For those wishing to produce PDF from Doxygen...

Hi Doxygen users,

Here is an alternative for those of us who need to produce both HTML and
PDF with Doxygen. Somebody turned me on to this other open-source tool,
htmldoc, which takes HTML as input and exports indexed PDF or
postscript.

http://freshmeat.net/projects/htmldoc

I realize that doxygen supports exactly these operations by first
exporting to LaTex and generating the PDF from there (which I have
done.) What I didn't like about it:

[*] I had very little knowledge and control of archiac LaTex.
[*] The generated PDF left much to be desired in terms of blank space
and layout.
[*] It would give me mostly garbage if the source system had
dependencies and used tag files.
[*] I couldn't manipulate these mechanisms easily to produce PDF from
HTML that didn't come from doxygen.

For other projects unrelated to doxygen, I was forced to look into a
tool to go from HTML to PDF. It turns out that htmldoc also works
admirably on doxygen output.

For those of you using a version of doxygen that still generates the
treeview.js and tree.js, the attached perl program will be handy. It
converts from the tree.js to a simple "batch" file that htmldoc uses as
input. Thereafter, the same script that lauches your doxygen build can
run this tool and then htmldoc.

Unfortunately, I have not upgraded to a version of doxygen that has the
improved javascript. I have many integrated HTML-systems that are
comprised of the output from many doxygen projects. The newer version
which does away with the tree.js breaks too many of my systems at the
moment.

[However, that newer javascript may prove better than the TOC Java
applet that I am using. I will be developing tools to parse that
javascript with the intent of creating much larger systems either from
the same javascript or for my TOC Java applet. At that time, I will also
develop the quick-and-dirty tools to convert this to the input required
for htmldoc.]

Also for the curious, I run post-processing tools (available in the
doxygen download section) on all of doxygen's HTML output so that I can
insert my navigation and copyright (and other bells-and-whistles). I
found it best to run my post-processing tools twice: once with a blank
master that strips out all navigation so that the htmldoc PDF output is
prettier, and then once with the HTML master navigation that I want the
HTML output to have.

Hope this information is useful to users of the great and wonderful
tool, doxygen.

Glenn Maxey
Technical Writer
Voyant Technologies, Inc.
1765 West 121st Avenue
Westminster, CO 80234-2301
Tel. +1 303.223.5164
Fax. +1 303.223.5275
glenn.maxey@xxxxxxxxxxxxxx

tree_js_2_book.pl
Description: tree_js_2_book.pl