-
Notifications
You must be signed in to change notification settings - Fork 1.2k
Checklist of pull request approval by Eclipse Che QE team
-
Definition-Of-Done (DoD) is clear: feature specification / bug use case described and available for review.
-
If DoD achievement verification requires creation of new E2E selenium tests or updating existed ones - they have to be created or updated.
-
Code should has latest changes from the master branch before start E2E selenium tests.
-
There are no changes in the code in pull request after E2E selenium tests execution was requested.
-
Coverage of new or updated functionality by unit-tests >= 80%. We can use Cobertura maven plugin to measure coverage - next command generate Cobertura report:
mvn cobertura:cobertura -
Build of whole project including unit and integration tests has passed successfully. it can be chacked by command
ci-buildin comment of pull reauest, which run build on [dedicated CI job]9https://ci.codenvycorp.com/view/pr-builds/job/che-pullrequests-build/). -
Execution of E2E selenium tests for Eclipse Che Multiuser on OCP (command
ci-testin comment to the pull request to run selenium tests on dedicated CI job) shows no regression, that is success rate about 100%. Such execution doesn't include tests which fail because of known issue. "flaky" tests has been restarted once. -
If pull request creator decides that changes influence assembly of Eclipse Che on Docker, E2E selenium tests should be executed against Eclipse Che on Docker as well - by using another the command
ci-test-docker-single, which run build on another CI job. -
Actually, E2E selenium tests cover about 50% of functionality of Eclipse Che upstream. There is a list of features which are not covered by E2E selenium tests.
-
There are some tests which require special configuration of product, or are executing to long, and so don't run by ci-test commands on Pull Requests: