[Fribidi-discuss] Some thoughts about BiDi Quirks Handling

[Dov Grobgeld]

Hi Behdad and everyone else,

First of all I must say that I agree with you that I don't think
that we should have any MicroSoft quirks hardcoded in the fribidi
core itself. Giving an API through which it is possible to change
the BiDi classification of any characters sounds very reasonable 
though. Could someone knowledgable explain if that is enough to
support the various historical M$ behaviours?

Regarding the support of different encodings, I sent a suggestion
a while ago that the only place that should need modification is
at the entry point to the algorithm in fribidi_analyse_string(). 
Appearently I had missed though that the FribidiChar type is queried
later on in the algorithm, but Behdad wrote that might be
solved by adding another flag in the bitfield currently used to 
hold the character properties. If this is this what you are planning 
to do, Behdad, then I don't understand where you would need the
macros?

Regards,
Dov