Github user mmiklavc commented on the issue:

    https://github.com/apache/incubator-metron/pull/463
  
    > We actually need it for all of the tests that ensure actual parallelism 
and I've strengthened the other test impacted. We had gotten around it by being 
too lenient, honestly. The rest of the test cases are non-parallel, so they are 
correct as implemented.
    
    > This is the correct fix, IMO. stream.parallel() does not give any 
explicit guarantees about parallelism, just that it can parallelize batches 
across a thread pool. This test essentially ensures that given enough tries, it 
does parallelize.
    
    Ok, I follow what's going on here and support the need to test for 
instances where you may end up with the same thread by looping. 
    
    Just for some clarification, what is the reason that we are not leveraging 
stream.isParallel() to check for parallelism? Would the concern here be that 
because we're returning an object the implements the Stream interface, the 
ReaderSpliterator could be changed to return something that  does not abide by 
the isParallel() contract?


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---

Reply via email to