>>> it has been fixed a long time ago to no longer mangle the high byte >>> of UTF-16 characters. Below you can see the output I get if I call >>> `ftdump -n -u simhei.ttf` >> >> This version of ftdump correctly returns the name "黑体" of the >> font. Which brings me back to my original question: what do I have >> to do to get this name too when using the API? > > You have to convert the UTF-16 output to UTF-8. Please look into the > source code of `ftdump`.
OK, so we are back at the very beginning:for this specific font, FT_Get_Sfnt_Name() does NOT return a valid UTF16-BE representation of the name "黑体". It seems to work properly for most of the other fonts I have but for this font the returned data are invalid!