@ngocsangairvds/vsaf 3.2.13 → 3.2.15

package/tools/vds-scripts/audit_orchestrator/tests/unit/state/test_state_loaders.py

@@ -1,2151 +0,0 @@
-from __future__ import annotations
-
-from pathlib import Path
-from unittest.mock import MagicMock, patch
-
-import pytest
-import vds_audit_orchestrator.state as state_module
-from vds_audit_orchestrator.state import (
-    doctor,
-    list_links,
-    load_manifest_payload,
-    load_registry_payload,
-    record_artifacts,
-    record_discovery,
-    repair,
-)
-
-
-def test_normalize_link_url_applies_phase47_rules() -> None:
-    normalize = state_module._normalize_link_url
-
-    assert normalize("HTTPS://Example.COM:443/a/../b/%7Euser/?b=2&a=1&b=1#Section") == (
-        "https://example.com/b/~user?a=1&b=1&b=2"
-    )
-    assert normalize("http://Example.COM:80/") == "http://example.com"
-    assert normalize("https://example.com/%41/%7E?q=%7E&z=%41#frag") == "https://example.com/A/~?q=~&z=A"
-
-
-def test_load_registry_payload_returns_bootstrap_compatible_payload() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 11,
-                "project_key": "project-a",
-                "display_name": "Project A",
-                "metadata": {
-                    "registry_url": "https://wiki.example/registry",
-                    "registry_page_id": "888",
-                    "mapping_version": "v2",
-                    "project": {
-                        "project_name": "Project A",
-                        "project_name_en": "Project A EN",
-                        "page_id": "101",
-                        "page_url": "https://wiki.example/pages/101",
-                        "department": "Department A",
-                        "department_page_id": "201",
-                        "root_page_id": "200",
-                        "root_title": "Registry Root",
-                        "root_page": {"page_id": "200", "title": "Registry Root"},
-                        "root_storage_key": "registry-root",
-                        "root_assignment_reason": "derived_from_registry_root_lineage_head",
-                        "department_storage_key": "department-a",
-                        "department_assignment_reason": "derived_from_registry_root_anchor",
-                        "confluence_ancestry_path": [
-                            {"page_id": "200", "title": "Registry Root", "page_url": "https://wiki.example/pages/200"},
-                            {"page_id": "201", "title": "Department A", "page_url": "https://wiki.example/pages/201"},
-                            {"page_id": "101", "title": "Project A", "page_url": "https://wiki.example/pages/101"},
-                        ],
-                        "documents": [{"title": "BRD"}],
-                        "confluence_links": ["https://wiki.example/x/1"],
-                        "bitbucket_links": ["https://bitbucket.example/projects/A/repos/r1"],
-                        "child_pages": [],
-                    },
-                },
-            }
-        ],
-        "repos": [],
-        "artifacts": [],
-    }
-
-    telemetry_events: list[tuple[str, dict[str, object]]] = []
-    with (
-        patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls,
-        patch("vds_audit_orchestrator.state._log.info") as mock_info,
-    ):
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_registry_payload(state_dsn="postgresql://state")
-        for call in mock_info.call_args_list:
-            if call.args:
-                telemetry_events.append((str(call.args[0]), dict(call.kwargs)))
-
-    assert payload["registry_url"] == "https://wiki.example/registry"
-    assert payload["registry_page_id"] == "888"
-    assert payload["mapping_version"] == "v2"
-    assert payload["checklist_profiles"] == []
-    migration_events = [
-        kwargs for event_name, kwargs in telemetry_events if event_name == "checklist_profile_migration_synthesized"
-    ]
-    assert migration_events == []
-
-    assert len(payload["projects"]) == 1
-    assert payload["projects"][0]["project_name"] == "Project A"
-    assert payload["projects"][0]["page_id"] == "101"
-    assert payload["projects"][0]["documents"] == [{"title": "BRD"}]
-    assert payload["projects"][0]["confluence_links"] == ["https://wiki.example/x/1"]
-    assert payload["projects"][0]["bitbucket_links"] == ["https://bitbucket.example/projects/A/repos/r1"]
-    assert payload["projects"][0]["department_page_id"] == "201"
-    assert payload["projects"][0]["root_page_id"] == "200"
-
-    store.close.assert_called_once()
-
-
-def test_load_registry_payload_preserves_bitbucket_quarantine_summary_shape() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 11,
-                "project_key": "project-a",
-                "display_name": "Project A",
-                "metadata": {
-                    "project": {
-                        "project_name": "Project A",
-                        "page_id": "101",
-                        "page_url": "https://wiki.example/pages/101",
-                        "department": "Department A",
-                        "department_page_id": "201",
-                        "root_page_id": "200",
-                        "root_title": "Registry Root",
-                        "department_assignment_reason": "derived_from_registry_root_anchor",
-                        "bitbucket_links": ["https://bitbucket.example/projects/A/repos/r1"],
-                        "bitbucket_quarantine": {
-                            "total_entries": 2,
-                            "quarantined_entries": 1,
-                            "samples": [{"repo_storage_key": "orphan-repo"}],
-                        },
-                    },
-                },
-            }
-        ],
-        "repos": [],
-        "artifacts": [],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_registry_payload(state_dsn="postgresql://state")
-
-    assert payload["projects"][0]["bitbucket_quarantine"] == {
-        "total_entries": 2,
-        "quarantined_entries": 1,
-        "samples": [{"repo_storage_key": "orphan-repo"}],
-    }
-
-
-def test_load_registry_payload_prefers_explicit_checklist_profiles_from_metadata() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 11,
-                "project_key": "project-a",
-                "display_name": "Project A",
-                "metadata": {
-                    "registry_url": "https://wiki.example/registry",
-                    "registry_page_id": "888",
-                    "checklist_page_id": "777",
-                    "checklist_profiles": [
-                        {
-                            "profile_id": "backend",
-                            "display_name": "Backend Checklist",
-                            "page_ids": ["11111"],
-                            "priority": 10,
-                            "repo_type_filters": ["backend"],
-                        }
-                    ],
-                    "project": {
-                        "project_name": "Project A",
-                        "page_id": "101",
-                        "page_url": "https://wiki.example/pages/101",
-                        "department": "Department A",
-                        "department_page_id": "201",
