[vaidikgarg-bit]

Purpose of the PR

• close [Bug] Store MultiListIterator leaks its sub-iterators when closed before the first hasNext() #3436

MultiListIterator.close() closes innerIterator and whatever remains in innerListIterator. Both fields are set on the first hasNext() or isValid() call, so an iterator closed before that has null in both. In that case close() skips both and clears the list, and none of the sub-iterators gets closed. AggregativeQueryObserver runs into this when a query is cancelled after getIterator() returns but before its read loop calls hasNext(). The scan iterators BusinessHandlerImpl already opened for the query are then left open.

Main Changes

• close() starts innerListIterator from the full list when iteration never began, so the existing loop closes every sub-iterator. When iteration has started, it keeps its current behavior: sub-iterators that were exhausted and closed are not closed again.
• FilterIteratorContractTest gains three tests. One closes a bare MultiListIterator before iteration and then closes it again, one closes it through the limit-dedup wrapper MapLimitIterator, and one closes it partway through. In each, every sub-iterator must be closed exactly once.

Verifying these changes

• Trivial rework / code cleanup without any test coverage. (No Need)
• Already covered by existing tests, such as (please modify tests here).
• Need tests and can be verified as follows:
  • on master the two close-before-iteration tests fail, since the sub-iterators are never closed; the partial-iteration test passes on master too and guards against double closing
  • store-server-test (192), store-core-test and store-rocksdb-test (19) pass locally, and mvn editorconfig:check reports nothing in tracked files

Does this PR potentially affect the following parts?

• Dependencies (add/update license info & regenerate_known_dependencies.sh)
• Modify configurations
• The public API
• Other affects (typed here)
• Nope

Documentation Status

• Doc - TODO: required documentation is pending; complete it before merging.
• Doc - Done: documentation is included here or linked below.
• Doc - No Need: no user-visible documentation is affected.

Visual overview

Closing before the first read now closes the registered child iterators instead of skipping cleanup. Scope: normal close path only; this illustration does not assert test/CI completion, exception safety or concurrency safety.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 57.73%. Comparing base (608cb1d) to head (f9a62d1).
⚠️ Report is 22 commits behind head on master.

Additional details and impacted files

@@             Coverage Diff              @@
##             master    #3437      +/-   ##
============================================
+ Coverage     57.27%   57.73%   +0.46%     
+ Complexity     6577     2589    -3988     
============================================
  Files           776      779       +3     
  Lines         68881    69103     +222     
  Branches       9431     9489      +58     
============================================
+ Hits          39449    39895     +446     
+ Misses        25804    25547     -257     
- Partials       3628     3661      +33     

Flag	Coverage Δ		*Carryforward flag
commons	96.37% <ø> (ø)		Carriedforward from 608cb1d
hstore	36.11% <ø> (+0.02%)	⬆️
pd	52.80% <ø> (+0.64%)	⬆️	Carriedforward from 608cb1d
server_hbase	42.18% <ø> (+<0.01%)	⬆️	Carriedforward from 608cb1d
server_memory	55.99% <ø> (+<0.01%)	⬆️	Carriedforward from 608cb1d
server_rocksdb	42.66% <ø> (+<0.01%)	⬆️	Carriedforward from 608cb1d
store	35.93% <ø> (+2.78%)	⬆️

*This pull request uses carry forward flags. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[bitflicker64]

Blocking: no. Score: 9/10. Summary: close() now walks the full iterators snapshot when iteration never started, so cancelling before the first hasNext() closes every sub-iterator, and the started path still closes each sub-iterator once; the three new FilterIteratorContractTest cases pin both paths.