Hi Duncan, your approach looks _quite_ useful to me! I'm a bit afraid it will be a very hard jump to get on that tool, but riding it must be fun.
Yes, I'm interested! Detlef On Wed, 17 Oct 2007 08:59:13 +1300 Duncan Temple Lang <[EMAIL PROTECTED]> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hi Doug. > > This is probably more than you want - either to know or to use > for your specific task, but I'll throw it out there for general > information. > > I write documents using XML, specifically an extended version > of Docbook with elements for describing R concepts (e.g. > code, plot, output, function name, argument, ...). Then, > I use XSL to transform this to either HTML, FO and on to PDF, > or to LaTeX (via db2latex). As for processing > the R code and inserting the output into the resulting view/document, > I use the Sxslt package from within R which allows me to combine XSL > rules with those that also call R functions. > I author the documents using the nxml mode in emacs > and have some basic "gestures" for sending code from the document > to R. > > > Deb Nolan and I use this for creating dynamic and interactive > documents which can be transformed to HTML with embedded > controls for the reader to control the R computations > interactively. > > Rather than thinking of the document as being one whose > primary purpose is to be displayed to readers, > the approach allows us to put arbitrary things into the > document as part of our work but render only the bits > we want for a particular audience, e.g. mix code, > pedagogical material, R documentation, data, code > from other languages. > > The XML/Docbook-XSL-R approach is very general and flexible > with possibly "too" many degrees of freedom. If the document is > destined only for HTML, then writing in HTML directly may be > best. The generality is useful when there are multiple targets > and one wants to extract information programmatically, e.g. > extract subsets of the code within the document such as > that in section 2, or only Matlab code. > > I will be packaging up all the material we have on this soon, > so if anyone wants a copy, let me know. > > > D. > > Douglas Bates wrote: > > My university provides me with a powerful course management system for > > the courses that I teach. Among other things I can create a wiki for > > the course, which is very convenient for cross-linking different bits > > of the course. > > > > Naturally I use R extensively in my teaching and I want to incorporate > > R code, output and graphics in such a wiki. If I were producing LaTeX > > sources instead of HTML sources I create .Rnw files for Sweave and I > > would edit them using ESS in emacs. > > > > What options do I have for producing HTML with embedded R content and > > what is a good, preferably emacs-based, way of editing the source > > code? > > > > One basic problem is trying to present mathematical expressions in HTML (see > > http://www.cs.tut.fi/~jkorpela/math/) but, aside from that, there are > > questions of presenting input R expressions and the corresponding > > output and of incorporating graphics files produced by R. I could try > > to use latex2html or texi2html but the output from latex2html at least > > would be quite inconvenient to use because it generates so many linked > > files. Once they are uploaded it would be horrible trying to get all > > the links straightened out. > > > > In a sense there are already tools for this type of output from .Rd > > files. Would it be best to use those tools or to use texinfo tools or > > ...? > > > > ______________________________________________ > > R-devel@r-project.org mailing list > > https://stat.ethz.ch/mailman/listinfo/r-devel > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.7 (Darwin) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > iD8DBQFHFRgR9p/Jzwa2QP4RAi2vAJ0S+Mnbjvt6z9pe1kPoIxHeaaZQkACggHOL > GflkuedfvPVQfm6fayigGK0= > =4fKz > -----END PGP SIGNATURE----- > > ______________________________________________ > R-devel@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel -- Somebody once said, "If you lend someone $10 and never see that person again, it was probably worth it". --- found in a mailing list ______________________________________________ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
