Date: Mon, 15 Nov 2004 16:30:24 +0100 From: Roman Neuhauser <neuhauser@chello.cz> To: Avleen Vig <lists-freebsd@silverwraith.com>, Kevin Lyons <klyons@corserv.com>, freebsd-hackers@freebsd.org Subject: Re: tcsh fix Message-ID: <20041115153024.GA2846@isis.wad.cz> In-Reply-To: <20041112031122.GA87071@falcon.midgard.homeip.net> References: <41940880.7070409@corserv.com> <20041112023023.GG19417@silverwraith.com> <20041112031122.GA87071@falcon.midgard.homeip.net>
next in thread | previous in thread | raw e-mail | index | archive | help
# ertr1013@student.uu.se / 2004-11-12 04:11:22 +0100:
> The only real problem with having bash as /bin/sh is that people tend
> to write scripts using bash-specific features and forget that such
> scripts are not portable to systems using a less powerful /bin/sh.
Or the other way around. Bash (at least the one from RHEL 3.0) is
seriously broken, these two things are more than it can handle (as
opposed to our /bin/sh):
$(case $foo in
x) do_x;;
y) do_y;;
esac)
out=`mktemp ...`
$otherscript >$out 2&1
where $otherscript contains
$(x | tee /dev/stderr | y)
I stumbled upon these when I was porting periodic(8) to RHEL. So, my
description would be that people run into problems when they write
scripts using sh-specific features and forget that such scripts are
not portable to systems using a less powerful bash. ;)
--
If you cc me or remove the list(s) completely I'll most likely ignore
your message. see http://www.eyrie.org./~eagle/faqs/questions.html
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041115153024.GA2846>