>> > ||text ||span rows ///||column 3|| >> > ||text || ||column 3|| >> > ||text || ||column 3|| >> >> Arguably, this is clearer, and pmwiki *should* remove the empty >> cells when generating HTML. However, I can't do that so easily as >> a local customisation. > > Indeed, the confusion surrounding the "skips" over rowspanned > cells is part of the reason I never worked too hard to accommodate > a rowspan markup. > > One could potentially come up with a markup that means "skip this > cell in output", as in: > > ||text ||span rows +++||column 3|| > ||text ||+++++++++++++||column 3|| > ||text ||+++++++++++++||column 3|| > > In this case, PmWiki would simply omit the cells containing > only ++'s from the HTML output, or otherwise convert those > into something that is innocuous to XHTML but perhaps useful > to other conversions such as LaTeX or the like. > On its own, this is an attractive idea. The trouble, I think, is that then simple tables and advanced tables would handle rowspan differently, unless the same convention was added, e.g. a (:cellrowspanned:) directive that is transparent to XHTML.
I have cracked the code to translate the simple table rowspan markup through to LaTeX, and think I can use the same logic in advanced tables so that they behave less badly. The following (somewhat pathological) example works correctly: ||aaaa||bbbb||cccc++|| ||dddd||eeee|| ||Data||Some text++|| A string|| ||More||Another string++++|| ||Not spanned ||String || ||Text 1++ ||Text 2 || ||xxxx|| I'll publish a working example in a day or so. The above is pretty opaque in wiki markup. JR _______________________________________________ pmwiki-users mailing list [email protected] http://www.pmichaud.com/mailman/listinfo/pmwiki-users
