Add agenteval-datasets, agenteval-reporting, and agenteval-junit5 modules (Phase 3)#4
Merged
Merged
Conversation
Add agenteval-datasets, agenteval-reporting, and agenteval-junit5 module POMs with dependency declarations. Update parent POM with new modules and managed dependencies for agenteval-metrics, agenteval-datasets, and agenteval-reporting.
Implement dataset loading infrastructure: - EvalDataset: immutable builder with Jackson support and save(Path) - DatasetLoader/DatasetWriter: pluggable format interfaces - JsonDatasetLoader: auto-detects envelope vs bare array JSON formats - JsonDatasetWriter: pretty-printed envelope format output - DatasetException: unchecked exception for load/write errors Includes 15 unit tests covering round-trip serialization, format auto-detection, error handling, and defensive copying.
Implement evaluation result reporting: - EvalReporter: interface for pluggable report formats - ConsoleReporter: formatted table with ANSI colors, PrintStream injection for testability, pass/fail icons, per-metric averages - JunitXmlReporter: standard JUnit XML via javax.xml DOM, compatible with Jenkins, GitHub Actions, and GitLab CI - ReportException: unchecked exception for report generation errors Includes 10 unit tests covering passing/failing results, ANSI toggle, XML structure validation, and empty result handling.
Implement JUnit 5 integration for agent evaluation: - @AgentTest: meta-annotation combining @test + @tag("eval") + @ExtendWith - @Metric/@metrics: repeatable metric declaration with threshold - @DatasetSource: parameterized test source from JSON dataset files - AgentEvalExtension: ParameterResolver + InvocationInterceptor + AfterEachCallback for full test lifecycle integration - MetricFactory: reflective metric instantiation probing 4 constructor signatures (JudgeModel+double, double, JudgeModel, no-arg) - DatasetArgumentsProvider: loads AgentTestCase args from classpath JSON - AgentAssertions/AgentTestCaseAssert: standalone fluent assertion API with meetsMetric, hasToolCalls, calledTool, outputContains, etc. Includes 28 unit tests covering parameter resolution, metric evaluation, dataset loading, factory instantiation, and assertion chain behavior.
Add exclusions for false positives in new modules: - EvalDataset: fields use List.copyOf/Map.copyOf (EI_EXPOSE_REP) - JsonDatasetLoader/Writer: ObjectMapper is thread-safe (EI_EXPOSE_REP2) - ConsoleReporter: PrintStream stored for DI/testing (EI_EXPOSE_REP2) - AgentTestCaseAssert: intentionally exposes mutable test case (EI_EXPOSE_REP)
This was referenced Apr 23, 2026
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
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.
Summary
EvalDataset,JsonDatasetLoader,JsonDatasetWriterConsoleReporter(ANSI-colored formatted table) andJunitXmlReporter(CI-compatible JUnit XML via javax.xml DOM)@AgentTest,@Metric,@DatasetSourceannotations,MetricFactory(reflective instantiation),DatasetArgumentsProvider, and standalone fluent assertion API (AgentAssertions)Completes the P0 MVP — all 7 modules are now implemented with 53 new tests across the 3 modules (15 + 10 + 28), all quality gates passing (Checkstyle, SpotBugs, -Werror).