>> ----
>> return $m['HTML']('Привет');
>> ----
>> now:
>> ----
>> return $m['HTML']('\xd0\x9f\xd1\x80\xd0\xb8\xd0\xb2\xd0\xb5\xd1\x82');
>> ----
> Ok, I looked in translator_proto.py sources and found following change:
> string constants now passed thru new 'uescape' function, which
> (intentionally) pass them thru repr().
more info: new code use 'pytranslator' instead of 'translator' module.
Looks like pytranslator doesn't process# -*- coding: utf-8 -*- line at the start of source file. Translator 'parseFile' call produces ast Const(u'\xd0\x9f\xd1\x80\xd0\xb8\xd0\xb2\xd0\xb5\xd1\x82') instead of (correct one) Const(u'\u041f\u0440\u0438\u0432\u0435\u0442') for unicode literal. First one is not actually a unicode, but undecoded utf-8 value. Seva
