pydoclint 0.0.3

A linter to check arguments in Python docstrings

pydoclint

A Python docstring linter to check whether a docstring's sections (arguments, returns, raises, ...) match the function signature or function implementation.

It runs really fast. In fact, it is at least ~1,475 times faster than darglint (another linter of the same purposes which is no longer maintained).

Here is a comparison of linting time on some famous Python projects:

	pydoclint	darglint
numpy	2.0 sec	49 min 9 sec (1,475x slower)
scikit-learn	2.4 sec	3 hr 5 min 33 sec (4,639x slower)

Currently, pydoclint supports two docstring styles:

• The numpy stlyle
• The Google style

Support for the Sphinx style may be added in the future if there are requests for it.

Another note: this linter and pydocstyle serves complementary purposes. It is recommended that you use both together.

1. Installation

pip install pydoclint

Note that pydoclint currently only supports Python 3.8 and above. (Python 3.7 support may be added if there are interests and requests.)

2. Usage

2.1. As a native command line tool

pydoclint <FILE_OR_FOLDER>

Replace <FILE_OR_FOLDER> with the file/folder names you want, such as ..

2.2. As a flake8 plugin

Once you install pydoclint you will have also installed flake8. Then you can run:

flake8 --select=DOC <FILE_OR_FOLDER>

If you don't include --select=DOC in your command, flake8 will also run other built-in flake8 linters on your code.

2.3. Native vs flake8

Should I use pydoclint as a native command line tool or a flake8 plugin? Here's comparison:

	Pros	Cons
Native tool	Slightly faster	No per-line or project-wide omission support right now [*]
flake8 plugin	Supports per-line or project-wide omission	Slightly slower because other flake8 plugins are run together

*) This feature may be added in the near future

3. Style violation codes

pydoclint currently has the following style violation codes:

3.1. DOC1xx: Violations about input arguments

Code	Explanation
DOC101	Docstring contains fewer arguments than in function signature
DOC102	Docstring contains more arguments than in function signature
DOC103	Docstring arguments are different from function arguments. (Or did you miss the space between the argument name and the ":" in the docstring?)
DOC104	Arguments are the same in the docstring and the function signature, but are in a different order.
DOC105	Argument names match, but type hints do not match

3.2. DOC2xx: Violations about return argument(s)

Code	Explanation
DOC201	Function/method does not have a return section in docstring
DOC202	Function/method has a return section in docstring, but there are no return statements or annotations

3.3. DOC3xx: Violations about class docstring and class constructor

Code	Explanation
DOC301	__init__() should not have a docstring; please combine it with the docstring of the class
DOC302	The docstring for the class does not need a "Returns" sections

3.4. DOC4xx: Violations about "yield" statements

Code	Explanation
DOC401	Function/method returns a Generator, but the docstring does not have a "Yields" section
DOC402	Function/method has "yield" statements, but the docstring does not have a "Yields" section
DOC403	Function/method has a "Yields" section in the docstring, but there are no "yield" statements or a Generator return annotation

3.5. DOC5xx: Violations about "raise" statements

Code	Explanation
DOC501	Function/method has "raise" statements, but the docstring does not have a "Raises" section
DOC502	Function/method has a "Raises" section in the docstring, but there are not "raise" statements in the body

4. Configuration options

There are several configuration options available. They can be used invidually or together.

4.1. --quiet (shortform: -q)

This flag activates the "quite mode", in which no output will be printed to the command line if there are no violations.

By default, this flag is not activated, so the files that are scanned are printed in the command line.

pydoclint --quiet <FILE_OR_FOLDER>

This option is only available in the "native" command-line mode, rather than in flake8. If you use pydoclint in flake8, please use flake8's own verbosity configuration instead.

4.2. --exclude

You can use this option to exclude files within the given folder. It is a regex pattern of full file paths.

For example:

pydoclint --exclude='\.git|\.tox|tests/data' <FOLDER_NAME>

This option is only available in the native command-line mode. If you use pydoclint within flake8, you can use flake8's --exclude option.

4.3. --style

Which style of docstring is your code base using. Right now there are two available choices: numpy and google. The default value is numpy.

pydoclint --style=google <FILE_OR_FOLDER>

or

flake8 --style=google <FILE_OR_FOLDER>

4.4. --check-type-hint (shortform: -th, default: True)

If True, the type hints in the docstring and in the Python code need to exactly match.

To turn this optoin on/off, do this:

pydoclint --check-type-hint=False <FILE_OR_FOLDER>

or

flake8 --check-type-hint=False <FILE_OR_FOLDER>

4.5. --check-arg-order (shortform: -ao, default: True)

If True, the input argument order in the docstring needs to match that in the function signature.

To turn this optoin on/off, do this:

pydoclint --check-arg-order=False <FILE_OR_FOLDER>

or

flake8 --check-arg-order=False <FILE_OR_FOLDER>

4.6. --skip-checking-short-docstrings (shortform: -scsd, default: True)

If True, pydoclint won't check functions that have only a short description in their docstring.

To turn this optoin on/off, do this:

pydoclint --skip-checking-short-docstrings=False <FILE_OR_FOLDER>

or

flake8 --skip-checking-short-docstrings=False <FILE_OR_FOLDER>

4.7. --skip-checking-raises (shortform: -scr, default: False)

If True, pydoclint won't report DOC501 or DOC502 if there are raise statements in the function/method but there aren't any "raises" sections in the docstring (or vice versa).