-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Sonntag, 1. Februar 2009 18:20:22 John Mandereau wrote:
> Hi Graham,
>
> Graham Percival a écrit :
> > How hard would it be to compile the CG with
> > --split=sections
> > instead of the current setting?
>
> I have the following in my compilation log on Jan 24th:
> texi2html -I /home/lilydev/git/lily/master/Documentation/user
> --I=/home/lilydev/git/lily/master/out/xref-maps --I=. --I=./out-www
> --output=out-www/contrib-guide/ --prefix=index --split=section
> --init-file=/home/lilydev/git/lily/master/lilypond-texi2html.init
> out-www/contrib-guide.texi
Yes, we run texi2html with --split=section for all texinfo files.
> I guess "--split=section" is ignored by our custom init file, so we should
> reenable Texi2HTML defaults there if this option is caught.
It's not ignored, but rather used to distinguish from big-file builds...
There is simply no --split value to split the doc at the points we want (all
sectioning commands, but not unnumbered), so we are kind of misusing section
to mean our splitting points here.
In our init file, we use our own algorithm to specify the resulting html page
for a given node ONLY for --split=section:
if ($Texi2HTML::Config::SPLIT eq 'section') {
$Texi2HTML::Config::element_file_name = \&lilypond_element_file_name;
}
Do I understand it correctly, that for the CG, you want e.g. all subsections
of "1.1 Gettin the source code" in the same html file, all subsections of "1.2
Updating the source code" in another html file, etc.?
The quick and dirty fix would be to add a check for the manual name in the if
clause quoted above. However, I'd like to keep the .init file rather clean and
not add such special cases...
Also, the $Texi2HTML::THISDOC{'input_file_name'} field is not set yet when the
.init file is loaded, so it's not that simple to determine the manual name.
The proper solution would be to add another possible value for --split to use
for our default splitting algorithm and have --split=section behave like in
vanilla texi2html... Unfortunately, this requires changes to texi2html, since
in this case
(1) with --split=section the html files will always be named index_19.html
etc. (the nice naming with the section title is done by our init file) and
(2) texi2html contains several checks that compare certain values of
$Texi2HTML::Config::SPLIT and if that value is not one of the know values,
certain functions are not called (most notably, there will be no TOC pane on
the left!)
However, the other part of our splitting algorithm is implemented in the
extract_texi_filenames.py script (which generates the file names for the
output html files), we might find a solution by tweaking that script, so it
places all subsections into the same html file...
Cheers,
Reinhold
- --
- ------------------------------------------------------------------
Reinhold Kainhofer, reinh...@kainhofer.com, http://reinhold.kainhofer.com/
* Financial & Actuarial Math., Vienna Univ. of Technology, Austria
* http://www.fam.tuwien.ac.at/, DVR: 0005886
* LilyPond, Music typesetting, http://www.lilypond.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iD8DBQFJhekETqjEwhXvPN0RAno2AJ48urpvrVCyITL5ywSMWbsRm14MdgCbBZPr
SZI4xlNsh9Haa5rVYEbHc88=
=VJGS
-----END PGP SIGNATURE-----
_______________________________________________
lilypond-devel mailing list
lilypond-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-devel
