[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

How to submit complex docbook document?

OK, I've converted my howto into docbook, and the result is three sgml
files, several png files, a modified collateindex.pl script, and a
mighty scary Makefile.

How do I submit this?  According to the HOWTO-squared, CVS isn't
really up, and I'd rather not use a remote CVS in the first place.
And there's some concern in my mind that the usual infrastructure
won't be able to build this properly - I had to fix assorted bugs in
the tools as I went:

 - The indexing process is rather half-baked, but works iff:

    * You put an id= for every sect containing any indexterms.
      Automagical AEN calculation doesn't appear to work for index
      purposes with the print backend.  The stylesheet seems to
      contain code for it, but it just didn't work for me in print
      until I id'd all my sects.

    * You either don't use the %use-id-as-filename% option, or you
      twiddle collateindex.pl to deal with it.  If use-id is on, then
      the role= attribute in the generate index will be the html
      filename rather than the actual id for setc1's.  I just
      subtracted the .html if there was a trailing .html in the $role
      right before printing it out in collateindex.pl.

    * Debian potato has collateindex.pl installed in /usr/bin, which
      is swell until you figure out that it's got # -*- perl -*- as
      the first line instead of #! /usr/bin/perl.

 - Pictures were fun.  I still haven't put in any vector drawings, but
   potato appears to ship dtd 3.0, which hasn't got the mediaobject
   stuff.  The regular old graphics tag works reasonably if you just
   leave off the extension, rather like you do in LaTeX.  It takes
   some Makefile trickery to have all the proper formats floating
   around, but this I just copied from Makefiles I had floating
   around.  Surprise du jour: pdfjadetex groks png files.

 - pdfjadetex produces unsuccessful results with my table, which
   features really tall rows that run off the bottom of the page.
   Hopefully htmldoc does better?  I suppose I should break up the big
   rows and eliminate the problem.

--
Grant Taylor - [email protected] - http://www.picante.com/~gtaylor/
    Linux Printing Website and HOWTO:  http://www.linuxprinting.org/
    I do consulting in most things Unix/Linux/*BSD/Perl/C/C++


--  
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]