On Sun, Oct 26, 2003 at 11:29:28AM +0100, Rein Klazes wrote: > On Sun, 26 Oct 2003 10:58:05 +0100, you wrote: > > > Hi, > > > > This fix supercedes Rein Klazes fix, since MS_MAKE_TAG is used in other places. > > Hmm thanks, I shoud have chacked that of course. > > Actually I was expecting the question why not use the definition in > freetype.h that comes with libfreetype? Or better use the > ready-to-be-used tags FT_ENCODING_MS_SYMBOL, FT_ENCODING_UNICODE and > FT_ENCODING_APPLE_ROMAN. > Makes it easier to read with the libfreetype's documentation as well. Because they changed during the freetype 2 release phase. I have freetype2-2.0.9, which has: FT_ENC_TAG( ft_encoding_symbol, 's', 'y', 'm', 'b' ), and with freetype2-2.1.4, which has: FT_ENC_TAG( FT_ENCODING_MS_SYMBOL, 's', 'y', 'm', 'b' ), hmm, and: #define ft_encoding_symbol FT_ENCODING_MS_SYMBOL Well, what about this patch instead: Ciao, Marcus Changelog: Use freetype 2.0 defines for font encodings instead of misuse of MS_MAKE_TAG. Index: dlls/gdi/freetype.c =================================================================== RCS file: /home/wine/wine/dlls/gdi/freetype.c,v retrieving revision 1.40 diff -u -r1.40 freetype.c --- dlls/gdi/freetype.c 15 Oct 2003 20:51:22 -0000 1.40 +++ dlls/gdi/freetype.c 26 Oct 2003 11:00:03 -0000 @@ -1249,16 +1249,16 @@ free_font( ret ); return 0; } - + if (ret->charset == SYMBOL_CHARSET && - !pFT_Select_Charmap(ret->ft_face, MS_MAKE_TAG('s','y','m','b'))) { + !pFT_Select_Charmap(ret->ft_face, ft_encoding_symbol)) { /* No ops */ } - else if (!pFT_Select_Charmap(ret->ft_face, MS_MAKE_TAG('u','n','i','c'))) { + else if (!pFT_Select_Charmap(ret->ft_face, ft_encoding_unicode)) { /* No ops */ } else { - pFT_Select_Charmap(ret->ft_face, MS_MAKE_TAG('a','r','m','n')); + pFT_Select_Charmap(ret->ft_face, ft_encoding_apple_roman); } ret->orientation = lf.lfOrientation;