Hello !

I would like to hear opnions from people using charset/collates in DBs with
portuguese language. I found many problems with the available collations...
explaining :

For charset ISO8859_1 the collate should be PT_PT, but in my view this is a
totally useless collate since it strips all the spaces from the string and
so you can get very strange results, like :

1) Josefa Joaquina
2) Jose Luiz

For WIN1252, the collate is PXW_INTL850. This solves the above problem, but
it is a pain when you expect equal comparisions with diferent chars, like :

João expected to be = Joao in comparisions
José expected to be = Jose in comparisions

if you need to do something like listing all the names starting with Jose
and you use STARTING WITH 'Jose' you miss all the José.

Also both collates seems to use 3 bytes (are 3 bytes really necessary for
portuguese ?) chars since I cannot make an index key > 83 chars.... too
bad, very annoying.

Any sugestions ?

[]s

Carlos
WarmBoot Informatica - http://www.warmboot.com.br
Interbase-BR - http://www.interbase-br.com