From: Andre Wobst <wobsta@us...> - 2006-02-20 21:21:00
On 20.02.06, Joerg Lehmann wrote:
> this point is probably better suited for the devel list, but anyway...
Right. So let's move to the dev-list.
> On 20.02.06, Andre Wobst wrote:
> Isn't it kind of strange that we need to set the font mapping file via
I don't think so. Note that the mapping files are something very TeX
specific. TeX just doesn't know what the fonts really are. It's
subject to the dvi reader to understand, what kind of typesetting TeX
want's us to do. TeX font names should *not* be passed to the backend.
> I know that we currently need the mapping file in the dvifile
> module, but we sure can get rid of it there. And somehow this sounds
> like being the right thing to do, at least from a conceptual point of
> view. And then only the writers would need to know about the mapping
I disagree. We should not get the map files out of the dvi, since
without TeX where would be no need for the mapping files at all.
(Sure, fonts needs to be reencoded to make the glyphs available in an
8-bit-encoding, but it's an artifact, that the final output represents
the same reencoding as declared in the mapping file. I don't argue to
remove this coincidence, but it clearly is a coincidence. At some
point we might even want to break that. But it's not an issue at the
We could move all the "TeX-fontnames to the final output" features
into our output devices, but I don't think it's The Right Thing to do.
What we really want to have is something like: For PostScript we can
take the builtin font XY and reencode it the following way, while for
PDF, where there is no builtin font XY and we need to do this and
that. *This* is a generic feature, which we need to implement, but
it's not at all generic, that the source of that information are some
stupid TeX mapping files.
by _ _ _ Dr. André Wobst
/ \ \ / ) wobsta@..., http://www.wobsta.de/
/ _ \ \/\/ / PyX - High quality PostScript and PDF figures
(_/ \_)_/\_/ with Python & TeX: visit http://pyx.sourceforge.net/