Add stream() method for convenient result traversal

[TomaszGaweda]

Checklist:

• Labels (Team:, Type:, Source:, Module:) and Milestone set
• Label Add to Release Notes or Not Release Notes content set
• Request reviewers if possible
• Send backports/forwardports if fix needs to be applied to past/future releases
• New public APIs have @Nonnull/@Nullable annotations
• New public APIs have @since tags in Javadoc

[TomaszGaweda]

I wonder what do you think of such thing, IMHO it makes API more user-friendly without much of a code (underlying interator is the same).
@k-jamroz @Fly-Style @ivanthescientist @burakgok

[TomaszGaweda]

It would close #22166 I think

[burakgok]

I don't understand why this default method doesn't exist in the Iterable interface.

[k-jamroz]

LMGTFY: https://stackoverflow.com/a/23177907

[k-jamroz]

argument with specialised iterator types does not apply to us. I wonder if adding this method would promote less efficient (but easier) usage. Also users will not be able to use this method in a fully fluent way because SqlResult has to be closed

[burakgok]

Java architects' defenses on these type of poor design choices are really funny. They are reluctant to get rid of primitive/boxed duality (by burying auto-(un)boxing deeper) and they use this as a counter-argument in order not to implement some convenience methods. By the way, I upvoted this and this.

To get rid of try-with-resources, we can return StreamSupport.stream(spliterator(), false).onClose(this::close), but try-with-resources is the only reliable way to release resources in case a RuntimeException is thrown. Closing SqlResult in the end of iterator(), i.e. when hasNext() returns false, is also a bad idea since the iterator doesn't have to be used to the end.

[k-jamroz]

to say "Java is not perfect" would be a huge understatement, but we have to live with it

[TomaszGaweda]

Yeah I know it's not perfect - still needs try-with-resources - but it's simpler for queries where you expect not many result rows (e.g. with limit), similar to Mongo's .into() function

[Patras3]

run-lab-run

[TomaszGaweda]

run-lab-run

[k-jamroz]

Because of https://github.com/hazelcast/hazelcast-enterprise/issues/5966 and time constraints postponing decision to 5.4 release.

[hz-devops-test]

The job Hazelcast-pr-builder of your PR failed. (Hazelcast internal details: build log, artifacts).
Through arcane magic we have determined that the following fragments from the build log may contain information about the problem.

Click to expand the log file

--------------------------
-------TEST FAILURE-------
--------------------------
[INFO] Results:
[INFO] 
[ERROR] Failures: 
[ERROR]   StreamKafkaPTest.when_processingGuaranteeNone_then_continueFromBeginningAfterJobRestart:249->testWithJobRestart:315->testWithJobRestart:350->HazelcastTestSupport.assertTrueEventually:1184->HazelcastTestSupport.assertTrueEventually:1165->lambda$testWithJobRestart$4:350 expected:<300> but was:<389>
[INFO] 
[ERROR] Tests run: 51, Failures: 1, Errors: 0, Skipped: 1
[INFO] 
[ERROR] There are test failures.

[TomaszGaweda]

run-lab-run