Much of the work of major software upgrades can now be automated via “codemods” - software that modifies other software’s source code (source code modification). Here is a collection of such codemods that either have saved us time in previous upgrades or seem likely to do so in future ones. For more context on many of these, see Herding Ponies: Coordinating and Automating Django Upgrades Across 100+ Repositories .

Applying Codemods Across Multiple Repositories

• https://github.com/edx/.github/blob/master/.github/workflows/bulk_repo_update.yml

Continuous Integration

• repo-tools/edx_repo_tools/codemods/django3 at master · openedx/repo-tools - Examples for auto-updating GitHub Actions workflows, tox.ini, and .travis.yml

Django

• GitHub - adamchainz/django-upgrade: Automatically upgrade your Django projects.

• GitHub - browniebroke/django-codemod: A tool to automatically fix Django deprecations.

• repo-tools/edx_repo_tools/codemods at master · openedx/repo-tools

• New options may appear in Django Packages : Auto-Upgrade

Python

• https://github.com/asottile/pyupgrade

Codemod Frameworks - Python

• Fixit 2: Meta’s next-generation auto-fixing linter (built on LibCST)

• https://github.com/Instagram/LibCST

• https://github.com/facebookincubator/Bowler - deprecated/inactive

• GitHub - google/pasta: Library to refactor python code through AST manipulation. - inactive

• GitHub - PyCQA/redbaron: Bottom-up approach to refactoring in python - inactive

Deprecation Warnings Classification - Python

• https://github.com/edx/edx-platform/blob/master/openedx/core/process_warnings.py

• GitHub - openedx/pytest-warnings-report: Create an aggregate report of the warnings seen during test runs. (incomplete refactoring of the edx-platform code above to make it usable in other repos)