--- Scott Reynen <[EMAIL PROTECTED]> wrote: > Wasn't there some talk of having the semantics of table > cells exist > in the table headers and propagate to the cells via > scope="col"? I > can't find it in the archive. Did I totally imagine > that?
Perhaps you are thinking of the "axis" attribute? More here: http://microformats.org/wiki/hcalendar-issues I recently suggested that scope also be treated in a similar way. Read here: http://microformats.org/discuss/mail/microformats-discuss/2006-June/004296.html > Right now I have a microformat parser that requires valid > XHTML but > doesn't do inclusion, and a proxy that does inclusion, > but may turn > valid XHTML into invalid XHTML in the process. I intend > to resolve > this by merging the two, but that's only an option > because I'm > rewriting the parser anyway. So I'm wondering how such > an issue > would best be resolved if merging weren't an option. > Should parsers > not require valid XHTML input, or is it just not possible > to cleanly > detach inclusion from the rest of the parsing process? As I understand it, well-formed XHTML is required when authoring content because it needs to be rendered by user agents (e.g., browsers) in a human-friendly way *and* parsable by XML tools (like Brians X2V parser, which uses XSLT to reformat the content). If these conditions were not required, we could author content in HTML or XML. Why does your parser need to have valid XHTML input instead of working with valid XML? If you loosen your input requirement for the parser, you can do your inclusions first and pass valid XML (but invalid XHTML) to your parser. -ml _______________________________________________ microformats-discuss mailing list [email protected] http://microformats.org/mailman/listinfo/microformats-discuss
