com Chain Link Technologies, Inc. http://www.clti.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From raf@comdyn.com.au Tue Dec 23 19:12:00 1997 From: raf@comdyn.com.au (raf) Date: Tue, 23 Dec 1997 19:12:00 -0000 Subject: Disappearing mount points with WinNT 3.51/4.0 Message-ID: <199712240311.OAA04652@mali.comdyn.com.au> >> >In win95 I don't need to remount my mount points >> >once they are set. In WinNT 4.0 and WinNT 3.51, >> >my mount points disappear if I close my bash >> >session. How do I make the mount points >> >persistent in WinNT? I know the mount points >> >live in the registry but I'm not sure how >> >to keep them there. Any help would be greatly >> >appreciated. >> >> you can place mount commands in /etc/bashrc or ~/.bashrc >> this is tacky and slow (to start bash) but it works. >> i hope you find a better method. >This is strange. I don't see my mount points disappear when I >close my bash session. I'm running WinNT Server 4.0 Service >Pack 3. >What variant of NT are you running, exactly? the same. raf - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jfmiller@polymail.cpunix.calpoly.edu Tue Dec 23 22:16:00 1997 From: jfmiller@polymail.cpunix.calpoly.edu (John Miller) Date: Tue, 23 Dec 1997 22:16:00 -0000 Subject: Windows 98 prospects Message-ID: <34A0A962.7057657@polymail.calpoly.edu> Dear All, I was reacently lucky enought to get my hands on a beta copy of Win98. I Was first wondering weather anyone else has tried Gnu-Win32 (ethier cygnus or mingw32) with Win98 and weather there would be any reason why either of these wouldn't work. (They outh to work fine IMOH). My second question is weather or not There are any plans to port (change, enhance) Gnu-Win32 to this "new" system. If so is there anything I (a novice programmer) can (legaly) do to help. sincerly, John Miller - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mason@primenet.com.au Tue Dec 23 22:59:00 1997 From: mason@primenet.com.au (Geoff Wing) Date: Tue, 23 Dec 1997 22:59:00 -0000 Subject: Disappearing mount points with WinNT 3.51/4.0 References: <349FD5B7.7C282413@cyberramp.net> Message-ID: Richard Henry Wood typed: :This is strange. I don't see my mount points disappear when I :close my bash session. I'm running WinNT Server 4.0 Service :Pack 3. :What variant of NT are you running, exactly? This person should probably also check whether or not he/she has permission to write to the appropriate part of the registry. I would assume yes if the mounts exist once created, so this person would also need to check what's happening with the registry file via auditing if possible. As a minimum the person: * would need permission to CreateSubkey on HKEY_CURRENT_USER/Software/ (or any of the appropriate subkeys if existing) * would need permissions of QueryValue in HKEY_CURRENT_USER/Software/Cygnus Solutions/CYGWIN.DLL setup/b15.0/mounts/ * probably would need permission to EnumerateSubkeys in the above path. Some mysterious unattributed person wrote: :> >In win95 I don't need to remount my mount points :> >once they are set. In WinNT 4.0 and WinNT 3.51, :> >my mount points disappear if I close my bash :> >session. How do I make the mount points :> >persistent in WinNT? I know the mount points :> >live in the registry but I'm not sure how :> >to keep them there. Any help would be greatly :> >appreciated. -- Geoff Wing [gcw@pobox.com] Phone : +61-3-9818 2977 Technical Manager: PrimeNet Computer Consultants Facsimile: +61-3-9818 5155 Mobile : 0412 162 441 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jeffdbREMOVETHIS@netzone.com Wed Dec 24 04:57:00 1997 From: jeffdbREMOVETHIS@netzone.com (Mikey) Date: Wed, 24 Dec 1997 04:57:00 -0000 Subject: multithread safe program References: <199712230535.OAA14498@kumando.lp.nm.fujitsu.co.jp> <199712240013.JAA15407@kumando.lp.nm.fujitsu.co.jp> Message-ID: <34a105e9.30879987@smtp.netzone.com> To use gld with vc++ objects you might want to try ftp://ftp.deninc.com/incoming B18-patched-link-compat-ld.tgz this should handle the comdat sections that are not being linked in by B18 gld. On Wed, 24 Dec 1997 09:13:06 +0900, you wrote: >Hi. >Thanks replying to my mail > Pat, Geoffrey. >I knew that the gnu-win32 cygnus library has not supported >multithread-safe yet. >And I have tried also with mingw32 and libcrtdll.a, but it resulted in >failure. I wonder that maybe crtdll.dll in $WINDOWS directory is not >multithread-safe. > >># like ld -e _mainCRTStartup mine.o libcmt.a libkernel32.a >>Several warnings are issued but executable is successfuly generated. >>But the executable is not able to run (DLL not found etc.). > >Using objects Microsoft VC++ generated with gld, gld often issues that >the object has duplicated .text section. nm (in gnuwin32) reports that >two .text sections exists in the object. >In gld (ldlang.c), if section is duplicated and multiple same section >exists, the later section is ignored. I think that the gld cannot link >with objects created by Microsoft C++ compiler in a few case. > >And I patched for gld that _imp__ is used for prefix of indirection of >symbols instead of __imp_. > >Best Regards. >--- >From: TANAKA Keishiro >Subject: multithread safe program >Date: Tue, 23 Dec 1997 14:35:39 +0900 >Message-ID: < 199712230535.OAA14498@kumando.lp.nm.fujitsu.co.jp > > >ksr> Hi. >ksr> >ksr> Can I create multithread program with gnu-win32 gcc ? >ksr> The library doesn't seem to be multithread safe. >ksr> >ksr> I tried that I linked my objects with libcmt.a imported from Microsoft >ksr> libcmt.lib. >ksr> >ksr> # like ld -e _mainCRTStartup mine.o libcmt.a libkernel32.a >ksr> Several warnings are issued but executable is successfuly generated. >ksr> But the executable is not able to run (DLL not found etc.). >ksr> >ksr> Best Regards. >ksr> - >ksr> For help on using this list (especially unsubscribing), send a message to >ksr> "gnu-win32-request@cygnus.com" with one line of text: "help". >ksr> >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > (jeffdbREMOVETHIS@netzone.com) delete REMOVETHIS from the above to reply Mikey Windows vs Linux is a no-win situation. Windows 95: -- 32-bit extensions and a GUI shell for a 16-bit patch to an 8-bit OS originally coded for a 4-bit processor written by a 2-bit company that can't produce 1 bit of quality. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Wed Dec 24 05:31:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Wed, 24 Dec 1997 05:31:00 -0000 Subject: Comment on the most recent dlltool and ld upgrades Message-ID: <199712241331.NAA14108@out2.ibm.net> The new versions of ld.exe and dlltool.exe seem to be an improvement over the previous. I tried generating an import library from wsock32.dll with the new version, and the resulting libsock.a generated link error messages in which the listed wsock32.dll entry points for the called functions were no longer ambiguous. In other words, the new versions still won't link a homemade wsock32.dll import library, but different functions list different (and correct) entry points during the attempted link process where this wasn't true of the previous version. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From arlindo@niteroi.gsfc.nasa.gov Wed Dec 24 07:23:00 1997 From: arlindo@niteroi.gsfc.nasa.gov (Arlindo da Silva) Date: Wed, 24 Dec 1997 07:23:00 -0000 Subject: EGCS 1.00: can't run hello.c on NT References: <9711130124.AA26539@modi.xraylith.wisc.edu> Message-ID: <199712241526.KAA02923@woodmore.gsfc.nasa.gov> Mumit, I installed egcs over a fresh b18 cdk distribution (Windows NT Version 4.0, 64 megs RAM) , did the collect-ld.exe trick, and installed Sergey's coolview, making sure to keep your libcygwin.a, not his. I can compile just about anything, but when I run: bash-2.01$ gcc -o hello hello.c bash-2.01$ ./hello bash: ./hello: Not owner However, "hello.exe" is world read-write-executable as reported by "ls -l". No such problem with the usual b18+coolview combo. Any clue? Thanks, Arlindo. PS: Below is how I set my environment: ------------------------------------------------------------------------- @echo off : : Sets path for GNU Win32 B18 system (with EGCS) : : : Path for compiler and X windows : Set PATH=c:\egcs\b18\H-i386-cygwin32\bin;%path% Set PATH=c:\egcs\b18\tcl\bin;%path% Set PATH=c:\bin\;%path% Set PATH=c:\usr\local\bin\;%path% Set PATH=c:\usr\X11R6.3\bin;%path% Set PATH=c:\winnt\system32;c:\winnt;%path% : : GNU Win32 environment : Set GCC_EXEC_PREFIX=c:\egcs\b18\H-i386-cygwin32\lib\gcc-lib\ Set TCL_LIBRARY=//c/egcs/b18/tcl/lib/tcl7.6 Set GDBTK_LIBRARY=//c/egcs/b18/share/gdbtcl Set LIBRARY_PATH=//C/egcs/b18/H-i386-cygwin32/lib/gcc-lib/i386-cygwin32/egcs-2.90.21://C/egcs/b18/H-i386-cygwin32/lib://C/egcs/b18/H-i386-cygwin32/i386-cygwin32/lib Set C_INCLUDE_PATH=//C/egcs/b18/H-i386-cygwin32/lib/gcc-lib/i386-cygwin32/egcs-2.90.21/include://C/egcs/b18/H-i386-cygwin32/i386-cygwin32/include://C/egcs/b18/include Set CPLUS_INCLUDE_PATH=%C_INCLUDE_PATH%://C/egcs/b18/include/g++ Set OBJC_INCLUDE_PATH=%C_INCLUDE_PATH% : : Other environment : set HOME=c:\ set DISPLAY=localhost:0.0 -- Arlindo da Silva dasilva@alum.mit.edu - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From weiqigao@a.crl.com Wed Dec 24 08:08:00 1997 From: weiqigao@a.crl.com (Weiqi Gao) Date: Wed, 24 Dec 1997 08:08:00 -0000 Subject: Windows 98 prospects References: <34A0A962.7057657@polymail.calpoly.edu> Message-ID: <34A13324.9FFAC256@a.crl.com> John Miller wrote: > > I was reacently lucky enought to get my hands on a beta copy of Win98. You call that lucky? I'm still dizzy from the little tooltip boxes that scrolls out in slow animation when I move the mouse over the "x". It says "close"! -- Weiqi Gao weiqigao@a.crl.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fac@neurocom.com Wed Dec 24 08:35:00 1997 From: fac@neurocom.com (Fabrice ANNIC) Date: Wed, 24 Dec 1997 08:35:00 -0000 Subject: bash & french keyboard in b18 under W95/NT4 Message-ID: <34A13939.F1768D09@neurocom.com> I got the cygnus NT port of the GNU unix tools and I am very happy with them. However there is one very annoying problem: I am using a french keyboard where I have to enter some special characters with the [Alt Gr] key (|, [, ], {, }, @). But in the bash prompt this does not work. That means that I can't even pipe two commands together, which makes the shell nearly useless. Do you know any fix/workaround for this? Thanx a lot Fabrice - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Wed Dec 24 10:06:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Wed, 24 Dec 1997 10:06:00 -0000 Subject: EGCS 1.00: can't run hello.c on NT References: <199712241526.KAA02923@woodmore.gsfc.nasa.gov> Message-ID: <9712241806.AA19090@modi.xraylith.wisc.edu> Hi Arlindo, All your settings looks good, so I'm hoping it's due to the bug in collect2 that causes it to exec itself instead of the real linker. Let's try the following test: copy the GNU ld executable (ld.exe in the H-i386-cygwin32/bin directory) to a file named collect-ld.exe in the gcc compiler directory (.../i386-cygwin32/egcs-2.90.21/). If you still have problems, please email me the following info: 1 the source code to hello.c 2 the output of the following command: % gcc -o hello -v hello.c -Wl,-verbose Happy holidays Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From chuck@ethereal.ru Wed Dec 24 12:34:00 1997 From: chuck@ethereal.ru (Chuck Bogorad) Date: Wed, 24 Dec 1997 12:34:00 -0000 Subject: anyone tried to build UUCP (Taylor) and smail ? Message-ID: <01bd10ab$38437940$0100a8c0@hysteria.spb.ru> Mike Bernson (mike@mbsun.mlb.org) wrote: >I have a version of uucp-1.06 that works under win'95 i tried building "UUCP 1.06.1". it worked out of the box - after ./configure i just added `#define HAVE_SOCKET 1"` to config.h and did "mount -b /dev/serial2 \\\\.\\com2" in bash. my mount table (relevant part - everything else is mounted as binary) looks now like this: [chuck@HYSTERIA]mount|sort|grep serial \\.\com1 /dev/serial1 native no-mixed,text=binary \\.\com2 /dev/serial2 native no-mixed,text=binary \\.\com3 /dev/serial3 native no-mixed,text=binary \\.\com4 /dev/serial4 native no-mixed,text=binary the only slight problem i'm having with it, is that the first packet is always rejected. the session, however, goes on. it looks like this (/usr/spool/uucp/Log): Protocol 'g' packets: sent 2, resent 1, received 1 Errors: header 0, checksum 0, order 0, remote rejects 1 also (/usr/spool/uucp/Debug): Handshake successful (protocol 'g' sending packet/window 1024/7 receiving 1024/7) DEBUG: floop: No work for master DEBUG: fgsendcmd: Sending command "H" DEBUG: fgsenddata: Sending packet 1 (32 bytes) DEBUG: fconn_io: Wrote 38 of 38, read 6 of 16307 DEBUG: floop: Waiting for data DEBUG: fgprocess_data: Got control RJ 0 DEBUG: fgprocess_data: Remote reject: next 2 resending 1 did you get similar results? did you look into the problem of rejects? i'm also very interested in what do you plan to use as local mail delivery agent. i tried sendmail, but it didn't compile. i tried smail - it compiled ok, but refuses to work (could be configuration issue) - it just crashes. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hgfernan@usp.br Wed Dec 24 13:43:00 1997 From: hgfernan@usp.br (hgfernan@usp.br) Date: Wed, 24 Dec 1997 13:43:00 -0000 Subject: rx.h missing ? References: <349855E5.4C4F33B@eml.co.uk> <6k5gFkj$pfB@mike.franken.de> Message-ID: <9712242146.AA22770@spider.uspnet.usp.br> On 18 Dec 97 at 23:53, Michael Hirmke wrote: > Hi malcolm, > > Indeed you have to get the sources to have rx.h - Cygnus forgot to > supply this file with the cdk.exe package. > > I would send it to the list, but it is about 3500 lines in size - so I > won't do it, if there are not at least four or five people, who'd like > to get it. Hi, Michael Another idea is to upload them to an anonymous ftp site that permits it. Maybe you could contact Jason L. Esman, aka _Jman, emailing him at jman@lx.net or root@lx.net. The site is ftp://ftp.deninc.com , and it is dedicated to Cygnus Win32. Regards, ++Hilton P.S.: Merry Christmas for you all. ------------------- Hilton Fernandes email: hgfernan@usp.br www: http://www.lsi.usp.br/~hilton.html (inactive) M. Sc. Student of Parallel Distributed Applications at Escola Politecnica (Polytechic School) University of S. Paulo - Brazil - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hgfernan@usp.br Wed Dec 24 13:43:00 1997 From: hgfernan@usp.br (hgfernan@usp.br) Date: Wed, 24 Dec 1997 13:43:00 -0000 Subject: Diff of executable files References: Message-ID: <9712242146.AA38180@spider.uspnet.usp.br> On 23 Dec 97 at 12:15, Martin Portmann wrote: > I want to do a 'diff' of binary files on the Win32 > platform (*.exe and *.dll). The problem is that > the binary files seems to contain timestamps so > that if I compile the same source twice I get > different binaries. > Where can I get information about the binary > format used on Win32 platforms? > Hi! A diff for executables? That seems a very useful idea. Maybe the compiler experts in this list could help you with this task. Please take a look at the Microsoft site, at http://www.microsoft.com And issue two searches: one for "COFF" -- the new executable format, used in Windows NT and 9X. And the other for "PE", the old format, for Win16. A good link is the documentation of the two formats: http://premium.microsoft.com/msdn/library/specs/f11/s9865.htm You need to subscribe to this "premium" site, but this is free. Microsoft web site is a good place to look for information about Win32 pgming. Some would say: they created the mess, they must help solve it. :-) HTH ++Hilton ------------------- Hilton Fernandes email: hgfernan@usp.br www: http://www.lsi.usp.br/~hilton.html (inactive) M. Sc. Student of Parallel Distributed Applications at Escola Politecnica (Polytechic School) University of S. Paulo - Brazil - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tjump@cais.com Wed Dec 24 13:50:00 1997 From: tjump@cais.com (Theodore Jump) Date: Wed, 24 Dec 1997 13:50:00 -0000 Subject: Windows 98 prospects References: <34A0A962.7057657@polymail.calpoly.edu> <34A13324.9FFAC256@a.crl.com> Message-ID: <34a18381.20023390@smtp> >You call that lucky? I'm still dizzy from the little tooltip boxes that >scrolls out in slow animation when I move the mouse over the "x". It >says "close"! You know, you can turn most of these things off via the View-Options menu ... - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From lizt@synopsys.com Wed Dec 24 15:49:00 1997 From: lizt@synopsys.com (B. Elizabeth Trojan) Date: Wed, 24 Dec 1997 15:49:00 -0000 Subject: installing ftp://ftp.cygnus.com/pub/gnu-win32/latest/CDK.exe Message-ID: <34A19F66.64880EEB@synopsys.com> The last message of the install process warns the user to read the file called README.txt. I could not find this file. What happenned to it? Is it critical? What steps in the install process am I missing? -- Liz Trojan Synopsys (503)531-2297 Office Beaverton (503)690-6906 FAX - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From lizt@synopsys.com Wed Dec 24 16:08:00 1997 From: lizt@synopsys.com (B. Elizabeth Trojan) Date: Wed, 24 Dec 1997 16:08:00 -0000 Subject: cygwin32 latest appears to be DOA Message-ID: <34A1A3CB.6F5992E1@synopsys.com> I just installed the latest cygnus stuff for win32 and it appears to be DOA. When I run a command I get the following: uname -a (unknown) In cygwin_except_handler (unknown) Error while dumping state (probably corrupted stack) (unknown) In cygwin_except_handler (unknown) Error while dumping state (probably corrupted stack) (unknown) In cygwin_except_handler . . . Has anyone else run into this? -- Liz Trojan Synopsys (503)531-2297 Office Beaverton (503)690-6906 FAX - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jstevens@samoyed.ftc.nrcs.usda.gov Wed Dec 24 17:23:00 1997 From: jstevens@samoyed.ftc.nrcs.usda.gov (John Stevens) Date: Wed, 24 Dec 1997 17:23:00 -0000 Subject: HELP: Relocatable DLL's Not Working Message-ID: <199712250125.SAA08387@samoyed.ftc.nrcs.usda.gov> Hi, I've tried every example I could find on how to use gnu-win32 to create dll's. Nothing I've tried works. I continue to get the same error message: Invalid format. Does anybody have a working example, or an explanation of why I get this error message? