Bash bug?

[Jeff Vian]

On Mon, 2005-09-26 at 14:42 -0700, BRUCE STANLEY wrote:
> Then this is a departure from original/standard System V
> behaviour. Again, all System V based Unixes I have programmed
> under did have this behaviour.
> 
> Anyone converting scritps (as I had to do) will run into
> this problem/behaviour.
> 
> 
I think the key here is the subshell behavior which is to be expected.
Scope of variables is important, and a subshell (as has already been
stated) cannot affect the values of variables in it's parent.

This behavior is posix compliant, as well as unix standard. (I tried the
same on an AIX machine with the same results).

People may run into problems at any time when the original is not 100%
posix compliant and/or has been written to take advantage of sometimes
lax standards in the OS.

Think of it this way.  The fixes you do will make the original better
and more standardized in the long run.  You are killing bugs.  I know
the work now is frustrating, but the end product is improved.


> 
> --- Tim Waugh <twaugh at redhat.com> wrote:
> 
> > On Mon, Sep 26, 2005 at 06:15:21AM -0700, BRUCE STANLEY wrote:
> > 
> > > This seems to be a Linux issue with the shells.
> > 
> > It is correct behaviour, and POSIX-compliant to boot.
> > 
> > Please see the bash FAQ, question E4.
> > 
> >   /usr/share/doc/bash-3.0/FAQ
> > 
> > Tim.
> > */
> > > -- 
> > fedora-list mailing list
> > fedora-list at redhat.com
> > To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
>