[pdftex] AR 5 not working with pdftex?
tjk at ams.org
Mon Jun 4 20:38:44 CEST 2001
pdftex alaread has *some* support for CFF fonts, but only by fully embedding
OpenType/CFf fonts. You are correct -- it is not an easy task to write a Type 1
-> CFF converter. Well, it might be "easy" if one writes an encoder that does
not take full advantage of what CFF has to offer.
I am trying to come up with a work around for the fully embedded font fiasco.
It has something to do with the built-in vs. custom encodings, and I haven't
found the right tack to take just yet.
Tomorrow will bring some kind of result, I hope.
On Tue, 5 Jun 2001 rhowlett at mail.usyd.edu.au wrote:
> It seems to me that the reason these problems don't show up with
> distiller produced pdf is because distiller embeds type 1 fonts
> in the pdf file in CFF form, and then Acrobat Reader manages to
> correctly unconvert them from CFF to the original type 1 format
> when sending the file to a postscript printer. (I've only got
> distiller 3.01, but that's how things appeared to me.) I presume
> that Adobe are using CFF in all their products; so the safest
> way for pdftex to go is to also use CFF. It has, of course,
> the added advantage of taking less space. I realize that it may
> be a difficult programming task to incorporate into pdftex a
> component that converts standard type 1 fonts to CFF for embedding
> purposes, but if there is anyone out there with the time and the
> knowledge needed for the task, that would be the best thing to do.
> Bob Howlett
> Quoting Tom Kacvinsky <tjk at ams.org>:
> > Well, that is no small job. xref table rebuilt, the size tags in the
> > dictionary
> > for the stream of the fontfile obect changed, the font disassembled, and
> > the
> > subrs array parsed *recursively* (subrs can call subrs, etc...) to catch
> > all
> > subrs references, etc...
> > Ugh...
> > Better for Thanh to "fix" this in pdftex. I really don't think pdftex
> > is broken
> > w.r.t. to gaps in the Subrs array, because the spec. is unclear on this
> > issue.
> > But die to the reality of the situation, its best if Thanh accomodates
> > what
> > Adobe is now doing. And I'll ask the Adobe font guys to fix their
> > documentation
> > in the Type 1 spec.
> > Tom
> > On Mon, 4 Jun 2001, Hans Hagen wrote:
> > >
> > > i can imagine a tool that postprocesses the pdf file by fixing the
> > arrays
> > > [adding dummies / replacing entries by short ones] with keeping the
> > > nofbytes the same.
> > >
> > _______________________________________________
> > pdftex mailing list
> > pdftex at tug.org
> > http://tug.org/mailman/listinfo/pdftex
> This mail sent through IMP: www-mail.usyd.edu.au
More information about the pdftex