Re: [tlug] Search MySQL for Japanese Names [SOLVED... presumably]

[Curt Sampson <cjs@example.com>]

On 2009-10-20 17:59 +0900 (Tue), Dave M G wrote:

> Okay then.  The system I'm working on has limited users on a private
> network, so I'll just make a kana input for now, and hope that MySQL
> collates and handles it properly.

You really ought to check that; I would be far from surprised if it
didn't do that correctly.

If I'm recalling correctly (it's been a while since I've touched I18N
stuff), one sorts characters with (the little "ten-ten" mark indicating
you voice the consonant) after those without dakuten, and handaukuten
(the little circle) after those, but only when the remainder of the two
strings are identical. Otherwise, one ignores the diacritical signs and
sorts based on the rest of the string. This can lead to a code point
with dakuten appearing both before and after the code point without, in
the same character position.

Here's an example of "ku" appearing both before and after "gu":
[ くどう, ぐどう, くに, ぐに ].

cjs
-- 
Curt Sampson       <cjs@example.com>        +81 90 7737 2974
           Functional programming in all senses of the word:
                   http://www.starling-software.com