Re: Accents bug ?

dbucher(at)niftycom(dot)com (Denis Bucher) writes:

> At 11:48 02.10.01 +0200, Rene Pijlman wrote:
>
> >I've heard of conversion problems before, but not of chunking
> >the data though.
>
> Yes, that's the most strange I think :-)

I have seen the same thing. It happened to Norwegian characters like
æøå. Strings read through the JDBC driver were truncated at the first
non-ASCII character. I found that it worked using the jdbc7.0-1.1.jar
version of the driver though. I don't remember all of the details
right now, but the database encoding was UNICODE and I had inserted
the data from an earlier database dump from a Latin-1 database. The
dump file used for INSERTs is in the Latin-1 character set.

If I can reproduce the problem I will try the getBytes() method on the
result set to get a byte array back and verify if the bytes returned
are legal UTF-8 sequences. But ... looking at the source ... that
doesn't appear to be supported by the driver(?). Maybe I'll have to
try to compile the driver and put some debug statements in it.

The database server was 7.1.2 an I think I reproduced it using a 7.1.3
backend as well. Is there a good way to get some exact version
numbers from a precompiled driver jar file in case I find a good way
of reproducing the error and want to give a better bug report?

Pardon the bad excuse for a bug report, I just wanted to let Denis
know that he is not the only one who has seen this bug.

-Knut