Minimum version requirements for sanity tests have been standardized:
- All single version sanity tests now require Python 3.5 or later.
- All multiple version sanity tests continue to use all supported Python versions.
- All version neutral sanity tests continue to work on any supported Python version.
Previously some tests required 3.5 or later with most of the remaining tests requiring 2.7 or later.
When using the `--python` option to specify a Python version:
- Tests which do not support the specified Python version will be skipped with a warning.
- If the specified Python version is not available, any test attempting to use it will generate an error.
When not using the `--python` option to specify a Python version:
- Multiple version tests will attempt to run on all supported versions.
- Single version tests will use the current version if supported and available, or if no supported version is available.
- Single version tests will use the lowest available and supported version if the current version is not supported.
- Any versions which are not available or supported will be skipped with a warning.
Unit tests automatically skip unavailable Python versions unless `--python` was used to specify a version.
* Initial ansible-test support for collections.
* Include cloud config in delegation payload.
* Add missing types import and fix `t` shadowing.
* Fix plugin traceback when config_path not set.
* Fix encoding issues.
* Remove unused imports.
* More encoding fixes.
* Handle delegation outside exception handler.
* Inject ssh keys only if not already in place.
* More defensive approach to getting remote pwd.
* Add missing string format var.
* Correct PowerShell require regex.
* Rename `is_install` and `INSTALL_ROOT`.
* Add timeout support to ansible-test.
* Fix ansible-test tar filename filter bug.
* Update timeouts used on Shippable.
* Kill subprocesses when parent process terminates.
* Require explicit use of env --show option.
Track the interpreter for each copy of the injector by the interpreter
path instead of the interpreter version. This avoids the possibility
of mixing different interpreters with the same version.
Inject a symlink to the correct python into the copied injector
directory instead of altering the shebang of the injector. This
has the side-effect of also intercepting `python` for integration
tests which simplifies cases where it needs to be directly invoked
without collecting code coverage.
* win_exec: refactor PS exec runner
* more changes for PSCore compatibility
* made some changes based on the recent review
* split up module exec scripts for smaller payload
* removed C# module support to focus on just error msg improvement
* cleaned up c# test classifier code
Now that we don't need to worry about python-2.4 and 2.5, we can make
some improvements to the way AnsiballZ handles modules.
* Change AnsiballZ wrapper to use import to invoke the module
We need the module to think of itself as a script because it could be
coded as:
main()
or as:
if __name__ == '__main__':
main()
Or even as:
if __name__ == '__main__':
random_function_name()
A script will invoke all of those. Prior to this change, we invoked
a second Python interpreter on the module so that it really was
a script. However, this means that we have to run python twice (once
for the AnsiballZ wrapper and once for the module). This change makes
the module think that it is a script (because __name__ in the module ==
'__main__') but it's actually being invoked by us importing the module
code.
There's three ways we've come up to do this.
* The most elegant is to use zipimporter and tell the import mechanism
that the module being loaded is __main__:
* 5959f11c9d/lib/ansible/executor/module_common.py (L175)
* zipimporter is nice because we do not have to extract the module from
the zip file and save it to the disk when we do that. The import
machinery does it all for us.
* The drawback is that modules do not have a __file__ which points
to a real file when they do this. Modules could be using __file__
to for a variety of reasons, most of those probably have
replacements (the most common one is to find a writable directory
for temporary files. AnsibleModule.tmpdir should be used instead)
We can monkeypatch __file__ in fom AnsibleModule initialization
but that's kind of gross. There's no way I can see to do this
from the wrapper.
* Next, there's imp.load_module():
* https://github.com/abadger/ansible/blob/340edf7489/lib/ansible/executor/module_common.py#L151
* imp has the nice property of allowing us to set __name__ to
__main__ without changing the name of the file itself
* We also don't have to do anything special to set __file__ for
backwards compatibility (although the reason for that is the
drawback):
* Its drawback is that it requires the file to exist on disk so we
have to explicitly extract it from the zipfile and save it to
a temporary file
* The last choice is to use exec to execute the module:
* https://github.com/abadger/ansible/blob/f47a4ccc76/lib/ansible/executor/module_common.py#L175
* The code we would have to maintain for this looks pretty clean.
In the wrapper we create a ModuleType, set __file__ on it, read
the module's contents in from the zip file and then exec it.
* Drawbacks: We still have to explicitly extract the file's contents
from the zip archive instead of letting python's import mechanism
handle it.
* Exec also has hidden performance issues and breaks certain
assumptions that modules could be making about their own code:
http://lucumr.pocoo.org/2011/2/1/exec-in-python/
Our plan is to use imp.load_module() for now, deprecate the use of
__file__ in modules, and switch to zipimport once the deprecation
period for __file__ is over (without monkeypatching a fake __file__ in
via AnsibleModule).
* Rename the name of the AnsiBallZ wrapped module
This makes it obvious that the wrapped module isn't the module file that
we distribute. It's part of trying to mitigate the fact that the module
is now named __main)).py in tracebacks.
* Shield all wrapper symbols inside of a function
With the new import code, all symbols in the wrapper become visible in
the module. To mitigate the chance of collisions, move most symbols
into a toplevel function. The only symbols left in the global namespace
are now _ANSIBALLZ_WRAPPER and _ansiballz_main.
revised porting guide entry
Integrate code coverage collection into AnsiballZ.
ci_coverage
ci_complete
- Works with the --remote option.
- Can be disabled with the --disable-httptester option.
- Change image with the --httptester option.
- Only load and run httptester for targets that require it.
* Fix type hint typos.
* Add one-time cloud env setup after delegation.
* Add generate_password to util.
* Add username/password support to HttpClient.
* Avoid pip requirement for ansible-test shell.
* Support provisioning Tower instances.
* allow ANSIBLE_KEEP_REMOTE_FILES for local test runner
* add ANSIBLE_KEEP_REMOTE_FILES to tox.ini, update docs
* Clarify handling of environment variables.
* Use correct pip version in ansible-test.
* Add git fallback for validate-modules.
* Run sanity tests in a docker container.
* Use correct python version for sanity tests.
* Pin docker completion images and add default.
* Split pylint execution into multiple contexts.
* Only test .py files in use-argspec-type-path test.
* Accept identical python interpeter name or binary.
* Switch cloud tests to default container.
* Remove unused extras from pip install.
* Filter out empty pip commands.
* Don't force running of pip list.
* Support delegation for windows and network tests.
* Fix ansible-test python version usage.
* Fix ansible-test python version skipping.
* Use absolute path for log in ansible-test.
* Run vyos_command test on python 3.
* Fix windows/network instance persistence.
* Add `test/cache` dir to classification.
* Enable more python versions for network tests.
* Fix cs_router test.
* Move sanity into directory.
* Omit abstract classes from returned subclass list.
* Split sanity tests out into plugins.
* Fix abstract class handling for Python 3.
