Thread: New Translation
hello, we are starting a new language translation in postgresql. I send the pg_controldata message translation in Farsi language. I sended it to the "Peter Eisentraut" with the name fa_IR.po. He said that "Is there any use of "fa" outside of "IR"? Else I would just call the translation files "fa.po"." there are some countries that uses this language. I see that there is a farsi translation in pgadmin and you named it fa_IR. But it's up to you what name you choose for it, some GNU packages named it fa and some fa_IR. It is the translation file attached.
Attachment
eetemadi@ce.sharif.edu wrote: > hello, > we are starting a new language translation in postgresql. > I send the pg_controldata message translation in Farsi language. > I sended it to the "Peter Eisentraut" with the name fa_IR.po. He said that > "Is there any use of "fa" outside of "IR"? Else I would just call the > translation files "fa.po"." > there are some countries that uses this language. > I see that there is a farsi translation in pgadmin and you named it fa_IR. This isn't comparable. In pgadmin, we're bound to the canonical names defined in wxWidgets, which is fa_IR (and there's no other Farsi version). Thus it seems reasonable to use .fa only if you have a choice. Regards, Andreas
Andreas Pflug <pgadmin@pse-consulting.de> writes:
> eetemadi@ce.sharif.edu wrote:
>> I see that there is a farsi translation in pgadmin and you named it fa_IR.
> This isn't comparable. In pgadmin, we're bound to the canonical names
> defined in wxWidgets, which is fa_IR (and there's no other Farsi
> version). Thus it seems reasonable to use .fa only if you have a choice.
Nonetheless, it would also be good to have some consistency between the
core PG server and related projects. I tend to agree that we should
honor pgadmin's precedent here; it's not a strong argument but the
argument for fa over fa_IR seems even weaker.
regards, tom lane
Tom Lane wrote: > Nonetheless, it would also be good to have some consistency between > the core PG server and related projects. I tend to agree that we > should honor pgadmin's precedent here; it's not a strong argument but > the argument for fa over fa_IR seems even weaker. "fa_IR" would uselessly restrict the range of users that would be able to use this translation. Translations are primarily targeted for a language, not a country. If wxWidgets is broken, that's not our problem. -- Peter Eisentraut http://developer.postgresql.org/~petere/
eetemadi@ce.sharif.edu wrote: > hello, > we are starting a new language translation in postgresql. > I send the pg_controldata message translation in Farsi language. Installed. -- Peter Eisentraut http://developer.postgresql.org/~petere/
Peter Eisentraut wrote: > Tom Lane wrote: > >>Nonetheless, it would also be good to have some consistency between >>the core PG server and related projects. I tend to agree that we >>should honor pgadmin's precedent here; it's not a strong argument but >>the argument for fa over fa_IR seems even weaker. > > > "fa_IR" would uselessly restrict the range of users that would be able > to use this translation. Translations are primarily targeted for a > language, not a country. If wxWidgets is broken, that's not our > problem. Please note that wxWidgets also depends on available locales. On Debian, locale -a returns fa_IR only, and guessing that from fa would be quite painful. OTOH, usual locale algorithms would strip the country fa_IR if not found. Regards, Andreas