Re: [PATCH] Fix buglet in code searching for a good shell (tests/init.sh).

[Stefano Lattarini]

At Friday 11 June 2010, Jim Meyering wrote:
> > diff --git a/tests/init.sh b/tests/init.sh
> > index ef0957c..4eb4a95 100644
> > --- a/tests/init.sh
> > +++ b/tests/init.sh
> > @@ -97,12 +97,15 @@ else
> >      for re_shell_ in "${CONFIG_SHELL:-no_shell}" /bin/sh bash
> > dash zsh pdksh fail do
> >        test "$re_shell_" = no_shell && continue
> > -      test "$re_shell_" = fail && skip_ failed to find an
> > adequate shell +      if test "$re_shell_" = fail; then
> > +        echo "$0: skipped test: failed to find an adequate
> > shell" 1>&2 +        exit 77
> 
> Did you deliberately choose not to use the "(exit 77); exit 77"
> idiom from Exit()?  If so, why?
I thought  the '(exit ...); exit ...' idiom to be necessary only when 
an exit trap is in place, to ensure that the code in the trap is 
passed the correct exit status in $?.  Did I miss something?

> Did you consider simply hoisting the definitions
> of Exit and/or ME_ to precede their first use?
No, I just tried to fix the problem with the minimal amount of changes.
Anyway, the absence of a decent shell and the failure of the exec call 
are very very unlikely situations, so IMHO it's not a big deal if the 
program name (`$0' instead of `$ME_') is suboptimal in the error 
messages associated with those situations.

Of course, feel free to edit my changes as you see fit.

Thanks for your review,
    Stefano