You're right. The LIKE comparison only works properly with the Latin
letters in utf8. MySQL hasn't yet fixed it to work with the rest of Unicode
yet. I believe there is a valid LIKE comparison for ucs2. From what I
remember, however, it has problems too if you start putting wildcards at the
Hello and sorry for a dumb question, but this is my third day of experimenting
and surprisingly no luck with archives, so:
Is UTF-8 case insensitive search OK?
set @@character_set_client=utf8;
set @@character_set_server=utf8;
set @@character_set_connection=utf8;
set @@character_set_results=utf8;