Re: [ft-devel] Experimental: v38 interpreter with minimal backwards compatibility mode and linear advance widths

[Werner LEMBERG]

>> A misunderstanding.  I believe that Alexei meant
>>
>>   #define TT_CONFIG_OPTION_SUBPIXEL_HINTING   1   /* infinality */
>>   #define TT_CONFIG_OPTION_SUBPIXEL_HINTING  40   /* new code */
>>   #define TT_CONFIG_OPTION_SUBPIXEL_HINTING  41   /* both */
>>
>> or something similar.
> 
> I.. don't get it. How is that a practical improvement over splitting
> TT_CONFIG_OPTION_SUBPIXEL_HINTING into
> 
> #define TT_CONFIG_OPTION_SUBPIXEL_HINTING_V38
> #define TT_CONFIG_OPTION_SUBPIXEL_HINTING_V40
> 
> ? That would remove the currently used macro and more accurately
> describe what it's doing. Or does the current macro have to stay?

I think we would have the following benefits.

. There is a lot of code that is identical to the Infinality and lean
  code, which could be then simply tested with

    #ifdef TT_CONFIG_OPTION_SUBPIXEL_HINTING

. In the future, if we are going to remove the Infinality code, we
  could again have a simple

    #define TT_CONFIG_OPTION_SUBPIXEL_HINTING

  in the configuration file, ignoring the macro's actual value.


     Werner