I’m assuming you’re talking about the new RunRelevantTests option for test-on-deploy? If so, are you also using the new TestFor and/or Critical options for the @IsTest annotation, or are you just letting Salesforce determine which tests are relevant for the deployed metadata. A concrete example that includes what you’d expect to have been run — and why — and what was actually run would be useful. It would likely need to go to Salesforce since IC is just setting that flag on the deployment request, but I’d certainly be curious to hear the details as well.
I like the idea of another pseudo-option for RunDeployedTests or perhaps RunTestsForDeployed. Do you mind logging an enhancement request in the public issue tracker?
Regards,
Scott Wells
You don’t have to use TestFor to specify explicit dependency relationships, though it can help if Salesforce isn’t finding some relationships that do exist, e.g., due to reflection-based access. It should find the obvious ones. And IC has an option to populate TestFor using its Apex transitive dependency analysis capabilities as well. But if you’re seeing it missing obvious relationships even without that annotation attribute, it’d be good to raise that to Salesforce.
And thanks for logging the enhancement request. I’m pretty deep into another integration right now but can hopefully look at it pretty soon.
Regards,
Scott Wells