From: Junio C Hamano <gitster@pobox.com>
Date: Sat, 22 Sep 2007 02:33:21 GMT
Subject: Re: [PATCH] Allow shell scripts to run with non-Bash /bin/sh
Message-ID: <7vlkazh1ji.fsf@gitster.siamese.dyndns.org>
In-Reply-To: <20070921214346.GF97288@void.codelabs.ru>

Eygene Ryabinkin <rea-git@codelabs.ru> writes:

> Good day.
>
> I had found that FreeBSD's /bin/sh refuses to work with git 1.5.3.2.
> correctly: no flags are recognized.

> @@ -109,7 +109,7 @@ dotest=.dotest sign= utf8=t keep= skip= interactive= resolved= binary=
>  resolvemsg= resume=
>  git_apply_opt=
>  
> -while case "$#" in 0) break;; esac
> +while case "$#" in 0) break;; *) : ;; esac
>  do

I am assuming that this works around _a_ bug in that /bin/sh; I
would make sure I understand the nature of the bug.  Is it Ok to
understand that with that shell, after this construct runs:

	case <some word> in
        <case arm #1>)
        	something ;;
	<case arm #2>)
        	something else ;;
	esac

the status from the whole case statement is false, when <some word>
does not match any of the glob patterns listed in any of the case arm?

That is, what does the shell say if you do this?

	case Ultra in
        Super)
        	false ;;
	Hyper)
        	true ;;
	esac &&
        echo case returned ok

The reason I ask is because

	while case $# in 0) ... esac
        do
        	...
	done

is not the only place the status from "case" itself matters in
our scripts.  There are places that do

	something &&
	case ... in
        ...
        esac &&
        something else

and we would need to add no-op match-everything arm to all of
such case statements in our scripts.

Besides test scripts, there is one in git-ls-remote.sh which you
seem to have missed.