* ci: improve Azure test reliability
Azure has intermittent failures, often the macOS framework-Python
`posix_spawn: Undefined error: 0` spawn race under load, plus transient
download flakiness. Address both:
- Retry integration tests via pytest-rerunfailures (already required):
add `--reruns=2 --reruns-delay=5` to the serial and non-serial runs so a
single transient flake self-heals instead of failing the whole job.
- Cap Azure at `--num-processes 3` to reduce concurrent process spawning
(the macOS runner was using 4 xdist workers), lowering the spawn-race
probability and peak resource pressure.
- Cache downloads across runs: relocate CIBW_CACHE_PATH to a stable path
and add a Cache@2 task (interpreter downloads + per-worker test pip
caches live under it), cutting the network-download flake surface.
- Bump the Azure host interpreter to 3.13.
Assisted-by: ClaudeCode:claude-opus-4.8
* ci: stabilize iOS tests, bound rerun cost
An iOS run on Azure (build 9043) hung for the full 40-min pytest timeout
on the second build config of test_ios_platforms, then reran twice
(reruns=2), blowing past the 180-min job cap with no useful signal. The
hang was a stuck simulator/xcodebuild on the second config, after the
first config had run.
- Shut down running simulators before each parametrized config via a
`clean_ios_simulators` fixture, so a simulator left booted/wedged by the
previous config isn't reused.
- Drop iOS reruns from 2 to 1. One retry still covers the documented
"fails the first time" simulator-boot flake, while halving the
worst-case retry cost (3x40=120min -> 2x40=80min) on a hang.
Assisted-by: ClaudeCode:claude-opus-4.8
* ci: don't limit num processes
Signed-off-by: Henry Schreiner <henryfs@princeton.edu>
* ci: rerun OCI unit tests that pull from Docker Hub
test_local_image et al. occasionally fail on a transient Docker Hub
anonymous-pull blip ("unauthorized: authentication required") when
pulling debian:trixie-slim. The integration runs already self-heal via
pytest-rerunfailures, but the unit run had no reruns, so a single flake
failed the whole job. Mark the three network-dependent OCI tests
(test_local_image, test_enter_error, test_multiarch_image) flaky so they
retry instead of blanket-rerunning the suite.
Assisted-by: ClaudeCode:claude-opus-4.8
* Drop iOS clean step
Co-authored-by: Henry Schreiner <HenrySchreinerIII@gmail.com>
* ci: scope integration reruns to test_0_basic::test
Drop the blanket --reruns from the serial and non-serial integration
runs and instead mark test_0_basic.py::test as flaky, so only the known
flaky basic test is retried.
Assisted-by: ClaudeCode:claude-opus-4.8
---------
Signed-off-by: Henry Schreiner <henryfs@princeton.edu>
* Revert Azure iOS runners to macos-14.
* Add warning about known macos-15 issues on Github Actions and Azure.
* Update workflow reference with actual ticket.
* ci: rework tests to break out ios/android tests
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
* Apply suggestions from code review
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Apply suggestions from code review
* ci: fix check for wheels
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
* Update .github/workflows/test.yml
* tests: pass via flag instead of envvar
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
* tests: apply to serial/parallel
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
* tests: join together marks
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
* tests: join together marks
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
* ci: break up azure job
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
* ci: use matrix for azure pipelines
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
* Try installing Java on macos as well
* Revert "Try installing Java on macos as well"
This reverts commit aedf0f5106d2467938705b85180a0b249d49b99d.
* Try to get the most basic android test running first
* Rename platform to test_select, move headers in run_tests
Maybe it's a personal thing, but the header location was really making
me have to think more than I needed to look at that file!
