> I've just been asked how to store unicode text in a postgresql database. The
> problem as I understand it is that unicode strings may contain binary 0s
> which might break string handling. Since I never tried, I think it's better
> to ask here before answering that question.
I guess you are talking about UCS encoding. There is another encoding
for Unicode, called UTF-8. It does not 0s, so you could use it with
PostgreSQL. Actually we have some unicode(utf-8) regression tests in
src/test/mb and they seem working.
> The application uses ecpg so it's not only a backend question.
I guess ecpg is ok as long as using UTF-8.
--
Tatsuo Ishii
