utf8_unicode_ci column accent sensitive

[Nils Meier]

Hi

I've been trying to figure this out, reading the docs and searching the forum. It seems to me that utf8_unicode_ci should be accent insensitive (from what I read).

I have a table with a column and its collation of utf8_unicode_ci. I can put in utf8 data, select it again and e.g. 'München' will match 'München' nicely (all done through php serving the utf8 back to the browser successfully).

If I look at it through mysql command line i see:

+----------+
| city |
+----------+
| München |
| Muenchen |
+----------+

Apparently there's a double byte character which I guess is as expected from the latin1 client's point of view. No problem.

But when doing a

select * where city = 'Munchen'

I'm not getting the result i'd expect - instead of the collation giving me the first entry I get none. Shouldn't the 'u' of 'Munchen' match the u-umlaut in the utf8 column?

What am I missing here?

Thanks a lot
Nils


---
show full columns gives
| city | varchar(64) | utf8_unicode_ci | ...
--
status gives

mysql> status
--------------
mysql Ver 11.18 Distrib 3.23.58, for redhat-linux-gnu (i386)

Server version: 4.1.12
Protocol version: 10
Connection: mysql4-g via TCP/IP
Client characterset: latin1
Server characterset: latin1
TCP port: 3306
Uptime: 11 days 10 hours 50 min 33 sec