I've applied the patch below to a temporary bug-fixing branch, merged
it to maint, merged maint to master, and pushed. The testcase still
works with GNU make, FreeBSD make and Solaris make, but I couldn't
test if it really fixes the problem with HP-UX make and IRIX make.
Regards,
Stefano
-*-*-*-
distlinksbrk.test: Work around botched "make -k".
* tests/distlinksbrk.test: Run "make" multiple times and grep
its output each time for a single error message, rather than
running "make -k" one single time and grepping its output for
all the expected error messages. This should work around make
implementations with limited (broken?) `-k' support; for more
information, see these subthreads on the automake-patches list:
- 2010-11-15, "Testsuite failures on HP-UX 11.23",
<http://lists.gnu.org/archive/html/automake-patches/2010-11/msg00162.html>
- 2010-11-15, "Testsuite failures on IRIX 6.5",
<http://lists.gnu.org/archive/html/automake-patches/2010-11/msg00166.html>
---
ChangeLog | 14 ++++++++++++++
tests/distlinksbrk.test | 31 +++++++++++++------------------
2 files changed, 27 insertions(+), 18 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index cd301e3..cc6b22a 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,17 @@
+2010-12-22 Stefano Lattarini <stefano.lattar...@gmail.com>
+
+ distlinksbrk.test: Work around botched "make -k".
+ * tests/distlinksbrk.test: Run "make" multiple times and grep
+ its output each time for a single error message, rather than
+ running "make -k" one single time and grepping its output for
+ all the expected error messages. This should work around make
+ implementations with limited (broken?) `-k' support; for more
+ information, see these subthreads on the automake-patches list:
+ - 2010-11-15, "Testsuite failures on HP-UX 11.23",
+
<http://lists.gnu.org/archive/html/automake-patches/2010-11/msg00162.html>
+ - 2010-11-15, "Testsuite failures on IRIX 6.5",
+
<http://lists.gnu.org/archive/html/automake-patches/2010-11/msg00166.html>
+
2010-06-21 Stefano Lattarini <stefano.lattar...@gmail.com>
Add a test checking that distributed broken symlinks cause
diff --git a/tests/distlinksbrk.test b/tests/distlinksbrk.test
index 1e478ab..2314408 100755
--- a/tests/distlinksbrk.test
+++ b/tests/distlinksbrk.test
@@ -49,30 +49,25 @@ test -h $lnk2
test -h $lnka
test -h $lnkb
-cat >>configure.in <<END
+cat >> configure.in <<'END'
AC_OUTPUT
END
-cat > Makefile.am <<END
-EXTRA_DIST = $lnk1 $lnk2 $lnka $lnkb
-END
-
-ls -l
-
$ACLOCAL
$AUTOCONF
-$AUTOMAKE
-./configure
-# Distribution must fail.
-$MAKE distdir && Exit 1
+ls -l # for debugging
-# Names of distributed broken symlinks should be reported in make output.
-$MAKE -k distdir >out 2>&1 || : # don't trust the exit status of make -k
-cat out
-$FGREP $lnk1 out
-$FGREP $lnk2 out
-$FGREP $lnka out
-$FGREP $lnkb out
+# Don't try to use "make -k", because some botched make implementations
+# (HP-UX, IRIX) might still exit on the first error in this situations.
+for lnk in $lnk1 $lnk2 $lnka $lnkb; do
+ echo "EXTRA_DIST = $lnk" > Makefile.am
+ $AUTOMAKE
+ ./configure
+ # Distribution must fail, with a decent error message.
+ $MAKE distdir >out 2>&1 && { cat out; Exit 1; }
+ cat out
+ $FGREP $lnk out
+done
:
--
1.7.1
