Re: [Devel] Re: extra metric points

[Antoine Leca]

Halo Werner,

On Tuesday, March 09, 2004 7:59 AM, Werner LEMBERG va escriure:
>
> :-) Let me reformulate.  Maybe four years ago I was aware of the
> problem, but meanwhile I completely forgot it.

:-)
There was no pun intended. In fact, the pun was to me, in that I forgot to
make clear this was something we have to address some day or another.

By the way, I believe we have to do a trip among all the XXX that are along
the code, to see if some are obsolete and if other need some degree of
attention.


>> I must add that obviously, the "n+2" and "n+3" do not come neither
>> from the 'glyf' nor the 'hmtx' tables. Since it is not required, the
>> 'vmtx' table is certainly not alone involved in this process.
>
> A good observation.  We could simply use the top and bottom of the
> local bounding box.

I believe from memory this is what is done. But I did not found it
immediately yesterday while grepping the source, and so I mention it for the
record!


>> When this table is missing, I do not know from where come the
>> informations. I remember Werner implementing some sort of hack to
>> deal with such fonts, and comenting the hack was not really
>
> Uh, oh, I don't remember at all.  What are you referring to?

I would have to dig the archives (at home). Probably this before 2000...
Former century...


>> The important point here, is that if we implement these vertical
>> metrics, we really should update the values returned by GETINFO[]
>> from the current 3 (meaning Windows 3.1, no vertical support) to
>> something more in line with current reality!  Since we do support
>> gray rendering, I believe 35 (Rasterizer 1.7) is a good value;
>
> Hmm, I'm not sure whether this is correct.  AFAIK, we don't have
> special code in the bytecode interpreter for gray rendering.

Here I really do not know what is the correct value, I never inspected the
real output from Windows rasterizer to see how it differs when using gray
rendering or not.

Only thing I understand, is that some fonts (probably Arial Times and the
like) have been enhanced to provide better result with grayscale
rendering/rasterizing. At least this is what I deduced from the different
announcements from Microsoft. I might easily be wrong.


Antoine