PrinceXML indexer

Automatically generates your index or table of contents for PrinceXML. This is a Bash script designed for injecting index references into a PrinceXML file and prompt the correspondent index lines. You can then copy paste those lines into your index file and regenerate your pdf.

What for?

I wrote my thesis using XML files. I generated the final pdf using PrinceXML and a CSS file based on Håkon Wium Lie's thesis CSS files. I split my work into several files, one for each chapter. I included all them in an index file using xi:include doing:

<xi:part>
    <xi:include href="introduction.xml"/>
    <xi:include href="technology.xml"/>
    <xi:include href="prototype.xml"/>
    [...]
</xi:part>

When I wrote the chapters, I didn't tag manually any of the headers, as I had in mind doing a script that did the dirty work for me. That's how PrinceXML indexer was born, as a fast hack done in a couple hours, so be gentle. My only regret, was not doing this script in Python.

Usage

First, It's recommendable to make a backup of the file. Then give indexer executable rights:

chmod +x indexer.sh

After doing so, you can run indexer doing:

./indexer.sh prince_file.xml

Indexer will find all headers (<h[1-6]>) and tag them using $filename-h$counter. So your first header will be id tagged id="prince_file-h1", the second one id="prince_file-h2" and so forth. Finally the script prompts the index lines that you can copy/paste in your index file:

$ ./indexer.sh prototype.xml 
<li class="chapter"><a href="#prototype-h1">Prototype</a><ul>
<li><a href="#prototype-h2">Design</a></li>
<li><a href="#prototype-h3">Landing page</a></li>
<li><a href="#prototype-h4">Authentication and user registration</a></li>
<li><a href="#prototype-h5">Social login</a></li>
<li><a href="#prototype-h6">Guided tour</a></li>
<li><a href="#prototype-h7">Start</a></li>
<li><a href="#prototype-h8">Publish a trip</a></li>
<li><a href="#prototype-h9">Tripboard</a></li>
<li><a href="#prototype-h10">Search system</a></li>
<li><a href="#prototype-h11">Warnings</a></li>
<li><a href="#prototype-h12">User profile</a><ul>
<li><a href="#prototype-h13">User profile edition</a></li>
<li><a href="#prototype-h14">Feedback</a></li>
<li><a href="#prototype-h15">Account configuration</a></li>
<li><a href="#prototype-h16">Inviting new users</a></li>
</ul></li>
<li><a href="#prototipo2-h17">Feedback</a></li>
</ul></li>

If you are using Håkon's CSS files, your index should then be something like this:

<div class="toc" id="toc">
    <ul class="toc">
        <!-- This is where you should put the index lines generated by indexer -->    
            
    </ul>
</div>

• Don't run indexer on the file that contains the index or table of contents, you need to put that aside.
• You can run indexer on an already tagged file, for example if you edit the xml files and you need to regenerate the index.
• You can run indexer in a file that already has some ids in headers, but beware, that those ids will be overriden by indexer.
• You can run indexer in the same file several times without side effects.