Thread: Locale bug?

Hi,
if I choose in new version 8.0 for win32 czech locales then lc-collate is czech_czech republic.1250. So I suppose that
stringsare compared by WIN1250 charset, but database encoding can't be set as WIN1250 (only ISO8859-2=LATIN2) and then
ORDERBY don't work correctly. I try make initdb myself, but locales cs_CZ.ISO8859-2 is wrong name. Where I find
suportedlocales with corect names? Or where is problem?
 
Thx
Vaclav Vozar

Václav Vozár wrote:
> if I choose in new version 8.0 for win32 czech locales then
> lc-collate is czech_czech republic.1250. So I suppose that strings
> are compared by WIN1250 charset, but database encoding can't be set
> as WIN1250 (only ISO8859-2=LATIN2) and then ORDER BY don't work
> correctly. I try make initdb myself, but locales cs_CZ.ISO8859-2 is
> wrong name. Where I find suported locales with corect names? Or where
> is problem?

Did you set your client encoding?  If not, you are probably inserting
WIN 1250-encoded characters without the system knowing about it.  If
that happens, the ordering can be incorrect.

--
Peter Eisentraut
http://developer.postgresql.org/~petere/

>Did you set your client encoding?  If not, you are probably inserting 
>WIN 1250-encoded characters without the system knowing about it.  If 
>that happens, the ordering can be incorrect.

Yes, I set client encoding to WIN1250 and the ordering is incorect. I try set server encoding to MULE_INTERNAL, but
thisdont work to.
 

Vaclav Vozar