[Localization] [etoys-dev] Sugar release and new Etoys translations

[Clytie Siddall]

Hi everyone :)

On 10/03/2010, at 1:12 AM, K. K. Subramaniam wrote:

> On Tuesday 09 March 2010 03:32:35 pm Bert Freudenberg wrote:
>> Err, the point is these entries should not be translated.
>> 
>> E.g., "Language-Name" should be "Hindi" in Hindi script, CJL suggested
>> हिन्दी or हिंदी. This is how the language name will appear in the language
>> menu - the comments are not stellar but they do say that specifically.
> My misunderstanding, Sorry. I would have expected meta-data like language 
> names or fonts to be built into the application (locale support library to be 
> specific) rather than pulled from a po file. In any case, the preferred display 
> name would be "Hindi (हिन्दी)". This allows readers to know which language is 
> involved when a proper font or rendering engine is missing on the host.
> 
>> The other entries should be left empty, unless you need to set them
>> specifically to make it work. E.g. Linux-Font should be left empty, or the
>> name of a specific font name to use. Better leave empty.
> I would say leave it out for now. There are multiple rendering engines on 
> Linux and it is best to pull supported fonts at run-time.

Also, if we _are_ going to have meta-data strings presented to translators, please add contextual comments (msgctxt [1] or dot comments). As usual, the Debian Installer project sets a good example (from Level 3, win-32-loader):

#. translate:
#. This must be the string used by GNU iconv to represent the charset used
#. by Windows for your language.  If you don't know, check
#. [wine]/tools/wmc/lang.c, or http://www.microsoft.com/globaldev/reference/WinCP.mspx
#.
#. IMPORTANT: In the rest of this file, only the subset of UTF-8 that can be
#. converted to this charset should be used.
#: win32-loader.sh:52
msgid "windows-1252"
msgstr "windows-1258"

#. translate:
#. Charset used by NTLDR in your localised version of Windows XP.  If you
#. don't know, maybe http://en.wikipedia.org/wiki/Code_page helps.
#: win32-loader.sh:57
msgid "cp437"
msgstr "cp1258"

#. translate:
#. The name of your language _in English_ (must be restricted to ascii)
#: win32-loader.sh:67
msgid "English"
msgstr "Vietnamese"

#. translate:
#. IMPORTANT: only the subset of UTF-8 that can be converted to NTLDR charset
#. (e.g. cp437) should be used in this string.  If you don't know which charset
#. applies, limit yourself to ascii.
#: win32-loader.sh:81
msgid "Continue with install process"
msgstr "Tiê'p tu.c cài đa(.t"

#. translate:
#. The nlf file for your language should be found in
#. /usr/share/nsis/Contrib/Language files/
#.
#: win32-loader.c:68
msgid "English.nlf"
msgstr "English.nlf"

Without those instructions, I would really have messed those pages up. :S

Note: this is the base PO file. In Pootle, the comments shown here after "#. " (or using msgctxt) will appear underneath the original string.

from Clytie 

Vietnamese Free Software Translation Team

[1] * GUI program support:
  - PO files can now contain messages constrained to a certain context.
    Most often such a context is a menu, dialog or panel identification.
    The syntax in the PO file is
      msgctxt "context"
      msgid "original"
      msgstr "translation"
  - The xgettext program can be told through the --keyword flag which
    function/macro argument has the role of a context.
  - The (non-public) include file gettext.h defines macros pgettext, dpgettext
    etc. that take a context argument.
  For more information, see the node "Contexts" in the manual.

http://www.gnu.org/software/gettext/manual/html_node/Contexts.html#Contexts