RE: Find: missing alphabetically last dirtree

["Hannu E K Nevalainen \(garbage mail\)" <garbage_collector at telia dot com>]

> From: cygwin-owner@cygwin.com [mailto:cygwin-owner@cygwin.com]On Behalf
> Of Kleven Bingham

> Hannu E K Nevalainen (garbage mail) wrote:
> >>From: cygwin-owner@cygwin.com [mailto:cygwin-owner@cygwin.com]On Behalf
> >>Of Christopher Faylor
> >
> <SNIP>
> >
> >>Well, I'm fresh out of ideas, then.  I'm back to being unable
> to fix what
> >>I can't duplicate.
> >
> >
> >  :?7  sorry to hear that...
> >
> > I did try another thing; copy the contents of the CD to a
> FAT32(D:) and a
> > NTFS(F:) partition and then try my script on the result, using the
> > 2003-09-12 snapshot. Dunno if this tells us anything, but I was a bit
> > curious.
> > Outcome: FAT32 AND NTFS - find works as it should.
> >
> > Guess: there is something peculiar with the CDROM fs or contents!
>    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> You may be on to something here...
>
> I've been watching this thread the last couple of days and refrained
> from jumping in since I'm not entirely clear on the specifics of this
> issue, but this latest message threw up another red flag that made me
> think I might want to chime in.
>
> I've recently been playing with CD images and ISO image programs and
> found out that Microsoft often uses an internal app called cdimage.exe
> to generate the ISO images they use for CD distributions.  This program
> has some 'extended features' that are supposedly NOT ISO-9660 compliant.
>   One in particular allows you to 'compress' the image by scanning for
> identical files in the potential image and then only inserting the file
> data into the image once, while creating directory pointers to it 'from
> multiple locations.'  Basically, something that sounds a lot like hard
> linking.  The documentation I read stated that this DEFINITELY was not
> standard ISO-9660.  I was a bit surprised at that (not knowing much
> about ISO-9660), and I don't know how reliable that doc was, so I'm
> taking it with a grain of salt.

 This seems just to be a border case to multisessions on CD's for me, I'm
not convinced that it is 'standards-legal' to have it this way. It will most
likeley work in most cases.
 Here it seems as find and/or cygwin has problems in some undefined way. I
suspect that Chris probably has a great deal more clues about it ;-)

> If you haven't already Hannu, I would suggest you find a CD (or few)
> from another manufacturer or that you know is ISO-9660 compliant and
> test against that.  You very well may find some different results.

 I already know that the problem is _this particular CD_, as I said before;
it is the first in a set of three, on the other two 'find' does what is
expected.

 Come to think of it; might be something to try - if the above is indeed
true.
 I also have an idea on how to check for files with same contents (find,
md5sum, uniq - in combination).

> Or perhaps I don't have any clue what I'm talking about with respect to
> your actual problem. ;-)  Wouldn't be the first time!

 I have recollections of someone here... ahh... It _might_ be ME ;-)

My opinion and experience is that even the clueless can *at times* come up
with things that brings the discussion forth.

/Hannu E K Nevalainen, B.Sc. EE - 59?16.37'N, 17?12.60'E

Yet another sad "9/11" story
 -> http://cnn.com/2003/WORLD/europe/09/11/sweden.stabbing/index.html

--END OF MESSAGE--


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/