- Add support for module_utils packages for better organization.
- Add support for "virtual" module_utils packages such as `six`.
(cherry picked from commit 4fdeade389)
* Add unified git diff parser.
* Add metadata and diff handling.
* Add test confidence/verification to bot output.
(cherry picked from commit 869449e288)
* Refactor sanity classes for use in all tests.
* Use lint/junit output for compile test.
* Add missing options for compile test.
* Fix early bailout on requirements install.
Without this, changing a large number of files results in target
processing taking a very long time due to repeatedly compiling
the same patterns in a loop over many targets.
- Tests are run to completion instead of stopping on first failure.
- Test results are now parsed instead of passing through to the console.
- Test results can be saved in junit xml format.
- Test results will show up on the Shippable "Tests" result tab.
- Added an experimental --lint option for easier integration with other tools.
- Code smell tests are now usable with the --list-tests, --test and --skip-test options.
- Code split out from executor.py into sanity.py.
- Rename download-logs to download.py and add support for test and coverage results.
- Miscellaneous improvements.
* Re-enable module comparisons, specifically for new module detection and for finding new options/arguments
* Only do new module checks in shippable, local will display warning
* Skip pep8 analysis when --explain is used.
* Fix return type annotations.
* Match line length requirement of PEP 8 config.
* Consider module_utils deps when running tests.
* updates all ios modules to support persistent socket
* adds ios action plugin to connect to device
* adds exec_command() to ios shared module
* fixes ios_config and ios_template local action
* update all unit test cases
* adds base test module for ios module testing
This reverts commit 91526cd9f2.
Removing this feature primarily because it interferes with
collecting proper code coverage results. I may restore the
feature later if that can be resolved.
* Unittests for some of module_common.py
* Port test_run_command to use pytest-mock
The use of addCleanup(patch.stopall) from the unittest idiom was
conflicting with the pytest-mock idiom of closing all patches
automatically. Switching to pytest-mock ensures that the patches are
closed and removing the stopall stops the conflict.
* Only start platform instances with tests selected.
* Enable ios on Shippable.
* Show inventory in explain mode.
* Fix indentation of generated network inventory.
* Update classification of network module_utils.