"""v7 机制第 7 项 user_side_items 定向测试。 钉住的验收（只测本项，不碰其它 7 项）： - B/C 全 pending 时：7 维 overall 与代码表仍可 pass， 生产表强制 not_pass，原因明确指向用户侧。 - user_side_items 为空/缺失时：标 user_side_unknown=True， 绝不写成用户侧已完成。 - 非法项不静默丢弃、不编造状态：记入 invalid_items。 - 阈值仍为 test_default_unapproved；本项不做 B/C 拍板。 """ import sys import unittest from pathlib import Path try: from cloud.audit_scorecard import ( DEFAULT_THRESHOLDS, USER_SIDE_ITEM_IDS, apply_user_side_three_table_rule, build_scorecard, normalize_user_side_items) except ImportError: sys.path.insert(0, str(Path(__file__).resolve().parents[2])) from cloud.audit_scorecard import ( DEFAULT_THRESHOLDS, USER_SIDE_ITEM_IDS, apply_user_side_three_table_rule, build_scorecard, normalize_user_side_items) def _full_evidence(): return { "delivery_quality": { "first_pass_rate": 0.96, "defect_rate": 0.01, "rework_rate": 0.04, "first_pass_known": True, "defect_known": True, "rework_known": True, "attempts_present": True, "evidence_kind": "real", }, "gate_audit": { "hard_gate_enforced": True, "gate_items_passed": 5, "gate_items_total": 5, "merge_blocked_verified": True, "loop_end_to_end_verified": True, "sampling_misjudgment_rate": 0.0, }, "arch_contract": { "contract_frozen": True, "arch_violations": 0, "directory_violations": 0, "import_violations": 0, "stack_changed_without_approval": False, }, "test_coverage": { "prd_coverage_rate": 1.0, "cross_matrix_coverage_rate": 1.0, "coverage_pct": 0.92, "cross_matrix_kind": "real", "real_pg": True, "real_service": True, "agent_six_complete": True, "lint_items": 0, "mypy_items": 0, "bandit_high": 0, }, "security_ledger": { "ledger_verify_ok": True, "ledger_coverage_rate": 1.0, "branch_protection": "protected", "shell_whitelist_pass_rate": 1.0, "anchor_check_recent": True, }, "cost_quota": { "cost_records_complete": True, "unpriced_as_zero": False, "per_creator_present": True, "quota_cutoff_verified": True, }, "ops_scale": { "ci_green_rate": 1.0, "deploy_success_rate": 1.0, "rollback_drill_done": True, "ha_status": "second_host", "parallel_scale_proven": 40, }, } # v7 审计 B/C 清单全 pending 证据（blocked_by 按阻塞方示例分三组）。 def _all_bc_pending(): blocked_by = { "B1": "user", "B2": "user", "B3": "contract_decision", "B4": "user", "B5": "host_env", "B6": "host_env", "B7": "user", "B8": "contract_decision", "B9": "user", "C1": "contract_decision", "C2": "host_env", "C3": "contract_decision", "C4": "contract_decision", } return [{"id": item_id, "status": "pending", "blocked_by": blocked_by[item_id]} for item_id in USER_SIDE_ITEM_IDS] class UserSideItemsTestCase(unittest.TestCase): def test_bc_list_is_the_v7_audit_set(self): self.assertEqual(USER_SIDE_ITEM_IDS, ["B1", "B2", "B3", "B4", "B5", "B6", "B7", "B8", "B9", "C1", "C2", "C3", "C4"]) self.assertEqual(len(_all_bc_pending()), 13) def test_bc_all_pending_code_passes_production_not_pass(self): ev = _full_evidence() ev["user_side_items"] = _all_bc_pending() report = build_scorecard(ev) # 验收核心：代码表仍可 pass，生产表 not_pass 且原因指向用户侧。 self.assertEqual(report["overall"], "pass") self.assertEqual(report["code_overall"], "pass") self.assertTrue(report["code_unchanged_by_user_side"]) self.assertEqual(report["production_overall"], "not_pass") self.assertEqual( report["production_reason"], "user_side_pending: " + ",".join(USER_SIDE_ITEM_IDS)) self.assertTrue(report["user_side_forced_production_not_pass"]) us = report["user_side_items"] self.assertEqual(us["pending_ids"], USER_SIDE_ITEM_IDS) self.assertEqual(us["done_ids"], []) self.assertEqual(us["blocked_ids"], []) self.assertTrue(us["all_pending"]) self.assertFalse(us["user_side_unknown"]) self.assertEqual(us["invalid_items"], []) self.assertEqual(us["count"], 13) # blocked_by 汇总：三组阻塞方覆盖全部 13 个 id，不多不少。 