[Devel] Auto-fitter nearly completed

[David Turner]

Hello,

  I have just commited the latest round of fixes and enhancements to the
  "autofit" module. The latter is a replacement for the "autohint" that
  should sport better code, and more importantly the ability to differentiate
  the hinting algorithm based on scripts/languages. The choice is made by
  parsing the Unicode charmap when there is one (otherwise, everything defaults
  to "Latin" hinting, as usual)

  The idea is to be able to use different algorithms to hint Latin, Arabic
  or CJK fonts, for example. This would allow the inclusion of Firefly's
  patches within the main FreeType sources without distorting the quality of
  Latin letters (like the current patches do). In other words, to get the
  best of all worlds. I also hope that the code's new structure will help
  in making experimentation with alternative algorithms easier as well.

  The code seems rather solid now, and should produce output that is 100%
  equivalent to the current auto-hinter. Believe me, this has been hard :-)

  I'd appreciate if some of you could start testing it. To do so, you simply
  need to change the line in <freetype/config/ftmodule.h> that says:

     FT_USE_MODULE(autohint_module_class)

  with

     FT_USE_MODULE(autofit_module_class)

  then recompile FreeType. I strongly suggest only running the ft2demos
  programs with it, since I didn't have the time to use valgrind with a
  big number of font files (i.e don't install the modified FreeType unless
  you know what you're doing)

  Voila, I hope to get this code rock solid soon, in order to incorporate
  other people's patches.

Best Regards,

- David Turner
- The FreeType Project  (www.freetype.org)