Review Request 120353: Add timezones i18n stuff
Chusslove Illich
caslav.ilic at gmx.net
Wed Sep 24 17:44:48 UTC 2014
> On Септ. 24, 2014, 2:42 по п., Sebastian Kügler wrote:
> > I wonder if it's enough to just have a file with the i18n("City") in there, so it sends up in the .po file, and then wrap the untranslated city name into it. This way, we can avoid putting a huge Hash into the .cpp file, and just get everything from the .po file. As long as the same catalog is used, we don't actually have to have every single string in the code. This could reduce the size of the binary by quite a bit.
>
> Martin Klapetek wrote:
> Dunno, it could work. Let's hear from someone from i18n team.
>
> Martin Klapetek wrote:
> (also for the record, the difference in the binary is exactly 53 629 bytes, ie. 52 kB)
>
> Sebastian Kügler wrote:
> Hm, that's pretty significant. Would be nice if we could save that.
>
> I suppose the i18n could even go into a file that isn't built, javascript, or cpp-not-mentioned-in-cmake, so it doesn't even affect build time.
>From the translators' point of view it doesn't matter how this is implemented, as long as the strings find their way into the PO file and are properly used at runtime. In code, in dummy code, in special format file, at maintainer's discretion.
What does matter are strings themselves. Here I would add context to each, so that translators know what this list of names is used for. E.g.
i18nc("referent timezone location", "Someplace")
> On Септ. 24, 2014, 2:42 по п., Sebastian Kügler wrote:
> > applets/digital-clock/plugin/timezonesi18n.cpp, line 418
> > <https://git.reviewboard.kde.org/r/120353/diff/1/?file=315266#file315266line418>
> >
> > Does it really make sense to have translations added for UTC offsets, and UTC itself?
>
> Martin Klapetek wrote:
> I thought for some localizations it could be useful...like we put i18n around names in KAboutData.
>
> Sebastian Kügler wrote:
> Which ones do you have in mind? I don't know of any translation for UTC, but I'm not that well-informed.
>
> Lukáš Tinkl wrote:
> I would leave this as-is, although RTL langs might want to translate it somehow :)
>
> Martin Klapetek wrote:
> Mostly the part after UTC, which should be localized.
The rule is really simple: every piece of text that users may see must be exposed to translators. No exceptions, nothing further to think about.
- Chusslove
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/120353/#review67347
-----------------------------------------------------------
On Септ. 24, 2014, 2:02 по п., Martin Klapetek wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/120353/
> -----------------------------------------------------------
>
> (Updated Септ. 24, 2014, 2:02 по п.)
>
>
> Review request for Localization and Translation (l10n) and Plasma.
>
>
> Repository: plasma-workspace
>
>
> Description
> -------
>
> So in Plasma 5 we're adding support for timezones into the clock applet. It's based on QTimeZone and that uses IDs which are consisting of "Continent/City". Qt has no accessor methods for getting the city of the given timezone so we're parsing the timezone ID and using the part after "/" as the city and we display that in the UI.
>
> We want those cities translated however so it can display eg. "Praha" instead of "Prague" for Czech users. So I've created this helper class which has a dictionary of timezone-id-city<-->i18n(city) and this is then used in the UI. I'm not sure where exactly to put it, I'd prefer KI18n framework but as we would like to have this in Plasma 5.1 release, I put it together with the applet as Plasma 5.1 does not (and probably will not) depend on KI18n 5.3.
>
> I'm adding the l10n group to the reviewers - guys can you please comment/+1 this?
>
> If this is deemed good enough, I'd like to also add the country and continent i18n functions into that class afterwards.
>
>
> Diffs
> -----
>
> applets/digital-clock/plugin/timezonemodel.cpp PRE-CREATION
> applets/digital-clock/plugin/timezonesi18n.h PRE-CREATION
> applets/digital-clock/plugin/timezonesi18n.cpp PRE-CREATION
> applets/digital-clock/package/contents/ui/DigitalClock.qml 614e172
> applets/digital-clock/plugin/CMakeLists.txt 0590e6e
> applets/digital-clock/plugin/digitalclockplugin.cpp PRE-CREATION
> applets/digital-clock/plugin/timezonemodel.h PRE-CREATION
>
> Diff: https://git.reviewboard.kde.org/r/120353/diff/
>
>
> Testing
> -------
>
>
> Thanks,
>
> Martin Klapetek
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/plasma-devel/attachments/20140924/ae378f67/attachment-0001.html>
More information about the Plasma-devel
mailing list