Fix flaky UnresolvedNamePatternTests equals check#143191
Open
ebarlas wants to merge 2 commits intoelastic:mainfrom
Open
Fix flaky UnresolvedNamePatternTests equals check#143191ebarlas wants to merge 2 commits intoelastic:mainfrom
ebarlas wants to merge 2 commits intoelastic:mainfrom
Conversation
elasticsearchmachine
added
the
needs:triage
ebarlas
added
:Analytics/ES|QL
Collaborator
|
Pinging @elastic/es-analytical-engine (Team:Analytics) |
elasticsearchmachine
added
the
Team:Analytics
alex-spies
requested changes
Feb 27, 2026
Contributor
alex-spies
left a comment
alex-spies
left a comment
There was a problem hiding this comment.
Thanks a lot @ebarlas , you've uncovered an old hidden trap.
It needs fixing more than the test, though. I'll open a separate PR with a proposed fix and plan to absorb your test fix into it if that's ok.
| switch (between(0, 1)) { | ||
| case 0 -> name = randomValueOtherThan(name, () -> randomAlphaOfLength(4)); | ||
| case 1 -> pattern = randomValueOtherThan(pattern, () -> randomAlphaOfLength(4)); | ||
| String pattern = randomValueOtherThan(instance.pattern(), () -> randomAlphaOfLength(4)); |
Contributor
There was a problem hiding this comment.
This doesn't follow the normal pattern of "retrieve all relevant instance attributes, randomly mutated one". The pattern is always mutated. We need to test the case when the pattern stays the same and only the name is mutated.
Changing just the name should lead to inequality. I think this PR uncovered a genuine bug in UnresolvedNamePattern that needs fixing instead.
| name = randomValueOtherThan(name, () -> randomAlphaOfLength(4)); | ||
| } | ||
| return new UnresolvedNamePattern(source, null, name, pattern); | ||
| return new UnresolvedNamePattern(source, null, pattern, name); |
Contributor
There was a problem hiding this comment.
Swapping the arguments is a correct test fix. We should only merge this part and leave the mutation logic as it was before.
elasticsearchmachine
pushed a commit
that referenced
this pull request
Mar 2, 2026
Thanks @ebarlas for the find! Supersedes #143191. * Fix `UnresolvedNamePatternTests` by actually ensuring that the equality is sensitive to changes of the `name`. * Fix `UnresolvedNamePattern#equals` and `#hashCode` to account for differences in the `name` We probably never saw problems from this in production because the name is derived from the pattern. Basic equality methods should still work correctly, however.
alex-spies
added a commit
to alex-spies/elasticsearch
that referenced
this pull request
Mar 2, 2026
Thanks @ebarlas for the find! Supersedes elastic#143191. * Fix `UnresolvedNamePatternTests` by actually ensuring that the equality is sensitive to changes of the `name`. * Fix `UnresolvedNamePattern#equals` and `#hashCode` to account for differences in the `name` We probably never saw problems from this in production because the name is derived from the pattern. Basic equality methods should still work correctly, however.
elasticsearchmachine
pushed a commit
that referenced
this pull request
Mar 2, 2026
Thanks @ebarlas for the find! Supersedes #143191. * Fix `UnresolvedNamePatternTests` by actually ensuring that the equality is sensitive to changes of the `name`. * Fix `UnresolvedNamePattern#equals` and `#hashCode` to account for differences in the `name` We probably never saw problems from this in production because the name is derived from the pattern. Basic equality methods should still work correctly, however.
tballison
pushed a commit
to tballison/elasticsearch
that referenced
this pull request
Mar 3, 2026
Thanks @ebarlas for the find! Supersedes elastic#143191. * Fix `UnresolvedNamePatternTests` by actually ensuring that the equality is sensitive to changes of the `name`. * Fix `UnresolvedNamePattern#equals` and `#hashCode` to account for differences in the `name` We probably never saw problems from this in production because the name is derived from the pattern. Basic equality methods should still work correctly, however.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
UnresolvedNamePatternTests.mutateInstancehad two bugs:Swapped constructor arguments — the constructor takes
(source, automaton, patternString, name)butmutateInstancepassed(source, null, name, pattern), swapping the two. This was masked when name and pattern differed (the common case).Mutating a non-equality field —
case 0only mutatedname, butUnresolvedNamePattern.innerEqualsonly checkspattern. Thenamefield is invisible to equality becauseUnresolvedNamedExpressionpasses"<unresolved>"as the superclassnametoNamedExpression, and the overriddenname()return value is never checked. When the random instance happened to generate the same value for bothnameandpattern, the swap became visible andcase 1produced a mutation with the samepatternas the original.The fix ensures
patternis always mutated (guaranteeing inequality) and corrects the constructor argument order.Reproduces with: