Unicode is hard – harder than I assumed

In my last post I announced that the GExperts code formatter is now Unicode aware. Little did I know. Mohamed Kamel sent me a source file with Arabic strings which still got converted to question marks. So today I have dived into the source code again. I nailed the problem with Arabic strings (and added a new unit test). But I am pretty sure I missed some more.

I’ll release a new version as soon as I got feedback from Mohamed.

Anybody got some source code with Japanese and Chinese strings for me?