Am 07.04.2016 um 11:11 schrieb Carl Sorensen: > On 4/7/16 12:11 AM, "lilypond-devel-bounces+c_sorensen=byu....@gnu.org on > behalf of Urs Liska" <lilypond-devel-bounces+c_sorensen=byu....@gnu.org on > behalf of > u...@openlilylib.org> wrote: > >> However, when *deleting* the .py and .pyc files an error was triggered, >> when lilypond-book tries to import book_html. >> >> Looking further into it I realized: >> - book_base keeps an array with "formatters" >> - upon importing any specific book_X it registers an instance of the >> formatter class in book_base >> - So if book_html is empty no HTML formatter is registered with >> lilypond-book >> >> This is why there is no error with the empty file. >> However, this isn't an answer to my question yet. If the HTML formatter >> is *not* used to produce the HTML included in the doc pages, what else? >> >> Any further ideas? > What about if you delete python/out and redo make doc? > > Just wondering if there's something wrong with the build script that > python/out isn't properly recreated when python/*.py changes. In my > setup, python/*.py is not executable, but python/out/*.py is. So make is > apparently copying the .py files, setting them executable, and creating > the .pyc files.
Hm, I've did a rebuild from scratch and immediately emptied the two book_html.py files, and *now* I got an error upon trying to create the HTML files. So while I think it wasn't related to the scripts itself it had to do with the state of my build directory. > Again, untested. But I have done a grep for blockquote and it only shows > up in book_html.py as far as I can see. So I think it must be > book_html.py. I finally found out where the code in book_html.py actually leads to. With some brute force I took care of replacing the snippet and image with a paragraph that I could then grep in the rendered docs. And it ended up exclusively in the lilypond-book regtests. So obviously the book_XXX.py files are really for lilypond-book and don't have anything to do with the building of LilyPond docs. Which brings us back to the original issue: Where is the code that generates the blockquotes with the LilyPond code snippets and images? Probably David's pointer to texi2html is plausible, but I must say reading through lilypond-texi2html.init doesn't really help. The references to <blockquote> that David found obviously don't point to LilyPond examples but rather to @example. And the rest of that file is definitely not related to LilyPond code examples. So it is really strange, I'm completely at a loss to find out where to start looking further ... Urs > > Thanks, > > Carl > _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
