Executable bit check

[Illia Bobyr]

Hi,

It seems that Cygwin does not check for executable bit before it allows 
a script to run, does it?

$ rm abc

$ cat <<EOF >abc
  > #!/bin/sh
  > echo Hm...
  > EOF

$ ll abc
-rw-r--r--+ 1 ibobyr Domain Users Nov 21  4 16:44 abc

$ test -x abc && echo Executable
Executable

$ ./abc
Hm...


I was able to find some references to a "patch for this [that is] ready" 
here: 
http://omgili.com/mailinglist/cygwin/cygwin/com/4A3D15705020506aimcom.html
Reply by Corinna Vinschen on Mon, 22 Jun 2009 15:11:44 +0200.

For some reason I do not see above mail in the cygwin.com mailing list 
archive (http://cygwin.com/ml/cygwin/2009-06/index.html).

For me the most annoying part is that "test -x" says that I can execute 
a file that I just did "chmod -x" on.

I'm looking at git-svn tests and they assume that "chmod -x abc && test 
! -x abc".  And it does not hold on Cygwin.

Accidentally I posted this on the cygwin-developers mailing list and 
already got the following reply (along with my answer):

On 11/04/2010 05:12 PM, Illia Bobyr wrote:
> On 11/4/2010 5:02 PM, Eric Blake wrote:
> > On 11/04/2010 03:57 PM, Illia Bobyr wrote:
> > > $ ll abc
> > > -rw-r--r--+ 1 ibobyr Domain Users Nov 21  4 16:44 abc
 > >
> > See that +?  That means there are ACLs in play.
> >
> > > $ test -x abc&&  echo Executable
> > > Executable
 > >
> > Likely, one of those ACLs is still making the file executable.
> >
> > What does 'getfacl abc' say?
 >
> $ getfacl abc
> # file: abc
> # owner: ibobyr
> # group: Domain Users
> user::rw-
> group::r--
> group:SYSTEM:rwx
> group:Administrators:rwx
> group:Users:r-x
> mask:rwx
> other:r--
>
> > > $ ./abc
> > > Hm...
> > One other thing to point out.  This will happen on FAT and other
> > brain-dead filesystems that lack true execute bits, where chmod -x has
> > no effect (since there's no bit to disable), such that cygwin always
> > represents the file as executable on those file systems if it has a
> > she-bang (well, depending on your mount options).  But given that FAT
> > also doesn't support ACLs, it's not the problem you are facing.
 >
> I'm on NTFS.  So, it is not that one.
> "chmod +x" and "chmod -x" do have visible effect:
>
> $ ll abc
> -rw-r--r--+ 1 ibobyr Domain Users Nov 21  4 16:44 abc
>
> $ chmod +x abc
>
> $ ll abc
> -rwxr-xr-x+ 1 ibobyr Domain Users Nov 21  4 16:44 abc*
>
> $ chmod -x abc
>
> $ ll abc
> -rw-r--r--+ 1 ibobyr Domain Users Nov 21  4 16:44 abc
>
> A newly created file is executable even without "chmod +x":
>
> $ echo "echo Hm... =)" > cba
>
> $ ./cba
> Hm... =)
>
> > > For me the most annoying part is that "test -x" says that I can execute
> > > a file that I just did "chmod -x" on.
> > There's more to permissions than just the 9 unix mode bits.  You really
> > CAN execute the file if test -x says you can; the question should rather
> > be why 'chmod -x' didn't remove all execute permissions, and that
> > usually boils down to ACL settings.