datadog/migrate-groovy-to-java
> Converts Spock/Groovy test files in a Gradle module to equivalent JUnit 5 Java tests. Use when asked to "migrate groovy", "convert groovy to java", "g2j", or when a module has .groovy test files that need to be replaced with .java equivalents.
npx skills add https://github.com/DataDog/dd-trace-java --skill migrate-groovy-to-java
Migrate test Groovy files to Java using JUnit 5
When converting Groovy code to Java code, make sure that:
@TableTest for data rows that are naturally tabular. See detailed guidance in the "TableTest usage" section.@TableTest and @MethodSource may be combined on the same @ParameterizedTest when most cases are tabular but a few cases require programmatic setup.@TableTest, and put only non-tabular/complex cases in @MethodSource.@TableTest is not viable for the test at all, use @MethodSource only.@TableTest was successfully used and if the @ParameterizedTest is not used to specify the test name, @ParameterizedTest can then be removed as @TableTest replace it fully.@MethodSource, name the arguments method <testMethodName>Arguments (camelCase, e.g. testMethodArguments) and return Stream<Arguments> using Stream.of(...) and arguments(...) with static import.Arguments.arguments(...) value or index the test caseassertTrue(a.equals(b)) or assertFalse(a.equals(b)), use assertEquals(expected, actual) and assertNotEquals(unexpected, actual)finalMock(...) usage, use libs.bundles.mockito instead of writing manual recording/stub implementationsinjectSysConfig(key, value) calls with @WithConfig when the key and value are static literals. Put it on the test method for per-test config, or on the class when every test needs it. The dd. prefix is added automatically — use the bare key (e.g. "trace.scope.strict.mode", not "dd.trace.scope.strict.mode"). For dynamic or parameterized values, keep the imperative WithConfigExtension.injectSysConfig(key, value) call.asMap() / asLong() / asList(), check whether json-unit-assertj (libs.json.unit.assertj) is already in the module's build file. If it is, add a method that returns the raw JSON string and use assertThatJson(json).node("some.nested.field").isEqualTo(value) directly instead of the map traversal.[key: val] map literals use a LinkedHashMap. When the test doesn't care about insertion order, use singletonMap for a single entry or HashMap for two or more. If a helper method builds these maps, add a two-arg overload rather than scattering new LinkedHashMap<>() constructions through test bodies.TableTest usage
Import: import org.tabletest.junit.TableTest;
JDK 8 rules:
@TableTest({
"a | b",
"1 | 2"
})
Spock where: → @TableTest:
scenario column as first column (display name, not a method parameter).| delimiter; align columns so pipes line up vertically.'a|b', '[]').'' = empty string.[a, b] = List/array, {a, b} = Set, [k: v] = Map.Mixed eligibility:
@TableTest + @MethodSource on one @ParameterizedTest when only some cases are complex.@MethodSource only when tabular representation is not practical for the test.Do NOT use @TableTest when:
Before writing any Java output, read .claude/skills/migrate-groovy-to-java/QUALITY_RULES.md in full.
Apply all BLOCKER rules unconditionally.
Apply WARNING rules unless they require creating utility classes that do not yet exist.
Apply STYLE rules where they improve clarity without added complexity.
Take datadog/migrate-groovy-to-java from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.