From: Alan G Isaac <aisaac@am...>  20041119 04:32:30

This is very close to what I had in mind! http://www1.chapman.edu/~jipsen/mathml/asciimath.html This provides an immediate solution for reST+math > XHTML and, if I understand correctly, a substantial part of the solution for reST+math > LaTeX and other formats. Cheers, Alan Isaac 
From: Colin J. Williams <cjw@sy...>  20041119 13:05:03

Alan G Isaac wrote: >This is very close to what I had in mind! > >http://www1.chapman.edu/~jipsen/mathml/asciimath.html > >This provides an immediate solution for reST+math > XHTML >and, if I understand correctly, a substantial part of >the solution for reST+math > LaTeX and other formats. > >Cheers, >Alan Isaac > > > > This provides for sums, products, integrals and appears to include matrices. It would be nice to have a graphic composition tool, but this looks very good indeed. Colin W. 
From: Colin J. Williams <cjw@sy...>  20041119 13:29:15

Colin J. Williams wrote: > Alan G Isaac wrote: > >> This is very close to what I had in mind! >> >> http://www1.chapman.edu/~jipsen/mathml/asciimath.html >> >> This provides an immediate solution for reST+math > XHTML >> and, if I understand correctly, a substantial part of >> the solution for reST+math > LaTeX and other formats. >> >> Cheers, >> Alan Isaac > The matrix spec is almost pythonesque: `[(S_[11],...,S_[1n]),(,...,),(S_[m1],...,S_[mn])]` is acceptable. Parentheses, instead of brackets, appear to be required for rows. It would be nice if the underscores could be dropped. Colin W. 
From: Beni Cherniavsky <cben@us...>  20041119 11:53:47

Alan G Isaac wrote: > This is very close to what I had in mind! > > http://www1.chapman.edu/~jipsen/mathml/asciimath.html > > This provides an immediate solution for reST+math > XHTML > and, if I understand correctly, a substantial part of > the solution for reST+math > LaTeX and other formats. > Cute, I'll take a closer look. See also EzMath_ (I wrote about it already once); click the Details_ link on the top for syntax design details. EzMath is yetanother eqnbased syntax but it's different in expressing semantics rather than presentation. I think that semantical math is closer to what people want to really express in a document and closer to how they think of it and pronounce it so it's preferable for docutils (it also creates more accessible documents). .. _EzMath: http://www.w3.org/People/Raggett/EzMath/ .. _Details: http://www.w3.org/People/Raggett/EzMath/EzMathPaper.html The current implementation is closedsource freeware. Perhaps the author could be persuaded to release the source. Also, it's not so hard to clone the part we need (translation to other engines, not rendering and GUI editing) in Python.  Beni Cherniavsky <cben@...>, who can only read email on weekends. 
From: Alan G Isaac <aisaac@am...>  20041120 05:48:44

> Alan G Isaac wrote: >> http://www1.chapman.edu/~jipsen/mathml/asciimath.html >> This provides an immediate solution for reST+math > XHTML >> and, if I understand correctly, a substantial part of >> the solution for reST+math > LaTeX and other formats. On Fri, 19 Nov 2004, Beni Cherniavsky apparently wrote: > See also EzMath_ (I wrote about it already once) ... > .. _EzMath: http://www.w3.org/People/Raggett/EzMath/ ... > The current implementation is closedsource freeware. Perhaps the author > could be persuaded to release the source. I will be happy with any solution. That said: ASCIIMathML gets my vote. i. It is open source. ii. It works great. iii. It is easier to read; EZMath leans toward verbosity. iv. It appears to be active and easily extensible; EZMath looks relatively inactive v. It can be used *right now* to display math in HTML documents, just by writing a link to the JavaScript in the HEAD element! vi. It has already been successfully ported once and the port does not look difficult. http://www.jcphysics.com/ASCIIMath/ This should mean that the issues of parsing are already handled nicely enough that reST can quickly move to write native syntax to multiple formats: LaTeX, XHTML+MathML (to remove dependence on the script), and Lout. vii. Because of its support for native LaTeX syntax, there is an immediate "stop gap" available as well for the LaTeX and HTML writers: for the moment, reST support for ASCIIMathML can be limited to *only* its supported LaTeX syntax, which as a stopgap can simply be written raw (with $ delimiters added) by both the HTML writer and the LaTeX writer. This gives an *immediate* if slightly restricted solution, with a bright future on the near horizon. So, to the request: can we please have a math role that for the moment writes its contents raw between delimiters, so :math:``\lim_{x\to\infty} 1/x = 0`` would work *now* if written raw (inside $ delimiters) to both LaTeX and to HTML (with a link to the JavaScript in the HEAD element) and would work correctly in the future when the contents were actually parsed to whatever end. And can we have a math directive that works exactly the same way for now (i.e., writes raw while we wait for the parser) but a. in the HTML writer puts the math raw inside its own DIV, with the math delimited by $s. b. in the LaTeX writes the math raw inside an equation environment I also volunteer to test and even tinker with a port of the script to Python, with an eye to being able to write the full ASCIIMathML syntax to arbitrary formats. Thanks for considering this, Alan Isaac 