-                        "root_page_id": "200",
-                        "root_title": "Registry Root",
-                        "root_page": {"page_id": "200", "title": "Registry Root"},
-                        "root_storage_key": "registry-root",
-                        "root_assignment_reason": "derived_from_registry_root_lineage_head",
-                        "department_storage_key": "department-a",
-                        "department_assignment_reason": "derived_from_registry_root_anchor",
-                        "confluence_ancestry_path": [
-                            {"page_id": "200", "title": "Registry Root", "page_url": "https://wiki.example/pages/200"},
-                            {"page_id": "201", "title": "Department A", "page_url": "https://wiki.example/pages/201"},
-                            {"page_id": "101", "title": "Project A", "page_url": "https://wiki.example/pages/101"},
-                        ],
-                        "documents": [],
-                        "confluence_links": [],
-                        "bitbucket_links": [],
-                        "child_pages": [],
-                    },
-                },
-            }
-        ],
-        "repos": [],
-        "artifacts": [],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_registry_payload(state_dsn="postgresql://state")
-
-    assert payload["checklist_profiles"] == [
-        {
-            "profile_id": "backend",
-            "display_name": "Backend Checklist",
-            "page_ids": ["11111"],
-            "priority": 10,
-            "repo_type_filters": ["backend"],
-        }
-    ]
-    store.close.assert_called_once()
-
-
-def test_load_manifest_payload_merges_artifact_paths_from_snapshot() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 31,
-                "project_key": "project-b",
-                "display_name": "Project B",
-                "metadata": {
-                    "registry_url": "https://wiki.example/registry",
-                    "registry_page_id": "999",
-                    "checklist_page_id": "555",
-                    "project": {
-                        "project_name": "Project B",
-                        "page_id": "202",
-                        "page_url": "https://wiki.example/pages/202",
-                        "documents": [],
-                        "confluence_links": [],
-                        "bitbucket_links": [],
-                        "child_pages": [],
-                    },
-                },
-            }
-        ],
-        "repos": [
-            {
-                "id": 41,
-                "project_id": 31,
-                "repo_key": "project-b:project",
-                "repo_url": "",
-                "default_branch": "main",
-                "metadata": {},
-            }
-        ],
-        "artifacts": [
-            {
-                "id": 51,
-                "repo_id": 41,
-                "artifact_key": "project-b:extract_docs",
-                "artifact_type": "extract_docs_summary",
-                "status": "ready",
-                "payload": {
-                    "project_storage_key": "project-b",
-                    "manifest_path": "/tmp/evidence/projects/project-b/manifest.json",
-                    "crawl_manifest_path": "/tmp/evidence/projects/project-b/crawl_manifest.json",
-                    "corpus_path": "/tmp/evidence/content/project-b/corpus.jsonl",
-                    "tables_path": "/tmp/evidence/content/project-b/tables.jsonl",
-                    "mentions_path": "/tmp/evidence/content/project-b/mentions.jsonl",
-                    "link_graph_path": "/tmp/evidence/content/project-b/link-graph.json",
-                    "linked_page_concurrency": 4,
-                    "linked_page_max_depth": 2,
-                    "linked_page_max_pages": 50,
-                    "linked_page_max_links_per_page": 20,
-                    "linked_page_resolution": {
-                        "status": "ok",
-                        "issue_count": 0,
-                    },
-                    "crawl": {
-                        "include_linked_confluence_pages": True,
-                        "linked_page_max_depth": 2,
-                    },
-                    "fetch_policy": "if-changed",
-                    "not_modified_count": 2,
-                    "sync_metadata": {
-                        "fetch_policy": "if-changed",
-                        "not_modified_count_total": 2,
-                        "artifact_count": 1,
-                        "artifacts": [
-                            {
-                                "resource_type": "confluence_page",
-                                "resource_id": "202",
-                                "fetch_reason": "materialized_from_cache",
-                                "not_modified_count": 2,
-                            }
-                        ],
-                    },
-                },
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_manifest_payload(state_dsn="postgresql://state")
-
-    project = payload["projects"][0]
-    assert project["project_name"] == "Project B"
-    assert project["page_id"] == "202"
-    assert project["project_storage_key"] == "project-b"
-    assert "manifest_path" not in project
-    assert "crawl_manifest_path" not in project
-    assert project["corpus_path"] == "/tmp/evidence/content/project-b/corpus.jsonl"
-    assert project["tables_path"] == "/tmp/evidence/content/project-b/tables.jsonl"
-    assert project["mentions_path"] == "/tmp/evidence/content/project-b/mentions.jsonl"
-    assert project["link_graph_path"] == "/tmp/evidence/content/project-b/link-graph.json"
-    assert project["linked_page_concurrency"] == 4
-    assert project["linked_page_max_depth"] == 2
-    assert project["linked_page_max_pages"] == 50
-    assert project["linked_page_max_links_per_page"] == 20
-    assert project["linked_page_resolution"]["issue_count"] == 0
-    assert project["crawl"]["linked_page_max_depth"] == 2
-    assert project["fetch_policy"] == "if-changed"
-    assert project["not_modified_count"] == 2
-    assert project["sync_metadata"]["not_modified_count_total"] == 2
-
-    assert payload["registry_page_id"] == "999"
-    assert payload["checklist_profiles"] == []
-    store.close.assert_called_once()
-
-
-def test_load_manifest_payload_prefers_extract_docs_summary_over_later_repo_artifact() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 31,
-                "project_key": "project-b",
-                "display_name": "Project B",
-                "metadata": {
-                    "project": {
-                        "project_name": "Project B",
-                        "page_id": "202",
-                        "documents": [{"title": "BRD"}],
-                    }
-                },
-            }
-        ],
-        "repos": [
-            {
-                "id": 41,
-                "project_id": 31,
-                "repo_key": "project-b:project",
-                "repo_url": "",
-                "default_branch": "main",
-                "metadata": {},
-            }
-        ],
-        "artifacts": [
-            {
-                "id": 51,
-                "repo_id": 41,
-                "artifact_key": "project-b:extract_docs",
-                "artifact_type": "extract_docs_summary",
-                "status": "ready",
-                "payload": {
-                    "project_storage_key": "project-b",
-                    "sync_metadata": {
-                        "artifacts": [
-                            {
-                                "resource_type": "confluence_page",
-                                "resource_id": "202",
-                            }
-                        ]
-                    },
-                },
-            },
-            {
-                "id": 99,
-                "repo_id": 41,
-                "artifact_key": "project-b:readme",
-                "artifact_type": "repo_readme",
-                "status": "ready",
-                "payload": {
-                    "project_storage_key": "project-b",
-                    "repo_slug": "service-b",
-                },
-            },
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_manifest_payload(state_dsn="postgresql://state")
-
-    project = payload["projects"][0]
-    assert project["project_storage_key"] == "project-b"
-    assert isinstance(project.get("sync_metadata"), dict)
-    assert project["sync_metadata"]["artifacts"][0]["resource_type"] == "confluence_page"
-    store.close.assert_called_once()
-
-
-def test_load_manifest_payload_falls_back_to_project_preparation_material_paths() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 31,
-                "project_key": "project-prep",
-                "display_name": "Project Prep",
-                "metadata": {
-                    "project": {
-                        "project_name": "Project Prep",
-                        "page_id": "303",
-                        "documents": [{"title": "SRS"}],
-                        "confluence_links": ["https://wiki.example/x/prep"],
-                        "bitbucket_links": [],
-                        "child_pages": [],
-                    }
-                },
-            }
-        ],
-        "repos": [],
-        "artifacts": [],
-        "project_preparation": [
-            {
-                "id": 201,
-                "project_id": 31,
-                "status": "prepared",
-                "source": "record_artifacts",
-                "payload_kind": "project_artifacts",
-                "project_storage_key": "project-prep",
-                "fetch_policy": "if-changed",
-                "sync_metadata": {
-                    "corpus_path": "/tmp/evidence/content/project-prep/corpus.jsonl",
-                    "tables_path": "/tmp/evidence/content/project-prep/tables.jsonl",
-                    "mentions_path": "/tmp/evidence/content/project-prep/mentions.jsonl",
-                    "link_graph_path": "/tmp/evidence/content/project-prep/link-graph.json",
-                    "materialized_content_root": "/tmp/evidence/docs/project-prep",
-                    "cache_project_root": "/tmp/evidence/projects/project-prep",
-                    "artifacts": [{"resource_type": "confluence_page", "resource_id": "303"}],
-                },
-                "prepared_at": "2026-02-23T00:00:00Z",
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_manifest_payload(state_dsn="postgresql://state")
-
-    project = payload["projects"][0]
-    assert project["project_storage_key"] == "project-prep"
-    assert project["fetch_policy"] == "if-changed"
-    assert project["materialized_content_root"] == "/tmp/evidence/docs/project-prep"
-    assert project["cache_project_root"] == "/tmp/evidence/projects/project-prep"
-    assert project["corpus_path"] == "/tmp/evidence/content/project-prep/corpus.jsonl"
-    assert project["tables_path"] == "/tmp/evidence/content/project-prep/tables.jsonl"
-    assert project["mentions_path"] == "/tmp/evidence/content/project-prep/mentions.jsonl"
-    assert project["link_graph_path"] == "/tmp/evidence/content/project-prep/link-graph.json"
-    assert project["sync_metadata"]["artifacts"][0]["resource_type"] == "confluence_page"
-    store.close.assert_called_once()
-
-
-def test_load_manifest_payload_merges_split_project_rows_using_inferred_storage_key() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 2,
-                "project_key": "88718856",
-                "display_name": "Mini App",
-                "metadata": {
-                    "project_name": "3.2.1.1 Mini App",
-                    "page_id": "88718856",
-                    "page_url": "http://confluence.digital.vn/pages/viewpage.action?pageId=88718856",
-                    "documents": [{"title": "BRD"}],
-                    "bitbucket_links": ["http://bitbucket.digital.vn/projects/PAR"],
-                    "confluence_links": ["http://confluence.digital.vn/display/TTCN24/3.2.1.1+Mini+app"],
-                },
-            },
-            {
-                "id": 33,
-                "project_key": "par",
-                "display_name": "PAR",
-                "metadata": {
-                    "project_storage_key": "par",
-                    "project_name": "PAR",
-                    "bitbucket_links": ["http://bitbucket.digital.vn/scm/par/merchant-profile-service.git"],
-                    "sync_metadata": {"source": "sync-repos", "operation": "pull"},
-                },
-            },
-        ],
-        "repos": [
-            {
-                "id": 41,
-                "project_id": 33,
-                "repo_key": "par-merchant-profile-service",
-                "repo_url": "http://bitbucket.digital.vn/scm/par/merchant-profile-service.git",
-                "default_branch": "main",
-                "metadata": {
-                    "repo_slug": "merchant-profile-service",
-                    "status": "not_modified",
-                    "local_code_path": "/tmp/PAR/merchant-profile-service",
-                },
-            }
-        ],
-        "artifacts": [],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-        payload = load_manifest_payload(state_dsn="postgresql://state")
-
-    par_projects = [project for project in payload["projects"] if project.get("project_storage_key") == "par"]
-    assert len(par_projects) == 1
-    merged_project = par_projects[0]
-    assert merged_project["page_id"] == "88718856"
-    assert merged_project["documents"] == [{"title": "BRD"}]
-    assert merged_project["sync_metadata"]["source"] == "sync-repos"
-    assert "http://bitbucket.digital.vn/projects/PAR" in merged_project["bitbucket_links"]
-    assert "http://bitbucket.digital.vn/scm/par/merchant-profile-service.git" in merged_project["bitbucket_links"]
-    assert len(merged_project["bitbucket"]) == 1
-    assert merged_project["bitbucket"][0]["repo_slug"] == "merchant-profile-service"
-
-
-def test_load_manifest_payload_preserves_project_and_department_page_metadata() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 91,
-                "project_key": "88718977",
-                "display_name": "pay",
-                "metadata": {
-                    "project": {
-                        "project_name": "pay",
-                        "page_id": "88718977",
-                        "page_url": "https://wiki.example/pages/88718977",
-                        "project_page": {"page_id": "88718977", "title": "pay"},
-                        "department": "3.2.4 Nền tảng thanh toán",
-                        "department_page_id": "88716694",
-                        "department_page": {"page_id": "88716694", "title": "3.2.4 Nền tảng thanh toán"},
-                        "root_page_id": "88716675",
-                        "root_title": "3.2 Các hệ thống nền tảng",
-                        "root_page": {"page_id": "88716675", "title": "3.2 Các hệ thống nền tảng"},
-                        "root_storage_key": "3-2-cac-he-thong-nen-tang",
-                        "root_assignment_reason": "derived_from_lineage_head",
-                        "department_storage_key": "3-2-4-nen-tang-thanh-toan",
-                        "department_assignment_reason": "derived_from_projects_root_lineage",
-                        "confluence_ancestry_path": [
-                            {
-                                "page_id": "88716675",
-                                "title": "3.2 Các hệ thống nền tảng",
-                                "page_url": "https://wiki.example/pages/88716675",
-                            },
-                            {
-                                "page_id": "88716694",
-                                "title": "3.2.4 Nền tảng thanh toán",
-                                "page_url": "https://wiki.example/pages/88716694",
-                            },
-                            {
-                                "page_id": "88718977",
-                                "title": "pay",
-                                "page_url": "https://wiki.example/pages/88718977",
-                            },
-                        ],
-                        "documents": [],
-                        "confluence_links": [],
-                        "bitbucket_links": [],
-                        "child_pages": [],
-                    }
-                },
-            }
-        ],
-        "repos": [],
-        "artifacts": [],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_manifest_payload(state_dsn="postgresql://state")
-
-    project = payload["projects"][0]
-    assert project["project_page"] == {"page_id": "88718977", "title": "pay"}
-    assert project["department_page_id"] == "88716694"
-    assert project["department_page"] == {"page_id": "88716694", "title": "3.2.4 Nền tảng thanh toán"}
-    assert project["root_page_id"] == "88716675"
-    assert project["root_page"] == {"page_id": "88716675", "title": "3.2 Các hệ thống nền tảng"}
-    assert project["root_storage_key"] == "3-2-cac-he-thong-nen-tang"
-    assert project["root_assignment_reason"] == "derived_from_lineage_head"
-    assert project["department_storage_key"] == "3-2-4-nen-tang-thanh-toan"
-    assert project["department_assignment_reason"] == "derived_from_projects_root_lineage"
-    assert [node["page_id"] for node in project["confluence_ancestry_path"]] == [
-        "88716675",
-        "88716694",
-        "88718977",
-    ]
-    store.close.assert_called_once()
-
-
-def test_load_registry_payload_preserves_lineage_page_metadata() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 92,
-                "project_key": "payment-core",
-                "display_name": "Core Payment",
-                "metadata": {
-                    "registry_url": "https://wiki.example/registry",
-                    "registry_page_id": "88716675",
-                    "project": {
-                        "project_name": "Core Payment",
-                        "page_id": "88718943",
-                        "page_url": "https://wiki.example/pages/88718943",
-                        "department": "P.NTTT",
-                        "department_page_id": "88716694",
-                        "root_page_id": "88716675",
-                        "root_title": "3.2 Các hệ thống nền tảng",
-                        "root_page": {"page_id": "88716675", "title": "3.2 Các hệ thống nền tảng"},
-                        "root_storage_key": "3-2-cac-he-thong-nen-tang",
-                        "root_assignment_reason": "derived_from_registry_root_lineage_head",
-                        "department_storage_key": "p-nttt",
-                        "department_assignment_reason": "derived_from_projects_root_lineage",
-                        "confluence_ancestry_path": [
-                            {
-                                "page_id": "88716675",
-                                "title": "3.2 Các hệ thống nền tảng",
-                                "page_url": "https://wiki.example/pages/88716675",
-                            },
-                            {
-                                "page_id": "88716694",
-                                "title": "P.NTTT",
-                                "page_url": "https://wiki.example/pages/88716694",
-                            },
-                            {
-                                "page_id": "88718943",
-                                "title": "Core Payment",
-                                "page_url": "https://wiki.example/pages/88718943",
-                            },
-                        ],
-                        "documents": [],
-                        "confluence_links": [],
-                        "bitbucket_links": [],
-                        "child_pages": [],
-                    },
-                },
-            }
-        ],
-        "repos": [],
-        "artifacts": [],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_registry_payload(state_dsn="postgresql://state")
-
-    project = payload["projects"][0]
-    assert project["project_page"] == {"page_id": "88718943", "title": "Core Payment"}
-    assert project["department_page"] == {"page_id": "88716694", "title": "P.NTTT"}
-    assert project["department_page_id"] == "88716694"
-    assert project["root_page"] == {"page_id": "88716675", "title": "3.2 Các hệ thống nền tảng"}
-    assert project["root_page_id"] == "88716675"
-    assert project["root_assignment_reason"] == "derived_from_registry_root_lineage_head"
-    assert project["department_storage_key"] == "p-nttt"
-    assert project["department_assignment_reason"] == "derived_from_projects_root_lineage"
-    assert [node["title"] for node in project["confluence_ancestry_path"]] == [
-        "3.2 Các hệ thống nền tảng",
-        "P.NTTT",
-        "Core Payment",
-    ]
-    department = payload["departments"][0]
-    assert department["department_page_id"] == "88716694"
-    assert department["root_page_id"] == "88716675"
-    assert department["root_page"] == {"page_id": "88716675", "title": "3.2 Các hệ thống nền tảng"}
-    store.close.assert_called_once()
-
-
-def test_load_registry_payload_filters_container_and_unassigned_rows() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 1,
-                "project_key": "par-project",
-                "display_name": "PAR",
-                "metadata": {
-                    "registry_url": "https://wiki.example/registry",
-                    "registry_page_id": "88716675",
-                    "project": {
-                        "project_name": "PAR",
-                        "page_id": "88727304",
-                        "page_url": "https://wiki.example/pages/88727304",
-                        "department": "3.2.1 Sản phẩm quản trị",
-                        "department_page_id": "88716677",
-                        "root_page_id": "88716675",
-                        "root_title": "3.2 Các hệ thống, nền tảng sản phẩm",
-                        "root_page": {"page_id": "88716675", "title": "3.2 Các hệ thống, nền tảng sản phẩm"},
-                        "root_storage_key": "3-2-cac-he-thong-nen-tang-san-pham",
-                        "root_assignment_reason": "derived_from_registry_root_lineage_head",
-                        "department_storage_key": "3-2-1-san-pham-quan-tri",
-                        "department_assignment_reason": "derived_from_projects_root_lineage",
-                        "confluence_ancestry_path": [
-                            {"page_id": "88716675", "title": "3.2 Các hệ thống, nền tảng sản phẩm"},
-                            {"page_id": "88716677", "title": "3.2.1 Sản phẩm quản trị"},
-                            {"page_id": "88727304", "title": "PAR"},
-                        ],
-                        "documents": [],
-                        "confluence_links": [],
-                        "bitbucket_links": ["https://bitbucket.example/projects/PAR/repos/merchant-report"],
-                        "child_pages": [],
-                    },
-                },
-            },
-            {
-                "id": 2,
-                "project_key": "88716677",
-                "display_name": "par",
-                "metadata": {
-                    "registry_url": "https://wiki.example/registry",
-                    "registry_page_id": "88716675",
-                    "project": {
-                        "project_name": "par",
-                        "page_id": "88716677",
-                        "page_url": "https://wiki.example/pages/88716677",
-                        "department": None,
-                        "department_page_id": None,
-                        "root_page_id": "88716675",
-                        "root_title": "3.2 Các hệ thống, nền tảng sản phẩm",
-                        "root_page": {"page_id": "88716675", "title": "3.2 Các hệ thống, nền tảng sản phẩm"},
-                        "root_storage_key": "3-2-cac-he-thong-nen-tang-san-pham",
-                        "root_assignment_reason": "derived_from_lineage_head",
-                        "department_storage_key": "unassigned",
-                        "department_assignment_reason": "unassigned:lineage_missing_department_after_candidate_root",
-                        "confluence_ancestry_path": [
-                            {"page_id": "88716675", "title": "3.2 Các hệ thống, nền tảng sản phẩm"},
-                            {"page_id": "88716677", "title": "3.2.1 Sản phẩm quản trị"},
-                        ],
-                        "documents": [],
-                        "confluence_links": [],
-                        "bitbucket_links": ["https://bitbucket.example/projects/PAR"],
-                        "child_pages": [],
-                    },
-                },
-            },
-        ],
-        "repos": [],
-        "artifacts": [],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_registry_payload(state_dsn="postgresql://state")
-
-    assert [project["page_id"] for project in payload["projects"]] == ["88727304"]
-    assert len(payload["departments"]) == 1
-    assert payload["departments"][0]["department_page_id"] == "88716677"
-    store.close.assert_called_once()
-
-
-def test_record_artifacts_persists_sync_metadata_fields() -> None:
-    payload = {
-        "kind": "project_artifacts",
-        "registry_page_id": "999",
-        "projects": [
-            {
-                "project_name": "Project C",
-                "project_storage_key": "project-c",
-                "page_id": "303",
-                "project_dir": "/tmp/evidence/projects/project-c",
-                "cache_project_root": "/tmp/evidence/projects/project-c",
-                "materialized_content_root": "/tmp/evidence/docs/project-c",
-                "manifest_path": "/tmp/evidence/projects/project-c/manifest.json",
-                "crawl_manifest_path": "/tmp/evidence/projects/project-c/crawl_manifest.json",
-                "fetch_policy": "if-changed",
-                "linked_page_concurrency": 6,
-                "linked_page_max_depth": 3,
-                "linked_page_max_pages": 120,
-                "linked_page_max_links_per_page": 30,
-                "linked_page_resolution": {
-                    "status": "ok",
-                    "issue_count": 0,
-                },
-                "crawl": {
-                    "include_linked_confluence_pages": True,
-                    "linked_page_max_depth": 3,
-                },
-                "sync_metadata": {
-                    "fetch_policy": "if-changed",
-                    "not_modified_count_total": 1,
-                    "artifact_count": 1,
-                    "artifacts": [
-                        {
-                            "resource_type": "confluence_page",
-                            "resource_id": "303",
-                            "fetch_reason": "materialized_from_cache",
-                            "not_modified_count": 1,
-                        }
-                    ],
-                },
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 31}
-        store.upsert_repo.return_value = {"id": 41}
-        store.upsert_artifact.return_value = {"id": 51}
-        mock_store_cls.return_value = store
-
-        result = record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    assert result["recorded"] is True
-    summary_call = store.upsert_artifact.call_args_list[0]
-    summary_payload = summary_call.kwargs["payload"]
-    assert summary_payload["fetch_policy"] == "if-changed"
-    assert summary_payload["not_modified_count"] == 1
-    assert summary_payload["project_dir"] == "/tmp/evidence/projects/project-c"
-    assert summary_payload["cache_project_root"] == "/tmp/evidence/projects/project-c"
-    assert summary_payload["materialized_content_root"] == "/tmp/evidence/docs/project-c"
-    assert summary_payload["linked_page_concurrency"] == 6
-    assert summary_payload["linked_page_max_depth"] == 3
-    assert summary_payload["linked_page_max_pages"] == 120
-    assert summary_payload["linked_page_max_links_per_page"] == 30
-    assert summary_payload["linked_page_resolution"]["issue_count"] == 0
-    assert summary_payload["crawl"]["linked_page_max_depth"] == 3
-    assert summary_payload["sync_metadata"]["fetch_policy"] == "if-changed"
-    assert summary_payload["sync_metadata"]["artifacts"][0]["fetch_reason"] == "materialized_from_cache"
-    assert "manifest_path" not in summary_payload
-    assert "crawl_manifest_path" not in summary_payload
-
-    project_call = store.upsert_project.call_args_list[0]
-    project_metadata = project_call.kwargs["metadata"]
-    assert project_metadata["project_dir"] == "/tmp/evidence/projects/project-c"
-    assert project_metadata["cache_project_root"] == "/tmp/evidence/projects/project-c"
-    assert project_metadata["materialized_content_root"] == "/tmp/evidence/docs/project-c"
-    assert project_metadata["linked_page_concurrency"] == 6
