Re: [docbook-apps] Source of Docbook + MathML -> PDF examples?

["Chris Chiasson" <chris@chiasson.name>]

I wouldn't suggest converting through Mathematica either. I was just
telling you how I did it because it might seem weird to have three
sources for each equation (PNG raster, EPS vector, MathML) if one
didn't know it came from a common source.

After seeing your PDF, I will say the following:

To get the same kind of results you presently have, you need an XSL-FO
engine that natively handles MathML, some combination of DocBook and
LaTeX (dblatex), some multi-stage toolchain, or your present tools.

You seem interested in FOP, because it is free. FOP (and XEP) are only
going to output something that looks like math if a program in your
tool chain either rasterizes the math (as PNG) or vectorizes it into
SVG (and even then I dunno if FOP handles SVG - I think it has to go
through rasterization in Batik first). SVG Math, an open source python
script, can turn MathML into (basline adjusted) SVG.

It would be good to turn this thread into a MathML "state of the
DocBook". If someone has info on, for instance, MathML -> (SVG Math)
-> SVG -> (Batik) -> PNG -> (FOP) -> PDF (a completely open source
toolchain), then it would be good to get it down in here for the
benefit of future users.

On 9/19/06, Jon Leech <jon@alumni.caltech.edu> wrote:
> On Tue, Sep 19, 2006 at 05:18:34PM -0500, Chris Chiasson wrote:
> > How are you creating your math? I'm using a computer algebra system
> > (which is what I use to generate multiple forms of the same equation).
> > The reason I ask is that if you don't use Mathematica or if you don't
> > run it on 64 bit Gentoo, then you probably won't experience the same
> > incompatibilities that I have. Even if you do use a proprietary XSL-FO
> > processor, you can always switch back to FOP once FOP improves enough.
>
>     Up to this point, all we've put into Docbook is our man pages. I
> wrote some Perl to compile the eqn source directly into MathML as part
> of the nroff->Docbook conversion.
>
>     For the API specification documents, we'll be dealing with
> conversion of math in LaTeX or Word docs, and writing expressions of
> comparable complexity in new documents. Mostly they are just linear
> algebra - you can get a sense of how complex from pp. 41-50 of
>
>     http://www.opengl.org/documentation/specs/version2.1/glspec21.pdf
>
>     I don't think we'd be using Mathematica for this. itex or something
> similar seems better matched to the task.
>
>     Jon
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: docbook-apps-unsubscribe@lists.oasis-open.org
> For additional commands, e-mail: docbook-apps-help@lists.oasis-open.org
>
>


-- 
http://chris.chiasson.name/