RE: [Devel] unpatented hinting system for Dynalab fonts, etc

[Graham Asher]

David,

thanks for your reply.

<<<<<<
I'm however surprised that this change would require a significant
change in freetype.h (apart
from the definition of a new FT_LOAD_XXX flag or an additional API).
>>>>>>

It doesn't, now - I found a way of simplifying it.

<<<<<<
Have you experimented with the NO_APPLE_PATENT macro in "ttinterp.c" ?
>>>>>>

No. Raph Levien told me that in his opinion these alternative pieces of code
didn't do the job of avoiding the patent properly, I believe because freedom
and projection vectors are still stored. The system he has commissioned from
me doesn't store these permanently, or calculate a dot-product. However, I
shall ask discuss this further with the Artifex people (and anyone else who
has any ideas).

<<<<<<
Could you provide an example of
the new public
types you're expecting and relative changes ?
>>>>>>

Luckily, no new public types. At present I expect to do everything in
ttinterp.c except for the addition of

#define FT_LOAD_UNPATENTED_HINTING           0x10000U

to freetype.h

and the new configuration macro

//#define TT_CONFIG_OPTION_UNPATENTED_HINTING

in ftoption.h

and some very minor changes to a few other files (ttgload.c sets an
interpreter flag saying whether unpatented hinting only is to be done,
etc.).

<<<<<<
PS: sorry if I seem so snotty sometimes :-)
>>>>>>

Not at all - it is your right to oversee changes. I am far, far worse when
somebody suggests changes to my software. In my younger days I tried not to
let people see it, which may well have been an unintended kindness.

Best regards,

Graham