dep-check

dep-check is a Python dependency checker. It lets you write rules to define what each module can import, dep-check can then parse each source file in your project to verify everything is in order. You can also use it to draw a dependency graph for your project.

Free software: MIT license

Supported languages

• Python

By default, the tool assumes it's operating on a Python project.

Installation

To install dep-check, run this command:

pip install dep-check

This is the preferred method to install dep-check, as it always installs the most recent stable release.

If you don't have pip installed, this Python installation guide can guide you through the process.

You can also see other installation methods.

Usage

Build your configuration file

dep_check build <ROOT_DIR> [-o config.yaml] [--lang LANG]

This command lists the imports of each module in a yaml file. Use this file as a starting point to write dependency rules on which module can import what, using wildcards.

Here is an example of additional rules added to the initial yaml file:

---

dependency_rules:
'*':
    - dep_check.models
    - dep_check.dependency_finder
    - dep_check.checker

dep_check.infra.io:
    - dep_check.use_cases%
    - jinja2
    - yaml

dep_check.infra.std_lib_filter:
    - dep_check.use_cases.interfaces

dep_check.use_cases%:
    - dep_check.use_cases.app_configuration
    - dep_check.use_cases.interfaces

dep_check.main:
    - '*'

lang: python
local_init: false

• Use * to include any string, even an empty one.
• Use % after a module name (e.g. my_module%) to include this module along with its sub-modules.

Here, for instance, the configuration file defines that dep_check.infra.io can import dep_check.use_cases, along with dep_check.use_cases.build, dep_check.use_cases.check, and so on.

To see all the supported wildcards, check the User Manual.

Check your configuration

Once your config file is ready, run

dep_check check <ROOT_DIR> [-c config.yaml] [--lang LANG]

The command reads the configuration file, and parses each source file. It then verifies, for each file, that every import is authorized by the rules defined in the configuration file.

When it's done, it writes a report on the console, listing import errors by module and unused rules:

Draw a dependency graph

You need to have graphviz installed to run this command

dep_check graph <ROOT_DIR> [-o file.svg/dot] [-c config.yaml] [--lang LANG]

Note : if you generate a svg file, a dot file is also created in /tmp/graph.dot

A lot of options are available to customize your graph (hide some modules, add layers etc.). Check the User Manual for more information.

Contributing

If you want to make a contribution, be sure to follow the Contribution guide.

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template

Authors & contributors

Check out all the Authors and contributors of this project.

CHANGELOG

3.0.1(2023-03-28)

BUGFIX

• Use OrderedSet instead of set, in order to have a deterministic runs

3.0.0(2022-12-21)

MAJOR

• Update python support, tested on 3.9, 3.10, 3.11.
• Add support of dynamic rules (see doc)
• Support multiple files and directories in parameters.
• Drop error_on_unused option (replaced by unused_level).
• Add unused_level option to choose the raise an error when unused rules are detected.

2.0.1(2022-06-02)

MAJOR

• Drop go compatibility.
• Update python support, tested on 3.8, 3.9, 3.10.
• Add error_on_unused option to raise an error when unused rules are detected.

1.0.3(2019-08-26)

BUGFIX

• The tool now works correctly with Go.

1.0.0 (2019-08-20)

Modified

• The tool now writes a full report, displaying the number of errors, warnings and files.
• The commands have been simplified, to avoid too long command lines

Added

• The tool now supports Go language
• Better documentation

0.2.0 (2019-07-16)

Added

• You can now add layers to your graph.
• The tool now warns you if a rule in your configuration file is not used.

0.1.0 (2019-07-09)

• First release on PyPI.