2006/10/13, Martijn van Oosterhout <kleptog@svana.org>:
> While sorting for multiple languages simultaneously is an issue, that's
> not the problem here. Linux/GLibc *does* support correct sorting for
> all language/charset combinations, and that's what he's using. Just for
> the hell of it I setup lv_LV.utf8 on my laptop and verifed that it
> sorts just fine:
...
> Similarly, upper/lower are also supported, although postgresql doesn't
> take advantage of the system support in that case.
I think this is the crux of the problem. Not supporting uppercase and
lowercase makes an e.g. generic people search dialog not malfunction:
searching for "Müller" will not find him if he is stored as "OTTO
MÜLLER" in the database. Certainly not if I have to make sure the
search finds one "Zvonimir Šimić" stored as "ZVONIMIR ŠIMIĆ". Whats
more, if the user gives up on the integrated search and tries to list
all the people in such a database ordered by their last names, he
probably won't find Šimić because the user expects him to be between S
and T, not after 'Z' (where he ends up beacuse the letter code of 'Š'
is greater than that of 'Z').
As for Martins' problem, he needs to support 2 non-english languages
which means he's stuck with the same problem of one language being
semi functional.
As an aside, why doesn't pg take advantage ot the underlying system's
support of upper/lower case?
t.n.a.