[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: -perm +... What should still work, what should not?
From: |
James Youngman |
Subject: |
Re: -perm +... What should still work, what should not? |
Date: |
Sun, 2 Oct 2005 20:31:05 +0100 |
User-agent: |
Mutt/1.5.9i |
On Sun, Oct 02, 2005 at 07:20:56PM +0200, Andreas Metzler wrote:
> My problem is in deciding which kind of strings "has a
> meaning in POSIX". e.g. +u+x or +g=w,u+x or ++s.
I also. The real answer is that if gnulib/lib/modechange.c
(mode_compile) likes it, then we believe that it has a POSIX meaning
as-is. Otherwise, we interpret an initial '+' as being equivalent to
'/'. So we rely on mode_compile() to be a faithful implementation of
the POSIX mode string. Paul (re-?) wrote that implementation and
suggested this strategy for backward compatibility. It seems
reasonable to me, since "find -perm" does need to be POSIX-compatible.
However, I accept that the current documentation could be more
helpful. It just codifies my current understanding of the state of
things. Sadly, like the documentation, that could be improved :(
Regards,
James.