Heya, been over a year since xLua's been brought up, but now that MMF2.5 is running on a Unicode base, I noticed some difficulties with strings in xLua. It appears that if a Unicode string is passed via parameters in an MMF2 function call it will normalize it into ascii for Lua's strings and be able to compare, take lengths of them, etc, just fine, but if you directly pull a string from an object via the MMFI, ie, object.getstring(id,0), it will pull the 2-byte unicode string array instead of the normalized 1-byte string, and any comparisons to it with static strings will fail.
I'm trying to think whether there is any way this could or should be fixed on either on MMFUnicode's end or on the xLua object, or if its worth it to port over one of the unicode normalization libraries (compile a .dll and use it as an external library?). I could probably just create a table look-up for my own purposes.