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From marc@watson.ibm.com Wed Dec 24 17:43:00 1997 From: marc@watson.ibm.com (Marc Auslander) Date: Wed, 24 Dec 1997 17:43:00 -0000 Subject: xargs Message-ID: <9712241917.AA47150@marc.watson.ibm.com> I have hit a limit in xargs - the total size of the passed in environment. Has anyone build xargs with a larger buffer? Marc Auslander 914 945-4346 (Tieline 862 Fax x4426) - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From nordwick@graft.XCF.Berkeley.EDU Thu Dec 25 02:54:00 1997 From: nordwick@graft.XCF.Berkeley.EDU (Jason Alan Nordwick) Date: Thu, 25 Dec 1997 02:54:00 -0000 Subject: Diff of executable files References: <9712242146.AA38180@spider.uspnet.usp.br> Message-ID: <199712251058.CAA23266@graft.XCF.Berkeley.EDU> hgfernan@usp.br, on Tue 12/24/1997, wrote the following: > > A diff for executables? That seems a very useful idea. Maybe the > compiler experts in this list could help you with this task. > There is a fairly good one for being alpha, called Xdelta. Look for it at htp://www.xcf.berkeley.edu/~jmacd > Please take a look at the Microsoft site, at > http://www.microsoft.com > > And issue two searches: one for "COFF" -- the new executable format, > used in Windows NT and 9X. And the other for "PE", the old format, > for Win16. > Amazing how COFF is called "new" > Microsoft web site is a good place to look for information about > Win32 pgming. Some would say: they created the mess, they must > help solve it. :-) > I have been looking for a C API, but cant find one. I can only find C++ level API documentation. (For the win32 subsystem) > HTH > jay -- Join the FreeBSD Revolution. Support the FSF, buy GNU. http://xcf.berkeley.edu/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From arlindo@niteroi.gsfc.nasa.gov Thu Dec 25 06:37:00 1997 From: arlindo@niteroi.gsfc.nasa.gov (Arlindo da Silva) Date: Thu, 25 Dec 1997 06:37:00 -0000 Subject: installing ftp://ftp.cygnus.com/pub/gnu-win32/latest/CDK.exe References: <34A19F66.64880EEB@synopsys.com> Message-ID: <199712251440.JAA04139@woodmore.gsfc.nasa.gov> Liz, > > The last message of the install process warns the user to read the > file called README.txt. I could not find this file. What happenned to > it? Is it critical? What steps in the install process am I missing? > You bet, it is critical to set the environment variables properly. Here is a BAT file I use on my system. Change "c:" and "//c" if have installed it on a drive other than "c:". Happy Holidays. Arlindo. --- cut --- cut --- cut --- cut --- cut --- cut --- cut --- cut --- cut --- @echo off : : Sets environment for GNU Win32 B18 system (without EGCS) : : : Path for compiler/TCL : Set PATH=c:\gnuwin32\b18\H-i386-cygwin32\bin;%path% Set PATH=c:\gnuwin32\b18\tcl\bin;%path% : : GNU Win32 environment : Set GCC_EXEC_PREFIX=C:\gnuwin32\b18\H-i386-cygwin32\lib\gcc-lib\ Set TCL_LIBRARY=//c/gnuwin32/b18/tcl/lib/tcl7.6 Set GDBTK_LIBRARY=//c/gnuwin32/b18/share/gdbtcl : : If you have X11 installed you will need this : Set PATH=c:\usr\X11R6.3\bin;%path% Set DISPLAY=localhost:0.0 : : Other environment : set HOME=c:\ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From weiqigao@a.crl.com Thu Dec 25 07:51:00 1997 From: weiqigao@a.crl.com (Weiqi Gao) Date: Thu, 25 Dec 1997 07:51:00 -0000 Subject: cygwin32 latest appears to be DOA References: <34A1A3CB.6F5992E1@synopsys.com> Message-ID: <34A28107.6BC6F606@a.crl.com> B. Elizabeth Trojan wrote: > > I just installed the latest cygnus stuff for win32 and it appears > to be DOA. When I run a command I get the following: > > uname -a > (unknown) In cygwin_except_handler > (unknown) Error while dumping state (probably corrupted stack) > (unknown) In cygwin_except_handler > (unknown) Error while dumping state (probably corrupted stack) > (unknown) In cygwin_except_handler > . > . > . > > Has anyone else run into this? No. D:\TEMP>uname -a CYGWIN32_NT WEIQIGAO 4.0 17.6 i586 -- Weiqi Gao weiqigao@a.crl.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Thu Dec 25 15:02:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Thu, 25 Dec 1997 15:02:00 -0000 Subject: bash & french keyboard in b18 under W95/NT4 References: <34A13939.F1768D09@neurocom.com> Message-ID: <6kYI45O$pfB@mike.franken.de> Hi Fabrice, >I got the cygnus NT port of the GNU unix tools and I am very happy with >them. However there is one very annoying problem: >I am using a french keyboard where I have to enter some special >characters with the [Alt Gr] key (|, [, ], {, }, @). But in the bash >prompt this does not work. That means that I can't even pipe two >commands together, which makes the shell nearly useless. Get the newest patches for cygwin.dll from Sergey's web site: http://miracle.geol.msu.ru/sos/ Please read and understand ;) the instruction on this site, too. > >Do you know any fix/workaround for this? > >Thanx a lot >Fabrice Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From BSimon@randata.com.au Thu Dec 25 17:18:00 1997 From: BSimon@randata.com.au (Brendan Simon) Date: Thu, 25 Dec 1997 17:18:00 -0000 Subject: Compiling "V" 1.16 - Specific Instructions? Message-ID: > >> Please be warned that V is mainly a Unix GUI library; its >> support to Windows 95/NT is still to be done. People wanting to > program for Win95/NT should take a look at wxWindows. >> Maybe wxWindows should be added to Cygnus page. > THIS IS DEFINATELY INCORRECT !!! V has both X Windows AND MS-Windows support. The MS-Windows version comes with make files for common compilers such as Borland etc. I have V version 1.16 working on both Windows 95 and X Windows (Linux) on a 486DX2/66. I compiled the MS-Win version with CYGWIN32. It runs about HALF the speed of the X Windows compiled version. I believe the performance hit is the CYGWIN stuff. I don't have any commercial compilers handy to compile V with so I can't be absolutely sure, but since other non GUI utilities such as "ls" are very slow I think I am right. Brendan Simon brendan@dgs.monash.edu.au > - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Fri Dec 26 10:12:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Fri, 26 Dec 1997 10:12:00 -0000 Subject: egcs plans Message-ID: <9712261812.AA21848@modi.xraylith.wisc.edu> Here're some near-future plans I have for egcs for cygwin32/mingw32. Comments welcome. - gcc-2.8.0 merge: Now that egcs has merged in 2.8.0 features and I have things running quite well, the next release will have the best of both worlds. Hopefully the new code to use the Win32 Registry will solve some of the installation woes. - dlltool/ld/gas fixes from Mikey: Hopefully I can merge in the fixes from Mikey that allows one to link with MS LIB files. I don't understand some of the changes yet, but I'll be in touch with Mikey to clear things up. - (mingw32) dlltool enhancements Pedro Pedro A. Aranda Guti\irrez: This allows synonyms and constants in .def files and obviates the need for libmoldname.a, and hence multiple imports for crtdll. - (very tentative) Installer: I don't know if this is going to happen, but it sure would be nice. If you can help in this, please let me know. I only have InstallShield Express, which is a pain to use when you tons of directories, and doesn't support scripting. - unpacked tree via ftp? I'd really like to have the unpacked directory hieararchy available via ftp, but I can't do it here due to resource constraints. I'm sure someone else will volunteer ;-) When? Definitely not before Jan 7 since I'll be few timezeones away (and fortunately +22 degrees C) from my home machine. I can still remotely build the snapshots and fix problems however. Looks like sometime around Jan 15. Why not just use FSF-2.8.0 when released instead of egcs? What benefits do you get over the soon-to-be-released gcc-2.8.0? - Crucial c++ fixes in egcs that affect the C++ code on PE-COFF. - g77 is "integrated" into egcs. Some of the "experimental" features such as alias analysis are in egcs, but not in FSF. - other misc x86 optimizations that are not in FSF. Happy Holidays to all. Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dlundy@primenet.com Fri Dec 26 16:51:00 1997 From: dlundy@primenet.com (Dan Lundy) Date: Fri, 26 Dec 1997 16:51:00 -0000 Subject: Status Access Violation Message-ID: <001901bd1254$122c7c20$c632a5ce@primenet.primenet.com> I have scaned the archives extensively without success.???? You know the problem, ???? STATUS_ACCESS_VIOLATION. ???? I have read a couple of potential fixes, and have attempted them without success. ???? If you have had similar problems under Win95, and have found an appropriate method to cure this problem, please provide me with the details. ???? dlundy@primenet.com From rdmiller@execpc.com Sat Dec 27 04:42:00 1997 From: rdmiller@execpc.com (Rick Miller) Date: Sat, 27 Dec 1997 04:42:00 -0000 Subject: egcs plans Message-ID: <199712271241.GAA20593@core0.mx.execpc.com> I might be able to help with the installation... I've been using InstallShield 5 Pro. but I haven't tried a "self extracting" thingamabob. I usually make a directory tree with "disk1", "disk2" etc. directories in it so that it can be installed from there or put onto floppies. I've done some pretty extensive scripting for the wierd ways my employer likes to have its software installed and have some experience modifying the registry from the install script. I could probably learn to create whatever kind of install you're looking for, except International... that's a different InstallShield "product". Rick Miller http://www.execpc.com/ ---------- > From: Mumit Khan > To: gnu-win32@cygnus.com > Subject: egcs plans > Date: Friday, December 26, 1997 12:12 PM > > Here're some near-future plans I have for egcs for cygwin32/mingw32. > Comments welcome. > > - gcc-2.8.0 merge: Now that egcs has merged in 2.8.0 features and I have > things running quite well, the next release will have the best of both > worlds. Hopefully the new code to use the Win32 Registry will solve > some of the installation woes. > > - dlltool/ld/gas fixes from Mikey: Hopefully I can merge in the fixes > from Mikey that allows one to link with MS LIB files. I don't > understand some of the changes yet, but I'll be in touch with Mikey > to clear things up. > > - (mingw32) dlltool enhancements Pedro Pedro A. Aranda Guti\irrez: This > allows synonyms and constants in .def files and obviates the need for > libmoldname.a, and hence multiple imports for crtdll. > > - (very tentative) Installer: I don't know if this is going to happen, > but it sure would be nice. If you can help in this, please let me know. > I only have InstallShield Express, which is a pain to use when you > tons of directories, and doesn't support scripting. > > - unpacked tree via ftp? I'd really like to have the unpacked directory > hieararchy available via ftp, but I can't do it here due to resource > constraints. I'm sure someone else will volunteer ;-) > > When? Definitely not before Jan 7 since I'll be few timezeones away (and > fortunately +22 degrees C) from my home machine. I can still remotely > build the snapshots and fix problems however. Looks like sometime around > Jan 15. > > Why not just use FSF-2.8.0 when released instead of egcs? What benefits > do you get over the soon-to-be-released gcc-2.8.0? > > - Crucial c++ fixes in egcs that affect the C++ code on PE-COFF. > - g77 is "integrated" into egcs. Some of the "experimental" > features such as alias analysis are in egcs, but not in FSF. > - other misc x86 optimizations that are not in FSF. > > Happy Holidays to all. > > Regards, > Mumit -- khan@xraylith.wisc.edu > http://www.xraylith.wisc.edu/~khan/ > > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sat Dec 27 08:43:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sat, 27 Dec 1997 08:43:00 -0000 Subject: how to get source of resource compiler and ld that supports .res f Message-ID: <199712271643.QAA42804@out1.ibm.net> >> I could build gnu-win32 cross on FreeBSD2.2.2. (gcc, as, ld etc) But I cannot link my objects with .res file(resource object file). How to get source of resource compiler and ld that supports .res file? It causes error. I'd like to build cross development environment on FreeBSD and get exe(executalbe file) on FreeBSD without using Windows. << In theory, you convert .res files to .o files and link them to the executable. In practise, these executables don't register under Windows. What I mean is that an icon .res file doesn't show up as an icon when the Windows file explorer displays the name of the executable. While you're at it, you might consider using libwine.so under FreeBSD to generate FreeBSD Windows executables with resources compiled in. The makefiles should be almost identical to `real' Windows makefiles in your cross-compiling environment. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From victor@tir.com Sat Dec 27 09:01:00 1997 From: victor@tir.com (victor) Date: Sat, 27 Dec 1997 09:01:00 -0000 Subject: Installation and running Message-ID: <004d01bd12e9$03331040$b955dacd@vic-and-michele> Title: Untitled Normal Page I followed the installation process as best I could.???? I was unsure of the exact syntax to use in the autoexec.bat???? Is it ???? SET GCC_EXEC_PREFIX=D:\gnuwin32\H-i386-cygwin32\tcl\bin ???? I set the program up on the D drive and eliminated the b18 sub-folder. I did the same with the other lines.???? Do I have to do anything with the "mount"???? if so, what do I do? (In Windows 95 please) Another problem I am having is the line when I launch base.exe is the???? request for /tmp.???????? What do I do and where do I do it????? In Windows95 you don't create a /tmp.???? I created a TMP folder and added a line TMP=J:\tmp to the auto exec.bat.???? Please explain these things in a manner that people with no experience in programming and working???? in Win 95 can understand.???? I have signed up for an online C++ class and they recommended your program.???? I want to have it set up before the class starts on the 5th.???? Vic???? Peters???????????? victor@tir.com From cgf@bbc.com Sat Dec 27 13:16:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Sat, 27 Dec 1997 13:16:00 -0000 Subject: Status Access Violation References: <001901bd1254$122c7c20$c632a5ce@primenet.primenet.com> Message-ID: In article < 001901bd1254$122c7c20$c632a5ce@primenet.primenet.com >, Dan Lundy wrote: >-=-=-=-=-=- > >I have scaned the archives extensively without success. You know the problem, > >STATUS ACCESS VIOLATION. > >I have read a couple of potential fixes, and have attempted them >without success. > >If you have had similar problems under Win95, and have found an >appropriate method to cure this problem, please provide me with the >details. > >dlundy@primenet.com Compile the program with the the -g option and use gdb to debug it just like you would any other program. STATUS ACCESS VIOLATION means that you are accessing memory that you shouldn't be. Possibly you are accessing a pointer with garbage in it. This is no mystery. The answer on how to debug problems with your program is almost always going to be "run it under a debugger". That's why debuggers were invented. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Sat Dec 27 16:38:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Sat, 27 Dec 1997 16:38:00 -0000 Subject: Installation and running References: <004d01bd12e9$03331040$b955dacd@vic-and-michele> Message-ID: <6kjQKdMppfB@mike.franken.de> Hi Victor, [...] >SET GCC_EXEC_PREFIX=3DD:\gnuwin32\H-i386-cygwin32\tcl\bin SET GCC_EXEC_PREFIX=3DD:\gnuwin32\H-i386-cygwin32\tcl\bin\ ^ this is important [...] >Vic Peters =20 Bye. Michael. P.S.: Please stop sending mails in HTML format. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From victor@tir.com Sat Dec 27 17:58:00 1997 From: victor@tir.com (victor) Date: Sat, 27 Dec 1997 17:58:00 -0000 Subject: install Message-ID: <000901bd1334$068be320$9c55dacd@vic-and-michele> I did add the trailing slash. Sorry about the HTML format. I switched now. I still have questions about the 1) /tmp how and where do I apply this? I made a set tmp=J:\Temp in the auto exec.bat. 2)the drive mount that is refereed to?? how and where do I do that? 3) what is bin/sh and how to create it and add sh.exe to it? This type of thing is very confusing to those of us only using Win 95 and no programming experience. Vic Peters victor@tir.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sat Dec 27 21:00:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sat, 27 Dec 1997 21:00:00 -0000 Subject: clear.c program that clears cygwin screen and disconnects keyboard Message-ID: <199712280500.FAA02704@out2.ibm.net> As the caption promised, the following is a supposedly correct clear.exe program that clears the bash$ window, but leaves the cygwin keyboard not working. Try it and see for yourself. clear.c: ========================================================================= #include int main (int argc, char **argv) { /* ansi.sys (and Linux) clear screen command: */ return printf("\e[2J"); } ========================================================================= - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From nordwick@graft.XCF.Berkeley.EDU Sat Dec 27 21:40:00 1997 From: nordwick@graft.XCF.Berkeley.EDU (Jason Alan Nordwick) Date: Sat, 27 Dec 1997 21:40:00 -0000 Subject: Anybody ever gotten dvips (TeX) to work ? Message-ID: <199712280542.VAA28785@graft.XCF.Berkeley.EDU> I have TeX/LaTeX working, generating dvis, and yap can view them. But when I run dvips, I get some font errors. I installed with all the defaults. Can anybody please help me, I really need to get this working soon. Here is the result of running dvips on a dvi file generated from LaTeX (MiKTeX). Also, what do pdftex and pdflatex do ? I thought that they might generate pdf files, but they still give me a dvi file and dvips chokes on. C:\jay\latex>dvips pp.dvi This is dvips 5.74 (MiKTeX 1.09) Copyright 1997 Radical Eye Software (www.radicaleye.com) ' TeX output 1997.12.21:1142' -> pp.ps - make_pk_font (cmr17, 600, 600, lwpro) Creating cmr17.pk... c:\texmf\miktex\bin\virmf.exe \mode:=lwpro; \mag:=magstep(0.0); batchmode; input cmr17 This is METAFONT, Version 2.718 (MiKTeX 1.09) Cannot access GF file cmr17.600gf! C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmr17.600pk not found, characters will be left blank. - make_pk_font (cmr12, 600, 600, lwpro) C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmr12.600pk not found, characters will be left blank. - make_pk_font (cmbx12, 720, 600, lwpro) C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmbx12.720pk not found, characters will be left blank. - make_pk_font (cmr10, 600, 600, lwpro) C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmr10.600pk not found, characters will be left blank. - make_pk_font (cmsy10, 600, 600, lwpro) C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmsy10.600pk not found, characters will be left blank. - make_pk_font (cmbx10, 600, 600, lwpro) C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmbx10.600pk not found, characters will be left blank. . [1] [2] [3] jay -- Join the FreeBSD Revolution. Support the FSF, buy GNU. http://xcf.berkeley.edu/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From emmess@bml.RMV.se.org Sun Dec 28 03:34:00 1997 From: emmess@bml.RMV.se.org (Martin 'EmmEss' Svenningsson) Date: Sun, 28 Dec 1997 03:34:00 -0000 Subject: bounce? Message-ID: <3.0.5.32.19971228123259.0091a3a0@Spam> Q: Why does http://www.deninc.com/~gnu-win32/ bounce to http://members.sexroulette.com/?acb=acb107314-d1150 ?!? I know deninc.com's http for gnu-win32 is no more, but why the bounce?? ---- Greetz from Helsingborg, SWEDEN | When replying to me, Martin Svenningsson, aka EmmEss | remove the RMV from http://www1.tripnet.se/~emmess/ | my mail address. ---- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Daniel@telaviv.ndsoft.com Sun Dec 28 06:04:00 1997 From: Daniel@telaviv.ndsoft.com (Daniel Narkis (req-telaviv)) Date: Sun, 28 Dec 1997 06:04:00 -0000 Subject: Operating bash/sh on bi-lingual Win95 Message-ID: <199712281505.RAA23207@proxy1.ndsoft.com> Hi, I have a problem with running bash/sh on my win95. This win95 is an English version that supports Hebrew. When I enter bash I can't type almost anything (because it refers to the Hebraic keyboard layout). I've tried to change the country & keyb in the autoexec.bat & config.sys to the default US English but nothing helped. Can anyone help ?? Best regards, Daniel. