[virt-tools-list] [libosinfo v2 2/2] Translate all (potentially) user visible strings

[Zeeshan Ali (Khattak)]

On Tue, Oct 30, 2012 at 4:26 PM, Christophe Fergeau <cfergeau at redhat.com> wrote:
> On Tue, Oct 30, 2012 at 02:50:16PM +0200, Zeeshan Ali (Khattak) wrote:
>> From: "Zeeshan Ali (Khattak)" <zeeshanak at gnome.org>
>>
>> Make some strings more translator-friendly at the same time.
>> ---
>>  osinfo/osinfo_deployment.c           |  4 +-
>>  osinfo/osinfo_devicelink.c           |  2 +-
>>  osinfo/osinfo_devicelinkfilter.c     |  2 +-
>>  osinfo/osinfo_entity.c               |  2 +-
>>  osinfo/osinfo_install_config_param.c |  4 +-
>>  osinfo/osinfo_install_script.c       | 42 +++++++++----------
>>  osinfo/osinfo_list.c                 |  2 +-
>>  osinfo/osinfo_loader.c               | 28 ++++++-------
>>  osinfo/osinfo_media.c                | 40 +++++++++---------
>>  osinfo/osinfo_os.c                   |  4 +-
>>  osinfo/osinfo_priv.h                 |  1 +
>>  osinfo/osinfo_product.c              | 12 +++---
>>  osinfo/osinfo_resources.c            | 10 ++---
>>  osinfo/osinfo_tree.c                 | 22 +++++-----
>>  po/POTFILES.in                       | 17 ++++++++
>>  tools/osinfo-db-validate.c           | 24 ++++++-----
>>  tools/osinfo-detect.c                | 35 ++++++++--------
>>  tools/osinfo-install-script.c        | 22 +++++-----
>>  tools/osinfo-query.c                 | 79 ++++++++++++++++++------------------
>>  19 files changed, 187 insertions(+), 165 deletions(-)
>>
>> diff --git a/osinfo/osinfo_deployment.c b/osinfo/osinfo_deployment.c
>> index 5f70e0d..c67456a 100644
>> --- a/osinfo/osinfo_deployment.c
>> +++ b/osinfo/osinfo_deployment.c
>> @@ -148,7 +148,7 @@ osinfo_deployment_class_init (OsinfoDeploymentClass *klass)
>>       */
>>      pspec = g_param_spec_object("os",
>>                                  "Os",
>> -                                "Operating system",
>> +                                _("Operating system"),
>
> spice-gtk does not translate any of the args to g_param_spec_object, gtk+
> does translate both the 2nd and 3rd arg. Dunno if you are doing something
> different from both on purpose?

There was lots of strings to decide about and it was a boring task so
I just made quick decisions: The 2nd argument is a 'nickname' so I
didn't see the point of translating that. No strong feelings either
way and I don't know if even description is something worth
translating.

>> diff --git a/tools/osinfo-db-validate.c b/tools/osinfo-db-validate.c
>> index 5b7274b..5a86af8 100644
>> --- a/tools/osinfo-db-validate.c
>> +++ b/tools/osinfo-db-validate.c
>> @@ -35,7 +35,7 @@ static gboolean verbose = FALSE;
>>
>>  static const GOptionEntry entries[] = {
>>      { "verbose", 'v', 0, G_OPTION_ARG_NONE, (void*)&verbose,
>> -      "Verbose progress information", NULL, },
>> +      N_("Verbose progress information"), NULL, },
>>      { NULL, 0, 0, 0, NULL, NULL, NULL }
>>  };
>
> Do these strings get translated without a call to
> g_option_context_set_translation_domain ?

Probably not, I'll add that call.

>>  static struct OsinfoLabel deployment_labels[] = {
>>      { OSINFO_ENTITY_PROP_ID,
>> -      "ID", TRUE, 40 },
>> +      N_("ID"), TRUE, 40 },
>>      { NULL, NULL, 0, 0 },
>>  };
>
> When you mark a string for translation with N_, you still need to use
> gettext() (which _() is #defined to) when you use it to get its
> translation, this does not seem to be done here.

Thanks for pointing it out, missed that from the docs.


-- 
Regards,

Zeeshan Ali (Khattak)
FSF member#5124