Re: testsuite failures when test scripts are run with zsh

I have rebased my private branch against master, and also amended the patch a bit (to deal with the test `cscope.test', which used a `$MAKE 2>stderr' redirection that I missed). The updated patch is attached. Regards, Stefano From e9666da90920c1373bfe6f9cbf8aa1e4e50c2b8a Mon Sep 17 00:00:00

Re: Remove dependency on gunzip

Hello Antonio, sorry for the delay. * Antonio Diaz Diaz wrote on Fri, Oct 30, 2009 at 09:09:47PM CET: > Ralf Wildenhues wrote: > >>As gunzip is a link to gzip, the only difference between using > >>"gunzip" and "gzip -d" is that the former adds a gratuitous > >>dependency on the gunzip name. > >

Use $DJDIR rather than $DJGPP to detect running on DJGPP.

I'm applying this patch to maint, in accordance with similar changes to Autoconf done a few weeks ago. I haven't actually tried to build and test git Automake on DJGPP, but $DJGPP is set in the environment even under my MinGW/MSYS and this caused bootstrap to fail. Cheers, Ralf Use $DJDIR ra

Fix requirements of silent8.test.

I'm applying this obvious fix to the jk-silent branch and merging to master. (I'll be pushing all of this out later today, I guess.) Cheers, Ralf Fix requirements of silent8.test. * tests/silent8.test: Require makeinfo --html, tex, texi2dvi -o, dvips. diff --git a/tests/silent8

Re: [Mingw-users] perl symlink oddity

[ adding automake-patches; please remove (subscription-only) mingw-users from replies unless it is on-topic there; this is ] * Ralf Wildenhues wrote on Fri, Dec 04, 2009 at 07:43:42AM CET: > * Charles Wilson wrote on Fri, Dec 04, 2009 a

Re: testsuite: fixed spurious failure in vala4.test (w.r.t. Solaris sh)

At Saturday 05 December 2009, Ralf Wildenhues wrote: > I've shortened the patch somewhat and adjusted the log entry > somewhat, hope you don't mind. No, it's OK. I wrote the extended description of the bug mostly to show that the patch weren't just papering over a not-understood problem. In fac

Re: testsuite: fixed spurious failure in vala4.test (w.r.t. Solaris sh)

[ adding autoconf-patches; this is ] Proposed patch for Autoconf. OK? Thanks, Ralf commit 4f89a570c7ccb07237d4a698a4e60db507880be0 Author: Stefano Lattarini Date: Sat Dec 5 12:56:30 2009 +0100 Document Solaris/Heirloo

Re: testsuite: fixed spurious failure in vala4.test (w.r.t. Solaris sh)

I've shortened the patch somewhat and adjusted the log entry somewhat, hope you don't mind. I've queued this for maint. (I would have queued the mode patches for maint as well, but they caused a conflict there, and I didn't figure porting was that important.) Thanks again, Ralf commit cce60a543

Re: testsuite: fixed spurious failure in vala4.test (w.r.t. Solaris sh)

Hi Stefano, thanks for working on this more. * Stefano Lattarini wrote on Fri, Dec 04, 2009 at 05:22:32PM CET: > I think I found the bug in heirloom-sh which was triggering the > spurious failure (it's related with `set -e', as you guessed). I also > found a simple workaround for it (caching t

Re: [PATCHES] Remove useless executable bits from two files

Hi Stefano, * Stefano Lattarini wrote on Fri, Dec 04, 2009 at 08:50:46PM CET: > I noticed that at least two files in the automake tree have the > executable bit on, without really needing it: > - the committed file `automake.in' (note that the similar file > `aclocal.in' is not executable)