gettext() cannot find translations for messages that have plural forms. I would expect that gettext(s) is equivalent to ngettext(s, s, 1) for such messages, as it is implemented in GNU gettext. >>> import gettext >>> with open('test.mo', 'rb') as mo: trans = gettext.GNUTranslations(mo) ... >>> trans.ngettext("egg", "eggs", 1) 'Ei' >>> trans.gettext("egg") 'egg'
Hi there, I worked on a patch for this issue (see attached pull request). I would be happy to discuss it with people involved in gettext module maintenance but I'm not sure how to contact them. Is there a dedicated mailing list or an IRC channel or something? I've found i18n-sig but it has been inactive for 5 years... Best regards Gilles