Cornershop Static Analysis

ast

Useful classes are under apps/patterns/analyzer.py. For examples on how to use, check the existing notebooks.

Configuration

Why you need a Github token?

We are going to analyse the backend repo, which is private. If you're in the organization, you can download it.

You must first create a .env file, within the analysis folder. Inside complete the following values

GITHUB_TOKEN=yourtoken

You can get your personal Github API token by visiting your profile -> Developer settings -> Personal access tokens.

• Generate a new token
• Mark the repo scope and the org:read scope.
• Copy the token and add it to .env.

Notebooks

check_catalogdb_services.ipynb

Use this notebook to:

• get a list of the services only present in your branch
• get a list of name collisions between your services and every other open PR labled with "catalogdb" in the backend repo.

Read the notebook for more information.

sample_prs.ipynb

Use this notebook to:

• get a list of open PR ordered by a custom score. You must set FROM_DATE, TO_DATE, SAMPLE_SIZE.

extract_services_documentation.ipynb

Use this notebook to analyse documentation on the services.py module of any app (technically any file will do)

Analysis is currently performed over every function inside the supplied file and the score is based on three elements:

• argument typing
• non-empty docstring
• return type present

score goes from 0 to 100.