Sorry I have to write "manual" replay since I've messed up mailing
list settings (got "Partial Digest"...).
John R Pierce wrote:
> your database encoding is UTF8 ? the language or environment you're using to generate those strings such as 'acząčž' is also UTF8 ?
Database created using:
initdb -D ../data -E utf-8 -U postgres
I have to take back statements about "lower()" and "~*" - they do not
work only for (strange) clusters created with "--no-locale".
But regexp_replace issue is still there. Regexp "\w" understands only
as "ascii word character" ?