Can packaging improve Django deployments?
summary
Argues that turning a Django site into a Python distribution makes deployments reproducible and faster: use setup.cfg + MANIFEST.in to build wheels, add an entry point for manage commands, and pin a full dependency tree with pip-compile into a constraints.txt. Build artifacts on CI and serve them via devpi or static HTTP, install with pip -c to avoid compiles on production, and manage runtime settings via environment-backed config libraries so identical releases run across many servers.
AI-written summary · about
chapters (8)
- 0:00Introduction and motivation
- 2:40Packaging a Django project
- 6:01Refining project configuration
- 12:23Manifests and project build
- 15:43Installing dependencies
- 20:10Deployment strategies
- 22:22Configuration and tools
- 26:59Summary and conclusion
related talks
53mDjango productivity tips and tricks
41mDjango Deployments Done Right
53mDjango productivity tips and tricks
26mWhat Django Deployment is Really About
36mPython Packaging from Init to Deploy
24mDjango in Production with PEX
34mContinuous Deployment for webapps based on Django
30m"What Django Deployment is Really About"
