Re: chmod RFE

[James Youngman]

On Sat, Jul 16, 2005 at 12:26:28PM -0700, Dave Yost wrote:
> At 10:28 AM +0100 2005-07-16, James Youngman wrote:

> >Thanks for your interest.  Maybe you're right.  Please explain in more
> >detail why you think that the MacOS X implementation of chmod(1) is
> >superior to the GNU coreutils implementation.  If you can be specific,
> >and we agree, perhaps we can improve things.

> The compelling examples for me is this:
> 
>   chmod g=u,o=u-w foo
> 
>   set the group bits equal to the user bits.
>   set the other bits equal to the user bits,
>   but clear the other write bit.

I think I understand what you mean.  In fact, coreutils already does
this (though obviously the version you tried must be older if it
doesn't do it):-

address@hidden:~/tmp$ sh example
+ rm -f foo
+ touch foo
+ chmod 300 foo
+ ls -l foo
--wx------  1 james users 0 2005-07-16 22:14 foo
+ chmod g=u,o=u-w foo
+ ls -l foo
--wx-wx--x  1 james users 0 2005-07-16 22:14 foo
+ chmod --version
chmod (coreutils) 5.2.1
Written by David MacKenzie and Jim Meyering.

Copyright (C) 2004 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.


> Also, the man page is much more detailed and has examples.

This criticism has been made of coureutils more generally before.
However, see the thread which starts at
  http://lists.gnu.org/archive/html/bug-coreutils/2005-03/msg00081.html
and also see
  http://www.gnu.org/prep/standards/standards.html#GNU-Manuals

> It would seem to me to be a good idea to start a project to evaluate
> all the features of the various BSD versions of core-utils for
> possible adoption and thereafter to track changes continuously.

I'm sure the coreutils maintainers (I'm not one of them) would find
this stimulating and often useful.  Please feel free to contribute and
assign patches.  If you are going to do this, you should start from
the current CVS code so that your patches can me more easily applied
(see http://savannah.gnu.org/cvs/?group=coreutils).

While I remember to mention it, other list members have responded to
your original email, but you won't have seen the responses since their
email was blocked by some kind of challenge-response thing.  Since you
are also not subscribed to this list, you won't have been able to see
those.  I suggest you fix one or the other, otherwise this gives the
appearance that you're just doing a drive-by flame.

Anyway, I look forward to seeing some interesting new technology in
some patches.  Thanks for your contribution.

Regards,
James.