Re: [Devel] Question about FreeType 2

[Vincent Caron]

address@hidden wrote:
> Can I access the glyphs in the TrueType face by unicode and not by
> character code?

Font files have 'charmaps' which provide a conversion from a given 
encoding (Unicode, KOI8, GB2312, etc) to its internal glyph indexes. 
FT_Load_Char() will load a char using the currently selected charmap. It 
means that you must know which charmap is currently used. Font files can 
embed multiple charmaps. It is useful to know that the TrueType loader 
(and maybe others ?) automatically select the Unicode charmap if there's 
one.

You can use the 'ftdump' program bundled in ft2demos to display the 
available charmaps from a font file (however it shows cryptic numbers). 
See FT_Select_Charmap() and the 'charmaps' field of FT_FaceRec.

http://www.freetype.org/freetype2/docs/reference/ft2-base_interface.html#FT_Select_Charmap
http://www.freetype.org/freetype2/docs/reference/ft2-base_interface.html#FT_FaceRec


> What encoding does FT_Load_Char(..) assume for character code? 7-bit
> ASCII/8-bit ASCII?

It uses the input encoding of the currently selected charmap. Since 
FT_ULong is 32bit under 32bit platforms, it expects UCS4 when the 
charmap is a Unicode one.