[Derby Documentation DITA] initial problems

[scott hutinger <s-...@wiu.edu>]

The major problems with the files, are the filesystem '/' vs '\' things.
I get output fine, after moving some of the \'s to /'s.   This is an 
issue that should be addressed.

I'll look at the xsl stylesheets a bit more.

Any good direction at what should be done first Jeff?  I think you know 
a lot more about this than anyone.   Possibly a list of what is broken 
currently; concerning the final output in html and pdf.  I think the pdf 
output will require a bit of work to get everything working correctly, 
so probably concentrate on correct html output.   It might be better to 
build some docs (such as pdf's that lack the final goal) and have the 
docs change in time due to the output getting more refined.  I think the 
preferred build is probably with ant(?) until a lot of issues get sorted 
out.  Although manually until that point.

Also, we might want to add something like [Derby Documentation] or what 
I have, so people not interested can filter e-mail.  Although I think 
everyone should be interested in the docs :-)

scott

[Derby Documentation DITA] PDF transform and problems, and eclipse plugin process

[Jeff Levitt <jl...@mutagen.net>]

OK, I sort of messed up my description of how to make
a PDF in my last post.  Here is what you do, and then
I'll get into the current problems with the PDF:

1.  Transform the dita files to one merged dita file
using the wacky topicmerge.xsl file.  The command I
used is:

saxon -o getstartmerge.ditamap -w1
"gettingstartedguide/getstartderby.ditamap"
"xsl/topicmerge.xsl"

Remember to have the xsl files from the dita package
available to saxon.

2.  Take the resulting merged dita file
(getstartmerge.ditamap) and move it to your FOP
install directory.  Make sure to also have the xsl
directory from the dita package in your FOP install
directory.  Finally, copy the gsconrefs.dita file to
the FOP install directory as well, as FOP will need to
refer to it to grab the names within it.

3.  Run the following command in FOP:

fop -xml getstartmerge.ditamap -xsl
xsl/dita2fo_shell.xsl -pdf getstart.pdf

This creates your PDF!  An ugly PDF, but a PDF
nonetheless.  I'd recommend creating batch files for
these commands as well to make things easy.

OK, so on to the issues.  Besides the issues I
mentioned previously in the html transforms, we have
additional issues in the PDF transform (note that the
parent-child linking issue is not as important in a
PDF, but would be nice to have, and will probably be
more crucial in the longer books).

1.  Cover page:  There is none.
2.  No Table of Contents
3.  None of the links work.  Links display topic id's
instead of the title of the linked topic.
4.  No index

I should add that the html transforms produced no
index either.  That is common to both the html and pdf
transforms.

So once again, we need to tweak these xsl files
somewhat.  But at least we can get from A to Z,
although we're skipping some letters on the way...

On a note unrelated to PDF transforms, is anyone
curious yet about making an Eclipse plugin from the
output of xhtml and the xml nav tree?  I hesitate to
post the process I followed for it now, because I feel
we have enough to discuss already without adding that
to the fold.  But if anyone wants to move to that now,
just let me know and I'd be happy to share my
experience with you.