In this PR, we update the skip strategy for ci-workflow to use the GitHub actions feature and simply filter out the changed paths (skip docs, changes).
* Add a retry on the setup test environment step
* Minor change to an e2e test to trigger CI
* Add changefile(s) from automation for the following project(s): woocommerce
---------
Co-authored-by: Jon Lane <jon.lane@automattic.com>
Co-authored-by: github-actions <github-actions@github.com>
In this PR, we are re-iterating the job speedup and leveraging single wp-instance, incremental builds, and baseline reports caching. The original benchmarking tool is still available but bypassed as we evaluate results for this iteration approach.
In this PR we refine cleaning build artifacts to build-folders only and leverage wireit integration in GitHub actions for leveraging incremental build approach in the job and improving the job running time.
* Add storybook workflow
* Update version and add trigger for push on trunk
* Add workflow_dispatch for manual trigger also
* Add permissions
* Add trigger for current branch
* Set install to true for monorepo setup
* Add write permissions
* Use hashes for GH action versions and add --quiet for building storybook
* Remove on push trigger for now
In this PR we spin off usage of recently introduced packageManager-property across CI workflows and consolidate the version of pnpm/action-setup action we are using.
* Use Playwright's last info run
* Add changelog
* Fix workflow syntax
* Fix workflow syntax
* Try with continue-on-error
* Try with merge-multiple
* Try with pattern instead of name
* Print the run command
* Print the run command
* Use LAST_RUN_FLAG
* Update outputDir
* Update upload paths and parse the file content
* Don't try to download lat run info if the test type is not e2e
* Rename the downloaded file
* Rename the downloaded file
* Rename the downloaded file
* Full path to the downloaded file
* Check download path content
* Check download path content
* Check the last-run file in the right path
* Fail different tests on different run attempts
* Fix artifact name
* Force test failures
* Cleanup console logs
* Cleanup console logs
* Cleanup console logs
* Log if last run info exists but no tests are found
* Revert test changes
In this PR, we use shallow checkout if the environment provides no base_ref, which is needed for identifying changed files. If no ref provided, we are not able to calculate the diff and therefore go with faster shallow copy.
Vladimir Reznichenko
Jonathan Lane
louwie17
Adrian Moldovan
nigeljamesstevenson
jonathansadowski
Bart Kalisz
Chi-Hsuan Huang
Jason Kytros