-    assert project_metadata["linked_page_max_depth"] == 3
-    assert project_metadata["linked_page_max_pages"] == 120
-    assert project_metadata["linked_page_max_links_per_page"] == 30
-    assert project_metadata["linked_page_resolution"]["issue_count"] == 0
-    assert project_metadata["crawl"]["linked_page_max_depth"] == 3
-    assert "manifest_path" not in project_metadata
-    assert "crawl_manifest_path" not in project_metadata
-    prep_kwargs = store.upsert_project_preparation.call_args.kwargs
-    prep_sync_metadata = prep_kwargs["sync_metadata"]
-    assert prep_sync_metadata["linked_page_concurrency"] == 6
-    assert prep_sync_metadata["linked_page_max_depth"] == 3
-    assert prep_sync_metadata["linked_page_max_pages"] == 120
-    assert prep_sync_metadata["linked_page_max_links_per_page"] == 30
-    assert prep_sync_metadata["linked_page_resolution"]["issue_count"] == 0
-    assert prep_sync_metadata["crawl"]["linked_page_max_depth"] == 3
-    store.link_project_hierarchy.assert_called_once()
-    store.link_repo_hierarchy.assert_called_once()
-    store.upsert_project_preparation.assert_called_once()
-    store.close.assert_called_once()
-
-
-def test_record_artifacts_skips_new_project_without_material_identity() -> None:
-    payload = {
-        "kind": "project_artifacts",
-        "projects": [
-            {
-                "project_name": "Core Accounting",
-                "project_storage_key": "91334048",
-                "page_id": "91334048",
-                "page_url": "http://confluence.digital.vn/display/TTCN24/Core+Accounting",
-                "documents": [],
-                "confluence_links": [],
-                "bitbucket_links": [],
-                "child_pages": [],
-                "bitbucket": [],
-                "bitbucket_projects": [],
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        mock_store_cls.return_value = store
-
-        result = record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    assert result["recorded"] is True
-    store.upsert_project.assert_not_called()
-    store.upsert_project_preparation.assert_not_called()
-    store.upsert_repo.assert_not_called()
-    store.upsert_artifact.assert_not_called()
-    store.close.assert_called_once()
-
-
-def test_record_artifacts_keeps_existing_project_when_sparse_payload_arrives() -> None:
-    payload = {
-        "kind": "project_artifacts",
-        "projects": [
-            {
-                "project_name": "Core Accounting",
-                "project_storage_key": "91334048",
-                "page_id": "91334048",
-                "documents": [],
-                "confluence_links": [],
-                "bitbucket_links": [],
-                "child_pages": [],
-                "bitbucket": [],
-                "bitbucket_projects": [],
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {
-            "projects": [
-                {
-                    "id": 91334048,
-                    "project_key": "91334048",
-                    "display_name": "Core Accounting",
-                    "metadata": {
-                        "page_id": "91334048",
-                        "project_name": "Core Accounting",
-                        "project": {
-                            "page_id": "91334048",
-                            "project_name": "Core Accounting",
-                            "documents": [{"title": "Audit Charter"}],
-                            "confluence_links": ["http://confluence.digital.vn/display/TTCN24/Core+Accounting"],
-                        },
-                    },
-                }
-            ],
-            "repos": [],
-            "artifacts": [],
-        }
-        store.upsert_project.return_value = {"id": 91334048}
-        store.upsert_repo.return_value = {"id": 91334049}
-        mock_store_cls.return_value = store
-
-        result = record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    assert result["recorded"] is True
-    assert store.upsert_project.call_args.kwargs["project_key"] == "91334048"
-    store.upsert_project_preparation.assert_called_once()
-    store.upsert_repo.assert_called()
-    store.close.assert_called_once()
-
-
-def test_record_discovery_persists_normalized_project_tables() -> None:
-    payload = {
-        "kind": "registry_discovery",
-        "registry_page_id": "1001",
-        "projects": [
-            {
-                "project_name": "Project D",
-                "project_storage_key": "project-d",
-                "page_id": "404",
-                "page_url": "https://wiki.example/pages/404",
-                "department": "Payments Department",
-                "department_page_id": "333",
-                "root_page_id": "200",
-                "root_title": "3.2 Các hệ thống",
-                "root_storage_key": "3-2-cac-he-thong",
-                "root_assignment_reason": "derived_from_lineage_head",
-                "department_storage_key": "payments-department",
-                "department_assignment_reason": "derived_from_projects_root_lineage",
-                "confluence_ancestry_path": [
-                    {
-                        "page_id": "200",
-                        "title": "3.2 Các hệ thống",
-                        "page_url": "https://wiki.example/pages/200",
-                    },
-                    {
-                        "page_id": "333",
-                        "title": "Payments Department",
-                        "page_url": "https://wiki.example/pages/333",
-                    },
-                    {
-                        "page_id": "404",
-                        "title": "Project D",
-                        "page_url": "https://wiki.example/pages/404",
-                    },
-                ],
-                "confluence_links": ["https://wiki.example/x/proj-d"],
-                "bitbucket_links": ["https://bitbucket.example/projects/D/repos/service-d"],
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 61}
-        mock_store_cls.return_value = store
-
-        result = record_discovery(state_dsn="postgresql://state", payload=payload)
-
-    assert result["recorded"] is True
-    assert store.upsert_project.call_count == 1
-    assert store.upsert_project_link.call_count == 3
-    store.link_project_hierarchy.assert_called_once()
-    store.upsert_project_preparation.assert_called_once()
-    prep_kwargs = store.upsert_project_preparation.call_args.kwargs
-    assert prep_kwargs["status"] == "discovered"
-    assert prep_kwargs["project_storage_key"] == "404"
-    assert prep_kwargs["source"] == "record_discovery"
-    project_metadata = store.upsert_project.call_args.kwargs["metadata"]
-    assert project_metadata["department"] == "Payments Department"
-    assert project_metadata["department_page_id"] == "333"
-    assert project_metadata["root_page_id"] == "200"
-    assert project_metadata["root_title"] == "3.2 Các hệ thống"
-    assert project_metadata["root_page"] == {"page_id": "200", "title": "3.2 Các hệ thống"}
-    assert project_metadata["root_storage_key"] == "3-2-cac-he-thong"
-    assert project_metadata["root_assignment_reason"] == "derived_from_lineage_head"
-    assert project_metadata["department_storage_key"] == "payments-department"
