Orchestrata Ultimate › Tests › Reference › scripts › test_error_report
Tests reference — tests/scripts/test_error_report.py
Module of tests/scripts, split out because the directory page would be too large.
Prose pending.
tests/scripts/test_error_report.py
source · 25 tests · 372 lines
Tests for scripts/error_report.py — the always-on red-list report.
| Kind | Name | Line | Description |
|---|---|---|---|
| class | TestFailingTestExtraction |
63 | A module count understates the damage — that was the original error. |
| method | TestFailingTestExtraction.test_pytest_summary_lines_are_read |
66 | |
| method | TestFailingTestExtraction.test_the_batched_runners_own_line_shape_is_read |
78 | The runner emits module::test — traceback, not pytest's form. |
| method | TestFailingTestExtraction.test_a_test_reported_in_both_shapes_is_counted_once |
85 | |
| method | TestFailingTestExtraction.test_clean_output_yields_nothing |
91 | |
| class | TestTheReportCountsTestsNotModules |
95 | |
| method | TestTheReportCountsTestsNotModules.test_three_modules_holding_eight_tests_report_eight |
96 | The exact shape of the 2026-08-09 miscount. |
| method | TestTheReportCountsTestsNotModules.test_a_module_with_no_parsable_ids_falls_back_to_its_count |
134 | Under-reporting is the failure mode; never report zero for a red |
| method | TestTheReportCountsTestsNotModules.test_a_passing_module_is_not_counted |
147 | |
| class | TestPartialRunsAreLabelled |
156 | A targeted run overwrites the suite's artifact. |
| method | TestPartialRunsAreLabelled.test_a_one_module_artifact_is_called_partial |
159 | |
| method | TestPartialRunsAreLabelled.test_a_green_partial_run_is_still_labelled |
175 | The dangerous direction: one green module reading as a green suite. |
| method | TestPartialRunsAreLabelled.test_a_full_run_is_not_labelled_partial |
192 | |
| method | TestPartialRunsAreLabelled.test_an_empty_artifact_is_not_labelled_partial |
205 | No modules at all is an absent run, already reported as such. |
| class | TestStalenessIsStated |
216 | "No failures recorded 4 hours ago" is not "nothing is red". |
| method | TestStalenessIsStated.test_a_fresh_artifact_carries_no_warning |
219 | |
| method | TestStalenessIsStated.test_an_old_artifact_is_marked_stale |
228 | |
| method | TestStalenessIsStated.test_an_absent_artifact_is_never_reported_as_green |
239 | |
| method | TestStalenessIsStated.test_ages_render_in_a_readable_unit |
253 | |
| class | TestGatesReportCachedVerdicts |
261 | |
| method | TestGatesReportCachedVerdicts.test_a_cached_pass_is_labelled_as_cached |
267 | A cached PASS says the inputs are unchanged, not that the gate was |
| method | TestGatesReportCachedVerdicts.test_a_failing_gate_is_counted |
283 | |
| method | TestGatesReportCachedVerdicts.test_a_skipped_gate_is_not_a_failure |
292 | |
| method | TestGatesReportCachedVerdicts.test_an_absent_cache_is_reported_not_assumed |
298 | |
| class | TestCorruptArtifactsDoNotCrashTheReport |
308 | A hook that raises gets disabled, and a disabled reporter reports |
| method | TestCorruptArtifactsDoNotCrashTheReport.test_unparsable_json_is_a_message_not_an_exception |
312 | |
| method | TestCorruptArtifactsDoNotCrashTheReport.test_the_whole_report_builds_against_the_live_repo |
325 | |
| class | TestExitStatus |
332 | |
| method | TestExitStatus.test_it_exits_zero_by_default_so_a_hook_never_blocks |
333 | |
| class | TestItIsWiredIntoSessionStart |
337 | A reporter nothing invokes is not a delivery (RUL-023). |
| method | TestItIsWiredIntoSessionStart.test_the_hook_script_exists_and_is_executable |
340 | |
| method | TestItIsWiredIntoSessionStart.test_session_start_invokes_the_hook |
346 | |
| method | TestItIsWiredIntoSessionStart.test_the_hook_does_not_pass_exit_code |
358 | --exit-code would let a red suite block session start, and a hook |