utf-8 coding detection is not done

[abeld@…]

andrewgodwin on IRC told me to open a ticket for this:

if a models.py is in utf-8, with a "# -*- coding: utf-8 -*-" at the beginning of the file to note this, and it uses accented chars in, for example default field values, the migration south generates will be missing the "# -*- coding: utf-8 -*-" comment and thus fail to run.

From IRC:
<andrewgodwin> also, we used to detect the whole utf-8 coding thing, but since we stopped parsing that's failed. Add it as a ticket, I'll get it fixed for the next release

[amiroff@…]

Please keep in mind that there are also other methods for specifying encoding like # coding=utf-8.

[andrew]

Since south now uses introspection entirely, and converts all strings into UTF-8 on the way out, and specifies # encoding:utf8 on the file, this is now a non-issue, so I'm closing as fixed (I should have done a while back...)