[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Axiom-developer] Re: new hyperdoc changes -- xhtml jenks book

From: root
Subject: [Axiom-developer] Re: new hyperdoc changes -- xhtml jenks book
Date: Sun, 14 Oct 2007 11:01:40 -0400

>The original book.pamphlet chapter 20 lacked any content.
>It's supposed to be a listing of operations but there's nothing
>there.  Do you know where the missing content is? 

Chapter 20, the operations, was automatically generated by a 
program that ran on the IBM/370 VM system. It was too large
to run on a PC at the time. The sources were never released
as far as I know. It was called "autodoc". I think Scott
Morrison wrote it.

I tried to write the equivalent program while editing the book
back to its original form but never got it finished.

>For writing out the mathml I wrote a "TeX to mathml" program
>in C.  I put that in quotes because it's just supposed to handle
>the TeX that Axiom produces between double dollar signs.
>It's about 95% finished.  Some things I still had to tweak by
>hand or substitute the output from Axiom.  A lot of the math
>TeX in the book has been hand altered and is not what
>Axiom produces in TeX mode.  The hand altered TeX tends
>to cause a lot of trouble.  I can post the C program if you're
>interested in it.

Yeah, the TeX in the book is hand modified Axiom output.
Axiom output needs tweaking to fit the format of the book.
It took several months of work to reconstruct the Jenks book.

I looked at the Xhtml book output and there are some things
that need minor tweaks (e.g. "f == " comes out as "f = ="
in some places. Unfortunately I don't know enough mathml yet
to hand-tweak the sources but I'll get there.

Also Axiom's mathml output is one long line. It needs to be
"pretty-printed" (e.g. single-space indent for each open element
and single-space outdent for corresponding close element) so it
can be printed in latex documents. I'm not sure if Axiom has
enough information to do this during generation so we might
need a pretty-printer postprocess.

>I also have a lot of emacs lisp macros for converting the rest
>of the book material, which you're also welcome to,  although
>they're not written with a view to somebody else using them.

That's ok. I generally have my own set of macros that fit well
"under my fingers" for these kinds of tasks.

>I can try running bookvol1.pamphlet through my emacs macros
>and see what happens.  In fact I've had this in mind after
>the Jenks-Sutor book but hadn't gotten around to it.

If you'll initially process it I'm willing to polish it.  It would be
easier to use the same process so it has the same look and feel.  I've
added a link to the new hyperdoc pages for it so it is certainly in
the queue of things to do.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]