Step 1: Getting started with RST

Now that we have our basic skeleton, let’s document the project. As you might have guessed from the name, we’ll be documenting a basic web crawler.

For this project, we’ll have the following pages:

  • Index Page
  • Support
  • Installation
  • Cookbook
  • Command Line Options
  • API

Let’s go over the concepts we’ll cover, and then we can talk more about the pages to create.


A lot of these RST syntax examples are covered in the Sphinx reStructuredText Primer.





Every Sphinx document has multiple level of headings. Section headers are created by underlining the section title with a punctuation character, at least as long as the text.

They give structure to the document, which is used in navigation and in the display in all output formats.

Code Samples

You can use ``backticks`` for showing ``highlighted`` code.

If you want to make sure that text is shown in monospaced fonts for code examples or concepts, use double backticks around it. It looks like this on output.

Code Example Syntax

A cool bit of code::

   Some cool Code

.. code-block:: rst

   A bit of **rst** which should be *highlighted* properly.

The syntax for displaying code is ::. When it is used at the end of a sentence, Sphinx is smart and displays one : in the output, and knows there is a code example in the following indented block.

Sphinx, like Python, uses meaningful whitespace. Blocks of content are structured based on the indention level they are on. You can see this concept with our code-block directive later.

Table of Contents Tree

.. toctree::
   :maxdepth: 2


Now would be a good time to introduce the toctree. One of the main concepts in Sphinx is that it allows multiple pages to be combined into a cohesive hierarchy. The toctree directive is a fundamental part of this structure.

The above example will output a Table of Contents in the page where it occurs. The maxdepth argument tells Sphinx to include 2 levels of headers in it’s output. It will output the 2 top-level headers of the pages listed. This also tells Sphinx that the other pages are sub-pages of the current page, creating a “tree” structure of the pages:

├── install
├── support


The TOC Tree is also used for generating the navigation elements inside Sphinx. It is quite important, and one of the most powerful concepts in Sphinx.


Create Installation page

Installation documentation is really important. Anyone who is coming to the project will need to install it. For our example, we are installing a basic Python script, so it will be pretty easy.

Include the following in your install.rst, on the same level as index.rst, properly marked up:


At the command line:

easy_install crawler

Or, if you have pip installed:

pip install crawler


Live Preview: Installation

Create Support page

It’s always important that users can ask questions when they get stuck. There are many ways to handle this, but normal approaches are to have an IRC channel and mailing list.

Go ahead and put this in your support.rst, but add the proper RST markup:


The easiest way to get help with the project is to join the #crawler
channel on Freenode.
We hang out there and you can get real-time help with your projects.
The other good way is to open an issue on Github.

The mailing list at!forum/crawler 
is also available for support.

Freenode: irc://


Live Preview: Support

You can now open the support.html file directly, but it isn’t showing on the navigation..

Add TocTree

Now you need to tie all these files together. As we mentioned above, the Table of Contents Tree is the best way to do this. Go ahead and complete the toctree directive in your index.rst file, adding the new install and support.

Sanity Check

Your filesystem should now look something like this:

├── src
└── docs
    ├── index.rst
    ├── support.rst
    ├── install.rst
    ├── Makefile

Build Docs

Now that you have a few pages of content, go ahead and build your docs again:

make html

If you open up your index.html, you should see the basic structure of your docs from the included toctree directive.

Extra Credit

Have some extra time left? Check out these other cool things you can do with Sphinx.

Make a manpage

The beauty of Sphinx is that it can output in multiple formats, not just HTML. All of those formats share the same base format though, so you only have to change things in one place. So you can generate a manpage for your docs:

make man

This will place a manpage in _build/man. You can then view it with:

man _build/man/crawler.1

Create a single page document

Some people prefer one large HTML document, instead of having to look through multiple pages. This is another area where Sphinx shines. You can write your documentation in multiple files to make editing and updating easier. Then if you want to distribute a single page HTML version:

make singlehtml

This will combine all of your HTML pages into a single page. Check it out by opening it in your browser:

open _build/singlehtml/index.html


You’ll notice that it included the documents in the order that your TOC Tree was defined.

Play with RST

RST takes a bit of practice to wrap your head around. Go over to, which is a live preview.


Use the Cheat Sheet for lots more ideas!

Looking for some ideas of what the syntax contains? The reStructuredText Primer in the Sphinx docs is a great place to start.

Moving on

Now it is time to move on to Step 2: Building References & API docs.