GNOME Bugzilla – Bug 334291
stupid I18N strings :)
Last modified: 2017-08-17 20:17:22 UTC
#: ../cut-n-paste/zoom-control/ephy-zoom.h:51 msgid "50%" msgstr "50%" #: ../cut-n-paste/zoom-control/ephy-zoom.h:52 msgid "75%" msgstr "75%" #: ../cut-n-paste/zoom-control/ephy-zoom.h:53 msgid "100%" msgstr "100%" #: ../cut-n-paste/zoom-control/ephy-zoom.h:54 msgid "125%" msgstr "125%" #: ../cut-n-paste/zoom-control/ephy-zoom.h:55 msgid "150%" msgstr "150%" #: ../cut-n-paste/zoom-control/ephy-zoom.h:56 msgid "175%" msgstr "175%" #: ../cut-n-paste/zoom-control/ephy-zoom.h:57 msgid "200%" msgstr "200%" #: ../cut-n-paste/zoom-control/ephy-zoom.h:58 msgid "300%" msgstr "300%" #: ../cut-n-paste/zoom-control/ephy-zoom.h:59 msgid "400%" msgstr "400%" Please merge all these strings. Thanks.
How? Those are needed because there are languages that do translate numbers.
i said 'merge'. "%d%%" is our friend.
%d%% won't put in localised digits. I guess %Iu%% would work. If you try that and it works, the master copy of ephy-zoom-control in epiphany should be updated too.
Give me an example, i've grep'ed over all po files and couldn't find any thing. I don't get your point. What are localised digits ? what's wrong with %d ? are every format string doomed ?
I see ... ' I For decimal integer conversion (i, d, u) the output uses the locale's alternative output digits, if any. For example, since glibc 2.2.3 this will give Arabic-Indic digits in the Persian (`fa_IR') locale. ' any example ? why use both %d and %Id ?
Closing as WONTFIX based on the need to localise digits in some languages. Feel free to reopen if you have a better way to tackle this issue.
what about %I !? and btw, just grep evince/po and let me know if you find a language that doesn't translate "100%" to "100%" ... nobody has been able to give me a real example up to now. So tell me why should i waste time translating every number in every gnome app if a) there's no real language that uses alternative digits b) if there was any, it could be worked around with %I. Does the HIG say something about using alternative digits ? What are alternative digits after all ? i've reported the same bug in #333039 . closing this bug as WONTFIX means to me that %d can't be used anymore.
Anyhow this belongs to ephy-zoom, probably we should reassign this bug to epiphany or libegg
(In reply to comment #7) > what about %I !? > >a) there's no real language that uses alternative digits There is, e.g. Arabic, Persian > b) if there was any, it could be worked around with %I. Does the HIG say > something about using alternative digits ? What are alternative digits after > all ? http://en.wikipedia.org/wiki/Hindu-Arabic_numeral_system Djihed
Now have a look at how these teams have translated these strings. I would be very curious to know if any gnome module has a translation for 100% that is not '100%' What's so special about % so that it needs special formatting ? I'm looking for a guideline that would tell us when %I should be used instead of %d. Btw, if %I fits, that would still kill a lot of strings.
Benoît, it's not up the language alone. In Arabic, locales like ar-ae (emirates) would write arab-hindu numbers, while ar-dz (Algeria) would have the normal numbers. If left to the Language translators, for the sake of consistency we translate it all to normal numbers. That doesn't make it correct. I don't see how the fact that it's not used ATM takes away any merit in having it properly.
Just tell me when to use %d or %I. If %d or %I is not correctly handled by your locale, then this is NOTGNOME.
As far as I understand, developer should use %d, and translator can switch to %Id (please confirm), but that's currently problematic because of this: Bug 375690 probably why not many translations used it. I'm not sure if that's the way it is everywhere :)
OK i understand. So we can merge all these string right now as "%Id%%" ?
I suspect we need to fix cvs.gnome.org first then.
These strings are still in the po files in git. See http://git.gnome.org/browse/evince/tree/po/ar.po so what is the fix for this? At least the version of this bug could be raised ...
Still valid in current git, has there been any conclusion?
ping. Can we close this bug as WONTFIX? Numbers are translated like in: https://git.gnome.org/browse/evince/tree/po/ps.po#n311
Closing as per comment #18.