Unlock a goldmine of knowledge on your Linux desktop.

This article will show you an easy way of quickly uncovering a lot of documentation that is sitting right on your system that I will bet you don’t even know that you have. Recently after installing a library and accompanying documentation, I had to dig into /usr/share/doc/lib-so and so only to find the docs were in HTML and had an INDEX.HTML launched. It was a Sunday and I had some time to burn so I surfed around a few more of the folders in doc and there were a number of applications and libraries that distributed their docs that way. It did seem rather random which ones fit this pattern and which ones did not so I wanted a cleaner way to view this documentation all in one place.

As an experiment I wrote a little Python script called Idxer (1.0 has been put under the GPL here) which does two simple things: it walked the /usr/share/doc looking for any sets of documentation with an INDEX.HTML file and it would output a simple HTML document with links to all of your documentation.

Installation:

This is my kind of installation:

1. Down load the latest source here: idxer1.0.tar.gz

2. Untar the source: tar -zxf ./idxer1.0.tar.gz

3. Done.

Use is equally simple:

1. Change to the idxer folder.

2. Run idxer.py, passing to it the name of the target output file. In my case I am doing mydocs.html

Idxer in action

When done I had a file called mydocs.html. When opened in FireFox it looked like this:

I was surprised to learn that I had over 120 documentation packages installed. There were game docs, lots of linux, unix and best of all, programmer docs just waiting to be devoured. Here are just a few I found on my box:

• Dive into Python
• Debian Reference
• Bazarre docs
• Monotone docs
• TV Time
• GDB
• GCC
• wxGlade
• Twisted lib
• LAME docs
• Image Magick
• libcurses, SDL, OGG, Glib, WWW, XML, Vorbis, TIFF and more
• espeak
• Valgrind
• XSane

I know, what this does it neither sexy (code and output are very plain for a reason) nor is it complex; it simply does something handy for me and I hope it does for you. If you want to add additional paths to the scan it should be a simple affair to code them into indexDocs(). If you have other mime-types installed for your browser you could modify the find call in indexDocs() to search for additional documentation types.

In any event, you can set up a cron job to run this periodically and drop a bookmark to it in Firefox or your favorite browser and always have your latest (HTML) docs at your fingertips.

Enjoy.

JeffC