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Sun Dec 28 07:54:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Sun, 28 Dec 1997 07:54:00 -0000 Subject: install References: <000901bd1334$068be320$9c55dacd@vic-and-michele> Message-ID: <6kjSPY6KpfB@mike.franken.de> Hi Victor, > >I did add the trailing slash. Sorry about the HTML format. I switched now. >I still have questions about the 1) /tmp how and where do I apply this? I >made a set tmp=J:\Temp in the auto exec.bat. 2)the drive mount that is >refereed to?? how and where do I do that? 3) what is bin/sh and how to >create it and add sh.exe to it? This type of thing is very confusing to >those of us only using Win 95 and no programming experience. 1) mkdir c:\tmp 2) don't know, what you are refferring to here 3) mkdir c:\bin copy \bash.exe c:\bin\sh.exe >Vic Peters Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jman@lx.net Sun Dec 28 09:26:00 1997 From: jman@lx.net (jman) Date: Sun, 28 Dec 1997 09:26:00 -0000 Subject: bounce? References: <3.0.5.32.19971228123259.0091a3a0@Spam> Message-ID: <3.0.5.32.19971228112714.007b0100@208.221.108.11> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 12:32 PM 12/28/97 -0100, Martin 'EmmEss' Svenningsson wrote: >Q: Why does http://www.deninc.com/~gnu-win32/ bounce to > http://members.sexroulette.com/?acb=acb107314-d1150 ?!? > >I know deninc.com's http for gnu-win32 is no more, but why the bounce?? It was an accident I appologize to anyone that went to the site, and got bounced to the wrong place, it will now bounce correctly. Once again sorry about that. Was a simple mistake on our Webmaster's part. -----BEGIN PGP SIGNATURE----- Version: PGP for Personal Privacy 5.0 Charset: noconv iQA/AwUBNKaL8Q6ne3t4b32aEQIN/gCeJ2d6oGYbdDWdlZvyvn5qsHiVqIoAoKt9 rGV6TwSrJJkcZz9BVVZGRsLt =4qe5 -----END PGP SIGNATURE----- ------------------------------------------------------- Jason L. Esman aka _Jman Owner Den Internet Services System Admin. Network Consulting http://www.deninc.com Email jman@lx.net or root@lx.net Finger jman@lx.net for PGP Public Keys... ------------------------------------------------------- - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kunglao@prairienet.org Sun Dec 28 10:23:00 1997 From: kunglao@prairienet.org (John Corey) Date: Sun, 28 Dec 1997 10:23:00 -0000 Subject: bounce? References: <3.0.5.32.19971228123259.0091a3a0@Spam> Message-ID: <34A6990E.BB002DF9@prairienet.org> Martin 'EmmEss' Svenningsson wrote: > > Q: Why does http://www.deninc.com/~gnu-win32/ bounce to > http://members.sexroulette.com/?acb=acb107314-d1150 ?!? > > I know deninc.com's http for gnu-win32 is no more, but why the bounce?? > ---- > Greetz from Helsingborg, SWEDEN | When replying to me, > Martin Svenningsson, aka EmmEss | remove the RMV from > http://www1.tripnet.se/~emmess/ | my mail address. > ---- > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". Apparently it has been set up to go to a cgi script that goes to different URLs each time it runs into a non-existent web page. How about just a 0 byte file, or a bounce to cygnus.com for anyone that tries to use the former ~gnuwin32 path? FYI, here's where it bounces you: The document has moved here - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mihailov@math.upenn.edu Sun Dec 28 11:37:00 1997 From: mihailov@math.upenn.edu (Aleksandrs Mihailovs) Date: Sun, 28 Dec 1997 11:37:00 -0000 Subject: Anybody ever gotten dvips (TeX) to work ? References: <199712280542.VAA28785@graft.XCF.Berkeley.EDU> Message-ID: > From: Jason Alan Nordwick , on 12/28/97 2:21 PM: > Also, what do pdftex and pdflatex do ? I thought that they might > generate pdf files, but they still give me a dvi file and dvips > chokes on. To produce pdf output from a latex file, you have to add in the beginning of the file two lines: \pdfcompresslevel=9 \pdfoutput=1 After that, if you type in the Run field of the Start menu pdflatex yourfile where yourfile is the name of your latex file, you get a pdf output. For more details, see David Meeker's http://members.aol.com/dcm3c/running.pdf Also, see the Miktex project page http://www.inx.de/~cschenk/miktex/ By the way, at the end of that page you can find the address of the MikTex mailing list, probably they can help to solve your dvips problem. My dvips (from the MiKTeX 1.09 installation) works excellent. Thank you, and have nice holidays, Alec Mihailovs http://www.math.upenn.edu/~mihailov/ -------------- next part -------------- A non-text attachment was scrubbed... Name: ATTRIBS.BND Type: application/octet-stream Size: 2350 bytes Desc: not available URL: From earnie_boyd@hotmail.com Sun Dec 28 13:50:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Sun, 28 Dec 1997 13:50:00 -0000 Subject: bounce? Message-ID: <19971228215008.12838.qmail@hotmail.com> For me it bounced to the Cygnus gnuwin32 project page. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Sun, 28 Dec 1997 12:32:59 -0100 >To: gnu-win32@cygnus.com >From: "Martin 'EmmEss' Svenningsson" >Subject: bounce? > >Q: Why does http://www.deninc.com/~gnu-win32/ bounce to > http://members.sexroulette.com/?acb=acb107314-d1150 ?!? > >I know deninc.com's http for gnu-win32 is no more, but why the bounce?? >---- > Greetz from Helsingborg, SWEDEN | When replying to me, > Martin Svenningsson, aka EmmEss | remove the RMV from > http://www1.tripnet.se/~emmess/ | my mail address. >---- >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From baeck@swipnet.se Sun Dec 28 14:33:00 1997 From: baeck@swipnet.se (Magnus Baeck) Date: Sun, 28 Dec 1997 14:33:00 -0000 Subject: install References: <000901bd1334$068be320$9c55dacd@vic-and-michele> Message-ID: <3.0.5.32.19971228233337.00a48440@janus.swip.net> At 20:57 1997-12-27 -0500, you wrote: >I still have questions about the 1) /tmp how and where do I apply this? I >made a set tmp=J:\Temp in the auto exec.bat. > That won't do, /tmp is referring to the directory \tmp which doesn't exist. Either create it, or mount your temporary directory as /tmp; mount j:\temp /tmp The mount command without parameters will show what directories you have mounted. >3) what is bin/sh and how to create it and add sh.exe to it? > Mount your GNU-Win32 bin-directory (normally \gnuwin32\b18\H-i386-cygwin32\bin) as /bin just like with the temporary directory. ___________________________________________ Magnus Baeck baeck@swipnet.se - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Sun Dec 28 16:08:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Sun, 28 Dec 1997 16:08:00 -0000 Subject: New alpha version of ntxlib on ftp.deninc.com Message-ID: <199712290008.AAA54454@out5.ibm.net> I posted a new alpha version of the ntxlibw32.tar.gz adaptation of ntxlib.zip on ftp.deninc.com. In theory, you could ultimately use Sergei's libX port minus libX11.a plus the emulated version below, and you could compile working programs. In practice, that may take filling in a few blanks as the linker informs you of missing or empty functions. But, it should be easier to get things like tcl/tk programs that call only XSync or whatever to run. Unfortunately, ntxlib's libX11.a (the simulated version) compiles with few problems, but the sample test program causes the following error message from Windows 95: TST caused an invalid page fault in module at 0000:8b26ddea. Registers: EAX=00400381 CS=0137 EIP=8b26ddea EFLGS=00010a96 EBX=0253fe24 SS=013f ESP=0253fe10 EBP=0253fe28 ECX=0253ff68 DS=013f ESI=8156c524 FS=19d7 EDX=0253fe24 ES=013f EDI=81571550 GS=0000 Bytes at CS:EIP: Stack dump: 000098c3 0253013f 00400381 0040667a 00000000 bffb89c0 0253fe38 0040103b 81e58955 037f0500 0253ff78 bff88f75 81571550 8156c524 00530000 00747354 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From tarheit@wcoil.com Sun Dec 28 20:09:00 1997 From: tarheit@wcoil.com (Tim Arheit) Date: Sun, 28 Dec 1997 20:09:00 -0000 Subject: undefined reference: gethostbyaddr, etc... Message-ID: <3.0.32.19971228231550.00747658@alpha.wcoil.com> I'm using b18 and am attempting to port a program. Compiling the parts of the program seems to work fine (I even had to grab and compile bison and it work just great), however when it gets to linking the program it reports several undefined references. Specifically gethostbyaddr, gethostbyname, srand48, drand48, tzset, major, minor and ualarm. I couldn't find these in any library (are they simply not implemented yet?) A search did show all the mentioned function listed in a file named SYSCALLS.c.X. Any ideas? Please forgive my ignorance if this is simple... it's the first time I've used this variant of GCC ;) -Tim - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net-nospam Sun Dec 28 20:47:00 1997 From: vischne@ibm.net-nospam (vischne@ibm.net-nospam) Date: Sun, 28 Dec 1997 20:47:00 -0000 Subject: install Message-ID: <199712290446.EAA64108@out2.ibm.net> >> Michael Hirmke (mh@mike.franken.de) 28 Dec 1997 16:28:00 +0100 > >I did add the trailing slash. Sorry about the HTML format. I switched now. >I still have questions about the 1) /tmp how and where do I apply this? I >made a set tmp=J:\Temp in the auto exec.bat. 2)the drive mount that is >refereed to?? how and where do I do that? 3) what is bin/sh and how to << You're getting piecemeal information. This system isn't for an inexperienced programmer. You need to know how to set up a Unix system, which means you need to know some version of Unix. It's a wonder you haven't had a serious accident yet, or that your other helpers haven't had one. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net-nospam Sun Dec 28 21:30:00 1997 From: vischne@ibm.net-nospam (root) Date: Sun, 28 Dec 1997 21:30:00 -0000 Subject: ntxlibw32 compiled using ming doesn't crash - tst.exe.gz [1/1] Message-ID: <199712290530.FAA36482@out5.ibm.net> I checked, and my cygwin version of tst.exe using ntxlib crashes before the first statement in WinMain. Out of curiosity, I reconfigured for mingw32 and recompiled, and the darned thing worked like a charm. Just to prove it, here is tst.exe, attached to this letter. Just get a uudecode program and decode it. begin 755 tst.exe.gz M'XL("#TUIS0``W1S="YE>&4`[7U]?%O%E>B5=&TKB1PIB0P"%"*"`TYBJ)TX M),H'R))E.V`EBN6O0MS@.')LX=C&EA('HL2I;,AP\:ZZ0!]0R@^OH?!:WC;[ M-D!*T\9.3)1T0]=DT]<`@7K;M'M=N:W;!F/`S=US9N[5A^V`6][O]_:/9SAW M9LZ0)(X[BC'_FS<%_]U`,Q=].9<[LBLMV\Z MJBIY^Z:R^H8V2TMK\\[6FEV6VIJFIF:_9;O7TAIHLC0T60HV>RR[FG=X;T]/ MGYTIU^%VB"P:=_;_\=[O?V^Z'L*]"9@C[RB?36#CN M_MNWM[5A?&LJ)_=]ZE\'5'/[CAI_#<1'903*AMN:3&=#N@9&2/L,?>=T`%^; M2F>QU.]IVN&N:?4V^==9TF=;+,4;-WG*\C0*FB)D10[\PN`D7K9RXR;[YLTEEFW;:OS^$Z2)/%_P4-:A55)JPST M22O[)S?'D9/B+X#.([[DX3BA>H*6%8\!_9F39,Q3LZNETY45GMSWK=7: M_68A:!`"NK!TA]A0"A6DA2)\Z!3O@+P#T6[OY?N^MNW,R>[='/(@5&L%IT[( MZ(3F?H"/KG/^16'KO7Q@GM6N#>@$-5&5BG7`0%1+WA'F=YT+_)(RD5#>I>NA M)?O\-X>M57S@6F$>U))B=6-YFY:HCJ*EB6H%(`E\F-N74-['B_.A;G(""EPG M.+7$S@,[Q*ZE-1(UH8PQ>89&^`J/$-!&G!,XVY0VEX:M+7P@0RB?4!J=A8TZ M)Z*SPK3%H6@JM!A=&:X0MV]!>7:][[=`P\^4R0TO(*T\N3>Y22GE24Q=H.)2 MVM?2]G/[B',0[+ MV'F2\1R5&J.#+($B?O`T&]];PE:U?Q:,9BD/P_EGP!@%*)3/S[#\C5/*ZUCY M4G&#)\X?K0WD<[W`23GB'S;#D"%&X`2;$1"_V:#0\9##H4YCF^6HEH"\ M*?30>E5`[UO@X^3_I&A&.)Z,IOH6N.'/=T5:(7YKLS+WZ$`)U3JAD(=*9GO< MXG_"A`D-:(542*_$$0BU0\5SDRHV)M?K@3^H-T=T0+W=WG$;$MM`F9?K-KP$P[W'M.%%"-IN MV_"/F+)NZ(&@=45T87C#RQC+VO`*!+OG;I@%2]>>ZS>H(&A=(-B-P$6*E!/5 MA$$R>S?%QR4TLJ\2^Z#+?5\*:,5S!:!#D=``WR7I.X-I''?P"FWVAH-7:+LK M#UZA#=]Y\`IM^8[H3>VC3RP]>H6T;#EZAC4,1VOH2H2BA]=7B9RZL%;/T MG7_$YZ-=4"QT9;Z^LS`%RJT-74G5=_X0E'O/M:$K:GUG3@IF:_2=3\#R/?QA M*J;T^LXM:HP8])T&R!^.`#H8#1J$\K!@A+35"8<7J MH@T*1]>%<<2WELA6Z!:P0D]L1A."PW_.?PW40^Z^BBWRB.M=N`RU?B9)D93[ M'P0\_%&+-'PO,(7JXX:0V8H?,ENA[]R`_%K7@KWX]YB]T'?>BH1@,QR3;,[P M7,R0YQ:3`9M?.BE@<(N+)V(S3-_Y*Q!'*&C@IIUG"Q/G&Z/"O."R%22K@\^AULEI2_HI:OIUU?U&("[1T) M]49GSZB<)B8L\=S='->+$AV>JT$-,L/X6\NU_@=!7Y:SRM*N5AFK8OA4O&_: MJY&6BK^^"QI">BE@!$VHQXF%([5W(X@GQ"DCE1IC#>LASD.L*J>Q7S00Y^.] M6%FW\Y`;BQ9"40$H#E&*QWL?5X9"Z]>$_I671P6B*AI5^34=_\:B!HRJ:52- M49XL#-/"6I*.$AA&BRGW24UY"\LZ!F5(^I/QSE]0*9S?A.-\1A4K]T5B>U[% M9NXW(1320N,0%2#JXWV<6TS[6)+(R="`#FS%(A\XI>CA:CR-6S_&Q6#WZSFP5EN970F:LQM\6(^XB/,7FKV#TURM` MC0K"R!DZ>W+E?\0H9W7""O`;3JE\.ZT\7M?SM*XPUG4SU@5E8;H?`ZKA?X[K MR#5TA/U,*$*^EDWM?!W50B'?P.25;^QE^OJX+'CFR39U2?[ZJ?5T3JFG*)/5 M4Y0UJ9[HT@0^HN:X/65Z68&*]5$A9&G"N7W=+[5#!GKM)PR;RY`]DI:;-!#VD2H;^+NO9=!4W+?CZ2X6Y@)B*1P$'L!E*O#ROD-CZG6 M6_S\^EF!U'Q(!T8[K(L#J;@.?/Z7F[#U8F.BZL'&),D/>0>IYX2_@ MX%`]+;(*BX@O.-@&"*10L-'C+LG_*K?-M;EL8^&V2M3CF)YO!OV&F MO_[`0BCX!$B2.$=`RBW,.D12=L0$4@^QEA8<''QR%.>/Q=I;F*&*I!R*E7T< M<^F,VX?,45P'XB2,==*R$N=SBZN+Z?;MR;6XQ!*/5ESFQ)C@T!(#/"(V+9W, M&,-E!TS*J%M,I32E-!2!5H6Z%A@F$,#(V;V8`6DT!0I-+-,"PLR69#%@FP6Y+`@CP5K M6+">!386%+"@F#51PM/`S9!E+*ABP586W,^"'2RH9T$C"UI8X&=!.POV8="# M/26/="#FD4Z._44>.<2Q(S;R_.,T'8[E/!F+/1V+/4>5!F,OQ'`OQF*OQ&*O MQF*'8[$CL=C16.R8'`L-6&`:/`K3H$)IB&2JD12 M:.14@3J-SV>E#OPK0]EX3HZD*I$4):*5Y,AL)3)+DE9A3/Q'.V/M/#7"0_)Z MZN-*L2N[[*BP(V*U'2?<2&\6.J2CW=X+2>6L@#7O_Z:3(8&E(S0C^ZQ]K0?BWG!SEC M!I^&&<::$$V76^R0U[D_+)CVQ M>.]S:-D2Y-#[0C)"U_MB,L+0^THRPMC[:C+"W'LX&6'J/0*(-\.)V%Y,$?U1 M#)0#.:H7;.T27#KANHASA!J7V"E3%3UE&DD^91I)/&4*C(`?[A'WW8FVO6O, M?S/L[_XE/[Z_6T#NY4GK=)L[7WJ%V'47KI*7H'']&_E\.EHO6`,0&:FL$/?0 M7+&[>A39DW+%_=!*I:>"AF#]*\1RH*@4*^_"Y*4*\0Z(5##2NWXG2>%F-&C.BT>-0%`]1*HO M`1_UW`%T\V;=25=V.:6AJ5']&Q&9^S"BF!=U9T,BS4([_V5$^OKZ1)$>7C])I#LW3"O2\/J82)WKJ4COHR(MR[FZ2%]=047ZTHJ_ M4J2Y83$_=\8R+5SQ-\K4@^>'CCS8R_CGN,'J[_TMV^@X\%RR37#GA3Z%G%+8 M7&R+Y20?E%HFGY/JE7/2C#MQ\>63CTC/KIUT1'H2$%*;3OPXAU)[Q-_GR`L, M\W,-..:PR@2,0M`D=##CHD59P*XR*Z(R<7B6`1.V7">M$%];BX(8):D"#%\: MQ(3R\:4G!)>9S`.,4#Z!"0N9)\_OYQDY)2*#0";,PTB_")O73*+J]EY*FL^4 MEZ!.J#;@V>W795Z(PPP>*^4C,%H!#D:.S(0&F7#PL08\6H4'XM$AX:+DYCU: MN7'J`UN8/YLYFP99*31@0^I@?J&#^87,Y8TXF%_HL+%R!70W[2ACFB-S2QQ5 M='>G="QIG88].;I6:F&+=NR$VI\"B0[UV`G>GY*.KQ_'3FC]:1#KXZ*(X#HH MG8'F:B'70F-JB-EH3`5^@+Q7R9;G;3K=RKG.^U34/Y-667`N'5S)K+<.EG%. MWUFNQDW"17>WF M^C#5!SM+H`K9LBQ1;9C%M#3(Q.!@>^:G4-7NV1$;'3:@@1@.C<^`E;=D22V9 M4HM%:C'[<-J\M!HX^)K.IQ4C:^-KG8% MGEK856A8[[Z0,RJ!/+1.P%+@/`[&)%8KZ^W+<%C1:/HXL64UW0D6ZF`G6)@- M)F".FPEN6ZPWGU@_KS>4>"T0PR!?XTMAZ16K.3DGA\6F.1E-3S@9I2>AWCM` M6?:;I<`=H?T6*9"#NU!6R;?N8).UT(RSM=""\]$UOK2?;,K$V>J:P&@6ZD'@ M1EE8/H.T.TO:G2GMMDB[<7SN$%.PEE0\$NHZA[(OS)-E)!1JI9Q$6I/H#3UTE:/N M^&'2ECMP$73@$WJRYQ*.+^R@\+P>T@XZS;00^V>YD<`R7PHDO[N*YG\/`GJ4 MLWA,DCP>I9B.%4,Y0^K&56AS3L=M#MT3R!N">A&$*OZ!*DL$K18'+ES$SK;0 M=KKT?7P*;Q[XM4)1%K&OCZ9%[%0C/SZ%1/X4H4@7U8'Y$@XRDYQ'[):/3Z$O M[)\E%)F(/1NUV)Y-=ZJC2W#6^S5"D9O8=]"DFB:+B7TK3?(T64+L]V.[%FCW ML:*J0_;&*'_*WLA3$@,E*2/V>DP"TYC,)/8UF%2QI(78\Z`"&^-B/;&[*1?N M_]0@Y[S,N3:)\S7$;O[X%/+I3Q/2B-V(?35J9!&H:$T%Q%Y%:ZJB_?ECYYD@ M8,W$GD.Q=&T@H["#02:,Q)Z%"0,F#,2>^?$IM2P7GMA-M(1)8O6#,/VI0E$. ML1?4=8T%0=HV8B]##LKD0>`913:QV["C!M91$'&.(#"J> M/]TKZI@+TU%LX)-].EA35T><`\G:S]XX#R1K_T"2]@^`]E>(+2L2M/_E58G: M?[67SCZ51WR&.J(&_1LV,_/[/.+?K<2UYS3$'E^)+CZ8"C`+IN7.LVXI8/2( M^U>R\^#2_T!G_^C4-30T8J3],>#V)G?2]B;V%O_S]C?5([1/S^4F].G?\V;2 MITJ/^&86[E?$=D1 M>["I4[)'U#XD25##+BPXVNT=2O)_+/*^;7+'LF;>KX&L1GJJQ`?S*6[/(_HSL4C]J`.*O[J M+W#@SU.IH&_C4\%_:,,;"KUS$!.BKR5$"OY=4[\W38.`CATN*\TZ(_?J]:_84]577#KC\^&&*\" M[=TB?@#T/K6[HE+\50[')=^'"8WD4'\LJVML7ZHFD(FCY,Q<&L@DIW!P9Q'7 MX*3K)\N$ZD'OP0GZ[G50'+H=M:Y+HN]>KUV1].YU$-^] M/C3=,$L!0Z5XYE;8-56*O_\*#NB@X+2`-ZQQFG+/E;I_D9V)^P^G)??KY"?">+[DCDBG?2BL]?=%Z$_*O6+04RQ6>2 MRMT:8^B+RM4DEW!XHM5%NZ"^9(;GS'4VVF42E@VE(I_O%VU-_!2O$% MB%2RZ?W,1=3?@9C^AD9*J+W*ZI)`%X**+@0ST6K0B?XWW4/"F?Y>=L),GYAI,;Q\9E".U=S^F[\O&UVXPTHE+,N#6F$=_Y"AL`YY?5 MB$KQ^"V)(U3^E9EJQ*-)Y3)B#'U1NY?DM@A(ZWXR$7G ML2\21&%2N?.W*0Q]4;FY2>5>O&TF[4TOP"/+L28=+#:ERW$UH+/IGG=Q-KV2 M8%N/%BP"7W8U2)N:6"-.(FNDZ]@0>D/Q>90;1M_^!N$IS$BZ1TA)E=GT'EC7 M;U/$JDKQL:4)\ZD_>R8^/IU/M]],Y].WEJ%1SCT'7;>6F_8O"NN/IY+.G_5! M[?KCA6J(#D`T=(JW0_:!7T*_N[LP$XKV+:->Q5>7Q3I>>@$Z_M+A(?;&=I+O M8Z*;*TIWYU+4.&-77U"--PK!0(U;+^S7D&HC&9>1$3`FUI_O5Y.?@\8HA&9& M!;Z714%9K(.`*K=\2?^C("M!B@\OG[%5RKF)2K$!.]`!RA MPNY[^*ZQP'S]<55LG`J>4$=OA)%3:3I_=A0PB1F`.L90="#[V4#^B/G6TVE@ MRU^G@16B]M8$V=VU;,:RN[R(RNXZ%/VI1,4ZD$5EQF7%_*M/?I:D6+(_QIP$,Y"[*@K_"M MU:;]VV"FP\Q!WQ+V"U,*%3R#I5*C6>'I\J[HCY<]HU8-NC$QH3]^#40KQ49< MP.A]9M.!E^7%^^)Y5*A193^ES#?CE]UKY"Q)D,W.K!F/Z_D;@30XJ@F:23!I M8^%+$8^P57P)BCU"J7^_D&(6W,*T8$(*C$N!RU)@U"/^_2W4P.`1%:C#'Y?$ MIM!O_ST^A61=T'YY7;!E)O1WSZTSTH4/\<:*N`S[4"%N1H$YQWVJ"K$,HKGO MHU$%/5@-@Q'V<6Q@W\5!3P-JLH134*DXR._R@-R'*U-$2\?W;1!%A?AI9NQ= MTI_/8;='$FPK;)SUK]_-ZU^WI_8/:6?U0T3;/V28U4]F)]V7QC>M4IM!?.86 M/--YXA;Y-D_">3I[AV+V58E!.B6PY^_[]^(+A05X5<+!SET<.CPF`&.N__Z5 M[KP@XBK$>["$PX!%^OR%W?QJM[CF%CR=\ULC#O;2P6%BUS.4.QL,2>]LT+<= M61%'-J,H.'R7C5[^/IJTAN"[P?@"\I?%*#L=2<_$\\`O8?=+%B?._R4S&O-J MV`I.W,S45107WXS&/EL*9&%J!/$1Y7:`$Z]MN,7E=")55M`UETHIL+!KS)]: M(;YU,[N"]G,\/Q:OA=3P62Y^.ZQ"/+L8!?QC>$8T6)G@RB+W\((KF]RCK8/1 M4''PG*L%N=$OE`1GEO[[_<1A!-ER<4EW>[)R^P1ZFX"X+EE/!W5U1!TZH;99 MG9<.7/:I/+`82RT&Z`8L817B:N"CWL*=A7Z9P2Y7*-W+I(N:*"ZZ&9F:=3/R M7+N8;H[E]7]0DF"MR\2[0#@RM&WYY:<\3XWR$5\PVQK,"LP*22I]YVFTEPNQ M@+])Q,N,A";H3GFE..^FA+-@U[F@ MIMNNR<0W*8-+^\GIJ)H,]F()_?=3D%35YQ:?![9[!Q^DBE(*_19:#+"U$6A5 M4.5XL9#!2IP`=T]8B)2JDZISI'PD]YRU/'OWC>'<,U97UN[T\&.IAS2%CGQ( M[+F\O'H$UM7L/;^:?#:">^K0!JR%TW?^BRI^?Y:>WU M^61=/I^DR^>I+G\21TRJ>JE%8-4B:P8CQ"'_2(WZ!\]RZ2 M=16/6SR)O"*B0FRC_:BVT#VO^!I05U!A2SGBBV_CAT=8KWCP)M3EAV]"Q1V4 M[RDFJO!`HLV5SROS.HKGJ2:=\TT.4=5]FM@K!Q6H\Y\6XK4W?(^Z("DG1[RP M4+;(\;+T*O7@I827KXF%L!.]4"@)M5(,)]?#OMD`IJM,H!OSI945XGZH4;B/ MATJ[W@_\7O[T0OZ.2B8.&H1R(SU&,*/"J6&DG^K3OSXH/**\WDR+J*B8Z">3 M\1-=*H_8-RB4YS293^#M.>3M/F1D01@Y`8,QE9/XMS?*>]U$'*847H/:W#-" MI0ET7)O/7C6/XS=4VJXS@128_LO#W69ZU'!9F*7!R$CT6GRO62KJ MH`QY3U,^"KN>Y>5#!$;!I"F_#!N>Y>67B$%3/4JJ#\.P:%R7B>L(S!JHOX+V MWR;WGSC/@EE((<[3T;QNMY8?W+1.<%ZS.P\'%UO(C^Q9:G4>"L'D^(@4."QK0C'?- MU%5($5:2P$#H)"\Y^UCC!QZ1`J?%WZ#&N`:$0":9*[^D?C&V=Z+S"_4J[P9E M5LVBTR+SABES"3M?;L1;Z*F"PTCL1KQZ#E$3L9OPFCI$S<1NQLOK$+40NP6O MJ4,TD]@SQ_HM&,TB]JRQ?AM&LXD]N_-,$&(YQ)XS]A:M-H_8\\;>HM6N(?8U M8V_1:FW$;AM[BU9;0.P%8V_1:HN)O7CL+5IM";&7C+U%JW43N[NNB]9;1NQE M'_?3USN0JB+V*DRI:6HKL6_%%$]3]^,[J7[Z_@I2.XA]!Z8LD'K,T7C(WH@I M&\U;3^SKY94DVV68M^,'<4&57P^4RP]3#2!GTCJ"]84M/6[U\"*^E[;=8++'*891"71"XITD_I> M:S04-*D"_2-D4J!W.X5O(Q;\$?E"Z6,%*_DU_-J` M^;&"M>HU!2M3`_,AIH)82D!+!%NKC8O.RC_XR16@W?^O.-5HM0=_A$^L??^1 M[I=&/I,WV>P]!OUL2(?+4`J7^!T=S]YGX5>WQOT^Q@\)3)!/R'M";C?_&K4A M@8DZ_>L%.7P=;$[J2,&U_)2L%"4K=4I6JI*E#?7K0B=TL"AOIX-R[4S!$2[S$!1PO8 MM=V",&:5RCE6$Q=3DE10DASP-D_A=QD5XD\&DK_+2+T6C?#X-?3.UVUO25*% M^'[&U/XHH46)G9R2<8$WM^U:L]]*JBP_&+P*(&X'YJ^6MNO:J,CAU$F7@$=6LUZL'L->_ M-F*O1V/Z%9^#^`X9=ZG)DQ#\/RPN_@R>;XX#ZBC>=`T7[_U+Z,3XWC\E$\L\ M_?/TM!053L:QF2G;J=P^QI,2'CW:8E/>W1W5/AB+Q_*KZ(WJ*ME=G-:6AD8* MZ#RQA#8<[:!;@$[J61Z5/4O1:$2WY&]_Q_WX@@2+@K*:@<^_T+8;;TCBD[W4 M?@-'Q7G,S>:Z1]R308]GV;]W.RW)_2\5'C=C-R_1H)5DOT8]%%]CZ;O`Z\I.#)XY!V3!^VC/G,?4A/M>> M;WUWW^@I/I=+6/^FKJ?@QW9RL??`,;_V*GN'T$BF,]-W#4?SPD" MM^(WGX$Y;#-./^D2RC-)^9!0GD7*+Z$'.8P5AZD'NEH\/@\)C)3`A`3@4)9? M+(6<;L@!AQM$.+Z\VM@U!K'`Q'+HJDL4G$/6_GW5L$CWMU4*U2."\Y+UY#Z7 M:M1ZLJW(@V_1LQ90AVTM908VSUD:YX@;]M`:X#2Z*)$C*7`1G`9IA;AP'GUG M+@6&Q._.0PT9C/O:RGR2[W\'=;EC8\[Y6O\ZGXK>&F)7AU9W5U_RB(%Y]-+7 M?)_*I_?II=4>L99B_'.ION(793]FMWF2[UQ?+80VZ[`;P9OB38&G(^7@#4-1 M/T^>!]C$/?4AR!S2TXTLIY&JU#E/3*-_D>O,M#=Q./ MY><=4@OY>403/X&;GC=VG@FKVQ8=>.>-@BHT+@4VA_::I,DD/U-P&$A^ M5O0:GTUJ,TAM.JE-*[7Q/LDMENIC'^-/SQ=Z@BY#[KE0NX[S7T.*LP6;CG[V MZ)^UQ2U^2X_SCPPFV]+I[BZDS^>2?F?!9IJ>5MZ+!73B]W[/[J#Y.*H,+7IV M`/$=-?TV6X]//SQ[\(-"DDJ#4-^5T/@B_:-?!QHILZNO#.;O8(6-6UMIXUZO M8#`*L+B2P=H$L)4GIZ\&KU3,C$Z!'OQDDA0:>]II:.K91T-S3P<-+3WT&]7" MS)X6&H)[0R.NLP+]V)($3POIM"CLBC;0LL$^@7Z)35S'A.!9O#,U.)TM'4RV MI8-)MG20VM+UZ0FVM,TPTWLXSKFXOSP*2T/"]Q-KYLKW<%;3W"/=P?,D>+@2 MB*1<<5TZ7L7IHY=N3I/@>=CRL4LW%T/!8Q0[0((7!-2L?#G]_0Y%&/^%8Z?55T%JS33W\H2<-/H!54Y#S<15.3I4N_(2[,BVY/ M_M:N)TSQ.;%\NS(.4VHX1,.LGL=IF*V4C)JN-I+*UE7YWL6CG-I0?7Y:1TV9 MB:$\8EB'&OTL?3ZG0ULQE'2F(]X%2AW[.,L<#CV\!N_P/FSD`FGL$ZW?=;VO M[SJ(SI%-_`X.PHF>#GJ3UA"QFVB?V;7`B'T-/9=7XRU>DH$T4,)/2Y!KE#)" M:L1&2V$D1XG0DAA9+V/HM[Q":B^6$FR&X<=0^KCC3BQN5"):)<*^94B56G+$ ME],FMZ%.;"/B8+PKKQ:4-Q,82&TY8GU:PEWXY*+X7D'YG1&/B&YC3'XW,OGI MJ/Q`BCJ\:DFE^)\`JT$>.]