Re: [ft-devel] CFF subfont matrix problem

[Werner LEMBERG]

> > Assuming that both those fonts have been extracted from documents
> > which correctly work with Acroread (Ding Li, Dirck, please
> > confirm) I assume that it has some heuristic code to either select
> > [1 0 0 1] or [0.001 0 0 0.001] as the top-level matrix.
>
> The original font in the PDF file worked correctly in Acrobat.

Thanks for confirmation.

> The PLRM mentions that when constructing a CID type 0 composite
> font, the upper level font matrix is set to .001 etc and the lower
> level matrices are multiplied by 1000.  Not sure how this relates to
> CFF fonts or how they are constructed by the relevant tools.

It's kind of relevant.  Dave Lemon from Adobe provided useful hints
how to handle it correctly in private conversation.

> Our system has some heuristics for selecting the 1 or .001 matrix,
> but I thought it was an artifact of some other impedance mismatch
> problem in our architecture.

What you've done is correct (and I'll do the same in my forthcoming
patch to FreeType).


    Werner