automake-patches
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH] Minor improvements and fixes in tests `depcomp*.test'.


From: Stefano Lattarini
Subject: Re: [PATCH] Minor improvements and fixes in tests `depcomp*.test'.
Date: Sun, 8 Aug 2010 14:23:22 +0200
User-agent: KMail/1.13.3 (Linux/2.6.30-2-686; KDE/4.4.4; i686; ; )

At Sunday 08 August 2010, Ralf Wildenhues wrote:
> 
> This one needs fixes.
> 
> > --- a/tests/depcomp2.test
> > +++ b/tests/depcomp2.test
> > @@ -43,11 +43,12 @@ END
> > 
> >  : > subdir/foo.c
> >  
> >  # Ignore user CFLAGS.
> > 
> > -CFLAGS=
> > -export CFLAGS
> > +unset CFLAGS
> 
> Unsetting a variable which is not set can cause nonzero exit
> status;
True; and I should have known better: even bash 2.x does this!  Fixed.
 
> > --- a/tests/depcomp6.test
> > +++ b/tests/depcomp6.test
> > 
> > @@ -20,11 +20,11 @@
> > 
> >  set -e
> > 
> > -cat >> configure.in << 'END'
> > +cat >>configure.in <<'END'
> 
> Tsk.
Sorry, changed slipped in while in "tweaking berserkr" mode.
Reverted.  Ditto for tests/depcomp7.test.

> > 
> > +  # Do not use `$MAKE && Exit 1' here, since even relatively-recent
> > +  # versions of the BSD shell wrongly exit when the `errexit' shell
> > +  # flag is active if a command within "&&" fails inside a compound
> > +  # statement.
> > 
> >    if $MAKE; then Exit 1; fi
> 
> I guess that is missing 'else :;' then, too, so that the result of
> the compound command is zero.  Likewise depcomp7.test.
Done, and ChangeLog updated accordingly.

Thanks,
  Stefano



reply via email to

[Prev in Thread] Current Thread [Next in Thread]