+_*@F2P>BJ\!?Q>BQU,J_DPJ3?&!+Q.TVPW?*/ MN3S/T1NT\O5^>OBQ>O;D]SE"N9;R"66N"5L-T!.;`2;Z1^BFA]K7:/QJM,RR M[:7.OU_KGX8QTWTN]#(!^_#YFG M?!_RT9S$WBCON>BAI,T8&M?K'WV4H\;\\7MMW%F`08#S`!<`=/?9N(L0#@%< M`A`!1@!&`2X#C`-,`'!`Q]^'%SZC=_AF15?X4J*W^]*CRWQSHK?XTJ*+??KH MC3YU]'K?;%@C4Z/S?7.C4<^HT%48/(;IV#AA7[H295NHB&?M3CX"7VNSD1YV6JZ^5:_1MV M]AZ8.,>5O:`]*S2N#FA_K('JHW/"H7%5(`7<'-J68,\.C6L"LP[^%.\71W60 MJ_:G0DK-ON^PYP"Y?SX,IQ:08.BBF&E@F0]E@E.9OH`!XH9`-[2$VM/5;89U^!1ZO`AP&.`(@`AP%.`;0!S``\TQG=H/]@,'EI2^]0><1,*([DG M*ZJA)ZWPV`WP$,!^@/\-\'6`1P`>`_@[@'\`^";`LP#/`_PCP'<`OEO]Y7KR MW52E)S8^[M>QO_!=6TH^Y%;^XAMW0=\^2!F`IMYS:3](.0N1#\\DW[/`WTRR M+CP/IF+_XHJ>`M@\2GE/BRU::HW0#<-UQL$H#GQ(?YQ&F]LW_`BU_AD7`&O- MP+Q]>[KYIUX`K%M:A;6(J[3,R]""$;B/9"`NE(+T6>3CI._N71-;Q"$8%](O MT)HH$P0T'AR4,X$YY'34'.YV%$2O"5=L$8\#80]2A?H*I!2,))X[8)_0;U%:9W"L:='87BG-%Z+C;.?B_+/5410*98A>AP/'?UPT%`I!2QN<01'')JVY^"'A_?F M1$O!)AQP=?7MVP@^OOA4ZE4;^KK2$"XV9E;;\ZGR*WHM:X.UEW!_GZX+76/^ MZX6O&:7=O+@J%5TS_"4H\B'6-!;X3=)^HC)VF!S,%@)Y0A5*2!>.B^5B_'.5M;FULM MS2W>IH:FG98=#6TMC35[9W/*.1:^O84>BG_`GS[JLH'ZX"\J"#1&NM:CU>_* M8Z>M7/U?[L=O1H^Q7X@!+0B8ZJ_)M^&O2`D"?1,6PG.$ M;KZC?X@7RE1KA`)5'NE:`]65EOK4=%]*=ZC2:_1[+/9\4.L1N_`P[E1H(+,^ MI4.-EVG,IR4)@@`$W5T%0.5FU(@[BYSFT(4I4WIM#:T)DV#??DHI-I\`!R_C M(OHX/Y_\RQBETFLV6L(6*P\_[X#^M"G M:GTG_@C$,/[*6CCTJ4[?]3_H=#@,9?^C_W-;6S^UM1^^37F/5@C./D(ECQ_! M4N'_%171#4<+K\'&;*89X<:N5PJQR:Y-`HAS8Y7"^')7)XOQSN MD\-Z.5PCARURV"Z'G!R:Y;!,#@OD,%L.L^30+X3D\ M*X>GY?""(@^YW!$Y?4GIKXP?E],Y!IN(?\1ZM=N0,>MU7!P&,/0+.XV#*/#2_LA3H;(64$+B[PTQ&Q1=&XX MU,?W7J+;L`EIL`=C(7!7AA4?/>-P'6`^E0+7YYX+;3@""8NI!4TTF#HY,3!SS`1^&/]BUXP4T\IKCJNJQN*857D`OI("D9PMHC? MH#_7"CS!G)#[5"*'Q7+8+H<[Y+!>#CODT"^'57*X3PY;Y+!1#N^7PZUR6,9" MV=PJH:>U5O[MX,V;BO"'>=M=-6T/)*/VQE$%E9M+"P#54-Y6L],[%=]<5V?? MU3`57[M]6C22-_C;IJ5/P`I*]]SC[VD+-NRS!U/T5+QG^9M\[<&:OT6 M?\U.)-EQ&` M^P;\R6[\J2*\3(*_YH6?0:;)]2`]OGQ$KT![M<83_EI:F[W)>WR>_=$2\[?C_[+;)CH.`C\@N*$I!;38N_L<#; MYK]O9?4DW9FB37NG8"RU4ZDLM7$Z-A86-AHQK&-SR>;24F>AI;;5_H"CN;&Y M-9:EJ(RLDU/Q3">GX*E*3L'*&CD==2(ZUF-_."ZO]F?@\0\`WXCCPH![''#& M!)SM68Y[`7#'$G"'`%?\Q`P&%&F![CF`PP!]`.P!9`'D` M!0!E`#L`V@$.`3P)\"+`$8#3`!`PP#&`TP#G M`42`RP`\^#Y&``M`-D`QP%:`>H!]`(<`7@`X#'`,X#3`18`1@`D`_FF.,P%D M`N0!V)Z.\^F&^%:`'0!^@`Z`0P!A@.<`7@0X#'`,X#3`(,!%`/'I_S\6_]W& M8@YG;X4]2&5#TX[F/67-9Y6U#^P%4KW)._.G^AB8_QQWG"AL:&TN]M1#_&5?D]3L:&\!,4L0' M-!UH;6MN=3>W<=R'F"YP<$,T]+8]X&]ND9OA1,05-M<&@(Y703S>''G?WZMOV&W5V&9.XFX MA/[]!--R'X8PKO`,Z=]BFI4L:6[:2>O\71Q'R_\AGB[SMOLIS4J.";2LV5/; MZO4V(>Y3E:>^>4^,#4ZO+FNM:6IKK/%[Y4Z`'JG+6W8`0J':PCE:O;&TLQTK MG\W9H9]-\DA2'?%Z02]R4;YE#;6PD`0PIY9SMC?X446\]!^+F,LYZIM;=W!_ MS\9Q=T.MUU'3T@9KL:-YU_:&)F_I3LKG:YCOKFGT^OU>9Y._M<';!OH".,BW M-[=S?1CW^)MK']B\W4)28L]T/HG(W M0W^P?^\H>)<7JJBE`_D!5P(,EC5C6T.XS[AW'Y%NS[E2KTUC0T/>64NL*Z_<*B*-4T[(5\'X]@(*:57Z7): M(5^`XVQ_`+TG[CJ,%VS$Q9.[41XCD"+,.&P*T+MJZ`\JWHITDR2RC.(:VKV- M%=QR&@=V<3K2FL$&(:YTLWL%Q!=-J1LF(NH7TJ",Y,EED>D*'/F<0X49FP,H MS%0NOZEA%V0D]'F53*N@ULEIE`2J0!J7WUK+X6U"AOIS?NK.-\ZB`&C06RV]5U>+4@OG/,,TJ*`[Z405M7KKA]1V,C%S9"6H$'O*U-WD8Y!_9968#[[PP[=S3(S)*Q]9#^ M6Z&VU0^5,'G$MT`>9XG34;;9?C<\I]_ZR#N=>J:;Z^2]3F(Y><.5A&+_A@M@ M"P!;YDPB3$1]WO:,5>C.!^HRI^7SN8,9&-N')95+8D[&?7ZC9?FECH*-I4"^ @6S+?\%NDXA8@!J``` ` end - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From SReddie@cybec.com.au Sun Dec 28 22:19:00 1997 From: SReddie@cybec.com.au (Steven Reddie) Date: Sun, 28 Dec 1997 22:19:00 -0000 Subject: beta18: building gdb gives incomplete import table??? Message-ID: <34A7414B.F9935C9E@cybec.com.au> I am attempting to build gdb with beta18. I have the "normal" source, not the cdksrc. [Is this my problem?] I have succeeded in compiling it, but not executing it. To get it to compile, I had to add "-D_MSC_VER -D__WIN32__ -DWIN32" to CFLAGS in Makefile. I also had to add four empty functions (I'll worry about their contents later) to gdb/win32-nat.c -- I think these were: void unix_to_dos_path( char*, char* ); void dos_to_unix_path( char*, char* ); void cygwin32_unix_to_dos_path( char*, char* ); void cygwin32_dos_to_unix_path( char*, char* ); gdb.exe is produced but crashes immediately. Debugging with the gdb supplied with beta18 shows me that the crash happens on a call to GetModuleHandle(). I think this was in the cygwin32 initialisation code (I don't have access to this machine at the moment so I can't check right now). I did a dumpbin (MSVC tool) and noticed that GetModuleHandle() was not listed in the kernel32 import table. I added a call to GetModuleHandle() [in unix_to_dos_path()] and the import table entry was then added by the linker. Executing it again crashed but seemed to make it slightly further, this time crashing on some other function that didn't have an import table entry either. I'm guessing that the linker is somehow screwing up the construction of the import table. Maybe the calls to these functions are in .a files that when linked-in don't bring their import declarations with them. If this has been fixed in a patch or someone knows of a work-around I would very much appreciate to hear about it. Sorry if this question has already been addressed -- I have only just returned to using the cygwin32 port after a length absence. Regards, Steven Reddie - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From colin@bird.fu.is.saga-u.ac.jp Sun Dec 28 23:24:00 1997 From: colin@bird.fu.is.saga-u.ac.jp (Colin Peters) Date: Sun, 28 Dec 1997 23:24:00 -0000 Subject: how to get source of resource compiler and ld that supports .res Message-ID: <01bd142a$85dbd300$fa173185@gbird0.fu.is.saga-u.ac.jp> From: vischne@ibm.net [original poster] >>> I could build gnu-win32 cross on FreeBSD2.2.2. >(gcc, as, ld etc) >But I cannot link my objects with .res file(resource object file). >How to get source of resource compiler and ld that supports .res file? >It causes error. > >I'd like to build cross development environment on FreeBSD and get >exe(executalbe file) on FreeBSD without using Windows. << [vischne@ibm.net] > >In theory, you convert .res files to .o files and link them to the >executable. In practise, these executables don't register under Windows. >What I mean is that an icon .res file doesn't show up as an icon when >the Windows file explorer displays the name of the executable. I don't find I have that problem, at least under Windows 95. I am using the latest version of windres and EGCS 1.0 from Mumit Khan, but I think it worked with rcl and res2coff as well. Is this an NT-only problem, or perhaps a problem with the tool set you are using to add the resources? Either way, the sources for ld and windres should be available from Mumit Khan's site ( http://www.xraylith.wisc.edu/~khan/software/gnu-win32/egcs.html ) somewhere. Colin. -- Colin Peters -- colin at fu.is.saga-u.ac.jp -- Saga University Dept. of Information Science -- http://www.fu.is.saga-u.ac.jp/~colin -- http://www.geocities.com/Tokyo/Towers/6162 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ivo.abeloos@skynet.be Mon Dec 29 01:37:00 1997 From: ivo.abeloos@skynet.be (Ivo Abeloos) Date: Mon, 29 Dec 1997 01:37:00 -0000 Subject: make problem Message-ID: <01bd143c$f19d67e0$620b0dc3@sky71189.skynet.be> Hi everybody, I have a little make problem. Consider the following file testfile.proto: ver XX.X ver XX.X When I perform a sed -e "" testfile.proto on it, it gives me the contents of the file on stdout, which is normal. When I try to do this from within a makefile, it refuses to do anything: testfile: testfile.proto echo 'Test the make' sed -e "" testfile.proto This is all, the echo does its thing but sed returns nothing! Anybody any suggestions? Thanx, Ivo - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From Axel.Riese@risc.uni-linz.ac.at Mon Dec 29 01:45:00 1997 From: Axel.Riese@risc.uni-linz.ac.at (Axel Riese) Date: Mon, 29 Dec 1997 01:45:00 -0000 Subject: strncat References: <199712232203.OAA26514@rtl.cygnus.com> Message-ID: <199712290944.AA14060@tiger.risc.uni-linz.ac.at> > > This has been fixed in more recent sources > -gjn > > Kees van Veen wrote: > > > > I have an application using strncat() which relies on NULL termination. > [...] > > -- > Geoffrey Noer But not in the coolview version available at Sergey's homepage... - Axel =========================================================================== Axel Riese Research Institute for Symbolic Computation J. Kepler University Linz A-4040 Linz e-Mail: Axel.Riese@risc.uni-linz.ac.at Austria URL: http://www.risc.uni-linz.ac.at/people/ariese/home/ =========================================================================== - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Mon Dec 29 02:27:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Mon, 29 Dec 1997 02:27:00 -0000 Subject: how to get source of resource compiler and ld that supports .res Message-ID: <199712291019.KAA15266@out4.ibm.net> > Colin Peters (colin@bird.fu.is.saga-u.ac.jp) > Mon, 29 Dec 1997 16:22:30 +0900 >[vischne@ibm.net] >> >>In theory, you convert .res files to .o files and link them to the >>executable. In practise, these executables don't register under Windows. >>What I mean is that an icon .res file doesn't show up as an icon when >>the Windows file explorer displays the name of the executable. > > >I don't find I have that problem, at least under Windows 95. I am using >the latest version of windres and EGCS 1.0 from Mumit Khan, but I think it >worked with rcl and res2coff as well. Is this an NT-only problem, or >perhaps a problem with the tool set you are using to add the resources? > I'd be more willing to believe you if you gave an example. As it is, I am using b18 with the newest versions of ld.exe, windres.exe and dlltool. And, unless the usual process of compiling and linking using gcc is wrong, you are lying. Comrade. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cvn@interchain.nl Mon Dec 29 02:27:00 1997 From: cvn@interchain.nl (Kees van Veen) Date: Mon, 29 Dec 1997 02:27:00 -0000 Subject: Error windows Message-ID: <34A77B3B.662C6024@interchain.nl> Hi, I sometimes get a window on the console of my NT server (4.0 SP3) telling me a DLL can't be found or something similar. Since I'm logged in using inetd/telnetd, I have to run to the console to click on a button in order to continue. Also. I sometimes get a window saying: The instruction at "....." referenced memory at "...." The memory location could not be "read". Click OK to terminate the application Click CANCEL to debug the application upon which I'm running down the hall to the console to click on the OK button. I'm sure the error is mine (though it's not a regular 'stack dump'), but I'd prefer to see the message on the screen I'm looking at. Does anyone know how I can redirect these messages to the screen? Thanks, Kees - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From boncz@wins.uva.nl Mon Dec 29 03:03:00 1997 From: boncz@wins.uva.nl (Peter Boncz) Date: Mon, 29 Dec 1997 03:03:00 -0000 Subject: No Subject Message-ID: <199712291103.MAA01533@carol.wins.uva.nl> This is a discussion between me and Bart Anderson (banders@ecd.rockwell.com) which I would like to bring back to the mailing list. The problem is compiling coolview on W95. Admittedly a crazy idea, but some people (including me) really want to do it. Problem #1 is that if you do a ./configure the following make chokes on the c++ files in the winsup/ directory. This problem I could resolve with a simple hack. My fix in the Makefile for the .cc files is in the .cc.o line. As a quick hack, I simply replaced the compiler variable by g++ the include variable by their actual values *minus* the -nostdinc Then on to Problem #2, which is more serious. It then compiles fine and produces a new-cygwin.dll. The trouble starts when you try to actually use it. All dependent programs that try to start fail with an error that there is no "r.dll". If you put a dummy r.dll in the directory, it goes on, but only to complain about a missing "|.dll". The most probable explanation is that the created cygwin.dll has a 'dependency dll section' or something like it, which is corrupted. It should be empty but it contains that characters 'r' and '|' (in my case, anyway). This may or may not have something to do with the seek trouble, signalled in the [Did _you_ ever try to compile..] thread on this list, also talked about earlier, in July 1997. In the mailing list archives I myself found a dllfix.exe program providid by a guy called Greathouse (july 1997). This program claims to fix the end part of my dll (it indeed says that mine wrong and claims to fix it), but it doesn't resolve my problem. Moreover, I inspected the end of the dll and it seems to contain routine names in the library, not a 'dependency section'. I could speculate that not only in the end, but also in other sections of the dll the 'garbage on seek past eof' occurs. If anybody knows anything, please help. I wish you a thread-safe 1998, Peter Boncz - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 29 05:28:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 29 Dec 1997 05:28:00 -0000 Subject: cygwin32 latest appears to be DOA (NBY) Message-ID: <19971229132731.452.qmail@hotmail.com> Hi Liz, The README file that you wanted to know about is at the ftp site. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Wed, 24 Dec 1997 16:07:39 -0800 >From: "B. Elizabeth Trojan" >To: gnu-win32@cygnus.com >Subject: cygwin32 latest appears to be DOA > >I just installed the latest cygnus stuff for win32 and it appears >to be DOA. When I run a command I get the following: > > >uname -a >(unknown) In cygwin_except_handler >(unknown) Error while dumping state (probably corrupted stack) >(unknown) In cygwin_except_handler >(unknown) Error while dumping state (probably corrupted stack) >(unknown) In cygwin_except_handler >. >. >. > > >Has anyone else run into this? > > > >-- >Liz Trojan >Synopsys (503)531-2297 Office >Beaverton (503)690-6906 FAX >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 29 05:44:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 29 Dec 1997 05:44:00 -0000 Subject: Installation and running (NBY) Message-ID: <19971229133808.4578.qmail@hotmail.com> Hi Vic, Let caution you that there are still some bugs in the compiler (hince, the BETA designation). If you are interested in GUI programming please review Colin Peters' and Mumit Khan's pages. Check the FAQ and README files found at the ftp site. Also, in your spare time, check out the mail archives. Sort by subject and look at the interesting ones. If the posts break down into discussions not on the subject, the go to the next subject. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >From: "victor" >To: >Subject: Installation and running >Date: Sat, 27 Dec 1997 12:01:01 -0500 > > >------=_NextPart_001_0048_01BD12BF.19952460 >Content-Type: text/plain; > charset="iso-8859-1" >Content-Transfer-Encoding: quoted-printable > >I followed the installation process as best I could. I was unsure of = >the exact syntax to use in the autoexec.bat Is it=20 > >SET GCC_EXEC_PREFIX=3DD:\gnuwin32\H-i386-cygwin32\tcl\bin > >I set the program up on the D drive and eliminated the b18 sub-folder. I = >did the same with the other lines. Do I have to do anything with the = >"mount" if so, what do I do? (In Windows 95 please) >Another problem I am having is the line when I launch base.exe is the = >request for /tmp. What do I do and where do I do it? In Windows95 you = >don't create a /tmp. I created a TMP folder and added a line = >TMP=3DJ:\tmp to the auto exec.bat. Please explain these things in a = >manner that people with no experience in programming and working in Win = >95 can understand. I have signed up for an online C++ class and they = >recommended your program. I want to have it set up before the class = >starts on the 5th. =20 >Vic Peters =20 >victor@tir.com > >------=_NextPart_001_0048_01BD12BF.19952460 >Content-Type: text/html; > charset="iso-8859-1" >Content-Transfer-Encoding: quoted-printable > > > > > >http-equiv=3DContent-Type>Untitled Normal Pagehref=3D"file://C:\Program Files\Common Files\Microsoft = >Shared\Stationery\"> > > >bgColor=3D#ffffff leftMargin=3D125> >
I followed the installation process as best I could.  I was = >unsure of=20 >the exact syntax to use in the autoexec.bat  Is it
>
 
>
SET GCC_EXEC_PREFIX=3DD:\gnuwin32\H-i386-cygwin32\tcl\bin
>
 
>
I set the program up on the D drive and eliminated the b18 = >sub-folder. I=20 >did the same with the other lines.  Do I have to do anything with = >the=20 >"mount"  if so, what do I do? (In Windows 95 = >please)
>
Another problem I am having is the line when I launch base.exe is = >the =20 >request for /tmp.   What do I do and where do I do it?  = >In=20 >Windows95 you don't create a /tmp.  I created a TMP folder and = >added a line=20 >TMP=3DJ:\tmp to the auto exec.bat.  Please explain these things in = >a manner=20 >that people with no experience in programming and working  in Win = >95 can=20 >understand.  I have signed up for an online C++ class and they = >recommended=20 >your program.  I want to have it set up before the class starts on = >the=20 >5th. 
>
Vic  Peters   
href=3D" mailto:victor@tir.com ">victor@tir.com