grouped = sorted(i for ids in us["blocked_by_summary"].values() for i in ids) self.assertEqual(grouped, USER_SIDE_ITEM_IDS) self.assertEqual(us["blocked_by_summary"]["user"], ["B1", "B2", "B4", "B7", "B9"]) def test_kwarg_user_side_items_overrides_evidence_key(self): report = build_scorecard(_full_evidence(), user_side_items=_all_bc_pending()) self.assertEqual(report["code_overall"], "pass") self.assertEqual(report["production_overall"], "not_pass") self.assertEqual(report["user_side_items"]["count"], 13) def test_passthrough_preserves_extra_fields_verbatim(self): items = [{"id": "B1", "status": "pending", "blocked_by": "user", "note": "branch protection not yet enabled"}, {"id": "B3", "status": "done", "blocked_by": None}] report = build_scorecard(_full_evidence(), user_side_items=items) self.assertEqual(report["user_side_items"]["items"], items) self.assertEqual(report["user_side_items"]["pending_ids"], ["B1"]) self.assertEqual(report["user_side_items"]["done_ids"], ["B3"]) def test_empty_list_is_unknown_not_done(self): report = build_scorecard(_full_evidence(), user_side_items=[]) self.assertTrue(report["user_side_unknown"]) self.assertTrue(report["user_side_items"]["user_side_unknown"]) self.assertEqual(report["code_overall"], "pass") self.assertEqual(report["production_overall"], "not_pass") self.assertIn("user_side_unknown", report["production_reason"]) # 否定表述本身含 user_side_done 一词，故断言否定语义而非子串缺席。 self.assertTrue(report["production_reason"].startswith( "user_side_unknown")) self.assertIn("not equivalent to user_side_done", report["production_reason"]) self.assertEqual(report["user_side_items"]["items"], []) self.assertEqual(report["user_side_items"]["pending_ids"], []) def test_missing_key_is_unknown_not_done(self): report = build_scorecard(_full_evidence()) self.assertTrue(report["user_side_unknown"]) self.assertEqual(report["production_overall"], "not_pass") self.assertIn("user_side_unknown", report["production_reason"]) self.assertNotIn("done", report["production_reason"].split(":")[0]) def test_invalid_items_are_recorded_not_silently_dropped(self): items = [ {"id": "B1", "status": "pending"}, # 缺 blocked_by {"status": "pending", "blocked_by": "user"}, # 缺 id {"id": "B2", "status": "whatever", "blocked_by": "user"}, {"id": "B3", "status": "pending", "blocked_by": "user"}, {"id": "B3", "status": "pending", "blocked_by": "user"}, "not-a-dict", ] norm = normalize_user_side_items(items) self.assertEqual(norm["items"], [{"id": "B3", "status": "pending", "blocked_by": "user"}]) reasons = [inv["reason"] for inv in norm["invalid_items"]] self.assertEqual(reasons, ["blocked_by_missing", "id_missing", "status_missing_or_unknown", "duplicate_id", "item_not_a_dict"]) self.assertEqual(len(norm["invalid_items"]), 5) # 非法项不编造状态：B1/B2 绝不在 items/pending_ids 里。 