[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Fri, 16 Jan 2009 19:38:59 -0500
Gnus/5.110011 (No Gnus v0.11) Emacs/23.0.60 (gnu/linux)
>>>>> "Handa" == Kenichi Handa <address@hidden> writes:
Handa> The rule Emacs currently has for choosing a font for Unicode
Handa> Plane 1 is to find a font supporting a scpecifc script. For
Handa> instance, U+1D400 belongs to `mathematical' script, so Emacs
Handa> searches fonts supporting that script. And, xft font-backend
Handa> uses `script-representative-chars' to do such searching. Ad
Handa> the representative charcters of `mathematical' is U+1D400,
Handa> xft font-backend searches fonts that support U+1D400.
Because of how the font-wide hinting works, font families like DejaVu
have added the mathematical characters -- at least initially -- to
the specific faces which have similar glyphs.
Ie, the MATHEMATICAL ITALIC characters were added to the Serif-Italic,
MATHEMATICAL SANS-SERIF ITALIC to Sans-Oblique, etc.
As such, it would probably be a good idea for emacs to consider each
range of MATHEMATICAL characters as its own script, for the purpose
of choosing a default font.
The (sub-)families in the MATHEMATICAL block would be:
MATHEMATICAL BOLD ITALIC
MATHEMATICAL BOLD SCRIPT
MATHEMATICAL BOLD FRAKTUR
MATHEMATICAL SANS-SERIF BOLD
MATHEMATICAL SANS-SERIF ITALIC
MATHEMATICAL SANS-SERIF BOLD ITALIC
It /may/ also be necessary to diferentiate between latin and greek in
those ranges, but I suspect not.
James Cloos <address@hidden> OpenPGP: 1024D/ED7DAEA6