|Subject||Re: ISO-8859-2 problems|
--- In email@example.com, "Gabor Kover" wrote:
> How can I setup a field (CHARSET, COLLATION, etc.) to
> make FB's UPPER function to work with character
> codes other than the ASCII?
Apart from a desperate hack for Cyrillic, all Firebird
charsets without a COLLATE clause only uppercase 'a'..'z'.
But with an explicit COLLATE, you even have some choice:
For ISO-8859-2, use COLLATE CS_CZ - you may not agree
with the Czech about sort order, but I assume there are
no disagreements over uppercasing.
You can use WIN1250, which is a superset of ISO-8859-2,
and use one of the two collations PXW_HUN or PXW_HUNDC,
which should implement "Hungarian" and "Hungarian Dictoniary"
sort order and uppercasing.
For unknown reasons, PXW_HUNDC has same strange ideas
about uppercasing (uppercases 'á' but not 'ä'), so either
hungarian dictionaries are somewhat strange or an
error occured in creating this collation.
If you have any feedback about the working of these collation,
please do so, as I'm in the process of reviewing the entire