Le Mar 19 mars 2013 09:31, Adam Williamson a écrit : > I don't object at all to fixing UTF-8 issues, but it seems needlessly > stressful to force ourselves to do so as a part of release validation, > with booby traps exploding all around us. It's the kind of thing that > can easily be worked on in a less stressful manner. And Fedora switched to UTF-8 by default when exactly (was it even Fedora or a RHL decision)? The less stressful manner seems to have failed utterly here, at some point some problems only get fixed (and not procastinated indefinitely) when software breaks. Moreover, experience shows any python tool (and we have tons of them system-side) will break on UTF-8 by default. And that python devs do not remember to test this case. Therefore, I doubt the problem could be realistically detected and fixed before branching point. Regards, -- Nicolas Mailhot -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel
