Quoting MJ Ray (firstname.lastname@example.org): > Question raised on -l10n-esperanto recently: can gettext be used for > localising a program with a utf-8 non-English source language? That is, > the thing in the _("...") has accents and isn't English. > > I hope I expressed the question adequately. The gettext manual seems to > stress using English everywhere, so the answer isn't clear to me. I > hope some people with relevant experience are reading -i18n. Others have anwered pretty clearly....there is no guarantee that this will work, in short... Allowing such ways to translate from another source than English is a topic that pops up here and now when talking about l10n tools. Offering the possibility to do such "derived" translations is among the TODO list of the Debian i18n server. Here the goal would be to allow translators to use another translation as the basis for their work. A use case was given by Rudy Godoy in a discussion I had once with him-->allow Quechua translators to work from the Spanish translation of programs instead of the English one because they speak Spanish very well while their knowledge of English is low. Of course, that example can be expanded to many other situations (translate from French for several African countries, etc.) This is maybe not the case you describe: another case would be translating an application orginally written in a non English language an,d just use gettxtfor its strings......Here, I'm not completely sure that there would be perfect solution.
Description: Digital signature