[Sugar-devel] [PATCH sugar-toolkit] Move the translation initialisation to an earlier stage, SL #3654

Simon Schampijer simon at schampijer.de
Tue Jun 5 02:28:55 EDT 2012


On 06/05/2012 08:21 AM, Simon Schampijer wrote:
> On 06/05/2012 08:06 AM, Simon Schampijer wrote:
>>
>>
>> Am 05.06.2012 um 07:03 schrieb Manuel Quiñones<manuq at laptop.org>:
>>
>>> 2012/6/4 Daniel Drake<dsd at laptop.org>:
>>>> On Mon, Jun 4, 2012 at 10:52 AM, Simon
>>>> Schampijer<simon at schampijer.de> wrote:
>>>>> To not depend on gconf at this stage the shell does set
>>>>> an env variable when reading the GConf key. This is handled
>>>>> in sugar-activity deployed by the sugar-toolkit-gtk3.
>>>>
>>>> Thanks Simon, these 3 patches look good to me. Unobtrusive way to
>>>> preserve the langpackdir functionality, and solves the issue in
>>>> question.
>>>
>>> Tested the patches in the XO with olpc build os12, compile details
>>> below. It fixes the issue for me in Physics, but not in Clock (I
>>> still see all tooltips in English, through I have the XO in Spanish).
>>>
>>> Compile details:
>>>
>>> I copied patched sugar, sugar-toolkit and sugar-toolkit-gtk. Then I
>>> satisfied dependencies:
>>>
>>> for sugar: yum install git make gcc intltool pygtk2-devel GConf2-devel
>>>
>>> for sugar-toolkit: yum install libICE-devel libSM-devel alsa-lib-devel
>>>
>>> for sugar-toolkit-gtk3: yum install gobject-introspection-devel
>>> librsgv2-devel gtk3-devel
>>>
>>> Then I run for each, ./autogen.sh; make; make install
>>>
>>> --
>>> .. manuq ..
>>
>> Thanks for testing Manuel.
>>
>> In the case of Clock actually the po file is borked. Have a look at
>> es.po for example, the strings like 'Nice Clock' are commented at the
>> end of the file, uncommenting them and rerunning 'setup.py build' does
>> bring them back for me. Except for the string 'Digital Clock' which
>> seem to have not the right context.
>
> Btw, I was not seeing this before as I was running the previous Clock
> version. So the issue has been introduced with the latest version.
>
> cc'ing Chris and Gary who might know where it broke and how to resolve it.
>
> Cheers,
> Simon

That looks suspicious: http://bugs.sugarlabs.org/ticket/3453#comment:9

Simon


More information about the Sugar-devel mailing list