jonnybot0 opened a new pull request, #299:
URL: https://github.com/apache/groovy-geb/pull/299

   This reverts the changes from https://github.com/apache/groovy-geb/pull/270. 
   
   Those changes were purportedly meant to fix the flakiness in 
`geb.spock.ParallelExecutionSpec` (per this comment from @Vampire). However, 
that test remains ignored. If I remove the `Ignore` annotation, then annotate 
the first method in that test spec with Spock's 
`@RepeatUntilFailure(maxAttempts = 200)`, the test fails around the 188th 
attempt on master. It's the same on this branch. As such, I don't believe that 
splitting the Global and Per thread drivers into distinct caches is worth the 
problems it is causing (such as 
https://github.com/apache/groovy-geb/issues/288), since it doesn't solve the 
problems it purported to solve.
   
   Having read over [the docs on implicit driver 
management](https://groovy.apache.org/geb/manual/current/#implicit-driver-management),
 the [docs on driver 
cacheing](https://groovy.apache.org/geb/manual/current/#driver-caching-configuration),
 and the code in and around 
`geb.Configuration#wrapDriverFactoryInCachingIfNeeded`, I just don't think 
consumers were really intended to have both global and per-thread driver caches 
at once as a going concern.
   
   I can agree that we need a solution that allows us to re-enable the 
`ParallelExecutionSpec`, but I'm not convinced these split caches are it. It 
seems best to me to revert the fix from 270 and go back to the drawing board on 
_that_, rather than the fix I'd originally proposed in 
https://github.com/apache/groovy-geb/pull/289.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to