Hi Karl. > I'm sorry, but what you want for @part is something completely different > than what I intended. Oddly enough, I implemented what I intended :), > which was essentially just standalone "part pages" that bear no > relationship to the rest of the document.
OK. We have a bad confluence of names. > (Aside: in general, I can't see how you're *ever* going to get the exact > output you want, in every detail, through two different markup systems, > each with their own crazy limitations. When I want to control my > output, which I very often do, I use TeX directly ... screw everything > else. Oh well. I know O'Reilly doesn't believe in TeX.) The languages are are not all that far apart, actually. What I want is valid docbook. (Well, there are times when I want to control my output, but that's not the main issue here.) OK. Let's retrench a bit. There are two orthogonal issues. 1. Karl is opposed to continued stretching of the language, including texinfo.tex, to provide features that map directly to Docbook but are of little or no value to TeX/HMTL/Info. Examples are full-fledged "part" pages and sidebars. (And better indexing, to be complete. :-) So, I finally get this and will do my best to stop asking for such. 2. For all current features of the Texinfo language, the docbook output generated by makeinfo should be valid docbook. I think we all agree on this as a principle. If we accept this, then for makeinfo --docbook, any text following @part and preceding the first @chapter should be wrapped in <partintro> tags, and Patrice should fix this. I am fine for makeinfo to issue a warning about such text if not in docbook mode. Karl, do you approve this docbook-specific fix? wrapping <partintro> tags in @docbook works, and I will do that for now so that I won't be blocked. But given #2, I would hope that it will be fixed in makeinfo. The other stuff I need/want I'll either make do with macros (not Karl's favorite thing, but it's my book, not his) or with pre- and post-processing scripts. Thanks, Arnold