12:50 a.m.
Tammy, you said you're working with Makefiles, stylesheets and all the
"background stuff" for docs, right?
May I ask exactly what you're doing and how may I help you?
For the next few months I'll have a lot of free time and I might be able to
help you out with your stuff too. It will also help me to gain full knowledge
of the docs-making process and even start a translation for one of the
current avaliable guides.
--
Luiz Rocha <lsdr(a)lsdr.net>
12:35 p.m.
On Thu, Oct 02, 2003 at 02:50:56AM -0300, Luiz Rocha wrote:
Tammy, you said you're working with Makefiles, stylesheets and
all the
"background stuff" for docs, right?
May I ask exactly what you're doing and how may I help you?
The biggest priority is getting xmlto to produce PDFs. Even with valid
DocBook XML, I can't get a PDF to generate for the Documentation or
the Developer's Guides. I have been commenting out sections to see
where the problems are, and so far I have found a few
workarounds. But, ideally we want it to work for valid DocBook XML. We
also want it to work without having to install additional packages on
top of Fedora Core. So, something like using some of the Java
solutions out there is out unless it is the last option.
For the next few months I'll have a lot of free time and I
might be able to
help you out with your stuff too. It will also help me to gain full knowledge
of the docs-making process and even start a translation for one of the
current avaliable guides.
Great. Sounds good.
--
Luiz Rocha <lsdr(a)lsdr.net>
Tammy
3:42 p.m.
On Thu, Oct 02, 2003 at 01:35:28PM -0400, Tammy Fox wrote:
The biggest priority is getting xmlto to produce PDFs. Even with
valid DocBook XML, I can't get a PDF to generate for the
Documentation or the Developer's Guides. I have been commenting out
sections to see where the problems are, and so far I have found a
few workarounds. But, ideally we want it to work for valid DocBook
XML. We also want it to work without having to install additional
packages on top of Fedora Core. So, something like using some of the
Java solutions out there is out unless it is the last option.
I already know about the <ulink url="...&..."> problem -- are
there more problems then?
(No progress on that by the way. Needs a TeX hacker, or a tool
written in a different language.)
Tim.
*/
3:53 p.m.
On Thu, Oct 02, 2003 at 09:42:56PM +0100, Tim Waugh wrote:
On Thu, Oct 02, 2003 at 01:35:28PM -0400, Tammy Fox wrote:
> The biggest priority is getting xmlto to produce PDFs. Even with
> valid DocBook XML, I can't get a PDF to generate for the
> Documentation or the Developer's Guides. I have been commenting out
> sections to see where the problems are, and so far I have found a
> few workarounds. But, ideally we want it to work for valid DocBook
> XML. We also want it to work without having to install additional
> packages on top of Fedora Core. So, something like using some of the
> Java solutions out there is out unless it is the last option.
I already know about the <ulink url="...&..."> problem -- are
there more problems then?
Yes. I tried this about a month and a half ago, so I don't remember
the specifics. Certain table attributes seemed to make it stop and if
I had:
<screen>
<command>
foo
</command>
</screen>
I had to change it to:
<screen>
<command>foo</command>
</screen>
I can't remember if that was because it gave it too much space or
because it stopped at that point when making the PDF.
I'll go through my log files this weekend and try to give you more
specifics.
If you have time, you can check out the fedora-docs module and run
make pdf in the developer-guide or documentation-guide directory and
see the error messages. Maybe they mean more to you than me.
Tammy
(No progress on that by the way. Needs a TeX hacker, or a tool
written in a different language.)
Tim.
*/
7:37 p.m.
> I already know about the <ulink
url="...&..."> problem -- are
> there more problems then?
Yes. I tried this about a month and a half ago, so I don't remember
the specifics. Certain table attributes seemed to make it stop and if
I had
I think this kind of problems could be solved preprocessing the file before
trying to do the transformation, don't you?
If the XML is preprocessed by a script (I thought about using python +
libxml, since both are part of Fedora Core), to clean excessive spaces,
resolv the entities (& and others) before trying to transform into PDF it
may solve this problem without making the XML a invalid DocBook.
Of course, it is just a first thought. It might be the wrong approach to the
problem.
--
Luiz Rocha <lsdr(a)lsdr.net>
11:56 p.m.
Better would be to fix the bugs. :-) I hadn't heard about this
problem before now.
I knew I missed something... :)
What I understood is that fixing the causes of the failure would result in a
non-valid DocBook XML, which is no good.
I might be wrong, though.
--
Luiz Rocha <lsdr(a)lsdr.net>
7515
days inactive
7517
days old
6 comments
3 participants
participants (3)
-
Luiz Rocha -
Tammy Fox -
Tim Waugh