>
> > IMHO, it would be a very good idea to have a 100% posix-compliant
> > /bin/sh without any extensions. /bin/sh should *not* be symlinked
> > to /bin/bash.
>
> bash(1):
>If bash is invoked with the name sh, it tries to mimic the
>startup behavior of historical versions of sh
Yes it tries to mimic the startup behavior, but that doesn't mean it
limits its own functionality strictly to the POSIX bourne shell subset.
I think Alan's comment that there should be a /bin/sh that behaves
strictly to POSIX compliance has some merits, particularly to force
coders to write portab
Thilo Mezger <[EMAIL PROTECTED]> writes:
> IMHO, it would be a very good idea to have a 100% posix-compliant
> /bin/sh without any extensions. /bin/sh should *not* be symlinked
> to /bin/bash.
bash(1):
If bash is invoked with the name sh, it tries to mimic the
startup behavior of
Hi!
There's a thread on the guinness list about sh/bash/bash2 and I
just wanted to tell you my $0.02 of this /bin/sh vs. bash1/bash2 story.
IMHO, it would be a very good idea to have a 100% posix-compliant
/bin/sh without any extensions. /bin/sh should *not* be symlinked
to /bin/bash.
The rea
