On Jul 13, 2008, at 10:31, Tom Lane wrote:
>> Grr. Kind of defeats the purpose. Is there no infrastructure for
>> testing multibyte functionality?
>
> There's some stuff under src/test/locale and src/test/mb, though it
> doesn't get exercised regularly. The contrib tests are a particularly
> bad place for trying to do any locale-dependent testing, because we
> only support "make installcheck" which means there is no way to set
> the database locale --- you *have to* work with whatever locale is
> predetermined by the postmaster you're pointed at.
>
> I don't recall the reason for not supporting "make check" in the
> contrib
> modules; maybe it was just that preparing a test installation for each
> contrib module sounded too slow? In any case, what you'd need to
> pursue
> is having some additional tests available that are not executed by the
> standard contrib regression test sequence.
>
> (If we get to having per-database collations in 8.4 then integrating a
> test with a non-default collation would get a lot easier, of course;
> but for the moment I'm afraid you've got to work with what's there.)
Could I supply two comparison files, one for Mac OS X with en_US.UTF-8
and one for everything else, as described in the last three paragraphs
here?
http://www.postgresql.org/docs/current/static/regress-variant.html
That way, I can at least make sure that the multibyte stuff *does*
work. Even if it's tested on only one platform, the purpose is not to
test a particular collation, but to test that CITEXT is actually
sensitive to locale.
Thanks,
David
