Closed D-Bullock closed 5 years ago
According to https://www.gnu.org/software/gettext/manual/gettext.html#Preparing-Strings there is nothing that prevents us from using Unicode in our msgId
and here https://www.gnu.org/software/gettext/manual/gettext.html#Names it is even encouraged.
Years ago, when source code would be ASCII, special characters would be written out in the msgId
, in your case the source code would be "say^TM". And the translater would then be responsible for using the ™ symbol if it was available in his particular character set. But nowadays everyone should use UTF-8, and at least some form of Unicode so that is not really an issue (outside East Asia at least).
The task is to change 'ascii' to 'utf-8' and make tests and examples that ensures the result works.
Failing tests are on the new https://github.com/robert-j-engdahl/ngettext-wpf/tree/msgid-unicode-support branch.
Currently if one is using unicode characters (say ™) then it's replaced with a ? in the xamlmessages.pot file
I think XGetText-Xaml.ps1:61 is to blame