Re: (patch) regression diffs on collate.linux.utf8 test

Jeff Davis <pgsql(at)j-davis(dot)com> writes:
> On Sun, 2011-10-16 at 16:00 -0400, Tom Lane wrote:
>> Hmm, yeah, I forgot to fix this regression test when I added that DETAIL
>> line. However, I don't see the need for fooling with the lc_time value?

> Here is the diff that I'm seeing on master right now with:
> make -s check EXTRA_TESTS=collate.linux.utf8

> If I qualify it as "tr_TR.UTF-8" it works. Perhaps I have something
> misconfigured on my system (Ubuntu 11.10)? I just installed:
> language-pack-de
> language-pack-tr
> language-pack-sv

That's very strange. It works as-is for me on Fedora 13, 14, and 15,
which you'd expect to have essentially the same I18N infrastructure as
Ubuntu (and I'm pretty sure I didn't install any optional language
support for Turkish). Anybody know what the problem is?

The reason I'm resisting just changing it is that I'd prefer to minimize
the number of dependencies this regression test has on the exact
spelling of "UTF-8", as that is not terribly well standardized. On my
Fedora boxes, for instance, locale -a says that "tr_TR.utf8" is the
name of that particular locale.

regards, tom lane