From: Johan De Meersman
Date: June 12 2012 1:37pm
Subject: Minor collation question
List-Archive: http://lists.mysql.com/mysql/227631
Message-Id: <8cbf4969-484c-4716-9b83-952f3f83ef7f@zimbra>
MIME-Version: 1.0
Content-Type: multipart/alternative;
 boundary="=_2f16861e-ded0-4460-85d8-48e7d85e2292"

--=_2f16861e-ded0-4460-85d8-48e7d85e2292
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

... but too lazy to look for it myself right now :-p 

I'm going to be running into (woot, scheduled problems :-p ) the 5.0 to 5.1=
 upgrade collation issue where German =CE=B2 is now collated as 's' instead=
 of as 'ss', causing duplicate key errors. 

The basic solution is to set the collation to utf8_bin - so far, so good - =
but I'm wondering wether the table collation impacts only sorting, or wethe=
r comparisons will also magically become binary and/or other symptoms might=
 pop up ? I suspect 'yes' based on the primary key issue, but I want to ver=
ify. 

-- 

Bier met grenadyn 
Is als mosterd by den wyn 
Sy die't drinkt, is eene kwezel 
Hy die't drinkt, is ras een ezel 

--=_2f16861e-ded0-4460-85d8-48e7d85e2292--