-    assert project_metadata["department_assignment_reason"] == "derived_from_projects_root_lineage"
-    assert len(project_metadata["confluence_ancestry_path"]) == 3
-    store.close.assert_called_once()
-
-
-def test_record_discovery_infers_project_storage_key_from_bitbucket_project_link() -> None:
-    payload = {
-        "kind": "registry_discovery",
-        "projects": [
-            {
-                "project_name": "3.2.1.1 Mini App",
-                "page_id": "88718856",
-                "bitbucket_links": ["http://bitbucket.digital.vn/projects/PAR"],
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 77}
-        mock_store_cls.return_value = store
-
-        result = record_discovery(state_dsn="postgresql://state", payload=payload)
-
-    assert result["recorded"] is True
-    assert store.upsert_project.call_count == 1
-    assert store.upsert_project.call_args.kwargs["project_key"] == "88718856"
-    assert store.upsert_project.call_args.kwargs["metadata"]["project_storage_key"] == "88718856"
-
-
-def test_record_discovery_normalizes_and_dedupes_project_links() -> None:
-    payload = {
-        "kind": "registry_discovery",
-        "projects": [
-            {
-                "project_name": "Project URL",
-                "project_storage_key": "project-url",
-                "page_id": "909",
-                "page_url": "HTTPS://Wiki.Example:443/pages/909/?b=2&a=1#top",
-                "confluence_links": [
-                    "https://wiki.example/pages/909?a=1&b=2",
-                    "https://wiki.example/pages/909/?a=1&b=2#section",
-                ],
-                "bitbucket_links": [
-                    "HTTPS://bitbucket.example:443/projects/A/repos/service?a=2&a=1",
-                    "https://bitbucket.example/projects/A/repos/service?a=1&a=2#frag",
-                ],
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 161}
-        mock_store_cls.return_value = store
-
-        record_discovery(state_dsn="postgresql://state", payload=payload)
-
-    assert store.upsert_project_link.call_count == 3
-    normalized_urls = [call.kwargs["normalized_url"] for call in store.upsert_project_link.call_args_list]
-    assert normalized_urls == [
-        "https://wiki.example/pages/909?a=1&b=2",
-        "https://wiki.example/pages/909?a=1&b=2",
-        "https://bitbucket.example/projects/A/repos/service?a=1&a=2",
-    ]
-    raw_urls = [call.kwargs["raw_url"] for call in store.upsert_project_link.call_args_list]
-    assert raw_urls == [
-        "HTTPS://Wiki.Example:443/pages/909/?b=2&a=1#top",
-        "https://wiki.example/pages/909?a=1&b=2",
-        "HTTPS://bitbucket.example:443/projects/A/repos/service?a=2&a=1",
-    ]
-    for call in store.upsert_project_link.call_args_list:
-        assert call.kwargs["metadata"]["source"] == "record_discovery"
-        assert call.kwargs["metadata"]["project_storage_key"] == "909"
-
-
-def test_record_discovery_preserves_bitbucket_projects_inventory() -> None:
-    payload = {
-        "kind": "registry_discovery",
-        "projects": [
-            {
-                "project_name": "PAR",
-                "page_id": "88727304",
-                "department": "3.2.4 Nền tảng thanh toán",
-                "department_page_id": "88716694",
-                "root_page_id": "88716675",
-                "root_title": "3.2 Các hệ thống, nền tảng sản phẩm",
-                "bitbucket_projects": [
-                    {
-                        "bitbucket_project_key": "PAR",
-                        "display_name": "PAR",
-                        "source": "registry_child_repository_pages",
-                        "repos": [
-                            {
-                                "repo_slug": "merchant-report",
-                                "repo_storage_key": "par:merchant-report",
-                                "repo_url": None,
-                                "path": "http://confluence.digital.vn/pages/viewpage.action?pageId=91337278",
-                                "status": "discovered",
-                            }
-                        ],
-                    }
-                ],
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 188}
-        mock_store_cls.return_value = store
-
-        record_discovery(state_dsn="postgresql://state", payload=payload)
-
-    metadata = store.upsert_project.call_args.kwargs["metadata"]
-    assert metadata["bitbucket_projects"] == payload["projects"][0]["bitbucket_projects"]
-    assert metadata["project"]["bitbucket_projects"] == payload["projects"][0]["bitbucket_projects"]
-
-
-def test_load_manifest_payload_rehydrates_bitbucket_repos_from_discovery_inventory(tmp_path, monkeypatch) -> None:
-    fake_home = tmp_path / "home"
-    repo_root = fake_home / ".vds" / "cache" / "repos" / "par-project" / "merchant-report"
-    repo_root.mkdir(parents=True)
-    monkeypatch.setattr(state_module.Path, "home", lambda: fake_home)
-
-    snapshot = {
-        "projects": [
-            {
-                "id": 77,
-                "project_key": "88727304",
-                "display_name": "PAR",
-                "metadata": {
-                    "project_storage_key": "88727304",
-                    "project_name": "PAR",
-                    "page_id": "88727304",
-                    "page_url": "http://confluence.digital.vn/pages/viewpage.action?pageId=88727304",
-                    "department": "3.2.4 Nền tảng thanh toán",
-                    "department_page_id": "88716694",
-                    "root_page_id": "88716675",
-                    "root_title": "3.2 Các hệ thống, nền tảng sản phẩm",
-                    "root_page": {"page_id": "88716675", "title": "3.2 Các hệ thống, nền tảng sản phẩm"},
-                    "project": {
-                        "project_name": "PAR",
-                        "page_id": "88727304",
-                        "page_url": "http://confluence.digital.vn/pages/viewpage.action?pageId=88727304",
-                        "department": "3.2.4 Nền tảng thanh toán",
-                        "department_page_id": "88716694",
-                        "root_page_id": "88716675",
-                        "root_title": "3.2 Các hệ thống, nền tảng sản phẩm",
-                    },
-                    "bitbucket_projects": [
-                        {
-                            "bitbucket_project_key": "PAR",
-                            "display_name": "PAR",
-                            "repos": [
-                                {
-                                    "repo_storage_key": "par:merchant-report",
-                                    "repo_slug": "merchant-report",
-                                    "repo_url": None,
-                                    "status": "discovered",
-                                }
-                            ],
-                        }
-                    ],
-                },
-            }
-        ],
-        "repos": [],
-        "artifacts": [],
-        "project_preparation": [],
-        "repo_preparation": [],
-        "project_links": [],
-        "repo_links": [],
-        "bitbucket_projects": [],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_manifest_payload(state_dsn="postgresql://state")
