Joe Conway <mail@joeconway.com> writes:
> I think the attached is what you had in mind. But I don't know right off
> how to trigger the failure (and therefore how to test the solution). A
> naive test with two databases, one LATIN2, the other UTF8 does not
> produce the error with simple text literals.
I can reproduce an error (and verify the patch corrects it) using this
test case:
select '�x�y'::text as x;
select * from dblink('dbname = u8', $$select '�x�y'::text$$)
as t1 (x text);
(The two non-ASCII characters are octal 340 and 367, if they don't come
through properly in your mail.) Execute in a LATIN1 database (being sure
client_encoding is also LATIN1), connecting to a database with encoding
UTF8. With the patch, both commands give the same results; without,
I get
ERROR: invalid byte sequence for encoding "UTF8": 0xe078f7
HINT: This error can also happen if the byte sequence does not match the encoding expected by the server, which is
controlledby "client_encoding".
CONTEXT: Error occurred on dblink connection named "unnamed": could not execute query.
Please get this committed soon, we have other stuff to get done
(like a pgindent run).
regards, tom lane
