As Dan has said more details are necessary. I have never seen any issues on the engine side with storing and retrieving unicode characters. The only issues I have seen is using the proper jdbc interfaces correctly to insert and retrieve the data.
The store system uses a modified UTF8 encoding to store characters to disk, but that in no way should affect how users see the data when they retrieve it back. The encoding preserves the exact unicode encoding, and will give it back exactly as it was input. The UTF8 is merely a lossless encoding scheme to save some space on disk, and is completely hidden from all layers above the store. Knut Anders Hatlen wrote: > Daniel John Debrunner <[EMAIL PROTECTED]> writes: > > >>Knut Anders Hatlen wrote: > > > [...] > > >>>What's your locale? I think Derby gets problems with unicode when >>>LC_ALL (or LC_CTYPE) is invalid. >> >>Are there Jira entries for these problems? I'm not sure what could be >>going on, because Derby's use of Unicode is through Java and independent >>of any local setting (at least for the engine). > > > No, I don't think it is in JIRA. I don't remember the details, but I > once experienced that non-ASCII characters were saved on disk (both > log and db) as the UTF-8 replacement character (which means "unknown > character" and is often printed as '?'). Unfortunately, I'm not able > to reproduce it now, and it might just have been an issue between my > terminal and ij, not a Derby issue. >
