Grouping by initial letter

[Glenn Nilsson]

Hi,

I have a database/table utf8_unicode_ci. The contents is in german, but I will have several installations of the app with different collations (french, russian as a start). In german the letter å, ä, ö and ü exists but are treated by the collation as a, a, o and u respectively, they're just a "normal" letter with a twist. This is all fine and dandy.

Österreich, Oman, Schweden, Ägypten, Argentinien
becomes:
Ägypten, Argentinien, Oman, Österreich, Schweden

Here the problem; I want in my front end group the data by initial letters, and a DISTINCT(LEFT(name, 1)) should give me good data, but in the special case above, I get "Ä", "O", "S" and not "A", "O", "S".

So, is there any way the fetch the data from a collation point of view, and not data point of view?

And since I want this general for many collation I can't (easily) handle this in the code in my app.

Thanks
/Glenn

PS: I'm using MySQL v5.0.45.



Edited 1 time(s). Last edit at 08/15/2007 05:41AM by Glenn Nilsson.