Univention Bugzilla – Bug 27283
Some UMC tracebacks are wrong encoded in UTF8 (errors are ISO8859-1)
Last modified: 2017-04-28 09:07:33 CEST
Tracebacks enthalten oft � anstatt Umlauten/Sonderzeichen/UTF-8. Es wäre schön, wenn das mal gefixt wird.
Ist das ggf. mit Bug 27256 behoben?
(In reply to comment #1) > Ist das ggf. mit Bug 27256 behoben? Nein
Given in latin-1 are: * Messages from PAM * Messages from python-apt * OSError exceptions (e.g. errno 12, cannot allocate memory)
The UMC-Moduleprocess/Server only sets LC_MESSAGES but not LC_ALL: >>> import os, locale >>> locale.setlocale(locale.LC_MESSAGES, 'de_DE.UTF-8') 'de_DE.UTF-8' >>> os.strerror(12) 'Nicht gen?gend Hauptspeicher verf?gbar' >>> locale.setlocale(locale.LC_ALL, 'de_DE.UTF-8') 'de_DE.UTF-8' >>> os.strerror(12) 'Nicht gen\xc3\xbcgend Hauptspeicher verf\xc3\xbcgbar'
I think it is the best to set: locale.setlocale(locale.LC_ALL, 'C.UTF-8') so, LC_COLLATE is still the same. (Just if scripts called by UMC depend on this).
Created attachment 6914 [details] patch small patch, only sets LC_CTYPE to minimize side effects. """ LC_CTYPE: This category determines the interpretation of byte sequences as characters (e.g., single versus multibyte characters), character classifications (e.g., alphabetic or digit), and the behavior of character classes. """
Could a similar issue be related to Bug 38370?
(In reply to Alexander Kläser from comment #7) > Could a similar issue be related to Bug 38370? I don't think so.
Works since UCS 4.2.