dep-appearances

A simple tool to see where your dependencies are imported. The dep-appearances CLI produces a report of which files import each of your dependencies. At this time, the CLI only works for projects that use pipenv, but support for any dependency management tool could be added.

Requirements

Python 3

How to Use dep-appearances

You can install dep-appearances via pip:

pip install dep-appearances

Installing the package provides a CLI:

dep-appearances --help
usage: dep-appearances [-h] [--underused_threshold UNDERUSED_THRESHOLD] [PATH]

Find dependencies that are unused and underused in your codebase.

positional arguments:
  PATH                  The path to your project's root (defaults to your
                        current working directory)

optional arguments:
  -h, --help            show this help message and exit
  --underused_threshold UNDERUSED_THRESHOLD
                        The threshold to set for marking dependencies as
                        underused (default: 2)

From the root of your project (i.e. wherever your Pipfile is) you can run dep-appearances and you will get a report of dependencies that don't appear to imported and a report of dependencies that my not be imported in very many places.

> dep-appearances
Unused dependencies:
	build
	pytest
	twine

Underused dependencies (usage threshold = 2):
	pipfile
		imported in:
		src/dep_appearances/appearances_report.py:3

Known Shortcomings

There are, unfortunately, packages that have a different name when importing them than when installing them. For example, the apache-airflow package shows up in a Pipfile as apache-airflow, but when it is used in a codebase you use

import airflow

dep-appearances currently does not account for such cases. Therefore you should not remove dependencies from your codebase without confirming that packages are unused.

How to contribute to dep-appearances

Pull requests are definitely welcome. Just fork this repo and open a PR against the main branch.

Useful Development Commands

Install dependencies

pipenv install

Install and use the package locally (to work on the CLI):

pipenv run pip install -e .
pipenv run dep-appearances

Running tests:

pipenv run pytest

Release Process

# Generate distribution archives:
python3 -m build
# => Should create files in dist/

# Push to pypi
python3 -m twine upload dist/*

Building for test.pypi

python3 -m build
python3 -m twine upload --repository testpypi dist/*

Install from test.pypi:

python3 -m pip install --index-url https://test.pypi.org/simple/ --no-deps dep-appearances

jdlubrano/dep-appearances