* Only run docker unit tests on machines that are testing linux
* Update the flag name in azure-pipelines too
* Update azure-pipelines.yml
---------
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Matthieu Darbois <mayeut@users.noreply.github.com>
Co-authored-by: Joe Rickerby <joerick@mac.com>
* Add Android to resource files
* Add Android to miscellaneous places
* Add Android documentation
* Docs cleanups
* Add Android platform module; implement top-level structure and target Python installation
* Implement setup_env and build_wheel
* lru-dict build working
* Alter prefix in sysconfigdata file; fix various issues with FLAGS variables
* Implement Android testing
* Add type annotations to _cross_venv
* Revert Python 3.8 to pip 25.0.1
* Make test-sources required on Android
* Add Android integration tests
* Test cleanups
* Add test of all available Python versions
* Update test-sources and test-command behavior to match iOS
* Documentation cleanups
* Replace Builder class with a set of global functions
* Rename "env" to "build_env"
* Remove Chaquopy repository from default pip command line
* Move native_platform to platforms module
* Fix parse_config_settings
Co-authored-by: Joe Rickerby <joerick@mac.com>
* Add unit tests for parse_config_settings and arch_synonym
* Make `shell_prepared` arguments keyword-only, and add tests for the commands that use it
* Replace `importlib.util.spec_from_file_location` with `runpy.run_path`
* Use python-build-standalone
* Update Android Python
* Enable KVM in Linux CI
* Move KVM code to test_android.py
* Use Java 17 on Azure
* Install emulator if necessary before running -accel-check
* Free up additional disk space on Linux runners
* Add sudo
* Skip emulator tests on CI platforms that don't support it
* Download Android Python from Maven Central
* Free up more disk space on Linux runners
* fix: minor fixups
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
* Set sysconfig._BASE_PREFIX to support sysconfig.get_path("include")
* Get ANDROID_API_LEVEL from the build environment, not cibuildwheel's own environment
* Correct relative path of test-sources
* Pass a CMake toolchain file to the build
* Add "repair" step which adds libc++ to the wheel when necessary
* Add missing needs_emulator decorator
* Provide useful error message if ANDROID_HOME is not set
* Remove use of HOST environment variable
* Update to Python 3.15.5
* Fix PyLint warnings, clarify comment
* Group common arguments into a dataclass
* Handle environment variables containing newlines
* Discourage the use of `pytest` test commands without `python -m`
* Use single quotes in user-visible messages
* Improve testing documentation
* Pass wheel filename to `log.build_end`
* In GitHub Actions example, skip Android tests on macOS
* Correct relative paths in `patchelf --set-rpath`
* Clarify `test-sources` docs
* Update to Python 3.13.5+20250722.214220
---------
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
Co-authored-by: Joe Rickerby <joerick@mac.com>
Co-authored-by: Henry Schreiner <henryschreineriii@gmail.com>
* Allow CIBW_ENABLE to control the wheels built in testing
* Set CIBW_ENABLE using PR labels
* Add docs
* Build everything on the main branch
* Add CIBW_ENABLE=all option
This was mostly for use in the `main` building case, because otherwise it's maybe a bit too easy to forget to update this file when adding an enable group
* Remove dead code
* Make unit tests robust to the value of CIBW_ENABLE
* Fix tests that explicitly choose pypy
* Fix test expectation
* Don't expect impossible wheels in expected_wheels
* Simplify logic in expected_wheels
* CircleCI- run with CIBW_ENABLE=all only on the main branch
* Azure pipelines - run with CIBW_ENABLE=all on main branch
* Update gitlab to run CIBW_ENABLE=all on main
* Set CIBW_ENABLE=all on travis - it only runs on main anyway
* Fix job name error on CircleCI
* Fix tests for graalpy
* Update the test configuration to use the label
* Remove duplication of default value. Make it affect sample build too
* Move the action to after deps are installed
* GraalPy workaround for this assumption
* Make unit test resilient to changing CIBW_ENABLE
* Properly close files used for testing
* Add support for GraalPy
* Help GraalPy discover build tools on Windows
* Expect manylinux-interpreters ensure graalpy* warning in pip
* Workaround GraalPy bugs on Windows
* Workaround oracle/graalpython#491 also when uv is not available
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update azure-pipelines.yml
* Update azure-pipelines.yml
* refacotor: use pathlib.write_text
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
* Include GraalPy in docker_warmup and remove workaround for installing it late
---------
Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Henry Schreiner <HenrySchreinerIII@gmail.com>
* chore: use SPEC 0 schedule for cibuildwheel
* remove support for {python} and {pip} in commands
* Remove specific Python versions from the update-dependencies job
The requirements pinning is done by uv now, so we don't need to
run the versions of Python to do the pinning anymore.
---------
Co-authored-by: Joe Rickerby <joerick@mac.com>
* fix: enforce minimum version of docker/podman
This allows to always pass `--platform` to the OCI engine
thus fixing issues with multiarch images.
* Allow older versions with warnings
* Upgrade docker on Travis CI
* fix: use `docker cp` instead of `tar`
* Enforce docker>=24.0
* move log to include container.copy_into
* fix: travis-ci, only update docker on aarch64
* skip test_multiarch_image on s390x / ppc64le
* skip flaky test
* chore: only install test deps on Travis CI
* fix: do not try to pull images tagged `cibw_local`
* use "--pull=never" for local images
* Use docker image inspect to check if an image needs to be pulled
Test errors in Azure are all printing KeyError: 'reason', it seems to be an incompatibility between pytest-azurepipelines and pytest-xdist. I've filed a bug here: https://github.com/Azure/pytest-azurepipelines/issues/71
in the meantime, I'll disable this plugin.
Henry Schreiner