This is the contributors guide for the documentation of Mitiq, the Python toolkit for implementing error mitigation on quantum computers.
Our documentation is generated with Sphinx. The necessary packages can be installed, from the root Mitiq directory
pip install -e .
pip install -r dev_requirements.txt
as they are included in the dev_requirements.txt
file.
Alternately, you can use the docker image provided in the repo and all requirements for working with the docs are already installed there.
myst-nb
andmyst-parser
allow both markdown and jupyter notebooks to be included and run by the Sphinx build. Also adds support for MyST markdown spec.sphinxcontrib-bibtex
allows to include citations in a.bib
file.pybtex
allows to customize how citations are rendered, e.g., APS-style.sphinx-copybutton
allows to easily copy-paste code snippets from examples.sphinx-autodoc-typehints
allows to control how annotations are displayed in the API-doc part of the documentation, integrating withsphinx-autodoc
andsphinx-napoleon
which handle specific styling requirements for the API documentation comments.
You can check that Sphinx is installed with sphinx-build --version
.
Since the documentation is already created, you need not to generate a
configuration file from scratch (this is done with sphinx-quickstart
).
Meta-data, extensions and other custom specifications are accounted for
in the conf.py
file.
To add specific feature to the documentation, Sphinx extensions can be added to the build.
As and example, to add classes and functions to the API doc, make sure that autodoc
extension is enabled in the conf.py
file, and for tests the doctest
one,
extensions = ['sphinx.ext.autodoc','sphinx.ext.doctest']
You need not to modify the docs/build
folder, as it is automatically generated. You should only modify the docs/source
files.
The documentation is divided into:
- a guide, whose content needs to be written from scratch,
- examples which can be either jupyter notebooks or MyST formatted notebooks, and
- an API-doc part, which is (mostly) automatically generated.
Information in the docs can be added as markdown (.md
/.myst
) files, since
the myst-parser
extension supports both basic markdown syntax as well as
the extended MyST syntax. Just add the file to source
directory and a TOC somewhere (if you want that).
Currently, `.rst` is supported for any of the files in the docs, but the current migration plan is to move everything to MyST serialization, to make it easier to include Jupyter notebooks and more consistent with documentation in the project root.
If you want a good intro to MyST and how it compares to `.rst` see [this guide](https://myst-parser.readthedocs.io/en/latest/sphinx/intro.html#intro-writing).
The main table of contents (TOC) file for the docs is index.myst
. It includes guide\guide.myst
and apidoc.myst
, among other files. To add a new file to the base TOC, make sure it gets listed in the toctree
directive like this:
```{toctree}
---
maxdepth: 2
caption: Contents
---
file.myst
```
If you use VS Code as your text editor there is a nice extension that does syntax highlighting for MyST: [https://marketplace.visualstudio.com/items?itemName=ExecutableBookProject.myst-highlight](https://marketplace.visualstudio.com/items?itemName=ExecutableBookProject.myst-highlight)
To include .md
files outside of the documentation source
directory, you can add a stub *.myst
file to the toctree inside the docs\source
directory that contains:
```{include} path/to/file.md
:relative-docs: docs/
:relative-images:
```
where file.md
is the one to be added. For more information on including files external to the docs, see the MyST docs.
To add information in the guide, it is recommended to add markdown (.md
) or MyST markdown files (.myst
) to the docs/guide/
directory.
Remember to add any files you add to the docs/guide/
directory to the guide TOC file docs/source/guide/guide.myst
.
All code examples, besides explanations on the use of core software package features, live in the examples
directory under docs/source
. You can add regular
Jupyter notebooks (.ipynb
) or MyST formatted markdown notebooks (.myst
or .md
) which you can think of as notebooks if you could write them in markdown.
In general MyST formatting will be preferred as it is much easier to diff in version control, but adds one additional step for folks contributing them directly.
If you have a notebook you want to add, and want to automatically convert it from the .ipynb
to the .myst
file format, you can use a great Python command line tool called jupytext.
Not only can jupytext
convert between the formats on demand, but once you install it, you can configure it to manage both a Jupyter and Markdown version of your file, so you don't have to remember to do conversions (for more details, see the jupytext
docs on paired notebooks.
Using the paired notebooks you can continue your development in the notebooks as normal, and just commit to git the markdown serialized version when you want to add to the docs.
You can even add this tool as a git pre-commit hook if you want!
There is a [sample markdown formatted notebook in the `examples` directory](./examples/template.myst) for you to take a look at as you write your own!
New modules, classes and functions can be added by listing them
in the appropriate .md or
*.mystfile (such as
apidoc.myst` or a child), e.g.,
## New Module
```{automodule} mitiq.new_module
:members:
will add all elements of the mitiq.new_module
module with a subtitle "New Module."
You can hand-pick classes and functions to add, to comment them, as well as exclude them.
If you are adding new features to Mitiq, make sure to add API docs in the
source code, and to the API page `apidoc.rst`.
The easiest way to build the docs is to just run make docs
from the project
root directory in bash, which by default builds the html docs output.
You can also use from root make pdf
to generate the PDF version.
If you want to remove previous builds and make the HTML docs fresh, try `make docs-clean`!
If you want to call sphinx directly, you can from bash move to the docs
folder and run
sphinx-build -b html source build
this generates the docs/build
folder. This folder is not kept track of in the
github repository, as docs/build
is present in the .gitignore
file.
The `html` and `latex` and `pdf` files will be automatically created in the
`docs/build` folder.
When writing a new code example in the docs, you can use different directives to include code blocks.
If you want to include a code snippet that doesn't get run (but has syntax
highlighting), use the code-block
directive:
```{code-block} python
1+1 # simple example
```
In order to make sure that the block is parsed with make doctest
, use the
testcode
directive. This can be used in pair with testoutput
, if something
is printed, and, eventually testsetup
, to import modules or set up variables
in an invisible block. An example is:
```{testcode} python
1+1 # simple example
```
with no output and
```{testcode} python
print(1+1) # explicitly print
```
```{testoutput} python
2 # match the print message
```
If you have code blocks you want to run, but not be displayed, use the
testsetup
directive:
```{testsetup} python
import numpy as np # this block is not rendered in the html or pdf
```
```{testcode} python
np.array(2)
```
```{testoutput} python
array(2)
```
There is also the doctest
directive, which allows you to include interactive
Python blocks. These need to be given this way:
```{doctest} python
>>> import numpy as np
>>> print(np.array(2))
array(2)
```
Notice that no space is left between the last input and the output when writing code blocks with interactive inputs and outputs.
In order to skip a test, if this is problematic, one can use the SKIP
and
IGNORE
keywords, adding them as comments next to the relevant line or block:
>>> something_that_raises() #: doctest: +IGNORE
Mitiq uses the doctest
extension to run and test code in the docs, which is configured in the conf.py
file. To execute the tests in bash, run:
make doctest
from the root directory.
This command tests the code examples in the documentation files, as well as testing the docstrings, since these are imported with the autodoc
extension.
One can also use various doctest
features by configuring them in the
docs/pytest.ini
file.
Here are some notes on how to build docs.
The MyST syntax guide is a cheat sheet for the extended Markdown formatting that applies to both Markdown files as well as Markdown in Jupyter notebooks.
The MyST-NB Notebook guide can help you get you write or convert your notebook content for the docs.