Re: ucs-2 server - yes. usc-2 client - no. What the ....?

[Randy]

Haven't checked on this thread for a while.

Cristi and Mike,
Thanks for responding to my point. Cristi, as Mike pointed out, UTF-16/UCS-2 uses a predictable, consistent 2 bytes for every unicode codepoint/char. UTF-8 can use from 1 to 4 bytes. That means UTF-8 could be using 4 bytes for codepoints that UCS-2 will still only use 2 bytes for. Thus, if you know you'll be dealing with data that would take 3 and 4 bytes to encode in UTF-8, it's a big savings to use UCS-2 instead.

Mike, you pointed that mysql does support UCS-2. Right, but the whole point of my post was the bummer that if I have an application that uses UCS-2 as its character set, I cannot connect to mysql as a UCS-2 client and send/receive string data encoded as UCS-2. I'll have to encode/decode to/from UTF-8. That's heavy price to pay, and... I must conclude that the jdbc driver does the same since java is natively a UTF-16 app.

Am I missing something? Mike, any idea when this will be fixed?

Thanks,
-randy