> >------=_NextPart_001_0048_01BD12BF.19952460-- > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 29 05:44:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 29 Dec 1997 05:44:00 -0000 Subject: Anybody ever gotten dvips (TeX) to work ? Message-ID: <19971229134414.21964.qmail@hotmail.com> Hi Jay, I don't have TeX working nor have I tried it as yet. However, make certain that your font input files don't contain the annoying \r\n combination. Are you directories mounted as "text=binary"? - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Sat, 27 Dec 1997 21:42:26 -0800 (PST) >From: Jason Alan Nordwick >To: gnu-win32@cygnus.com >Subject: Anybody ever gotten dvips (TeX) to work ? > > >I have TeX/LaTeX working, generating dvis, and yap can view >them. But when I run dvips, I get some font errors. I installed >with all the defaults. Can anybody please help me, I really need >to get this working soon. Here is the result of running dvips >on a dvi file generated from LaTeX (MiKTeX). > >Also, what do pdftex and pdflatex do ? I thought that they might >generate pdf files, but they still give me a dvi file and dvips >chokes on. > > >C:\jay\latex>dvips pp.dvi >This is dvips 5.74 (MiKTeX 1.09) Copyright 1997 Radical Eye Software (www.radicaleye.com) >' TeX output 1997.12.21:1142' -> pp.ps >- make_pk_font (cmr17, 600, 600, lwpro) >Creating cmr17.pk... >c:\texmf\miktex\bin\virmf.exe \mode:=lwpro; \mag:=magstep(0.0); batchmode; input cmr17 >This is METAFONT, Version 2.718 (MiKTeX 1.09) > >Cannot access GF file cmr17.600gf! >C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmr17.600pk not found, characters will be left blank. >- make_pk_font (cmr12, 600, 600, lwpro) >C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmr12.600pk not found, characters will be left blank. >- make_pk_font (cmbx12, 720, 600, lwpro) >C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmbx12.720pk not found, characters will be left blank. >- make_pk_font (cmr10, 600, 600, lwpro) >C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmr10.600pk not found, characters will be left blank. >- make_pk_font (cmsy10, 600, 600, lwpro) >C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmsy10.600pk not found, characters will be left blank. >- make_pk_font (cmbx10, 600, 600, lwpro) >C:\TEXMF\MIKTEX\BIN\DVIPS.EXE: Font cmbx10.600pk not found, characters will be left blank. >. [1] [2] [3] > > >jay >-- >Join the FreeBSD Revolution. >Support the FSF, buy GNU. > http://xcf.berkeley.edu/ >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 29 05:54:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 29 Dec 1997 05:54:00 -0000 Subject: undefined reference: gethostbyaddr, etc... (NBY) Message-ID: <19971229135351.8133.qmail@hotmail.com> Hi Tim, Have you installed Sergey's patches yet? I don't remember if I've sent you this list of URL's yet or not so please forgive me if you have already received these. Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Sun, 28 Dec 1997 23:15:54 -0500 >To: gnu-win32@cygnus.com >From: Tim Arheit >Subject: undefined reference: gethostbyaddr, etc... > >I'm using b18 and am attempting to port a program. Compiling >the parts of the program seems to work fine (I even had to grab >and compile bison and it work just great), however when it gets >to linking the program it reports several undefined references. >Specifically gethostbyaddr, gethostbyname, srand48, drand48, >tzset, major, minor and ualarm. > >I couldn't find these in any library (are they simply not >implemented yet?) A search did show all the mentioned function >listed in a file named SYSCALLS.c.X. > >Any ideas? > >Please forgive my ignorance if this is simple... it's the >first time I've used this variant of GCC ;) > >-Tim > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 29 06:14:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 29 Dec 1997 06:14:00 -0000 Subject: compiling coolview on W95. Message-ID: <19971229141412.3270.qmail@hotmail.com> Check out Mumit Khan's page for a newer version of dllfix. The problem with W95 is that the file cluster is not clean. Garbage exists beyond the end of file. This causes problems with binary mode opens because you end up with a physical end of file and not a logical end of file. This means that the garbage remaining in the file cluster is being read as part of the DLL. This problem doesn't exist on NT. Check out Sergey's page for more info. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Date: Mon, 29 Dec 1997 12:03:48 +0100 (MET) >From: boncz@wins.uva.nl (Peter Boncz) >To: gnu-win32@cygnus.com > >This is a discussion between me and Bart Anderson (banders@ecd.rockwell.com) >which I would like to bring back to the mailing list. > >The problem is compiling coolview on W95. Admittedly a crazy >idea, but some people (including me) really want to do it. > >Problem #1 is that if you do a ./configure the following make chokes >on the c++ files in the winsup/ directory. This problem I could resolve >with a simple hack. > >My fix in the Makefile for the .cc files is in the .cc.o line. >As a quick hack, I simply replaced the compiler variable by g++ >the include variable by their actual values *minus* the -nostdinc > >Then on to Problem #2, which is more serious. > >It then compiles fine and produces a new-cygwin.dll. The trouble starts >when you try to actually use it. All dependent programs that try to start >fail with an error that there is no "r.dll". >If you put a dummy r.dll in the directory, it goes on, but >only to complain about a missing "|.dll". > >The most probable explanation is that the created cygwin.dll has >a 'dependency dll section' or something like it, which is >corrupted. It should be empty but it contains that characters >'r' and '|' (in my case, anyway). > >This may or may not have something to do with the seek trouble, >signalled in the [Did _you_ ever try to compile..] thread on >this list, also talked about earlier, in July 1997. > >In the mailing list archives I myself found a dllfix.exe program >providid by a guy called Greathouse (july 1997). >This program claims to fix the end part of my dll (it indeed says >that mine wrong and claims to fix it), but it doesn't resolve my problem. > >Moreover, I inspected the end of the dll and it seems to contain >routine names in the library, not a 'dependency section'. >I could speculate that not only in the end, but also in other >sections of the dll the 'garbage on seek past eof' occurs. > >If anybody knows anything, please help. > >I wish you a thread-safe 1998, > >Peter Boncz >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Mon Dec 29 08:16:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Mon, 29 Dec 1997 08:16:00 -0000 Subject: make problem (NBY) Message-ID: <19971229161601.18879.qmail@hotmail.com> >Reply-To: "Ivo Abeloos" <@skynet.be> >From: "Ivo Abeloos" >To: "gnuwin32 (E-mail)" >Subject: make problem >Date: Mon, 29 Dec 1997 10:34:21 +0100 > >Hi everybody, > >I have a little make problem. Consider the following file testfile.proto: > >ver XX.X >ver XX.X > >When I perform a sed -e "" testfile.proto on it, it gives me the contents of >the file on stdout, which is normal. When I try to do this from within a >makefile, it refuses to do anything: > >testfile: testfile.proto > echo 'Test the make' > sed -e "" testfile.proto > testfile: testfile.proto; \ echo 'Test the make'; \ sed -e "" testfile.proto This worked for me on NT 3.51. Original CDK.EXE install (No Sergey Patches). I will try this on W95 later tonight. It also would only work this way on AIX unix. >This is all, the echo does its thing but sed returns nothing! > >Anybody any suggestions? > >Thanx, > >Ivo > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > Let me point you to: ftp://ftp.cygnus.com/pub/gnu-win32/latest/ (ftp site) http://www.cygnus.com/pubs/gnupro/ (Comercial Page) http://www.cygnus.com/misc/gnu-win32/ (Project Page) http://www.cygnus.com/ml/gnu-win32 (Mail Archives) http://www.itribe.net/virtunix/winhelp-man-pages/ (HTML Man Pages) http://www.lexa.ru/sos (Sergey Okhapkin) http://www.fu.is.saga-u.ac.jp/~colin/gcc.html (Colin Peters) http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ (Mumit Khan) - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 29 08:26:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 29 Dec 1997 08:26:00 -0000 Subject: compiling coolview on W95. References: <19971229141412.3270.qmail@hotmail.com> Message-ID: <9712291625.AA24017@modi.xraylith.wisc.edu> "Earnie Boyd" writes: > Check out Mumit Khan's page for a newer version of dllfix. The problem > with W95 is that the file cluster is not clean. Garbage exists beyond > the end of file. This causes problems with binary mode opens because > you end up with a physical end of file and not a logical end of file. > This means that the garbage remaining in the file cluster is being read > as part of the DLL. This problem doesn't exist on NT. Check out > Sergey's page for more info. Actually, the version of dllfix.exe on my web page is not a newer version, it's something I picked up from the mailing list. I'll update the verbiage to make sure there is no confusion. Has anybody built DLLs on W95 using egcs and encountered the same problem? I've built a few test ones and it seems to work, but my test cases were trivial at best and it might not show the problem. I'll try building coolview on W95 using egcs and let the mailing list know if I have similar problems or not. Regards, Mumit -- khan@xraylith.wisc.edu Gnu-Win32 EGCS: http://www.xraylith.wisc.edu/~khan/software/gnu-win32/ ftp://ftp.xraylith.wisc.edu/pub/khan/gnu-win32/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From PATRIK@Oslo4GW.Norway.NCR.COM Mon Dec 29 10:09:00 1997 From: PATRIK@Oslo4GW.Norway.NCR.COM (Patrik Wennberg - NCR) Date: Mon, 29 Dec 1997 10:09:00 -0000 Subject: looking for cygnus compatible perl (cygwin32 NT4.0 NTFS - problem - part #2) Message-ID: <34A7E700@Oslo4GW.Norway.NCR.COM> There is one thing that fails brutally (actually during the install part) when trying to compile the native perl5.004_04 under cygwin32-b18, NT4.0 and NTFS: Perl s -w test. Please see my previous mail, dated Mon, 01 Dec 97 09:29:00 MET for a full description of the "problem". I ve tried to point out this situation once before but I didn t get any response to it at. Anyone has any ideas about it now? Regards Patrik Wennberg _______________________________________________________________ | | Patrik Ericson Wennberg | NCR Norway, PS Division |--- | Email: Patrik.Wennberg@Norway.NCR.COM |--- | Phone: +47-22953770 | Cellphone (inc. SMS): +47-91321474 | MobileFax: +47-91240069 |--- | NCR Norge A.S. | Box 24, N-0409 OSLO | NORWAY |______________________________________________________________ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jeffdbREMOVETHIS@netzone.com Mon Dec 29 10:09:00 1997 From: jeffdbREMOVETHIS@netzone.com (Mikey) Date: Mon, 29 Dec 1997 10:09:00 -0000 Subject: compiling coolview on W95. References: <19971229141412.3270.qmail@hotmail.com> Message-ID: <34a7d34f.159010409@smtp.netzone.com> In a copy of your linker script, .../H-i386-cygwin32/i386-cygwin32/lib/ldscripts/i386pe.x replace the .idata{} section with .idata BLOCK(__section_alignment__) : { /* This cannot currently be handled with grouped sections. See pe.em:sort_sections. */ *(.idata$2) *(.idata$3) /* These zeroes mark the end of the import list. */ LONG (0); LONG (0); LONG (0); LONG (0); LONG (0); *(.idata$4) *(.idata$5) *(.idata$6) *(.idata$7) } and replace the .reloc{} section with .reloc BLOCK(__section_alignment__) : { *(.reloc) /* These zeroes mark the end of the reloc section. */ LONG (0); LONG (0); LONG (0); LONG (0); LONG (0); } name the copy i386pe.xfixed or something similar and use -T i386pe.xfixed on your link line to use that linker script. or you could download and install B18-patched-link-compat-ld.tgz available at ftp://ftp.deninc.com/pub/sdk which has these and several other fixes, (end of blatant plug ;^) but just these 2 fixes will eliminate the need for dllfix.exe and for fixup.o. to add an .idata$3 section to dll's. The first change has been made to current sources, and that is why your dll import list (.idata$2) isn't terminated properly. and the dllfix.exe problem does exist on some versions of NT according to comments posted to the list. Why you are having problems with gcc <-> g++ I have no Idea, have you make any changes to your specs file? On Mon, 29 Dec 1997 06:14:11 PST, you wrote: >Check out Mumit Khan's page for a newer version of dllfix. The problem >with W95 is that the file cluster is not clean. Garbage exists beyond >the end of file. This causes problems with binary mode opens because >you end up with a physical end of file and not a logical end of file. >This means that the garbage remaining in the file cluster is being read >as part of the DLL. This problem doesn't exist on NT. Check out >Sergey's page for more info. > >- \\||// >---o0O0--Earnie--0O0o---- >-earnie_boyd@hotmail.com- >------ooo0O--O0ooo------- > >>Date: Mon, 29 Dec 1997 12:03:48 +0100 (MET) >>From: boncz@wins.uva.nl (Peter Boncz) >>To: gnu-win32@cygnus.com >> >>This is a discussion between me and Bart Anderson >(banders@ecd.rockwell.com) >>which I would like to bring back to the mailing list. >> >>The problem is compiling coolview on W95. Admittedly a crazy >>idea, but some people (including me) really want to do it. >> >>Problem #1 is that if you do a ./configure the following make chokes >>on the c++ files in the winsup/ directory. This problem I could resolve >>with a simple hack. >> >>My fix in the Makefile for the .cc files is in the .cc.o line. >>As a quick hack, I simply replaced the compiler variable by g++ >>the include variable by their actual values *minus* the -nostdinc >> >>Then on to Problem #2, which is more serious. >> >>It then compiles fine and produces a new-cygwin.dll. The trouble starts >>when you try to actually use it. All dependent programs that try to >start >>fail with an error that there is no "r.dll". >>If you put a dummy r.dll in the directory, it goes on, but >>only to complain about a missing "|.dll". >> >>The most probable explanation is that the created cygwin.dll has >>a 'dependency dll section' or something like it, which is >>corrupted. It should be empty but it contains that characters >>'r' and '|' (in my case, anyway). >> >>This may or may not have something to do with the seek trouble, >>signalled in the [Did _you_ ever try to compile..] thread on >>this list, also talked about earlier, in July 1997. >> >>In the mailing list archives I myself found a dllfix.exe program >>providid by a guy called Greathouse (july 1997). >>This program claims to fix the end part of my dll (it indeed says >>that mine wrong and claims to fix it), but it doesn't resolve my >problem. >> >>Moreover, I inspected the end of the dll and it seems to contain >>routine names in the library, not a 'dependency section'. >>I could speculate that not only in the end, but also in other >>sections of the dll the 'garbage on seek past eof' occurs. >> >>If anybody knows anything, please help. >> >>I wish you a thread-safe 1998, >> >>Peter Boncz >>- >>For help on using this list (especially unsubscribing), send a message >to >>"gnu-win32-request@cygnus.com" with one line of text: "help". >> > > > >______________________________________________________ >Get Your Private, Free Email at http://www.hotmail.com >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". Linux a platform built by, and for users, standing on the firm legs of reliability, and speed. Microsoft Windows, a platform without a leg to stand on. (jeffdbREMOVETHIS@netzone.com) delete REMOVETHIS from the above to reply Mikey - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From derek@biddersedge.com Mon Dec 29 11:34:00 1997 From: derek@biddersedge.com (Derek Young) Date: Mon, 29 Dec 1997 11:34:00 -0000 Subject: login/passwd file Message-ID: <34A7F8AF.7DD96E28@biddersedge.com> I have not been able to get login from Sergey's page working. I downloaded crypt.exe from his page as well, and created an /etc/passwd file. Whenever I login I get "Login incorrect". At first I thought this was a problem with telnet, but even just straight login fails (thus causing the problem with telnet). I've tried setting up a passwd file with a null password like this: dyoung::1000:0:Derek Young:/derek:/usr/gnuwin32/b18/H-i386-cygwin32/bin/bash.exe I've also tried setting one up with a valid password: dyoung:$1$11$r8tCuNX00ZGmGVaCftwTe1:1000:0:Derek Young:/derek:/usr/gnuwin32/b18/H-i386-cygwin32/bin/bash.exe both passwords have the same effect. The shell executable is was changed from /bin/sh.exe, but this doesn't seem to matter. I'm trying to run CVS which uses the same password authentication, and this doesn't authenticate either. Am I doing something wrong? Do I need to do more set up? -- derek - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From kasper@altisimo.com Mon Dec 29 11:34:00 1997 From: kasper@altisimo.com (Richard Kasperowski) Date: Mon, 29 Dec 1997 11:34:00 -0000 Subject: GNU indent binary available References: <34A19D48.5313AFDA@equator.com> Message-ID: <34A7FB0F.17A1D1D7@altisimo.com> A binary of GNU indent 1.9.1, compiled on NT 4.0 for GNU-Win32 b18, is available at URL " http://www.altisimo.com/kasperowski.html ". -- Richard Kasperowski Altisimo Computing http://www.altisimo.com/ President Cambridge, MA, USA kasper@altisimo.com Tel: (617)576-1552 Fax: (617)576-2441 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mc4420@mclink.it Mon Dec 29 12:35:00 1997 From: mc4420@mclink.it (Giandomenico De Sanctis) Date: Mon, 29 Dec 1997 12:35:00 -0000 Subject: _G_config.h C++ header Message-ID: <199712292034.VAA18173@ammi.mclink.it> Where is it possible to unload the _G_config.h C++ header? It is not present in GNU-Win32 distribution files. <[GiDeSa]> - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From newsham@lava.net Mon Dec 29 14:35:00 1997 From: newsham@lava.net (Tim Newsham) Date: Mon, 29 Dec 1997 14:35:00 -0000 Subject: spawn and wait? Message-ID: Hi, I'm not too clear on the semantics of spawn since I haven't come across any documentation for it. I assume its a way to spawn a process without the fork/exec expenses (using just 1 CreateProcess instead of two). So I ran some simple tests and found what I think might be a bug: #include main() { int stat; stat = spawnl(_P_NOWAIT, "/gnuwin32/b18/H-i386-cygwin32/bin/sleep.exe", "sleep", "5", 0); printf("spawned (%d). waiting\n", stat); wait(&stat); printf("status %x\n", stat); } when I run this I get: spawned (40). waiting [pause] wait_found: CloseHandle thread failed status 0 I tried other spawn flags and I seem to get that error when I use NOWAIT or NOWAITO. Cygwin version is based off of a sept snapshot with some local changes (nothing involving spawn or wait). Tim N. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ivo.abeloos@skynet.be Mon Dec 29 14:35:00 1997 From: ivo.abeloos@skynet.be (Ivo Abeloos) Date: Mon, 29 Dec 1997 14:35:00 -0000 Subject: Fw: make problem (ctd.) Message-ID: <01bd14a9$7c39c030$2e0b0dc3@sky71189.skynet.be> Hi Ernie, Thanx for the ultra fast reply! I still have problems but I can narrow it down a bit. I'm on NT4.0 by the way... I have the following makefile: testfile: testfile.proto @(echo Start;\ sed -e "" testfile.proto;\ echo Done;\ ) When I open the bash shell and run the make, it works, it prints the contents of the testfile.proto file as it should. Running make twice in the same shell only gives Start and Done. I really don't have a clue... I play with it once more but if you know of a solution fast, let me know... Thanx, Ivo - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From banders@ECD.Rockwell.COM Mon Dec 29 15:44:00 1997 From: banders@ECD.Rockwell.COM (Bartlee A. Anderson) Date: Mon, 29 Dec 1997 15:44:00 -0000 Subject: Who's got source to windres? Message-ID: <34A83436.23F4@switch.rockwell.com> Getting back to Vischne's problem with locating the sources... Where is the source for windres? ... hidden in a snapshot? And Mumit where's the source for your egcs? ... build instructions? You don't have to answer till after you've fixed the recompilation of coolview with egcs. ;) Bart -- Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) Rockwell International Electronic Commerce Division 300 Bauman Ct. banders@ecd.rockwell.com MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 Wood Dale, IL 60191 FAX (630) 227-9771 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From SReddie@cybec.com.au Mon Dec 29 15:44:00 1997 From: SReddie@cybec.com.au (Steven Reddie) Date: Mon, 29 Dec 1997 15:44:00 -0000 Subject: beta18: building gdb gives incomplete import table [and other problems]??? References: <34A7414B.F9935C9E@cybec.com.au> Message-ID: <34A82DA4.742F3EDD@cybec.com.au> Further to yesterdays post, I have succeeded in getting gdb to compile and run but I am still confused about what's going on. The linker still seems faulty. The full steps are below. Note that this is using the "normal" 4.16 distribution of the gdb source. I first do a standard ./configure. To get it to compile, I had to add "-D_MSC_VER -D__WIN32__ -DWIN32" to CFLAGS in Makefile. I also had to add four empty functions (I'll worry about their contents later) to gdb/win32-nat.c -- I think these were: void unix_to_dos_path( char*, char* ); void dos_to_unix_path( char*, char* ); void cygwin32_unix_to_dos_path_keep_rel( char*, char* ); void cygwin32_dos_to_unix_path_keep_rel( char*, char* ); This would successfully compile but crash immediately when executed. The crash was on the call to GetModuleHandle() which is called from cygwin_crt0. To fix this crash, I removed the -lkernel32 from the gdb/Makefile and re-linked. It appears that the -lkernel32 somehow looses the import table from libcygwin32.a. Next, it crashes in ScreenRows() and ScreenCols(). I replaced the calls to these functions in readline/readline.c and gdb/utils.c with 24 and 80 respectively. I haven't yet fully investigated what the cause of this is. After recompiling these files and re-linking, it works great. I hard-coded unix_to_dos_path() to return "F:\gdb-4.16\gdb\test.exe" and it actually runs my test program correctly. I'm guessing that the linker is somehow screwing up the construction of the import table. Maybe the calls to these functions are in .a files that when linked-in don't bring their import declarations with them. If this has been fixed in a patch or someone knows of a work-around I would very much appreciate to hear about it. Additional info for bug report: Unpatched cygwin32 beta 18 distribution. Running under NT 4.0. Building gdb 4.16 with "./configure". Regards, Steven Reddie - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From wvd_vegt@knoware.nl Mon Dec 29 15:44:00 1997 From: wvd_vegt@knoware.nl (Wim van der Vegt) Date: Mon, 29 Dec 1997 15:44:00 -0000 Subject: Problem with patching unix files with patch b18 and earlier Message-ID: <000101bd14a1$250f6ce0$9e55b2c2@veg> Hi, I used gnu patch for some time now to patch sources of a unix based mud that can also be compiled under win32. What i noticed was that : 1. patch only likes files in unix format (ie lf only) for both file to be patched and the patch itself. It then patches correctly and signals plan A. If the file to be patched is saved in dos mode (ie crlf terminated) patch signals plan B and ruins the file. (i just use a simple commandline patch ). 