Description
What is the problem or limitation you are having?
At present, CI only tests iOS builds, using the Python-standalone-testbed project. Other platforms are, at present, untested release artefacts.
However, both iOS and visionOS provide a testbed project that can be invoked to trigger the CPython test suite.
Describe the solution you'd like
An additional CI stage that invokes the testbed for the given platform.
To save time/resources, it probably doesn't need to run the full test suite on every commit. A "representative subset" of the full CPython test suite, checking basic language features, some of the dynamic libraries (like ssl, hashlib and bzip2), and complex features like sockets would probably suffice, rather than doing a full ~20 minute CI run on every platform.
In a super ideal case, a representative subset would be run by default, but adding a tag to the PR (e.g., preview) would result in the full suite being executed, as would the release workflow.
Describe alternatives you've considered
Status quo
Additional context
The testbed is currently only available for iOS and visionOS; iOS is being tested with Python-support-testbed. That test should be retained; it checks a couple of additional features about the runtime that are useful to confirm.
Any solution should allow for other backends to be tested if/when a testbed is added for watchOS/tvOS.