[publican-list] Possible alternative to wkhtmltopdf

[Peter Moulder]

I said I'd try to get a git repository URL ready, so I'll give that in
this message.  However, remember that so far it has neither ToC page
numbers nor clickable links, so its output isn't yet satisfactory for
either on-screen or print use.  Apart from that, situations that aren't
handled yet trigger assertion failures, which are still relatively
common.

I'll post here again every once in a while as things progress.

(Speaking of which, there is just a minor update, a stylesheet thing: I
noticed that the rules for choosing a page heading weren't what FOP was
using, so I've gone through a bit more methodically to match things up,
and have updated the pdf accordingly.)

So the git repositories as promised are

  git://bowman.infotech.monash.edu.au/srv/git/libcroco.git
  git://bowman.infotech.monash.edu.au/srv/git/morp.git

but I expect most people might as well wait until there's news such as
page numbers in ToC & index.

The reason for posting a sample now is that I hope that seeing some
output now gives an idea of what the options will be later.  Of course
I'd like to think that the software I'm working on will become useful to
more people, so posting now before things are ready might help people
know how to get what they want sooner.  So if clickable links are
important to you, then consider helping the person who was looking at
giving cairo an interface accessing pdf features.  Otherwise, if you like
Inkscape or something else that uses libcroco, then maybe you'll want to
browse the libcroco git repository I'm using and see what's useful there
and in that way help to bring the various copies of libcroco a bit closer
together.  These are things that aren't needed by the research projects,
but they're things that might help you get a lot for a little effort, and
indirectly that might help me -- in job satisfaction if nothing else.

Oh, btw, re:

> (Currently I use curl to create blah.header and blah.data files somewhere
> under ~/url-cache, but I deliberately don't handle Max-Age stuff because
> re-fetching would interfere with the reproduceability of results of
> experiments associated with said research.)

Just in case anyone's greatly put off by the prospect of it not seeing
an updated version of a source file, I should add that the software does
have "reload" functionality, and there's also a list of URI patterns that
should always be re-fetched.

pjrm.