test_p_evidence_e2e.py
file-level
1
files
1
commits
0
hotspots
0
🧊 dead
0
💥 blast risk
| 1 | """End-to-end P-evidence CLI tests (§PE.10).""" |
| 2 | |
| 3 | from __future__ import annotations |
| 4 | |
| 5 | import json |
| 6 | from pathlib import Path |
| 7 | |
| 8 | import yaml |
| 9 | |
| 10 | from cli.kit_root import kit_root |
| 11 | from tests.fixtures.p_evidence import ( |
| 12 | FIXTURES, |
| 13 | copy_p_evidence_entries, |
| 14 | load_p_evidence_entry, |
| 15 | seed_p_evidence_repo, |
| 16 | ) |
| 17 | from tests.support import git_status_runner, muse_mirror_status_runner, run_cli, seed_muse_substrate |
| 18 | |
| 19 | |
| 20 | def _append_evidence(tmp_path: Path, body_name: str) -> int: |
| 21 | copy_p_evidence_entries(tmp_path) |
| 22 | payload = tmp_path / "payload.json" |
| 23 | payload.write_text(json.dumps(load_p_evidence_entry(body_name)), encoding="utf-8") |
| 24 | return run_cli( |
| 25 | ["ledger", "append", "--kind", "verification_evidence", "--file", "payload.json"], |
| 26 | cwd=tmp_path, |
| 27 | runner=git_status_runner(), |
| 28 | kit=kit_root(), |
| 29 | ) |
| 30 | |
| 31 | |
| 32 | def test_e2e_append_show_and_mode_b_match(tmp_path: Path, capsys) -> None: |
| 33 | seed_p_evidence_repo(tmp_path, require_verification_evidence="require") |
| 34 | assert _append_evidence(tmp_path, "verification-evidence-pass.json") == 0 |
| 35 | show_code = run_cli( |
| 36 | ["ledger", "show", "--last", "3"], |
| 37 | cwd=tmp_path, |
| 38 | runner=git_status_runner(), |
| 39 | kit=kit_root(), |
| 40 | ) |
| 41 | assert show_code == 0 |
| 42 | out = capsys.readouterr().out |
| 43 | assert "verification_evidence" in out |
| 44 | |
| 45 | code = run_cli( |
| 46 | [ |
| 47 | "honesty-status", |
| 48 | "--verification-evidence", |
| 49 | "Track P / P-evidence", |
| 50 | "--frozen-spec", |
| 51 | "docs/archive/phases/PHASE-TRACK-P-P-EVIDENCE.md", |
| 52 | "--json", |
| 53 | ], |
| 54 | cwd=tmp_path, |
| 55 | runner=git_status_runner(), |
| 56 | kit=kit_root(), |
| 57 | json_mode=True, |
| 58 | ) |
| 59 | assert code == 0 |
| 60 | payload = json.loads(capsys.readouterr().out) |
| 61 | assert payload["verification_evidence"]["matched_entry_hash"] is not None |
| 62 | |
| 63 | |
| 64 | def test_e2e_last_wins_findings_then_pass(tmp_path: Path, capsys) -> None: |
| 65 | seed_p_evidence_repo(tmp_path, require_verification_evidence="require") |
| 66 | assert _append_evidence(tmp_path, "verification-evidence-findings.json") == 0 |
| 67 | body = load_p_evidence_entry("verification-evidence-pass.json") |
| 68 | body["round"] = 2 |
| 69 | second = tmp_path / "second.json" |
| 70 | second.write_text(json.dumps(body), encoding="utf-8") |
| 71 | assert run_cli( |
| 72 | ["ledger", "append", "--kind", "verification_evidence", "--file", "second.json"], |
| 73 | cwd=tmp_path, |
| 74 | runner=git_status_runner(), |
| 75 | kit=kit_root(), |
| 76 | ) == 0 |
| 77 | code = run_cli( |
| 78 | [ |
| 79 | "honesty-status", |
| 80 | "--verification-evidence", |
| 81 | "Track P / P-evidence", |
| 82 | "--json", |
| 83 | ], |
| 84 | cwd=tmp_path, |
| 85 | runner=git_status_runner(), |
| 86 | kit=kit_root(), |
| 87 | json_mode=True, |
| 88 | ) |
| 89 | assert code == 0 |
| 90 | payload = json.loads(capsys.readouterr().out) |
| 91 | ledger = (tmp_path / ".overseer" / "honesty" / "VERDICT-LEDGER.jsonl").read_text(encoding="utf-8") |
| 92 | last_line = json.loads(ledger.strip().splitlines()[-1]) |
| 93 | assert last_line["bv_verdict"] == "pass" |
| 94 | assert payload["verification_evidence"]["matched_entry_hash"] == last_line["entry_hash"] |
| 95 | |
| 96 | |
| 97 | def test_skill_normative_pass_body_includes_test_output() -> None: |
| 98 | fixture = json.loads((FIXTURES / "skill-pass-body.json").read_text(encoding="utf-8")) |
| 99 | body = load_p_evidence_entry("verification-evidence-pass.json") |
| 100 | assert body["bv_verdict"] == fixture["bv_verdict"] |
| 101 | types = {item["type"] for item in body["artifacts"]} |
| 102 | assert "test_output" in types |
| 103 | for required in fixture["required_artifact_types_for_pass"]: |
| 104 | assert required in types |
| 105 | |
| 106 | |
| 107 | def test_e2e_git_only_and_muse_regime_unsigned(tmp_path: Path) -> None: |
| 108 | for regime in ("config-git-only.yaml", "config-muse-git-mirror.yaml"): |
| 109 | repo = tmp_path / regime |
| 110 | repo.mkdir() |
| 111 | seed_p_evidence_repo(repo, require_verification_evidence="require", regime_config=regime) |
| 112 | if "muse" in regime: |
| 113 | seed_muse_substrate(repo) |
| 114 | copy_p_evidence_entries(repo) |
| 115 | payload = repo / "payload.json" |
| 116 | payload.write_text(json.dumps(load_p_evidence_entry("verification-evidence-pass.json")), encoding="utf-8") |
| 117 | runner = muse_mirror_status_runner(repo) if "muse" in regime else git_status_runner() |
| 118 | assert run_cli( |
| 119 | ["ledger", "append", "--kind", "verification_evidence", "--file", "payload.json"], |
| 120 | cwd=repo, |
| 121 | runner=runner, |
| 122 | kit=kit_root(), |
| 123 | ) == 0 |
| 124 | assert run_cli( |
| 125 | ["honesty-status", "--verification-evidence", "Track P / P-evidence", "--json"], |
| 126 | cwd=repo, |
| 127 | runner=runner, |
| 128 | kit=kit_root(), |
| 129 | json_mode=True, |
| 130 | ) == 0 |