dependency-metrics 0.1.2

Python tool to track outdated dependencies.

Welcome to dependency-metrics

Summary

A command line tool to analyze and monitor how up-to-date dependencies of a project are. For more detail, see the Usage section below.

Supported Package Managers

• pip
• yarn

Supported Application Monitoring Platforms

• DataDog

Usage

Run the metrics tool from within the repository you wish to generate metrics for.

Default behavior

All that is needed is the package manager you want to generate metrics for, and metrics will output a table detailing how out-of-date each dependency is.

$ metrics pip
Behind   Package                      Latest       Version
0.1.0    my-depedency                 1.1.0        1.0.0
2.0.0    my-other-dependency          3.2.1        1.2.3

--stats option

Use the --stats option to generate a simple dictionary that displays the total number of outdated dependencies, as well as a breakdown detailing the number of outdated dependencies for each version type.

$ metrics pip --stats
Total: 5
Outdated: 2
Multi-Major: 1
Major: 0
Minor: 1
Patch: 0
Unknown: 0

--send option

NOTE: Additional setup is required to successfully post stats to a supported backend. See platform setup for further information.

Use the --send option to post the same metrisc generated by --stats to an application monitoring platform.

$ metrics pip --send

Application Monitoring Platform Setup

This is required to make use of the --send option, enabling the ability to send generated stats to a backend of your choice.

DataDog

Set the DATADOG_API_KEY and DATADOG_APP_KEY environment variables in the environment you will run metrics <package_manager> --send from.

Development

Requirements

pip install -e .  # installs dependencies defined in pyproject.toml
pip install -e .[test]  # installs test dependencies -- '.[test]' if using zsh

Tests

Must install test dependencies first. See requirements above.

nose2  # runs all tests
nose2 dot.path.to.file.class.or.test  # runs specific test

Publishing to PyPI

To publish a new release on pypi.org:

• Update __version__ in dependency_metrics/init.py.
• Create and push a git tag with the new version number.

  git tag vX.Y.Z
  git push --tags
  

• Wait for the pypi.yml workflow to build and publish the new release.

A dev release is published on test.pypi.org on each new push or merge to main. A dev release may also be published on-demand for any branch with workflow dispatch.