These are the tests that didn't fail in the release candidate because they were
covered by another 'REQUIRES' directive.
Details
- Reviewers
bcraig hans mclow.lists EricWF - Commits
- rG3906d4ae416c: Merging r258920: --------------------------------------------------------------…
rG90c86ea5f12a: [libcxx] Additional 'REQUIRE' directives for tests that require en_US.UTF-8.
rCXX258920: [libcxx] Additional 'REQUIRE' directives for tests that require en_US.UTF-8.
rL258920: [libcxx] Additional 'REQUIRE' directives for tests that require en_US.UTF-8.
Diff Detail
Event Timeline
There's only 7 instead of the 8 we were expecting because one match turned out to be the python script that checks for the availability of the locale.
+dim who said the locale tests were also failing for him on FreeBSD. Maybe this is relevant?
LGTM. May want to wait on mclow though considering the trouble I helped cause with the last one of these.
Thanks.
May want to wait on mclow though considering the trouble I helped cause with the last one of these.
Sure. That wasn't your fault though, it was mine. I should have asked for clarification first.
Thanks
@dsanders: What platform are you on that doesn't have en_US.UTF-8? If your serious about testing libc++ you should install the locale. (but I understand that's not always possible).
It's Debian Jessie but it was configured for the Europe/London region during installation so I had en_GB.UTF-8 instead.
I've installed en_US.UTF-8 (and the other missing locales like ru_RU.UTF-8) on all bar one of the machines I build releases on. The last machine lacks en_US.UTF-8 (but has the others) so I can test this patch before I commit it. After the commit I'll enable it.