Warehouse roadmap
As we develop Warehouse, here are the milestones along the way. To talk about them with us, please contact us.
1. Maintainer Minimum Viable Product milestone -- COMPLETE
The point of this first milestone is to give package maintainers a solid chance to try out Warehouse and report critical bugs early. This milestone is for necessary features so we can ask package maintainers to try, test, and report bugs against Warehouse.
We completed this milestone on Feb. 20th, 2018.
2. End User MVP -- COMPLETE
Necessary features for Python users (other than package maintainers) to try, test, and report bugs against Warehouse. Including infrastructure improvements tracked in the PyPI infrastructure repository.
We completed this milestone on Feb. 23rd, 2018.
3: Publicize beta -- in progress
Blog, email, etc. to invite Python community to test Warehouse. Redirect an increasing portion of pip install traffic to Warehouse, as we did during a load test in mid-March. This thus requires an infrastructure changeover.
4: Launch: redirect pypi.python.org to pypi.org
Replace the legacy code base powering PyPI with Warehouse, e.g., redirect web browser requests for pypi.python.org to pypi.org. Warehouse (as instantiated at PyPI.org) at this milestone is good enough to be the main entryway to Python packages. We will announce this on the pypi-announce mailing list.
This is not necessarily at feature parity with the old PyPI system. We auto deploy the master branch to both pypi.org and test.pypi.org and pypi.org and pypi.python.org share data stores (so they're really just two diffrent views over the same data) (and test.pypi.org and testpypi.python.org also share data store), so this is all live and able to be used right now, so launching is really just flipping the switch to make people go to the new site by default. We're going to have to keep the legacy code base running somewhere for people to be able to access it until we maintain feature parity, so that people depending on something that hasn't yet been implemented have some mechanism to continue to do that.
5: Shut Down Legacy PyPI
Prerequisites to shut down legacy PyPI. In general, we're aiming to do this once we reach feature parity, but some existing features may be acceptable to complete later.
We will announce this on the pypi-announce mailing list. Tentative schedule: the second half of April, 2018.
6. Post Legacy Shutdown
Issues that become unblocked once legacy is dead (RIP).
Cool but not urgent
Wishlist; not part of the critical path for shutting down legacy PyPI.