-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Edward Z. Yang wrote:
Brion Vibber wrote:
If MySQL supported UTF-8, we'd be happy to
make use of it. Using proper
character sets gives us warm, fuzzy feelings and makes it easier to work
in terminals and other direct-database tools, as well as potentially
making it easier to use built-in database support for case-insensitive
lookups and proper sorting. [snip]
On a related tack, why isn't MediaWiki using MySQL 5 yet? (I remember
there was a very good reason, but I can't find it now)
1) The biggest reason we'd want to upgrade would be to make use of
native unicode support, but it doesn't work ;)
There are a few other niceties in 5, but not a lot of incentive to
upgrade considering that...
2) To upgrade cleanly without using the (broken) unicode support, we'll
want to finish working out and testing a proper binary schema and a
migration to it.
The good news is that we've more or less got the hang of doing clean
schema upgrades using replication and switching masters around. So once
we get the schema details worked out, we *should* be able to do the
upgrade with minimal downtime.
In theory. :)
1) Does mysql support UCS-2, UTF-8, or UCS-4 beyond the BMP? If so we
can still use UTF-8 internally and rely on mysql's encoding conversion
for the rest.
2) Has a bug report or feature request been filed on mysql's site? I'm
fumbling through now trying to find one.