Follow-up Comment #4, patch #3481 (project freeciv): I've prepared an en_US translation (without diacriticals, but just `make ${PROJECT}.pot; msginit`). Unfortunately, the mentioned comments (I searched for `U+` appearing in comments) seem to assume there is a one-to-one relationship between a language domain and a character set, which isn't even true in Europe anymore (ISO-8859-1/ISO-8859-15/UTF-8/UCS2 are all very common), let alone places with more complex orthography.
Do we use bind_textdomain_codeset in a way that makes it safe for me to use UTF-8 representations in the translation (I'm happy to do en_GB at the same time, if this is the case)? Note that we appear to use UTF-8 in the rulesets for strings not subject to translation (e.g. city names, ruler names). Alternately, is my understanding of gettext outdated? Is it now safe to declare a character set for msgid so that we may just use UTF-8 there? _______________________________________________________ Reply to this item at: <http://gna.org/patch/?3481> _______________________________________________ Message sent via/by Gna! http://gna.org/ _______________________________________________ Freeciv-dev mailing list Freeciv-dev@gna.org https://mail.gna.org/listinfo/freeciv-dev