Smelly Python

A code smell reporting tool for Python projects, created for the Release Engineering for Machine Learning applications course at TU Delft.

Smelly Python generates an HTML report for easy viewing of Pylint errors.

How To

1. Run pip install smelly-python
2. Run smelly-python -d {src}. Replace {src} with the directory you want to analyse.
3. Open ./report/smelly_python/index.html to view the generated report.

Using Advanced Features of Pylint

Smelly Python will run the following command: pylint {src} --output-format:json:report.json,text:grade.txt --exit-zero. Therefore, in order to customize what settings Pylint runs with, use the .pylintrc file to configure it.

Using Smelly-Python in GitHub Actions

The tool has been designed to be run within a GitHub workflow using the smell-my-pr GitHub action. The action will automatically post the output of the tool to your GitHub pull request as a comment and add a summary and artifact to the job.