Talk:Features/PythonEncodingUsesSystemLocale

From FedoraProject

Revision as of 21:55, 21 January 2010 by Poelstra (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

The disadvantage, such as it is, of this is that it means python2 scripts and programs written on Fedora can work fine here but fail with UnicodeErrors when trying to run it on a non-Fedora machine. Note that upstream python has refused to make this change. AFAIK, we'd be the first to make this change (although some of the python-dev posters propose that the distributions should change on their own; while others say that would be wrong). --abadger1999 22:06, 20 January 2010 (UTC)

One thing I don't like about this is that we lose consistency with upstream and the other distributions without even gaining consitstency between our python3 and python2 handling. python3 hardcodes utf8 for the system locale. This feature would make the encoding dependent on locale everywhere (unless overridden by a third party module like pango). This means that porting code from python2 to python3 would be transitioning from a locale-dependent encoding to hardcoded utf8 which could lead to bugs that no one outside of Fedora has experience with. I'd rather that we either change python2 to hardcode utf8 like python3 does or also change python3 to use the locale. --abadger1999 22:06, 20 January 2010 (UTC)