Orchestrata Ultimate › Tests › Reference › orchestrator › test_pipeline_escalation
Tests reference — tests/orchestrator/test_pipeline_escalation.py
Module of tests/orchestrator, split out because the directory page would be too large.
Prose pending.
tests/orchestrator/test_pipeline_escalation.py
source · 32 tests · 801 lines
Tests for orchestrator.pipeline_escalation — the W3.1 escalation-as-
Covers: orchestrator.pipeline_escalation, orchestrator.pipeline_router, orchestrator.pipeline_runner, core.packet/ (package)
| Kind | Name | Line | Description |
|---|---|---|---|
| function | test_resolve_capability_via_builtin_map_when_provider_registered |
54 | |
| function | test_resolve_capability_returns_none_when_provider_not_registered |
61 | |
| function | test_resolve_capability_returns_none_for_empty_capability |
67 | |
| function | test_resolve_capability_via_volk_declared_capabilities |
73 | |
| function | test_resolve_capability_returns_none_for_unknown_capability |
84 | |
| function | test_classify_interrupt_none_response_is_none |
94 | |
| function | test_classify_interrupt_normal_handoff_is_none |
102 | A resolved-and-moving-on response (PENDING, addressed to the next |
| function | test_classify_interrupt_human_in_progress_escalates |
121 | The real signal ported volks use to say "I'm blocked / need human |
| function | test_classify_interrupt_human_resolved_is_none |
142 | The LAST stage of a run also addresses HUMAN (see compliance/volk.py) |
| function | test_classify_interrupt_human_pending_default_is_none |
161 | Existing tests build response packets without ever setting |
| function | test_classify_interrupt_capability_already_next_stage_is_noop |
179 | Architect's real needs_capability="implementation" packet ahead of |
| function | test_classify_interrupt_capability_not_next_inserts |
205 | |
| function | test_classify_interrupt_unknown_capability_escalates |
225 | |
| function | test_classify_interrupt_repeat_capability_request_escalates_cycle_guard |
242 | |
| function | test_build_capability_stage_maps_capability_and_artifacts |
266 | |
| function | test_build_capability_stage_falls_back_for_unmapped_mesh_string |
282 | |
| function | test_stage_escalates_run_suspends_with_checkpointed_state |
296 | |
| function | test_needs_capability_insertion_does_not_escalate |
334 | A resolved needs_capability insertion is NOT an escalation — the |
| function | test_needs_capability_repeat_request_escalates |
379 | |
| function | test_needs_capability_unknown_escalates_cleanly |
427 | |
| function | test_approve_resumes_from_next_stage |
460 | |
| function | test_approve_with_no_pending_escalation_returns_failed |
500 | |
| function | test_approve_with_checkpoint_unavailable_returns_failed |
509 | |
| function | test_reject_reenters_same_stage_with_feedback |
521 | |
| function | test_reject_with_no_pending_escalation_returns_failed |
563 | |
| function | test_respond_delivers_text_and_may_reescalate_before_resolving |
575 | |
| function | test_respond_with_no_pending_escalation_returns_failed |
634 | |
| function | test_escalation_round_trips_across_a_new_runner_instance |
646 | True interrupt semantics: the escalation must be resolvable by a |
| function | test_escalate_still_reports_escalated_when_checkpoint_unavailable |
688 | A checkpoint failure must never crash an already-computed escalation |
| function | test_save_and_resolve_escalation_are_noop_when_checkpoint_unavailable |
719 | Direct unit coverage of the defensive early-return in both |
| function | test_load_escalation_returns_none_for_malformed_record |
750 | A checkpoint record missing required EscalationState fields (e.g. |
| function | test_reenter_escalated_stage_with_stage_not_in_route_fails_cleanly |
770 | If the escalated stage alias somehow isn't part of the work type's |