FAQ

Does ReqCover replace code coverage tools like JaCoCo?

No. Code coverage measures which lines your tests execute. ReqCover measures which requirements your tests verify - a test suite can have 100% line coverage and 0% requirements coverage if it never asserts anything meaningful about the behavior a requirement describes. The two are complementary.

What counts as “covered”?

A requirement is covered when a test annotated @ForRequirement("id") for that ID passes. A failing, erroring, or skipped test does not count - see Core Concepts.

What languages does ReqCover support?

Any JVM language whose tests run on JUnit 5 Jupiter - Java, Kotlin, Scala, and Groovy are all common choices. reqcover-core’s annotations are defined in Kotlin Multiplatform, so non-JVM targets are technically possible in the future, but the engine, listener, sources, and reporters are JVM-only today.

Do I have to use OpenAPI as my requirements source?

No - reqcover-source-openapi is the one source ReqCover ships, but RequirementsSource is a public SPI. See Requirements Sources for how to write one against a plain requirement list, a different spec format, or an export from a requirements-management tool.

What happens if I annotate a test with a requirement ID that isn’t in my requirements source?

It shows up in unexpectedRequirements() and in the “unexpected” section of both built-in reporters. It won’t fail the build by itself, but it’s usually a sign of a typo in the ID or a stale reference to a requirement that was removed from the spec.

Can I use ReqCover without failing the build on low coverage?

Yes - set reqCover.failIfBelowMinimum=false. ReqCover will still log and report the shortfall; it just won’t exit with a non-zero status. See Configuration Reference.

Where are the coverage reports written?

build/reports/tests/reqcover/requirements-coverage.html and .xml, relative to the working directory the test JVM was started in - see Reporters.