On Thu, Mar 5, 2009 at 4:55 PM, Nils Kneuper <crazy-ivano...@gmx.net> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> David Philippi schrieb:
>> Am Donnerstag, 5. März 2009 13:05:00 schrieb Andrius Štikonas:
>>> does not have any means to display context information. Though the question
>>> is do we have to wait for them. After all, these are only about comments.
>>
>> Aren't those comments which differentiate otherwise identical strings? Those
>> are required to know the context of the string as a precondition for a
>> fitting translation. I guess most translators don't want to look up the
>> source reference to get the context.
>
> Hehe, those prefixes are for two things at the same time:
> 1) Make sure that the translator knows what is "special" about this string
> (like
> it is used for a female unit).
> 2) Make sure that a 2nd, different string *can* be used at all. If there was
> the
> string "General" for the unit name as well as the Menu entry, at least in the
> german translation this would make one of the two places have a wrong
> translation, since in gettext the string would only appear once. That is why
> the
> preferences string is "Prefs Section^General".
>
> So yeah, those "comments in string" blocks *do* make sense and something in
> the
> strings itself is required to have them differ from each other.
Example for Galician (old way):
msgid "General"
msgstr "Xeral"
msgid "Prefs Section^General"
msgstr "Xeral"
Example for Galician (with msgctxt):
msgctxt "Translators: this is a unit"
msgid "General"
msgstr "Xeral"
msgctxt "Translators: this is a menu entry"
msgid "General"
msgstr "Xeral"
I really think that the second way is the best one because the context
is clearly separated from the string to translate, making more
difficult to get wrong translators who don't understand that there is
a comment on the string to translate.
Bye,
Leandro Regueiro
_______________________________________________
Wesnoth-i18n mailing list
Wesnoth-i18n@gna.org
https://mail.gna.org/listinfo/wesnoth-i18n
