"ca" is already UTF-8 so there's no problem from this side ;) El dom, 15-07-2007 a las 00:17 +1000, Malcolm Tredinnick escribió: > Hi all, > > Short version: anybody unhappy if we *require* PO files to be encoded as > UTF-8? For almost everybody, this means no change at all. For "es" and > "es_AR", I will have to re-encode the current files (they are > ISO-8859-1) to UTF-8 and, in future, you will need to edit them as > UTF-8. I would like to make sure I'm not going to make things impossible > for anybody by asking for this. > > Longer version (with explanation): > > Working my way through thinking about a bunch of i18n tickets and > problems. One popular request is to allow non-ASCII msgid strings (the > original, untranslated strings). > > I'd always believed the xgettext man page that said this wasn't possible > for Python. Turns out that the docs are wrong and it is possible. So we > have a nice patch in #4734 that works well and I'm happy with it. > > The only catch is that we now have to agree on an encoding for PO files > because xgettext is going to be potentially putting non-ASCII in there, > which means an encoding is required. Typically (on most Open Source > projects I've worked on in the past, including big ones like GNOME), > UTF-8 has been the right choice. This isn't a big deal for core code, > although it might be useful in localflavors at some stage, but it will > help make the documentation easier. Note that if we can't agree, you > guys don't get to use non-ASCII msgid strings. > > Aside from "es" and "es_AR", the only other non-UTF-8 file is > djangojs.po in "fr", but that hasn't been updated for a while and than > main (django.po) "fr" file is in UTF-8, so I figure it's not an issue > there. > > Regards, > Malcolm >
signature.asc
Description: Esta parte del mensaje está firmada digitalmente
