🚀 go-pugleaf

Henry McGilton writes
> 
>                                               What's interesting
> is that the FontBBox entry in both of these fonts is completely
> bogus, although the FontBBox in their supplied AFM files appears
> to be correct.  Does anybody have any ideas why the font dictionary
> notion of FontBBox is whacko and the AFM file version is OK?.

You don't say how they're bogus, but is it possible that the scale
factor for the font is unusual, and that the FontBBox within the
font is interpreted in the context of the FontMatrix?  Adobe uses
[0.001 0 0 0.001 0 0] for their FontMatrix, which effectively means
that all the coordinates are 1000 times bigger than the 1-point master,
but that's not a requirement.  I've seen FontMatrices of [1 0 0 1 0 0]
that would make things significantly different.

I'm not sure if FontBBox inside the font is expressed in terms of
FontMatrix or not.  But it is used by the cacheing machinery (typically,
although it's up to BuildChar), so it shouldn't be completely bogus
(although it could be too big without too much ill effect)

--
 Glenn Reid        		NeXTmail: glenn@rightbrain.com
 RightBrain Software		415-326-2974 (NeXTfax 326-2977)
 Palo Alto, California