Closed shashank-elastic closed 15 minutes ago
Reconciling this PR design
What are the possibilities that we can have without skipping -- We could dynamically change the base branch ref on PR / Push event to use those. -- My Point of view on this is that the test on push events is invariably wont fetch any events. -- There is a diff of the rule on the PR stage, on Push events ur already looking at the modified file and comparing it with its base branch wont fetch any diff. -- On the other hand, back porting branches wont have any PR(s) and are always on direct commit mode. -- So even if we have this test case to have diff on dynamic branch there will be no result as u will be diffing on the latest branch contents.
opening this up to @Mikaayenson / @eric-forte-elastic / @traut to see if this is less disruptive for dev workflows.
Issues
https://github.com/elastic/ia-trade-team/issues/118
Summary
updated_date
before merging to main via pull_request.Additional Information
Code Changes
rules/
andrules_building_block
are considered.'--diff-filter=M'
where only modified files are considered for rule pathrules/
andrules_building_block/
Edge Case
Testing
Local Testing with following git hub events Event- pull_request and Branch = 8.14 --> Test Case Skipped.
Local Testing with following git hub events Event- push and Branch = main--> Test Case Skipped.
Local Testing with following git hub events Event- pull_request and Branch = main--> Test Case Executed
On Github PR Run. The testcase was executed - sample_run
Testing With Skip criteria for push
Locally this can be run without any additional settings
On PR the test is checked --> Refer Latest Unit Test Result.
When event is push the test case is skipped