Lars,
Can you submit a test case that demonstrates the problem? The only
situation I have seen this is when you are storing 8bit characters in a
database declared as ascii. I have not seen any reports of this
happening when the character set of the database was an 8bit character set.
--Barry
Lars Stenberg wrote:
> Hi!
>
> A couple of days ago i had som serius problems with an database, the
> decodeUTF8() method in the org.postgresql.core.Encoding crashed
> (java.lang.ArrayIndexOutOfBoundsException: 12, at
> org.postgresql.core.Encoding.decodeUTF8(Encoding.java:254)).
>
> After some searches in the archives i found that i should create my
> database with another encoding (im using scandinavian chars). So i tried
> with Latin1, but it didnt work.
>
> The database is now in Unicode and i patched the Encoding.java to bypass
> the decodeUTF8 and just convert it to ISO-8859-1, it works but i still
> wants to know what the problem was, so i anyone could tell me what im
> done wrong i´ll be happy.
>
> Cheers
> Lars Stenberg