Context
CI enforces ≥90% line coverage on a scoped subset of farm/ via .coveragerc ([report] omit). The modules below are currently excluded from that report so the gate can focus on the simulation core (environment, action, decision, database, observations, agent, etc.).
This issue tracks bringing test coverage to these analysis modules and, when ready, removing them from [report] omit in .coveragerc.
Scope (paths)
farm/analysis/** (orchestrators, loaders, processors, dominance/advantage/genesis/social/spatial/temporal pipelines, protocols, mocks, etc.)
Suggested approach
- Prefer unit tests with small fixtures and mocked I/O (filesystem, DB, matplotlib
plt.savefig where needed).
- Add
@pytest.mark.analysis where appropriate (see pytest.ini).
- After meaningful coverage, shrink
.coveragerc [report] omit for the files you cover and confirm pytest --cov=farm --cov-config=.coveragerc --cov-fail-under=90 still passes.
Acceptance criteria
- New tests under
tests/ covering the above behavior without requiring full production data.
- Optional: document in the PR which omit entries were removed and why.
Context
CI enforces ≥90% line coverage on a scoped subset of
farm/via.coveragerc([report] omit). The modules below are currently excluded from that report so the gate can focus on the simulation core (environment,action,decision,database,observations,agent, etc.).This issue tracks bringing test coverage to these analysis modules and, when ready, removing them from
[report] omitin.coveragerc.Scope (paths)
farm/analysis/**(orchestrators, loaders, processors, dominance/advantage/genesis/social/spatial/temporal pipelines, protocols, mocks, etc.)Suggested approach
plt.savefigwhere needed).@pytest.mark.analysiswhere appropriate (seepytest.ini)..coveragerc[report] omitfor the files you cover and confirmpytest --cov=farm --cov-config=.coveragerc --cov-fail-under=90still passes.Acceptance criteria
tests/covering the above behavior without requiring full production data.