Encoding Conversion
Hi,
When I´m doing a conversion PostgreSQL returns SCAPES to me besides the desire
caracter. Why ?
Using Postgresql 8.2:
SELECT CONVERT('Rejeição: Duplicidade de NF-e, com diferença na Chave de
Acesso [35110100608804000178550010001009471840996034]"','UTF8','LATIN1')
Result: "Rejeição: Duplicidade de NF-e, com diferença na Chave de Acesso
[35110100608804000178550010001009471840996034]: "
Using Postgresql 9.0:
SELECT CONVERT('Rejeição: Duplicidade de NF-e, com diferença na Chave de
Acesso [35110100608804000178550010001009471840996034]"','UTF8','LATIN1')
Result: "Rejei\347\343o: Duplicidade de NF-e, com diferen\347a na Chave de
Acesso [35110100608804000178550010001009471840996034]""
Both databases are "encoding latin1".
Both servers are "Linux".
Regards,
Wilton Ruffato Wonrath
Wilton Wonrath <wrwonrath@yahoo.com.br> writes:
When I�m doing a conversion PostgreSQL returns SCAPES to me besides the desire
caracter. Why ?
Somewhere between 8.2 and 9.0 convert() was redefined to take and return
bytea, not text, to reflect the fact that the data it deals in isn't
necessarily encoded in the database's character encoding. This was part
of a general tightening up to ensure that text that ought to be in the
database encoding really is.
There are some related functions now that convert between text (assumed
to be in the database encoding) and bytea strings in a specified
encoding. You probably want to switch over to using those.
regards, tom lane