Avoid final field modification in JUnit5 instrumentation#11752

Open

daniel-mohedano wants to merge 1 commit into

daniel.mohedano/jep-500-junit

daniel-mohedano commented Jun 26, 2026 •

edited

Loading

Contributor

What Does This Do

Avoids final field modifications in JUnit5 instrumentation by:
- Building a fresh TestTask for every retry, reusing the original parent context, instead of reusing and modifying the original test task. This approach is only supported for versions 5.3.1+.
- Introducing a new TestDescriptor cloning method based on JupiterTestDescriptor#copyIncludingDescendants(UnaryOperator<UniqueId>)}, which also avoids manually setting the generated uniqueId after the cloning. This approach is only supported for versions 5.13+.

Motivation

Avoid failures in future JDKs due to JEP 500: Prepare to Make Final Mean Final.

Additional Notes

Running an example project on JDK26 with JAVA_TOOL_OPTIONS=--illegal-final-field-mutation=debug produced the following logs:

Final field testDescriptor in class org.junit.platform.engine.support.hierarchical.NodeTestTask has been unreflected for mutation ...
Final field node in class org.junit.platform.engine.support.hierarchical.NodeTestTask has been unreflected for mutation ...
Final field uniqueId in class org.junit.platform.engine.support.descriptor.AbstractTestDescriptor has been unreflected for mutation ...
Final field interceptorCall in class org.junit.jupiter.engine.descriptor.TestMethodTestDescriptor has been mutated reflectively ...
Final field methodInfo in class org.junit.jupiter.engine.descriptor.MethodBasedTestDescriptor has been mutated reflectively ...
Final field configuration in class org.junit.jupiter.engine.descriptor.JupiterTestDescriptor has been mutated reflectively ...
...

The build containing the fix doesn't produce the logs.

Contributor Checklist

Format the title according to the contribution guidelines
Assign the type: and (comp: or inst:) labels in addition to any other useful labels
Avoid using close, fix, or any linking keywords when referencing an issue
Use solves instead, and assign the PR milestone to the issue
Update the CODEOWNERS file on source file addition, migration, or deletion
Update public documentation with any new configuration flags or behaviors
Add your completed PR to the merge queue by commenting /merge. You can also:
- Customize the commit message associated with the merge with /merge --commit-message "..."
- Remove your PR from the merge queue with /merge -c
- Skip all merge queue checks with /merge -f --reason "reason"; please use this judiciously, as some checks do not run at the PR-level (note: the PR still needs to be mergeable, this will only skip the pre-merge build)
- Get more information in this doc

Jira ticket: SDTEST-3867


          feat: avoid final field modifications in junit instrumentation

c73e0a4

daniel-mohedano added type: enhancement comp: ci visibility labels

cit-pr-commenter-54b7da Bot commented Jun 26, 2026

Test Environment - sbt-scalatest

Job Status: 🟢 success

Scenario	This PR (%)	7d median	Δ 7d	30d median	Δ 30d	runs (7d/30d)
agent	56.69	54.33	$\color{red}{\blacktriangle}$ +2.36	55.43	$\color{red}{\blacktriangle}$ +1.26	46/188
agentEvpProxy	55.01	n/a	n/a	n/a	n/a	-

Baseline: median of @test.tracer_overhead on main (gitlab) over the last 7/30 days, per OSS project & scenario. Δ = this PR − baseline median; red ▲ = more overhead, green ▽ = less overhead than baseline.

cit-pr-commenter-54b7da Bot commented Jun 26, 2026

Test Environment - nebula-release-plugin

Job Status: 🟢 success

Scenario	This PR (%)	7d median	Δ 7d	30d median	Δ 30d	runs (7d/30d)
agent	37.55	36.42	$\color{red}{\blacktriangle}$ +1.13	36.42	$\color{red}{\blacktriangle}$ +1.13	23/91
agentless	37.69	35.70	$\color{red}{\blacktriangle}$ +1.99	35.70	$\color{red}{\blacktriangle}$ +1.99	23/91
agentlessCodeCoverage	44.57	44.48	$\color{red}{\blacktriangle}$ +0.09	44.48	$\color{red}{\blacktriangle}$ +0.09	23/91
agentlessLineCoverage	76.92	74.82	$\color{red}{\blacktriangle}$ +2.10	74.82	$\color{red}{\blacktriangle}$ +2.10	22/90