2. if patch patches a unix file it returns a crlf (ie dos) file as a result that can't be patched again without first saving it again in unix format. I investigated this earlier for some other msvc based program and found that no matter what you write to stdout, win95 and winnt always expands a lf (\n) to crlf (\r\n). To correct this i found the following (msvc) code allows a console application (as the gnu tool is) to emit both \r and \n as they were intended. int result = _setmode( _fileno( stdout ), _O_BINARY ); if( result == -1 ) fprintf(stderr, "Cannot set mode\n" ); Any comments, tips or hints? ---------- Wim van der Vegt / Dutch Open University - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hgfernan@usp.br Mon Dec 29 20:04:00 1997 From: hgfernan@usp.br (hgfernan@usp.br) Date: Mon, 29 Dec 1997 20:04:00 -0000 Subject: Compiling "V" 1.16 - Specific Instructions? References: Message-ID: <9712300407.AA37836@spider.uspnet.usp.br> On 26 Dec 97 at 12:22, Brendan Simon wrote: > > > >> Please be warned that V is mainly a Unix GUI library; its > >> support to Windows 95/NT is still to be done. People wanting to > > program for Win95/NT should take a look at wxWindows. > >> Maybe wxWindows should be added to Cygnus page. > > > THIS IS DEFINATELY INCORRECT !!! > > V has both X Windows AND MS-Windows support. Hi, Brendan! I'm sorry if i was incorrect. I appreciate the V library and its community, and don't want to be unfair with any of them. Internet and computing in general changes very quickly, so it's very easy to have information no longer current. Maybe you can help enlighten me and this mailing list about V making the necessary corrections to what i'm going to relate. I first heard about V in the famous paper by Prof. Wampler, V's creator, published in C/C++ Users Journal (Aug/96). Prof. Wampler related that he designed a multiplatform class library and was using it to teach his students about object-oriented programming and interface design. I'm a C++ instructor and i'm constantly embarrassed by the "good" o-o design principles followed in the construction of MFC. So i was very interested when i read Prof. Wampler's paper. But prudence advised me to take some time before jumping into V and preparing classes to my students. To that purpose i subscribed V mailing list and read its messages for a while. >From what i could see, in that times (circa 1996) several people were having problems with V's window implementation. I'm not saying that V didn't compile, but people related that in some cases V didn't work correctly under Windows. Then in several messages, experienced users advised beginners that they should consider V a Unix-oriented class library. This wasn't emphatically contested then. I thought that was still the case when i read V's page: they say that Windows support is still missing. Maybe that was just a misunderstood, but i'd like to ask your opinion about it. Thank you for the attention, ++Hilton ------------------- Hilton Fernandes email: hgfernan@usp.br www: http://www.lsi.usp.br/~hilton.html (inactive) M. Sc. Student of Parallel Distributed Applications at Escola Politecnica (Polytechic School) University of S. Paulo - Brazil - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hgfernan@usp.br Mon Dec 29 20:05:00 1997 From: hgfernan@usp.br (hgfernan@usp.br) Date: Mon, 29 Dec 1997 20:05:00 -0000 Subject: Diff of executable files References: <9712242146.AA38180@spider.uspnet.usp.br> <199712251058.CAA23266@graft.XCF.Berkeley.EDU> Message-ID: <9712300407.AA23718@spider.uspnet.usp.br> On 25 Dec 97 at 2:58, Jason Alan Nordwick wrote: > There is a fairly good one for being alpha, called Xdelta. > Look for it at htp://www.xcf.berkeley.edu/~jmacd > Nordwick offers a link to a utility that he created, called Xdelta, an improvement of diff that will deal better with binary files. >From what i can read in its site, it is a welcome idea. However, this is not what this thread discusses: a diff for executables. That would be a useful tool. For a program that knows the structure of an executable file, or a library file would be far useful than a binary utility, like cmp, that equals any file format to an unstructured sequence of bytes. For instance, it could understand that an executable library files have sections; when comparing two such files, it would compare between section boundaries with the same name. That could even more useful in ELF's DWARF debugging format, where an executable file is marked by tags of several types. The idea of a diff program specialized in executables can be easily understood by an analogy: diff, a text utility is line oriented and doesn't know about the syntax of a program source: for instance, in the diff versions i have access to the next two programs are considered different: if (max < j) { max = j; } and if (max < j) { max = j; } In other words, diff doesn't know about the C language syntax. (Nor does Nordwick's Xdelta.) diff is usually very useful, but a syntax differencer that would know about C syntax would be even more useful. That way, one could compare two programs independently of its indentation, comments etc. > Amazing how COFF is called "new" What i tried to say is that COFF is newer in a Microsoft context, and PE is older. But... hey, don't be picky! :-) > I have been looking for a C API, but cant find one. I can only find C++ > level API documentation. (For the win32 subsystem) There must be a misunderstood here: Win32 API is a C API; in its beginnings (circa 1985), Windows API (then Win16) could only be programmed in C. If you want a link to Win32 API, it is "Introduction to the Win32 Application Programming Interface", at http://www.microsoft.com/WIN32DEV/APIEXT/INTRO32.HTM Maybe i'm being "unpolite" when i quote Microsoft's site pages. Well, people, i hate Microsoft just like the other guy; i don't use Office 97, i prefer TeX etc. However, i can't deny that i use Win95 / NT, like everyone else in this list. Like everyone else in this list, i'd like to use a better OS, preferably an Un*x-like one. For professional reasons we have to use M$'s Win32. But sometimes even M$ can do good things. :-) I'd like to talk more about this dilemma, but i'll leave this to other day, other people. > > jay > -- > Join the FreeBSD Revolution. > Support the FSF, buy GNU. > http://xcf.berkeley.edu/ > > Happy new year, everybody, ++Hilton ------------------- Hilton Fernandes email: hgfernan@usp.br www: http://www.lsi.usp.br/~hilton.html (inactive) M. Sc. Student of Parallel Distributed Applications at Escola Politecnica (Polytechic School) University of S. Paulo - Brazil - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From arlindo@niteroi.gsfc.nasa.gov Mon Dec 29 21:10:00 1997 From: arlindo@niteroi.gsfc.nasa.gov (Arlindo da Silva) Date: Mon, 29 Dec 1997 21:10:00 -0000 Subject: ntxlib works on win95 References: <199712290008.AAA54454@out5.ibm.net> Message-ID: <199712300513.AAA10528@woodmore.gsfc.nasa.gov> > > Unfortunately, ntxlib's libX11.a (the simulated version) compiles with few > problems, but the sample test program causes the following error message > from Windows 95: > > TST caused an invalid page fault in > module at 0000:8b26ddea. > ... In the makefile you have: $(cc) -o tst $(OBJECTS) -luser32 -lkernel32 -lgdi32 Remove "-lkernel32" and all works well on win95. -- Arlindo da Silva dasilva@alum.mit.edu - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 29 21:39:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 29 Dec 1997 21:39:00 -0000 Subject: Who's got source to windres? References: <34A83436.23F4@switch.rockwell.com> Message-ID: <9712300539.AA24626@modi.xraylith.wisc.edu> "Bartlee A. Anderson" writes: > Where is the source for windres? ... hidden in a snapshot? Yes, it is. I can't share the binutils snapshot sources, but I'm quite sure that the binutils maintainer(s) will give it out to folks who are willing to deal with snapshots (eg., be prepared to deal with bugs in snapshots, test/debug and possibly solve the bugs, etc). Ian Taylor , who I believe subscribes to this list, is the person you want to talk to. > > And Mumit where's the source for your egcs? ... build instructions? Check the egcs home page at http://www.cygnus.com/egcs/ . If you want to build egcs-1.00 release, then all you need is the patch on my ftp site ftp://ftp.xraylith.wisc.edu/pub/khan/gnu-win32/cygwin32\ /releases/patches/ to build from source. EGCS 971225 and will build w/out any patches. I have never built this thing natively (ie., on a machine running 95/NT), but Sergey has done it, so it does work. However, to build this beast, you do need a binutils snapshot (971023 or newer), otherwise you're out of luck. Until you get binutils snapshot sources, you can use my binary release to build newer releases of egcs, which is what I recommend. build instructions? You're joking, right ;-) It's actually quite simple, if you read the install docs that comes with egcs and also the FAQ at the egcs web site. No different than building egcs/gcc on other platforms. > You don't have to answer till after you've fixed the recompilation of > coolview with egcs. ;) > I believe Mikey's response answered it already. Did putting the ASM hack back (I believed it's ifdef'd out with OLDWAY in libccrt0.cc and dcrt0.cc) solve your problem? Of course, you can always get an Unix box and just cross-build everything instead of diddling with 95/NT ;-) Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 29 22:32:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 29 Dec 1997 22:32:00 -0000 Subject: _G_config.h C++ header References: <199712292034.VAA18173@ammi.mclink.it> Message-ID: <9712300632.AA24699@modi.xraylith.wisc.edu> "Giandomenico De Sanctis" writes: > Where is it possible to unload the _G_config.h C++ header? > It is not present in GNU-Win32 distribution files. > - _G_config.h is located in /H-i386-cygwin32/i386-cygwin32/include where is the root directory of the gnu-win32 installation (eg., c:/gnuwin32/b18). Regards, Mumit Cc: Gnu-Win32 mailing list, Giandomenico De Sanctis - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 29 22:46:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 29 Dec 1997 22:46:00 -0000 Subject: spawn and wait? References: Message-ID: <9712300646.AA24720@modi.xraylith.wisc.edu> Hi Tim, it's not a bug. Try the following: #include #include #include int main() { int stat, pid; pid = spawnl(_P_NOWAIT, "/gnuwin32/b18/H-i386-cygwin32/bin/sleep.exe", "sleep", "5", 0); printf("spawned (%d). waiting\n", stat); cwait(&stat, pid, WAIT_CHILD); printf("status %x\n", stat); exit (0); } and you should something like the following (if the spwan is successful): spawned (xx). waiting status 0 where "xx" is the pid. fyi, wait goes with [fork]+exec*, and cwait goes with spawn*. Regards, Mumit - khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Mon Dec 29 23:08:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Mon, 29 Dec 1997 23:08:00 -0000 Subject: how to get source of resource compiler and ld that supports .res References: <199712291019.KAA15266@out4.ibm.net> Message-ID: <9712300707.AA24842@modi.xraylith.wisc.edu> vischne@ibm.net writes: > > And, unless the usual process of compiling and linking using gcc is wrong, > you are lying. Comrade. ^^^^^^^^^^^^^^^^^^^^^^^^ I hope I'm not the only who finds such comments inappropriate, if not outright offensive. Regards, Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cvn@interchain.nl Tue Dec 30 00:01:00 1997 From: cvn@interchain.nl (Kees van Veen) Date: Tue, 30 Dec 1997 00:01:00 -0000 Subject: login/passwd file References: <34A7F8AF.7DD96E28@biddersedge.com> Message-ID: <34A8AA91.62B049B4@interchain.nl> Derek Young wrote: > > I have not been able to get login from Sergey's page working. I > downloaded crypt.exe from his page as well, and created an /etc/passwd > file. Whenever I login I get "Login incorrect". > > At first I thought this was a problem with telnet, but even just > straight login fails (thus causing the problem with telnet). > > I've tried setting up a passwd file with a null password like this: > > dyoung::1000:0:Derek > Young:/derek:/usr/gnuwin32/b18/H-i386-cygwin32/bin/bash.exe > > I've also tried setting one up with a valid password: > > dyoung:$1$11$r8tCuNX00ZGmGVaCftwTe1:1000:0:Derek > Young:/derek:/usr/gnuwin32/b18/H-i386-cygwin32/bin/bash.exe > > both passwords have the same effect. The shell executable is was > changed from /bin/sh.exe, but this doesn't seem to matter. > > I'm trying to run CVS which uses the same password authentication, and > this doesn't authenticate either. Am I doing something wrong? Do I > need to do more set up? > > -- derek > Are the permissions of the user executing login.exe sufficient to do a CreateProcessAsUser (see Sergey's homepage)? You should be able to run login.exe from the command line (I created an icon that does '...H-i386-cygwin32\bin\sh.exe -c "/usr/local/sbin/login.exe"', which works fine). Try running mkpasswd to generate a password file. Good luck, Kees - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hmwang@csit.edu.cn Tue Dec 30 01:26:00 1997 From: hmwang@csit.edu.cn (S.Q.Yang) Date: Tue, 30 Dec 1997 01:26:00 -0000 Subject: port postgresql to winNT/95 Message-ID: <9712300925.AA01603@ns> Is there anyone interested in porting postgresql to win? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From chuck@ethereal.ru Tue Dec 30 01:56:00 1997 From: chuck@ethereal.ru (Chuck Bogorad) Date: Tue, 30 Dec 1997 01:56:00 -0000 Subject: my page with binaries Message-ID: <01bd14ff$0a181b90$0100a8c0@hysteria.spb.ru> please take a look at binaries i have successfully compiled with b18. http://gnu-win32.paranoia.ru - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Tue Dec 30 03:44:00 1997 From: vischne@ibm.net (Victor Schneider) Date: Tue, 30 Dec 1997 03:44:00 -0000 Subject: ntxlib works on win95 Message-ID: <1.5.4.16.19971230064247.2e0f59d2@pop1.ibm.net> At 12:13 AM 12/30/97 -0500, you wrote: >> >> Unfortunately, ntxlib's libX11.a (the simulated version) compiles with few >> problems, but the sample test program causes the following error message >> from Windows 95: >> >> TST caused an invalid page fault in >> module at 0000:8b26ddea. >> ... > >In the makefile you have: > > $(cc) -o tst $(OBJECTS) -luser32 -lkernel32 -lgdi32 > >Remove "-lkernel32" and all works well on win95. > You're absolutely right. It does work. So now the Cygwin version of tst no longer crashes. Thanks. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Tue Dec 30 05:58:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Tue, 30 Dec 1997 05:58:00 -0000 Subject: how to get source of resource compiler and ld that supports .res References: <9712300707.AA24842@modi.xraylith.wisc.edu> Message-ID: <6kre0kl$pfB@mike.franken.de> Hi Mumit, >vischne@ibm.net writes: >> >> And, unless the usual process of compiling and linking using gcc is wrong, >> you are lying. Comrade. > ^^^^^^^^^^^^^^^^^^^^^^^^ >I hope I'm not the only who finds such comments inappropriate, if not >outright offensive. A few weeks ago, I asked, whether guys like this one can be excluded from the list. Nobody answered, but I'm not joking - even if he might be a genius (at least in his own opinion), it is not only inappropriate, but really offending for all the other ones on this list. Btw. this was not his first statement in this direction 8-(( > >Regards, >Mumit -- khan@xraylith.wisc.edu Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Tue Dec 30 05:58:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Tue, 30 Dec 1997 05:58:00 -0000 Subject: Problem with patching unix files with patch b18 and earlier Message-ID: <19971230135215.8340.qmail@hotmail.com> >From: "Wim van der Vegt" >To: >Subject: Problem with patching unix files with patch b18 and earlier >Date: Mon, 29 Dec 1997 22:31:31 +0100 > >Hi, > >I used gnu patch for some time now to patch sources of a unix based mud that >can also be compiled under win32. > >What i noticed was that : > >1. patch only likes files in unix format (ie lf only) for both file to be >patched and the patch itself. It then patches correctly and signals plan A. >If the file to be patched is saved in dos mode (ie crlf terminated) patch >signals plan B and ruins the file. (i just use a simple commandline patch > ). > >2. if patch patches a unix file it returns a crlf (ie dos) file as a result It will only return \n if your directories are mounted as text=binary. This sets the default output file mode. For further information check the mail archives as there has been lengthy discussions on this. You can find the mail archives at http:/www.cygnus.com/ml/gnu-win32. [snip] >Wim van der Vegt / Dutch Open University > - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Tue Dec 30 05:58:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Tue, 30 Dec 1997 05:58:00 -0000 Subject: how to get source of resource compiler and ld that supports .res Message-ID: <19971230132322.4035.qmail@hotmail.com> >To: gnu-win32@cygnus.com >Subject: Re: how to get source of resource compiler and ld that supports .res >Date: Tue, 30 Dec 1997 01:07:59 -0600 >From: Mumit Khan > >vischne@ibm.net writes: >> >> And, unless the usual process of compiling and linking using gcc is wrong, >> you are lying. Comrade. > ^^^^^^^^^^^^^^^^^^^^^^^^ >I hope I'm not the only who finds such comments inappropriate, if not >outright offensive. > Your not the only one Mumit. It is comments like this that tend to cause flames. >Regards, >Mumit -- khan@xraylith.wisc.edu > http://www.xraylith.wisc.edu/~khan/ >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From msphil@widomaker.com Tue Dec 30 05:58:00 1997 From: msphil@widomaker.com (msphil@widomaker.com) Date: Tue, 30 Dec 1997 05:58:00 -0000 Subject: [VGUI] RE: Compiling "V" 1.16 - Specific Instructions? References: <9712300407.AA37836@spider.uspnet.usp.br> Message-ID: <199712301345.IAA00123@mail.3gi.com> "Fernandes, Hilton" wrote: > On 26 Dec 97 at 12:22, Brendan Simon wrote: > > > V has both X Windows AND MS-Windows support. > > From what i could see, in that times (circa 1996) several people > were having problems with V's window implementation. I'm not saying > that V didn't compile, but people related that in some cases V didn't > work correctly under Windows. I've had better luck with V under Windows than V under RedHat 5.0 for some odd reason (some of the built-in X dialogs, for instance, file selection, misbehave). The biggest drawback to V is that the widgets available are currently limited (no multi-select comboboxes, no trees, and some limitations on creating lists on the fly, IIRC), but for simpler programs, it's an excellent framework to work with. > I thought that was still the case when i read V's page: they say > that Windows support is still missing. > > Maybe that was just a misunderstood, but i'd like to ask your opinion > about it. As of V 1.15 and 1.16, it works with both Win16 and Win32, at least with MS Visual C++ (1.52 and 5.0, respectively), and I have compiled a program I wrote with both (and, for that matter, Linux/X11). V is slightly persnickety on data alignment and pointer sizes in Win32, but the distribution comes with project files for MSVC and Makefiles for Borland C++ (I don't know which versions). There have been consistent and reported problems mixing V with the Cygnus windows stuff, though. I don't have details, since I haven't tried it myself. -- Mike Phillips, msphil@widomaker.com Home Page: http://www.widomaker.com/~msphil/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From earnie_boyd@hotmail.com Tue Dec 30 05:58:00 1997 From: earnie_boyd@hotmail.com (Earnie Boyd) Date: Tue, 30 Dec 1997 05:58:00 -0000 Subject: Fw: make problem (ctd.) Message-ID: <19971230135750.11703.qmail@hotmail.com> I haven't a clue. - \\||// ---o0O0--Earnie--0O0o---- -earnie_boyd@hotmail.com- ------ooo0O--O0ooo------- >Reply-To: "Ivo Abeloos" <@skynet.be> >From: "Ivo Abeloos" >To: "gnuwin32 (E-mail)" >Subject: Fw: make problem (ctd.) >Date: Mon, 29 Dec 1997 23:31:20 +0100 > >Hi Ernie, > >Thanx for the ultra fast reply! I still have problems but I can narrow it >down a bit. I'm on NT4.0 by the way... > >I have the following makefile: > >testfile: testfile.proto > @(echo Start;\ > sed -e "" testfile.proto;\ > echo Done;\ > ) > >When I open the bash shell and run the make, it works, it prints the >contents >of the testfile.proto file as it should. Running make twice in the same >shell >only gives Start and Done. I really don't have a clue... I play with it >once >more but if you know of a solution fast, let me know... > >Thanx, > >Ivo > > >- >For help on using this list (especially unsubscribing), send a message to >"gnu-win32-request@cygnus.com" with one line of text: "help". > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From mh@mike.franken.de Tue Dec 30 05:58:00 1997 From: mh@mike.franken.de (Michael Hirmke) Date: Tue, 30 Dec 1997 05:58:00 -0000 Subject: port postgresql to winNT/95 References: <9712300925.AA01603@ns> Message-ID: <6kre0ra$pfB@mike.franken.de> Hi S.Q.Yang, >Is there anyone interested in porting postgresql to win? AFAIK this has already been done or is in the works. Bye. Michael. -- Michael Hirmke | Telefon +49 (911) 557999 Georg-Strobel-Strasse 81 | FAX +49 (911) 557664 90489 Nuernberg | E-Mail mailto:mh@mike.franken.de | WWW http://minimike.franken.de/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 30 08:56:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Tue, 30 Dec 1997 08:56:00 -0000 Subject: spawn and wait? References: <9712300646.AA24720@modi.xraylith.wisc.edu> Message-ID: In article < 9712300646.AA24720@modi.xraylith.wisc.edu >, Mumit Khan wrote: >fyi, wait goes with [fork]+exec*, and cwait goes with spawn*. Actually if you are running a recent coolview, then I believe that Sergey has made cwait and wait interchangeable. Judging by the error message in the original mail, Tim was not running a coolview release. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From cgf@bbc.com Tue Dec 30 09:03:00 1997 From: cgf@bbc.com (Christopher Faylor) Date: Tue, 30 Dec 1997 09:03:00 -0000 Subject: how to get source of resource compiler and ld that supports .res References: <199712291019.KAA15266@out4.ibm.net> <9712300707.AA24842@modi.xraylith.wisc.edu> Message-ID: In article < 9712300707.AA24842@modi.xraylith.wisc.edu >, Mumit Khan wrote: >vischne@ibm.net writes: >> >> And, unless the usual process of compiling and linking using gcc is wrong, >> you are lying. Comrade. > ^^^^^^^^^^^^^^^^^^^^^^^^ >I hope I'm not the only who finds such comments inappropriate, if not >outright offensive. I 100% agree. I tried to respond to this message several times yesterday but couldn't find the words to express my dismay that an honest attempt to help could engender such a mean spirited reply. For a little background, however, it might be instructive to do a DejaNews search on the author 'vischne'. That might put this in perspective. -- http://www.bbc.com/ cgf@bbc.com "Strange how unreal VMS=>UNIX Solutions Boston Business Computing the real can be." - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From roland.schwarz@telecom.at Tue Dec 30 09:17:00 1997 From: roland.schwarz@telecom.at (Roland Schwarz) Date: Tue, 30 Dec 1997 09:17:00 -0000 Subject: gcc native build on NT? Message-ID: <01bd1547$1ef388e0$0100007f@fledermaus> Has anyone tried to rebuild the gcc compiler on a NT box ? I had to build gperf first, and than ran into an error that said something like: ... no spill registers available ... in a function of print-tree.c. Any experiences appreciated. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From ian@cygnus.com Tue Dec 30 10:01:00 1997 From: ian@cygnus.com (ian@cygnus.com) Date: Tue, 30 Dec 1997 10:01:00 -0000 Subject: Who's got source to windres? References: <34A83436.23F4@switch.rockwell.com> <9712300539.AA24626@modi.xraylith.wisc.edu> Message-ID: <199712301801.NAA09792@tweedledumb.cygnus.com> In gnu-win32 khan@xraylith.wisc.edu (Mumit Khan) writes: >Yes, it is. I can't share the binutils snapshot sources, but I'm quite >sure that the binutils maintainer(s) will give it out to folks who are >willing to deal with snapshots (eg., be prepared to deal with bugs in >snapshots, test/debug and possibly solve the bugs, etc). Ian Taylor >, who I believe subscribes to this list, is the person >you want to talk to. I do read this list, but I use a news reader and I skip most messages. I'm willing to tell people how to find the binutils snapshots, but for serious inquiries only. If you don't already know how to build GNU releases from source, or you don't want to deal with bugs, then please don't ask. I'm too swamped to help people figure these things out. I would much rather have people like Mumit build and distribute binaries, and filter bug reports before they get to me. It's very helpful. Of course, I also need to get an official GNU binutils release out. Ian - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From jordanh@ccia.com Tue Dec 30 11:48:00 1997 From: jordanh@ccia.com (Jordan Henderson) Date: Tue, 30 Dec 1997 11:48:00 -0000 Subject: port postgresql to winNT/95 References: <9712300925.AA01603@ns> Message-ID: <199712301950.OAA32406@jaguar.ccia.com> Yes, most certainly. I have been working alittle on this. Perhaps we can work together? Thanks Jordan Henderson - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From pgarceau@teleport.com Tue Dec 30 14:49:00 1997 From: pgarceau@teleport.com (Paul Garceau) Date: Tue, 30 Dec 1997 14:49:00 -0000 Subject: C++ conversions from Unix to NT Message-ID: <199712302249.OAA28731@mail1.teleport.com> Greetings, Would someone be willing to outline (or itemize) any problems that have come up in converting Unix C++ (Solaris) to WinNT (assuming Alpha) platforms using b18? I am in the process of evaluating how effective b18 is (or b19 will be) when it comes to converting Unix based C++ client/server environment to a WinNT client/server environment It is not clear at this time whether the target system is x86 based or Unix based (assuming Unix) platform. NT 4.0 or 5.0 is the target system. As this is a rare occurrence, personal email would be preferable to taking up bandwidth reserved for the mailing list. Thanks, Paul G. Information Systems Consultant NewDawn Productions http://www.teleport.com/~pgarceau/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From emmess@bml.RMV.se.org Tue Dec 30 15:52:00 1997 From: emmess@bml.RMV.se.org (Martin 'EmmEss' Svenningsson) Date: Tue, 30 Dec 1997 15:52:00 -0000 Subject: linking in DLLs? Message-ID: <3.0.5.32.19971231005203.0094d370@Spam> I may be a bit lame, but how do I link other libs in my DLL's? (non-reloc) I've tried ld foo.dll foo.exp foo.o C:\pathtolibs\libbar.a, but it doesn't seem to work. TIA and Happy New Year :) ---- Greetz from Helsingborg, SWEDEN | When replying to me, Martin Svenningsson, aka EmmEss | remove the RMV from http://www1.tripnet.se/~emmess/ | my mail address. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dbe@wgn.net Tue Dec 30 16:07:00 1997 From: dbe@wgn.net ($Bill Luebkert) Date: Tue, 30 Dec 1997 16:07:00 -0000 Subject: how to get source of resource compiler and ld that supports .res References: <199712291019.KAA15266@out4.ibm.net> <9712300707.AA24842@modi.xraylith.wisc.edu> Message-ID: <34A98C95.49CA2155@wgn.net> Christopher Faylor wrote: > > In article < 9712300707.AA24842@modi.xraylith.wisc.edu >, > Mumit Khan wrote: > >vischne@ibm.net writes: > >> > >> And, unless the usual process of compiling and linking using gcc is wrong, > >> you are lying. Comrade. > > ^^^^^^^^^^^^^^^^^^^^^^^^ > >I hope I'm not the only who finds such comments inappropriate, if not > >outright offensive. > > I 100% agree. I tried to respond to this message several times > yesterday but couldn't find the words to express my dismay that an honest > attempt to help could engender such a mean spirited reply. > > For a little background, however, it might be instructive to do a DejaNews > search on the author 'vischne'. That might put this in perspective. Make that 'vischne@ibm.net'. Sounds like this guy has extreme paranoia of "secret communists" hidden in our societies. -- ,-/- __ _ _ $Bill Luebkert (_/ / ) // // DBE Collectibles / ) /--< o // // http://www.wgn.net/~dbe/ -/-' /___/_<_ >> For a little background, however, it might be instructive to do a DejaNews search on the author 'vischne'. That might put this in perspective. << The above and all your mock indignation is just a public admission that the author in question, along with one or two others on this conference, have been stalking me on the internet and have deliberately posted to unrelated conferences where I have had reason to post. ???? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Tue Dec 30 17:34:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Tue, 30 Dec 1997 17:34:00 -0000 Subject: beta18: building gdb gives incomplete import table [and other problems]??? References: <34A7414B.F9935C9E@cybec.com.au> <34A82DA4.742F3EDD@cybec.com.au> Message-ID: <19971231123414.00894@mundook.cs.mu.OZ.AU> On 30-Dec-1997, Steven Reddie wrote: > Further to yesterdays post, I have succeeded in getting gdb to compile and > run but I am still confused about what's going on. The linker still seems > faulty. ... > This would successfully compile but crash immediately when executed. The > crash was on the call to GetModuleHandle() which is called from cygwin_crt0. > To fix this crash, I removed the -lkernel32 from the gdb/Makefile and > re-linked. It appears that the -lkernel32 somehow looses the import table > from libcygwin32.a. One bug with the linker that has been mentioned on this list previously is that if the same library name is mentioned twice on the link line, then the linker generates executables that crash. `-lkernel32' is linked in by default, so if you mention it explicitly then it will get linked in twice, causing the symptoms you observed. -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From newsham@lava.net Tue Dec 30 17:43:00 1997 From: newsham@lava.net (Tim Newsham) Date: Tue, 30 Dec 1997 17:43:00 -0000 Subject: signals from non-cygwin processes? Message-ID: Hi, I would like to be able to send a cygwin process a signal from a non-cygwin process. When one cygwin process wants to signal another it does it with a message as: SendMessage(hwnd, WM_USER, SIGNAL, signo); where hwnd is the hidden window (?) of the target process and SIGNAL is the constant 1 and signo is the signal number to send. So it seems that it should be possible for non-cygwin processes to signal a cygwin process if it can find out the value of hwnd. So my question is: Is there a way one process can find out the value of the hwnd belonging to a cygwin process? What if the first process was the entity that spawned the cygwin process? Tim N. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From chuck@ethereal.ru Tue Dec 30 17:43:00 1997 From: chuck@ethereal.ru (Chuck Bogorad) Date: Tue, 30 Dec 1997 17:43:00 -0000 Subject: binaried page - udpate Message-ID: <01bd158c$460e9f10$0100a8c0@hysteria.spb.ru> hi! i updated my page with the following stuff: - X-enabled version of Aladdin Ghostscript - X-enabled version of VIM - X-support DLL's compiled by Sergey Okhapkin - X-libs & X-includes to enable compiling X-enabled stuff (also by Sergey Okhapkin) the address is: http://gnu-win32.paranoia.ru/ p.s. Happy New Year everybody! - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Tue Dec 30 23:19:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Tue, 30 Dec 1997 23:19:00 -0000 Subject: spawn and wait? Message-ID: <01BD15D5.A72D67B0@gater.krystalbank.msk.ru> Christopher Faylor wrote: > Actually if you are running a recent coolview, then I believe that Sergey > has made cwait and wait interchangeable. Yes, you're absolutely right. Here is an extract from spawn code: else if ((mode == _P_NOWAIT) || (mode == _P_NOWAITO)) { res = pid; } CloseHandle (pi.hThread); return (int) res; And cwait implementation: extern "C" int cwait (int *result, int pid, int) { return waitpid (pid, result, 0); } In original b18 cygwin.dll spawn(_P_NOWAIT) returned win32 child process's handle and cwait() waited on that handle. -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hmwang@csit.edu.cn Tue Dec 30 23:49:00 1997 From: hmwang@csit.edu.cn (S.Q.Yang) Date: Tue, 30 Dec 1997 23:49:00 -0000 Subject: port postgresql to win Message-ID: <9712310748.AA03729@ns> Mr. Jordan Henderson: I want to port postgresql to win. But the developer of postgresql told me that postgresql can not be ported to win95.Do you think so? If you have made some progress,please tell me.I try to port postgresql-v6.1 to win95. But when I run configure(postgresql),it reports the following errors: loading cache ./config.cache checking host system type... i386-pc-cygwin32 ************************************************************* configure does not currently recognize your operating system, therefore you must do a manual configuration of: cygwin32 Please contact scrappy@hub.org to see about rectifying this, including the above 'checking host system type...' line ************************************************************* Then I change the configure file by adding line "cygwin32*) PORTNAME='cygwin32';;" and re-run configure with default argument. It then reports: -ASSERT CHECKING disabled checking for gcc... (cached) gcc checking whether the C compiler (gcc ) works... no configure: error: installation or configuration problem: C compiler cannot create executables. I have checked gnuwin32 and found gcc. I don't know why it report this error.May you help me? Thanks! - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Wed Dec 31 01:44:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Wed, 31 Dec 1997 01:44:00 -0000 Subject: how to get source of resource compiler and ld that supports .res Message-ID: <01BD15E9.CD3D6CC0@gater.krystalbank.msk.ru> $Bill Luebkert wrote: > Sounds like this guy has extreme paranoia of "secret communists" hidden in > our societies. This is about me?-) -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From sos@prospect.com.ru Wed Dec 31 01:51:00 1997 From: sos@prospect.com.ru (Sergey Okhapkin) Date: Wed, 31 Dec 1997 01:51:00 -0000 Subject: signals from non-cygwin processes? Message-ID: <01BD15EA.D8FFDF60@gater.krystalbank.msk.ru> Tim Newsham wrote: > I would like to be able to send a cygwin process a signal > from a non-cygwin process. When one cygwin process wants to > signal another it does it with a message as: > > SendMessage(hwnd, WM_USER, SIGNAL, signo); > It would be much better to call cygwin.dll-provided kill() export. Signal handling in b18, in coolview and in upcoming b19 differs very much. More precisely, nobody knows now exactly, how signals will be handled in b19 :-) -- Sergey Okhapkin, http://www.lexa.ru/sos Moscow, Russia Looking for a job. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Wed Dec 31 06:58:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Wed, 31 Dec 1997 06:58:00 -0000 Subject: egcs collect2/ld bug update Message-ID: <9712311458.AA25998@modi.xraylith.wisc.edu> I'm beginning to sound like a broken record ;-) Looks like collect2 (which is needed for C++ -frepo option, and that's why I had it in the egcs distribution) is rather hosed due to various problems. The best thing to do for now is to simply delete (or rename) the collect2 executable. You gain a functional system, but you lose -frepo. $ cd /H-i386-cygwin32/lib/gcc-lib/i386-cygwin32/egcs-2.90.21 $ mv ld.exe ld-orig.exe I suggest this for mingw32 users as well just to be consistent with the cygwin32 distribution. $ cd /lib/gcc-lib/i386-mingw32/egcs-2.90.21 $ mv ld.exe ld-orig.exe If you're building from source, please add the following line: USE_COLLECT2= to egcs/gcc/config/i386/x-cygwin32. In case you're curious what the problem really is, here's a recap: 1. Bug in fork/exec + freopen in winsup, where the child choses stderr. Collect2 uses this to collect error messages from linker, and that's why you don't see 'undefined references to xxx' messages when using collect2. 2. Bug in pathname handling. That's why you need the collect-ld.exe. Thanks to all of you for your informative bug reports and your patience. Happy New Year to all Mumit - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From dbe@wgn.net Wed Dec 31 06:58:00 1997 From: dbe@wgn.net ($Bill Luebkert) Date: Wed, 31 Dec 1997 06:58:00 -0000 Subject: how to get source of resource compiler and ld that supports .res References: <01BD15E9.CD3D6CC0@gater.krystalbank.msk.ru> Message-ID: <34AA5CFC.6623DC7A@wgn.net> Sergey Okhapkin wrote: > > $Bill Luebkert wrote: > > Sounds like this guy has extreme paranoia of "secret communists" hidden in > > our societies. > > This is about me?-) No, I said "secret communists" not "GNU writing russians". Although you might think of GNU as a communist effort. :) You should read some of his posts. PS: If you really are a communist, Sergey, you should come out of the closet and at least let all of your GNUist comrades know. :>) -- ,-/- __ _ _ $Bill Luebkert (_/ / ) // // DBE Collectibles / ) /--< o // // http://www.wgn.net/~dbe/ -/-' /___/_<_ <34A7414B.F9935C9E@cybec.com.au> <19971231123414.00894@mundook.cs.mu.OZ.AU> Message-ID: <3.0.5.32.19971231153549.0079dc70@mail.servisoft.pt> At 12:34 31-12-1997 +1100, Fergus Henderson wrote: >... >One bug with the linker that has been mentioned on this list previously >is that if the same library name is mentioned twice on the link line, >then the linker generates executables that crash. `-lkernel32' is >linked in by default, so if you mention it explicitly then it will >get linked in twice, causing the symptoms you observed. > >-- >Fergus Henderson | "I have always known that the pursuit >WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" >PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. I can't successfully link DLLs (relocatable or non-relocatable) if I don't include: -lc -lcygwin -lkernel32 -lc Also, this is somewhat consistent with your Makefile.DLLs, an excerpt of which I pasted next: (...) # The `sed' commands below are to convert DOS-style `C:\foo\bar' # pathnames into Unix-style `//c/foo/bar' pathnames. CYGWIN32_LIBS = $(shell echo \ -L`dirname \`gcc -print-file-name=libgcc.a | \ sed -e 's@^\\\\([A-Za-z]\\\\):@//\\\\1@g' -e 's@\\\\\\\\@/@g' \` ` \ -L`dirname \`gcc -print-file-name=libcygwin.a | \ sed -e 's@^\\\\([A-Za-z]\\\\):@//\\\\1@g' -e 's@\\\\\\\\@/@g' \` ` \ -L`dirname \`gcc -print-file-name=libkernel32.a | \ sed -e 's@^\\\\([A-Za-z]\\\\):@//\\\\1@g' -e 's@\\\\\\\\@/@g' \` ` \ -lgcc -lcygwin -lkernel32 -lgcc) # Making relocatable DLLs doesn't seem to work. # Note quite sure why. The --image-base values below # where chosen at random, they seem to work on my machine. RELOCATABLE=no LDFLAGS-libgc += --image-base=0x2345000 LDFLAGS-libmer += --image-base=0x1234000 LDFLAGS-libmercury += --image-base=0x3456000 ifeq "$(strip $(RELOCATABLE))" "yes" # to create relocatable DLLs, we need to do two passes # (warning: this is untested) %.dll: %.exp %.a %_dll.o dll_init.o dll_fixup.o $(LD) $(LDFLAGS) $(LDFLAGS-$*) --dll -o $*.base \ -e _dll_entry@12 \ $*.exp $*.a $*_dll.o \ dll_init.o dll_fixup.o \ $(LDLIBS) $(LDLIBS-$*) \ $(CYGWIN32_LIBS) (...) If I understand correctly, you use "-lgcc -lcygwin -lkernel32 -lgcc" among other obvious things. So, should we, or should we not repeat libs? By the way, is there a description of the available libs and their contents? -- JPSousa - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From banders@ECD.Rockwell.COM Wed Dec 31 11:03:00 1997 From: banders@ECD.Rockwell.COM (Bartlee A. Anderson) Date: Wed, 31 Dec 1997 11:03:00 -0000 Subject: port postgresql to win References: <9712310748.AA03729@ns> Message-ID: <34AA93E8.5F20@switch.rockwell.com> Last I heard there was a problem with make that didn't handle the postgresql package very well. Haven't heard anything in a while! How about an update from the guys working on this? S.Q.Yang wrote: > > Mr. Jordan Henderson: > I want to port postgresql to win. But the developer of postgresql told me > that postgresql can not be ported to win95.Do you think so? If you have > made some progress,please tell me.I try to port postgresql-v6.1 to win95. > But when I run configure(postgresql),it reports the following errors: > > loading cache ./config.cache > checking host system type... i386-pc-cygwin32 > > ************************************************************* > configure does not currently recognize your operating system, > therefore you must do a manual configuration of: > cygwin32 > Please contact scrappy@hub.org to see about rectifying this, > including the above 'checking host system type...' line > ************************************************************* > > Then I change the configure file by adding line "cygwin32*) > PORTNAME='cygwin32';;" and re-run configure with default argument. It then > reports: > -ASSERT CHECKING disabled > checking for gcc... (cached) gcc > checking whether the C compiler (gcc ) works... no > configure: error: installation or configuration problem: C compiler cannot > create executables. > > I have checked gnuwin32 and found gcc. I don't know why it report this > error.May you help me? > Thanks! > > - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". -- Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) Rockwell International Electronic Commerce Division 300 Bauman Ct. banders@ecd.rockwell.com MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 Wood Dale, IL 60191 FAX (630) 227-9771 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Wed Dec 31 13:20:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Wed, 31 Dec 1997 13:20:00 -0000 Subject: ntxlib works on win95 Message-ID: <199712312119.VAA61054@out4.ibm.net> >>> TST caused an invalid page fault in >>> module at 0000:8b26ddea. >>> ... >> >>In the makefile you have: >> >> $(cc) -o tst $(OBJECTS) -luser32 -lkernel32 -lgdi32 >> >>Remove "-lkernel32" and all works well on win95. >> >You're absolutely right. It does work. So now the Cygwin version of tst no >longer crashes. Thanks. > Of course, why removing lkernel32 causes it to work is still a mystery.... I checked ntxlib substituted into Sergei's X11R6 libs, and, to compile rxvt requires implementing about 8 more ntxlib functions, some of which are variants on already-implemented functions. That gives a measure of what work could be done on ntxlib. If you are not concerned about using your system as a remote X-server, than ntxlib is one way to go. If you want your system to act as an X-server, then the full X11R6 distribution, with a commercial X-server, is the way to go. If someone goes ahead and implements additional ntxlib functionality, why not upload it (with deninc.com permission) to deninc.com, and use some notation to identify it as an upgrade, along with an accompanying readme? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From shorti@apci.net Wed Dec 31 14:38:00 1997 From: shorti@apci.net (shorti) Date: Wed, 31 Dec 1997 14:38:00 -0000 Subject: take it outside (was: Re: how to source of resource...) References: <199712310044.AAA14786@out5.ibm.net> Message-ID: <3.0.3.32.19971231163538.007b6cc0@apci.net> hello all... well.. i'll admit that i'm the biggest newbie on this list, and that i don't understand 99.999% of what you guys(i midwestern - gimme a break and include females) talk about, so it's not my place to tell anyone to shut up, but WOULD YOU STOP BICKERING IN AN OPEN, PUBLIC LIST?? geez.. if you're gonna talk about something non-gnu related, take gnu-win32@cygnus.com out of the headers, please. as much as i find it slightly amusing, and as much as it is one of the few things here i _understand_ (), personal preference in politics, sexual partners, religion, etc, should not become a part of this mailing list. i have my opinions, you have your opinions, if you would like to express them: do it somplace else. heck, make an opinions page at your personal homepage; put a reference to it in your sig. good enough? sorry to all who were not involved in this lil incident.. i didn't want to increase the mail volume, but it had to be done. thanx for everyone's time... sorry to interrupt normally productive content, - josh, aka shorti -=- http://www.apci.net/~shorti/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From hop@elwood.pionet.net Wed Dec 31 14:48:00 1997 From: hop@elwood.pionet.net (Justin Hopkins) Date: Wed, 31 Dec 1997 14:48:00 -0000 Subject: Cygwin32 & C++ standard Message-ID: <3.0.1.32.19971231165103.0069dfec@elwood.pionet.net> Can someone give me an idea of how conforming cygwin32 is to the C++ standard? Or at least to the December '96 public draft? Thanks Justin Hopkins - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From banders@ECD.Rockwell.COM Wed Dec 31 14:48:00 1997 From: banders@ECD.Rockwell.COM (Bartlee A. Anderson) Date: Wed, 31 Dec 1997 14:48:00 -0000 Subject: how to get source of resource compiler and ld that supports .res References: <01BD15E9.CD3D6CC0@gater.krystalbank.msk.ru> <34AA5CFC.6623DC7A@wgn.net> Message-ID: <34AAC8B6.154C@switch.rockwell.com> $Bill Luebkert wrote: > > Sergey Okhapkin wrote: > > > > $Bill Luebkert wrote: > > > Sounds like this guy has extreme paranoia of "secret communists" hidden in > > > our societies. > > > > This is about me?-) > > No, I said "secret communists" not "GNU writing russians". Guffaw! > Although you might think of GNU as a communist effort. :) > > You should read some of his posts. > > PS: If you really are a communist, Sergey, you should come out of the closet > and at least let all of your GNUist comrades know. :>) Rotfl At least we get one good laugh (or two) out of that nasty post! I hope no one if offended by your limited attempt at levity ;) Bart > > -- > ,-/- __ _ _ $Bill Luebkert > (_/ / ) // // DBE Collectibles > / ) /--< o // // http://www.wgn.net/~dbe/ > -/-' /___/_<_ - > For help on using this list (especially unsubscribing), send a message to > "gnu-win32-request@cygnus.com" with one line of text: "help". -- Bartlee A. Anderson System Test (Interfaces-Tools-Automation-ISDN) Rockwell International Electronic Commerce Division 300 Bauman Ct. banders@ecd.rockwell.com MS 933-605 Opinions my own, not Rockwell's VOICE (630) 227-8975 Wood Dale, IL 60191 FAX (630) 227-9771 - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From khan@xraylith.wisc.edu Wed Dec 31 16:56:00 1997 From: khan@xraylith.wisc.edu (Mumit Khan) Date: Wed, 31 Dec 1997 16:56:00 -0000 Subject: Cygwin32 & C++ standard References: <3.0.1.32.19971231165103.0069dfec@elwood.pionet.net> Message-ID: <9801010055.AA26764@modi.xraylith.wisc.edu> Justin Hopkins writes: > Can someone give me an idea of how conforming cygwin32 is > to the C++ standard? Or at least to the December '96 public > draft? Thanks Depends on what you mean by cygwin32 C++. If you mean the original b18 distribution, it's horribly old, so all bets are off if you're looking at the latest draft that was approved as the standard. On the other hand, the egcs distribution is pretty close, and in fact, close enough to (almost) compete with some top quality commercial compilers such as Kai and others based on the EDG front end. Also note that G++ now uses unmodified SGI STL distribution, which means that as SGI updates its implementation to match the standard, G++ will follow automatically. Here're some quotes from the latest cp/NEWS (my commens are within [[ ]]): * g++ now uses a new implementation of templates. The basic idea is that now templates are minimally parsed when seen and then expanded later. This allows conformant early name binding and instantiation controls, since instantiations no longer have to go through the parser. What you get: + Inlining of template functions works without any extra effort or modifications. + Instantiations of class templates and methods defined in the class body are deferred until they are actually needed (unless -fexternal-templates is specified). + Nested types in class templates work. + Static data member templates work. + Member function templates are now supported. + Partial specialization of class templates is now supported. + Explicit specification of template parameters to function templates is now supported. [[ also, now you don't have to worry about instantiatng templates yourself on ELF and PE-COFF, since the linker will simply discard the duplicates. In fact, I've seen 25% size reduction in final code size that until now used explicit template instantiation in conjunction with -fno-impliction-template flag ]] Things you may need to fix in your code: + Syntax errors in templates that are never instantiated will now be diagnosed. + Types and class templates used in templates must be declared first, or the compiler will assume they are not types, and fail. + Similarly, nested types of template type parameters must be tagged with the 'typename' keyword, except in base lists. In many cases, but not all, the compiler will tell you where you need to add 'typename'. For more information, see http://www.cygnus.com/misc/wp/dec96pub/template.html#temp.res + Guiding declarations are no longer supported. Function declarations, including friend declarations, do not refer to template instantiations. You can restore the old behavior with -fguiding-decls until you fix your code. [[ The last one will catch lot of existing code, and using -fguiding-decls is really no good since that can cause trouble with library code, especially iostreams. The right solution is to convert your code correctly, and if you need backward compatibility with older compilers, you can use preprocess macros. For an example of how to do this using autoconf, take a look at my patch to GNU Octave 2.0.9 available at: ftp://ftp.xraylith.wisc.edu/pub/khan/gnu-win32/cygwin32/ports/ ]] Other features: + Default function arguments in templates will not be evaluated (or checked for semantic validity) unless they are needed. Default arguments in class bodies will not be parsed until the class definition is complete. + The -ftemplate-depth-NN flag can be used to increase the maximum recursive template instantiation depth, which defaults to 17. If you need to use this flag, the compiler will tell you. + Explicit instantiation of template constructors and destructors is now supported. For instance: template A::A(const A&); Still not supported: + Member class templates. + Template template parameters. + Template friends. [[ Not having Template friends stink, but since I'm not about contribute the code, I can't really complain. The next one in line that I'd like is the Template template parameters, since I have tons of code that uses this feature, and there is not easy work-around. I can do without the member class templates feature since there are easy work-arounds. ]] [[ Also not implemented is namespaces, but at least code that uses just std namespace will work. ]] * Standard usage syntax for the std namespace is supported; std is treated as an alias for global scope. General namespaces are still not supported. [[ Other things like RTTI, exception handling work quite well. On platforms where Dwarf2 unwinding works well, EH work like a charm; on others, such as win32, where it doesn't yet, g++ uses setjmp-longjmp, which imposes space and time penalty when using EH. ]] [[ Two more items that can potentially break existing code: ]] * The overload resolution code has been rewritten to conform to the latest C++ Working Paper. Built-in operators are now considered as candidates in operator overload resolution. Function template overloading chooses the more specialized template, and handles base classes in type deduction and guiding declarations properly. In this release the old code can still be selected with -fno-ansi-overloading, although this is not supported and will be removed in a future release. * (void *)0 is no longer considered a null pointer constant; NULL in is now defined as __null, a magic constant of type (void *) normally, or (size_t) with -ansi. Happy New Year Mumit -- khan@xraylith.wisc.edu http://www.xraylith.wisc.edu/~khan/ - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From liralen@geocities.com Wed Dec 31 19:09:00 1997 From: liralen@geocities.com (Kim) Date: Wed, 31 Dec 1997 19:09:00 -0000 Subject: Compiling errors Message-ID: <34AB0A2F.79D@geocities.com> when i try to compile using gcc... it says something about Error while dumping state (probably corrupted stack) and In cygwin_except_handler this is my first time posting on this list and i am also quite new at this gnu-win32 program... only had experience with linux... so thanks ahead of times =) Kim - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Wed Dec 31 21:16:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Wed, 31 Dec 1997 21:16:00 -0000 Subject: Linking with -lkernel32 [was: Re: beta18: building gdb gives incomplete import table [and other problems]???] References: <34A82DA4.742F3EDD@cybec.com.au> <34A7414B.F9935C9E@cybec.com.au> <19971231123414.00894@mundook.cs.mu.OZ.AU> <3.0.5.32.19971231153549.0079dc70@mail.servisoft.pt> Message-ID: <19980101161636.43667@mundook.cs.mu.OZ.AU> On 31-Dec-1997, Joao Pedro Sousa wrote: > At 12:34 31-12-1997 +1100, Fergus Henderson wrote: > >... > >One bug with the linker that has been mentioned on this list previously > >is that if the same library name is mentioned twice on the link line, > >then the linker generates executables that crash. `-lkernel32' is > >linked in by default, so if you mention it explicitly then it will > >get linked in twice, causing the symptoms you observed. > I can't successfully link DLLs (relocatable or non-relocatable) if I don't > include: > -lc -lcygwin -lkernel32 -lc Whether or not you need to specify `-lkernel32' etc. explicitly depends on whether you are using `gcc' to do the link, or whether you are invoking `ld' directly. If you're using `gcc' to do the link, it will pass `-lkernel32' etc., so you don't need to (and in fact must not) specify it yourself. If you're invoking `ld' directly, then yes you do need to specify it explicitly. > If I understand correctly, you use "-lgcc -lcygwin -lkernel32 -lgcc" among > other obvious things. > So, should we, or should we not repeat libs? I think repeating ordinary libraries (e.g. `-lgcc') is OK; I think it is only DLL import libraries (e.g. `-lcygwin' or `-lkernel32') that cause problems if they are linked twice. (However, this is basically just educated guesswork.) -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From fjh@cs.mu.OZ.AU Wed Dec 31 22:31:00 1997 From: fjh@cs.mu.OZ.AU (Fergus Henderson) Date: Wed, 31 Dec 1997 22:31:00 -0000 Subject: linking in DLLs? References: <3.0.5.32.19971231005203.0094d370@Spam> Message-ID: <19980101173110.25136@mundook.cs.mu.OZ.AU> On 31-Dec-1997, Martin 'EmmEss' Svenningsson wrote: > I may be a bit lame, but how do I link other libs in my DLL's? (non-reloc) Just add the approproate `-lfoo -lbar' options to the link line when creating your DLL; they should go after the `.o' files but before any `-l' options for system libraries (e.g. `-lkernel32'). If you're using my Makefile.DLLs, put these in the LDLIBS variable (or, if you are buidling multiple DLLs in the same directory, you can use the variables LDLIBS-foo and LDLIBS-bar). > I've tried ld foo.dll foo.exp foo.o C:\pathtolibs\libbar.a, but it doesn't > seem to work. Well, that would need to be ld --dll -o foo.dll foo.exp foo.o ... ^^^^^^^^ Also depending on which shell you're using, you may need to escape the \ characters or use Unix-style pathnames (e.g. //c/pathtolibs/libbar.a). In addition, I think you'll need to specify `-lkernel32' etc. Furthermore, I think cygwin.dll doesn't work if it gets relocated; this may be more likely if you're trying to link with multiple DLLs. If your program links fine but won't run, then try running your program under gdb -- this will give you some information about what's happening with DLLs. -- Fergus Henderson | "I have always known that the pursuit WWW: < http://www.cs.mu.oz.au/~fjh > | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp. - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help". From vischne@ibm.net Wed Dec 31 22:45:00 1997 From: vischne@ibm.net (vischne@ibm.net) Date: Wed, 31 Dec 1997 22:45:00 -0000 Subject: ntxlib-0.1alpha at deninc.com Message-ID: <199801010645.GAA30172@loopback> ntxlib-0.1alpha.tar.gz was posted in the /incoming directory of deninc.com. It has some additional functions and an updated xnt.h file that better matches current Xlib.h. The rclock program in the rxvt distribution is partly working under this release. ???? - For help on using this list (especially unsubscribing), send a message to "gnu-win32-request@cygnus.com" with one line of text: "help".