self.assertNotIn("B1", norm["pending_ids"]) self.assertNotIn("B2", norm["pending_ids"]) report = build_scorecard(_full_evidence(), user_side_items=items) self.assertEqual(report["code_overall"], "pass") self.assertEqual(report["production_overall"], "not_pass") self.assertIn("user_side_invalid", report["production_reason"]) self.assertIn("user_side_pending: B3", report["production_reason"]) def test_non_list_user_side_is_invalid_not_crash(self): norm = normalize_user_side_items("B1") self.assertEqual(norm["items"], []) self.assertEqual(norm["invalid_items"][0]["reason"], "user_side_items_not_a_list") self.assertFalse(norm["user_side_unknown"]) report = build_scorecard(_full_evidence(), user_side_items="B1") self.assertEqual(report["production_overall"], "not_pass") self.assertIn("user_side_invalid", report["production_reason"]) def test_mixed_pending_and_done_keeps_code_pass(self): items = [{"id": "B1", "status": "done", "blocked_by": None}, {"id": "B2", "status": "pending", "blocked_by": "user"}, {"id": "C1", "status": "blocked", "blocked_by": "contract_decision"}] report = build_scorecard(_full_evidence(), user_side_items=items) self.assertEqual(report["code_overall"], "pass") self.assertEqual(report["production_overall"], "not_pass") # 原因只列未完成项，不把已 done 的 B1 写进 pending。 self.assertEqual(report["production_reason"], "user_side_pending: B2,C1") self.assertEqual(report["user_side_items"]["blocked_ids"], ["C1"]) self.assertFalse(report["user_side_items"]["all_pending"]) self.assertFalse(report["user_side_items"]["all_done"]) def test_all_done_production_follows_its_own_overall(self): items = [{"id": item_id, "status": "done", "blocked_by": None} for item_id in USER_SIDE_ITEM_IDS] report = build_scorecard(_full_evidence(), user_side_items=items) self.assertTrue(report["user_side_items"]["all_done"]) self.assertEqual(report["code_overall"], "pass") self.assertEqual(report["production_overall"], "pass") self.assertIsNone(report["production_reason"]) self.assertFalse(report["user_side_forced_production_not_pass"]) def test_standalone_rule_never_changes_code_overall(self): rule = apply_user_side_three_table_rule( "pass", "pass", _all_bc_pending()) self.assertEqual(rule["code_overall"], "pass") self.assertEqual(rule["production_overall"], "not_pass") self.assertEqual(rule["production_reason"], "user_side_pending: " + ",".join(USER_SIDE_ITEM_IDS)) # 代码表本身不达标时，用户侧全 done 也不把它抬高。 rule2 = apply_user_side_three_table_rule( "not_pass", "not_pass", [{"id": "B1", "status": "done", "blocked_by": None}]) self.assertEqual(rule2["code_overall"], "not_pass") self.assertEqual(rule2["production_overall"], "not_pass") self.assertFalse(rule2["user_side_forced_production_not_pass"]) # dict 输入同样只取 overall，不改代码表结论。 rule3 = apply_user_side_three_table_rule( {"overall": "pass"}, {"overall": "pass"}, []) self.assertEqual(rule3["code_overall"], "pass") self.assertTrue(rule3["user_side_unknown"]) def test_thresholds_still_unapproved_no_decision_fabricated(self): for entry in DEFAULT_THRESHOLDS.values(): self.assertEqual(entry["source"], "test_default_unapproved") report = build_scorecard(_full_evidence(), user_side_items=_all_bc_pending()) self.assertEqual(report["thresholds_source"], "test_default_unapproved") # 7 维 overall 本身仍只由 7 维决定：用户侧项不进 dimensions。 self.assertEqual(len(report["dimensions"]), 7) self.assertEqual([d["id"] for d in report["dimensions"]], ["delivery_quality", "gate_audit", "arch_contract", "test_coverage", "security_ledger", "cost_quota", "ops_scale"]) def test_input_items_not_mutated(self): items = _all_bc_pending() snapshot = [dict(it) for it in items] build_scorecard(_full_evidence(), user_side_items=items) self.assertEqual(items, snapshot) if __name__ == "__main__": unittest.main()