Baseline: median of @test.tracer_overhead on main (gitlab) over the last 7/30 days, per OSS project & scenario. Δ = this PR − baseline median; red ▲ = more overhead, green ▽ = less overhead than baseline.

cit-pr-commenter-54b7da Bot commented Jun 26, 2026

Test Environment - netflix-zuul

Job Status: 🟢 success

Scenario	This PR (%)	7d median	Δ 7d	30d median	Δ 30d	runs (7d/30d)
agent	87.88	86.07	$\color{red}{\blacktriangle}$ +1.81	86.07	$\color{red}{\blacktriangle}$ +1.81	24/130
agentless	81.38	81.05	$\color{red}{\blacktriangle}$ +0.33	81.05	$\color{red}{\blacktriangle}$ +0.33	23/129
agentlessCodeCoverage	96.01	95.12	$\color{red}{\blacktriangle}$ +0.89	95.12	$\color{red}{\blacktriangle}$ +0.89	23/127
agentlessLineCoverage	111.63	111.62	$\color{red}{\blacktriangle}$ +0.01	111.62	$\color{red}{\blacktriangle}$ +0.01	22/126

Baseline: median of @test.tracer_overhead on main (gitlab) over the last 7/30 days, per OSS project & scenario. Δ = this PR − baseline median; red ▲ = more overhead, green ▽ = less overhead than baseline.

cit-pr-commenter-54b7da Bot commented Jun 26, 2026

Test Environment - pass4s

Job Status: 🟢 success

Scenario	This PR (%)	7d median	Δ 7d	30d median	Δ 30d	runs (7d/30d)
agent	15.39	9.92	$\color{red}{\blacktriangle}$ +5.47	9.92	$\color{red}{\blacktriangle}$ +5.47	24/94
agentless	11.90	10.13	$\color{red}{\blacktriangle}$ +1.77	9.92	$\color{red}{\blacktriangle}$ +1.98	24/95
agentlessCodeCoverage	18.25	17.03	$\color{red}{\blacktriangle}$ +1.22	17.73	$\color{red}{\blacktriangle}$ +0.52	22/93

Baseline: median of @test.tracer_overhead on main (gitlab) over the last 7/30 days, per OSS project & scenario. Δ = this PR − baseline median; red ▲ = more overhead, green ▽ = less overhead than baseline.

cit-pr-commenter-54b7da Bot commented Jun 26, 2026

Test Environment - reactive-streams-jvm

Job Status: 🟢 success

Scenario	This PR (%)	7d median	Δ 7d	30d median	Δ 30d	runs (7d/30d)
agent	19.69	21.65	$\color{green}{\blacktriangledown}$ -1.96	21.65	$\color{green}{\blacktriangledown}$ -1.96	26/143
agentless	17.38	18.82	$\color{green}{\blacktriangledown}$ -1.44	18.82	$\color{green}{\blacktriangledown}$ -1.44	25/140
agentlessCodeCoverage	18.13	19.59	$\color{green}{\blacktriangledown}$ -1.46	19.99	$\color{green}{\blacktriangledown}$ -1.86	25/140
agentlessLineCoverage	28.07	29.82	$\color{green}{\blacktriangledown}$ -1.75	29.82	$\color{green}{\blacktriangledown}$ -1.75	24/140

Baseline: median of @test.tracer_overhead on main (gitlab) over the last 7/30 days, per OSS project & scenario. Δ = this PR − baseline median; red ▲ = more overhead, green ▽ = less overhead than baseline.

cit-pr-commenter-54b7da Bot commented Jun 26, 2026

Test Environment - jolokia

Job Status: 🟢 success

Scenario	This PR (%)	7d median	Δ 7d	30d median	Δ 30d	runs (7d/30d)
agent	91.25	93.23	$\color{green}{\blacktriangledown}$ -1.98	93.23	$\color{green}{\blacktriangledown}$ -1.98	27/96
agentless	86.66	89.58	$\color{green}{\blacktriangledown}$ -2.92	89.58	$\color{green}{\blacktriangledown}$ -2.92	26/96
agentlessCodeCoverage	96.15	99.00	$\color{green}{\blacktriangledown}$ -2.85	99.00	$\color{green}{\blacktriangledown}$ -2.85	23/93
agentlessLineCoverage	98.49	99.00	$\color{green}{\blacktriangledown}$ -0.51	99.00	$\color{green}{\blacktriangledown}$ -0.51	23/93

Baseline: median of @test.tracer_overhead on main (gitlab) over the last 7/30 days, per OSS project & scenario. Δ = this PR − baseline median; red ▲ = more overhead, green ▽ = less overhead than baseline.

cit-pr-commenter-54b7da Bot commented Jun 26, 2026

Test Environment - okhttp

Job Status: 🟢 success

Scenario	This PR (%)	7d median	Δ 7d	30d median	Δ 30d	runs (7d/30d)
agent	17.94	19.20	$\color{green}{\blacktriangledown}$ -1.26	19.59	$\color{green}{\blacktriangledown}$ -1.65	26/96
agentless	19.33	18.82	$\color{red}{\blacktriangle}$ +0.51	19.20	$\color{red}{\blacktriangle}$ +0.13	25/98
agentlessCodeCoverage	23.41	22.09	$\color{red}{\blacktriangle}$ +1.32	22.09	$\color{red}{\blacktriangle}$ +1.32	25/94
agentlessLineCoverage	44.67	44.48	$\color{red}{\blacktriangle}$ +0.19	44.48	$\color{red}{\blacktriangle}$ +0.19	28/97

Baseline: median of @test.tracer_overhead on main (gitlab) over the last 7/30 days, per OSS project & scenario. Δ = this PR − baseline median; red ▲ = more overhead, green ▽ = less overhead than baseline.

cit-pr-commenter-54b7da Bot commented Jun 26, 2026

Test Environment - sonar-java

Job Status: 🟢 success

Scenario	This PR (%)	7d median	Δ 7d	30d median	Δ 30d	runs (7d/30d)
agent	-0.16	24.91	$\color{green}{\blacktriangledown}$ -25.07	16.04	$\color{green}{\blacktriangledown}$ -16.20	26/96
agentless	25.34	19.99	$\color{red}{\blacktriangle}$ +5.35	15.11	$\color{red}{\blacktriangle}$ +10.23	25/95
agentlessCodeCoverage	86.10	95.12	$\color{green}{\blacktriangledown}$ -9.02	87.80	$\color{green}{\blacktriangledown}$ -1.70	25/95
agentlessLineCoverage	95.98	133.64	$\color{green}{\blacktriangledown}$ -37.66	144.77	$\color{green}{\blacktriangledown}$ -48.79	24/94

Baseline: median of @test.tracer_overhead on main (gitlab) over the last 7/30 days, per OSS project & scenario. Δ = this PR − baseline median; red ▲ = more overhead, green ▽ = less overhead than baseline.

daniel-mohedano marked this pull request as ready for review

June 29, 2026 14:58

daniel-mohedano requested a review from a team as a code owner

June 29, 2026 14:58

datadog-official Bot reviewed

View reviewed changes

datadog-official Bot left a comment

Contributor

More details

Three critical correctness invariants hold: (1) ThrowableCollectorFactoryWrapper is installed on NodeTestTaskContext (shared across all retry tasks), so factory.setSuppressFailures() correctly controls each retry even though currentTask is now a fresh NodeTestTask; (2) AbstractTestDescriptor.setParent() only sets the child's parent reference without addChild(), so the snapshot and retry descriptors are not injected into the parent's children list; (3) legacyCopy's new conditional uniqueId setter is correct — tryShallowClone already copies all fields including uniqueId, so the setter is only needed when the ID actually changes.

Was this helpful? React 👍 or 👎

_{Open Bits AI session}

_{🤖 Datadog Autotest · Commit c73e0a4 · What is Autotest? · Any feedback? Reach out in #autotest}

juan-fernandez approved these changes

View reviewed changes

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

comp: ci visibility type: enhancement