Opp Repl Feature And Release Tests

Feature, release, and all-tests suites

Three convenience entry points that sit above the specific test types (opp-repl-smoke-tests, opp-repl-fingerprint-tests, etc.).

Upstream reference: https://github.com/omnetpp/opp_repl/blob/main/doc/feature_tests.md

Feature tests

Check that the project builds and simulations set up correctly with each documented combination of optional features (OMNeT++ "features" as declared in .oppfeatures).

run_feature_tests(simulation_project=inet_project)

CLI:

opp_run_feature_tests --load inet.opp -p inet

Release tests

A curated suite suitable for validating a release build — combines smoke, fingerprint, statistical, feature, and chart tests (exact composition is project-defined).

run_release_tests(simulation_project=inet_project)

CLI:

opp_run_release_tests --load inet.opp -p inet

Run all tests

Sequentially run every configured test type:

run_all_tests(simulation_project=inet_project)

CLI:

opp_run_all_tests --load inet.opp -p inet

When to use which

• run_smoke_tests() only -- fast feedback on feature branches.
• run_feature_tests() -- before merging changes that touch build configuration, NED features, or optional modules.
• run_release_tests() -- the official gate for pushing a release.
• run_all_tests() -- use during maintenance or when you want a single verdict; overlap with run_release_tests() is intentional in many projects. When module-image tests are configured for the project, run_all_tests() includes them automatically (via run_module_image_tests()).
• Module-image tests (opp-repl-module-image-tests) -- visual regression of module/network rendering; run independently with run_module_image_tests() / opp_run_module_image_tests CLI.

Pitfalls

• These entry points aggregate long-running tests; expect tens of minutes to hours depending on project size. Pin sim_time_limit and use --no-concurrent sparingly.
• Because each sub-suite has its own baselines, any ONE stale baseline can fail the whole aggregate. Fix baselines incrementally using the individual update_*_test_results entry points.
• Results are reported per sub-suite, not merged into a single MultipleTaskResults. Inspect each section for drill-down.

See also

• Every other opp-repl-*-tests skill.
• opp-repl-module-image-tests — visual regression of module/network rendering (distinct from chart/analysis-plot tests).
• opp-repl-tasks-and-results — per-sub-suite drill-down.
• opp-repl-github-actions — dispatch suites remotely.