> I'll report the results, for the record. Okay, for the record, all went well. I re-initialise my PostgreSQL 7.4 database cluster using the following command:
initdb --locale=C --encoding UNICODE Then, after defining the relevant groups and users, I used pg_restore to restore my data from a dump file. And now everything works as expected with both European (Latin-1) and Japanese names and words. Sorting with ORDER BY on European names and words is adequate for my requirements, even though the C locale ordering is case sensitive. As Tatsuo Ishii pointed out, I can use lower/upper to get case insensitive ordering if I ever need it. The bottom line seems to be: if you're planning to store data with languages from across the world, initialise your database cluster with the C locale and Unicode encoding. Thanks again to everyone who replied! Kind regards Harry Mantheakis London, UK ---------------------------(end of broadcast)--------------------------- TIP 2: Don't 'kill -9' the postmaster
