autowisp.tests.test_error_persistence module

Class Inheritance Diagram

Unit tests for error persistence (row + JSON sidecar).

These need a real (throwaway) project database, created once per class in a temporary directory.

class autowisp.tests.test_error_persistence.TestCleanupErrors(methodName='runTest')

Bases: TestCase

wisp-cleanup-errors prunes aged rows, orphan files, dangling rows.

Uses a fresh project per test, since cleanup mutates global DB state.

_age_error(error_id, *, days)

Backdate an error’s created so it counts as old.

setUp()

Hook method for setting up the test fixture before exercising it.

test_aged_rows_and_sidecars_removed()

An old error and its sidecar are deleted; a recent one is kept.

test_dangling_row_reference_cleared()

A row whose sidecar vanished keeps the row but clears the path.

test_delete_all_sidecars_keeps_foreign_files()

Project-deletion sidecar purge removes tracked files only.

Every recorded error’s sidecar is removed, but an unrelated file placed under the errors directory survives.

test_orphan_files_removed()

A stray file with no owning row is swept away.

test_sweep_keeps_valid_sidecar()

A healthy error’s sidecar is not touched by a plain sweep.

class autowisp.tests.test_error_persistence.TestParseDuration(methodName='runTest')

Bases: TestCase

parse_duration accepts compact durations and rejects junk.

test_invalid_raises()

test_units()

class autowisp.tests.test_error_persistence.TestPersistError(methodName='runTest')

Bases: _PersistenceTestCase

persist_error writes the queryable row and the detail sidecar.

test_artifact_fks_resolved_from_related_files()

A related file matching a known image/master sets the FK.

test_delete_missing_error_is_noop()

Deleting a non-existent error returns False, does not raise.

test_delete_removes_row_and_sidecar()

delete_error removes both the row and its sidecar file.

test_large_payload_is_gzipped()

A payload over the threshold is written as .json.gz and reloads.

test_missing_sidecar_degrades_to_none()

Reading an error whose sidecar is gone returns None, no crash.

test_row_and_sidecar_are_complementary()

Inline fields on the row; the heavy remainder in the sidecar.

test_sidecar_failure_leaves_valid_row()

A sidecar-write failure still leaves a row, sidecar_path NULL.

test_worker_crashed_persists_step_name()

A synthesised worker-death error lands its step on the row.

End-to-end regression for the crash-report “no matching logs found” gap: a WorkerCrashedError is synthesised by the parent (so it is not a per-stage StepError subclass), but it must still persist step_name into the queryable column – that is what crash-report log-collection resolves the run/step logs from. It is a step-component error because the failure is in the algorithm running inside the step.

class autowisp.tests.test_error_persistence.TestRunPipelineHandler(methodName='runTest')

Bases: _PersistenceTestCase

run_pipeline.main records an escaping error and re-raises it.

test_handler_persists_and_reraises()

An AutoWISPError out of the run is persisted, then re-raised.

test_handler_records_plain_exception()

A non-AutoWISP exception is recorded; the original re-raises.

Mirrors a real failure: a bad config expression raises a plain NameError in the orchestration layer, outside any step’s capture boundary. The handler records it (wrapped as a PipelineError) but re-raises the original so its true traceback reaches the log.

class autowisp.tests.test_error_persistence._PersistenceTestCase(methodName='runTest')

Bases: TestCase

Base creating one throwaway project database for the class.

classmethod setUpClass()

Hook method for setting up class fixture before running tests in the class.

classmethod tearDownClass()

Hook method for deconstructing the class fixture after running all tests in the class.