David Crossley wrote: > I revised the list, taking the replies into account > and adding some new queries. > >>- agree the subset of XHTML2 to be used >>and document that via example. > > See other thread. > >>- create DTD's > > Why do we need DTDs for use with internal structure?
For using XHTML2 as an input format. In fact it will be a modularization of XHTML2 via RelaxNG validation. >>- Decide how to make the pipeline process. >>Now we have body-*.html and stuff, but a simpler process >>should be devised. >> >>- Convert all sitemaps and transformers to utilise >>that new process. >> >>- add a sitemap that processes *.xhtml2 files in a >>parallel section > > Does this refer to XHTML2 as input source files? It refers to all processing. > What processing is needed? That's what has to be decided ;-) Whoever does it gets to try a simpler pipeline (without the body, header, etc doc matches, but just a single transformation). >>- convert all forrest:contracts to accept XHTML2 >>as input rather than XDoc while keeping the old >>ones in the meantime >> >>- convert forrest:views to work with XHTML2 >>and make them work in the extra pipeline. >>Here we should be able to accept XHTML2 input >>and produce the usual output. >> >>For legacy: >> >>- xdoc to XHTML2 stylesheet and create XDoc input plugin >> >>- document migration process -- Nicola Ken Barozzi [EMAIL PROTECTED] - verba volant, scripta manent - (discussions get forgotten, just code remains) ---------------------------------------------------------------------