I suspect this got added because of `eautoconf` (skipping the test
confused by that now) but not sure. Passes for me now, anyway.
It was added by:
commit 01ba51b126378518559db6a0188773d918036753
Author: Diego Elio Pettenò <flameeyes@gentoo.org>
Date: Sun Aug 19 02:23:59 2012 +0000
Version bump; remove old. This version closes bug #209667 (disabled ucd-snmp compatibility); bug #328177 (lm_sensors 3 is supported by default now); #425990 (added GPL-2 to LICENSE);
should close bug #351051 as well, for the most part; add new init scripts to use the new runscript features; restrict tests (tests that might or might not fail are useless); build in pa
rallel (upstream fixed it); drop USE flags that added no dependencies.
(Portage version: 2.2.0_alpha120/cvs/Linux x86_64)
It's very possible that it got fixed substantially since then upstream
though. In any case, all fine now.
Signed-off-by: Sam James <sam@gentoo.org>
* Fix build w/ C23
* Unrestrict tests by dropping dmalloc requirement
It dates back to:
commit a46f63115ab7e5db64009173f0b8662f0965f943
Author: Samuli Suominen <ssuominen@gentoo.org>
Date: Thu Jun 4 19:02:15 2009 +0000
RESTRICT test because it needs too old dmalloc we don't ship anymore.
(Portage version: 2.1.6.13/cvs/Linux x86_64)
But tests for memory leaks aren't really something we're interested in
from a packaging PoV, plus dmalloc was removed a little while ago. There's
tests which work fine w/o dmalloc so let's allow running those.
Closes: https://bugs.gentoo.org/946659
Signed-off-by: Sam James <sam@gentoo.org>
Tests in the past were automagic based on whether dev-util/cppcheck was
installed. That's long-fixed since 0.5.2 which moves that into a separate
target.
Unfortunately, filtering LTO for now though as we can't test it effectively
b/c of ODR issues in tests.
Bug: https://github.com/tdf/libcmis/issues/12
Bug: https://bugs.gentoo.org/577926
Signed-off-by: Sam James <sam@gentoo.org>
The restriction has been there since the package was added in
a58112e0200336c6563cbaf17a4d9b52e970e815 but the tests pass for me
w/o liblangtag installed. Fedora has a workaround I may pull in later
though (setting LD_LIBRARY_PATH for tests).
I can't immediately see what would've fixed this in liblangtag,
but perhaps it was https://bugs.gnu.org/71489 (https://gray.gnu.org.ua/public/ltbug).
Signed-off-by: Sam James <sam@gentoo.org>
Checked the resulting hash of files, no diff at all, so I feel safe with
straight to stable revbump.
Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>