-
-    project = payload["projects"][0]
-    repo = project["bitbucket"][0]
-    assert repo["repo_slug"] == "merchant-report"
-    assert repo["repo_storage_key"] == "par:merchant-report"
-    assert repo["source"] == "bitbucket_projects_fallback"
-    assert repo["local_code_path"] == str(repo_root)
-    assert repo["readiness"] == "ready"
-
-
-def test_list_links_emits_normalized_link_payload_without_legacy_fields() -> None:
-    project_payload = {
-        "project_name": "Project List Links",
-        "project_name_en": "Project List Links EN",
-        "page_id": "777",
-        "page_url": "https://wiki.example/pages/777",
-        "confluence_links": ["https://wiki.example/pages/777?a=2&a=1"],
-        "bitbucket_links": ["HTTPS://bitbucket.example:443/projects/Z/repos/service-z?a=2&a=1#fragment"],
-        "bitbucket": [
-            {
-                "project": "Z",
-                "repo": "service-z",
-                "repo_url": "HTTPS://bitbucket.example:443/projects/Z/repos/service-z?a=2&a=1#fragment",
-            }
-        ],
-    }
-    selected_repos = [project_payload["bitbucket"][0]]
-
-    with patch("vds_audit_orchestrator.state._resolve_state_project_and_repos") as mock_resolver:
-        mock_resolver.return_value = ({}, project_payload, selected_repos)
-        result = list_links(state_dsn="postgresql://state", project="project-list-links")
-
-    project_link = result["project_links"][0]
-    assert "link_url" not in project_link
-    assert project_link["raw_url"] == "https://wiki.example/pages/777?a=2&a=1"
-    assert project_link["link_role"] == ""
-    assert project_link["source_page_id"] == "777"
-    assert project_link["source_page_url"] == "https://wiki.example/pages/777"
-    assert project_link["source_context"] == "list_links"
-
-    repo_link = result["repo_links"][0]
-    assert "link_url" not in repo_link
-    assert repo_link["raw_url"] == "HTTPS://bitbucket.example:443/projects/Z/repos/service-z?a=2&a=1#fragment"
-    assert repo_link["normalized_url"] == "https://bitbucket.example/projects/Z/repos/service-z?a=1&a=2"
-    assert repo_link["link_role"] == ""
-    assert repo_link["source_page_id"] == "service-z"
-
-
-def test_list_links_emits_auxiliary_repo_role_when_available() -> None:
-    project_payload = {
-        "project_name": "Project Auxiliary",
-        "project_name_en": "Project Auxiliary EN",
-        "page_id": "778",
-        "page_url": "https://wiki.example/pages/778",
-        "confluence_links": ["https://wiki.example/pages/778"],
-        "bitbucket_links": ["https://bitbucket.example/projects/DOPS/repos/config-repo"],
-        "bitbucket": [
-            {
-                "project": "DOPS",
-                "repo": "config-repo",
-                "repo_url": "https://bitbucket.example/projects/DOPS/repos/config-repo",
-                "role": "auxiliary_config",
-            }
-        ],
-    }
-    selected_repos = [project_payload["bitbucket"][0]]
-
-    with patch("vds_audit_orchestrator.state._resolve_state_project_and_repos") as mock_resolver:
-        mock_resolver.return_value = ({}, project_payload, selected_repos)
-        result = list_links(state_dsn="postgresql://state", project="project-auxiliary")
-
-    repo_link = result["repo_links"][0]
-    assert repo_link["link_role"] == "auxiliary_config"
-    assert repo_link["repo_slug"] == "config-repo"
-
-
-def test_load_manifest_payload_includes_auxiliary_repo_role_from_state_metadata() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 301,
-                "project_key": "project-aux",
-                "display_name": "Project Aux",
-                "metadata": {
-                    "project": {
-                        "project_name": "Project Aux",
-                        "page_id": "aux-1",
-                    }
-                },
-            }
-        ],
-        "repos": [
-            {
-                "id": 401,
-                "project_id": 301,
-                "repo_key": "project-aux:config-repo",
-                "repo_url": "https://bitbucket.example/projects/DOPS/repos/config-repo",
-                "default_branch": "main",
-                "metadata": {
-                    "repo_slug": "config-repo",
-                    "status": "not_modified",
-                    "local_code_path": "/tmp/workspace/DOPS-project/config-repo",
-                    "role": "auxiliary_config",
-                },
-            }
-        ],
-        "artifacts": [],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_manifest_payload(state_dsn="postgresql://state")
-
-    project = payload["projects"][0]
-    assert project["bitbucket"][0]["repo_slug"] == "config-repo"
-    assert project["bitbucket"][0]["role"] == "auxiliary_config"
-    store.close.assert_called_once()
-
-
-def test_record_artifacts_repo_sync_sets_repo_preparation_deterministically() -> None:
-    payload = {
-        "kind": "repo_sync",
-        "projects": [
-            {
-                "project_name": "Project E",
-                "project_storage_key": "project-e",
-                "page_id": "606",
-                "bitbucket_links": ["https://bitbucket.example/projects/E/repos/service-e"],
-                "bitbucket": [
-                    {
-                        "repo": "service-e",
-                        "repo_url": "https://bitbucket.example/projects/E/repos/service-e",
-                        "default_branch": "main",
-                        "status": "success",
-                        "local_code_path": "/tmp/workspace/service-e",
-                        "last_commit": {"id": "abc123"},
-                    }
-                ],
-                "sync_metadata": {
-                    "source": "sync-repos",
-                    "operation": "fetch",
-                },
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 71}
-        store.upsert_repo.side_effect = [{"id": 81}, {"id": 82}]
-        mock_store_cls.return_value = store
-
-        result = record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    assert result["recorded"] is True
-    store.upsert_repo_preparation.assert_called_once()
-    repo_prep_kwargs = store.upsert_repo_preparation.call_args.kwargs
-    assert repo_prep_kwargs["status"] == "ready"
-    assert repo_prep_kwargs["local_code_path"] == "/tmp/workspace/service-e"
-    assert repo_prep_kwargs["sync_status"] == "success"
-    assert repo_prep_kwargs["sync_decision"] == "not_modified"
-    assert repo_prep_kwargs["readiness"] == "ready"
-    repo_keys = {str(call.kwargs.get("repo_key") or "") for call in store.upsert_repo.call_args_list}
-    assert "e-service-e" in repo_keys
-    repo_upsert_kwargs = next(
-        call.kwargs for call in store.upsert_repo.call_args_list if call.kwargs.get("repo_key") == "e-service-e"
-    )
-    assert repo_upsert_kwargs["head_commit_sha"] == "abc123"
-    store.close.assert_called_once()
-
-
-def test_record_artifacts_persists_grouped_bitbucket_contract_and_quarantine() -> None:
-    payload = {
-        "kind": "project_artifacts",
-        "projects": [
-            {
-                "project_name": "Project Grouped",
-                "project_storage_key": "project-grouped",
-                "page_id": "707",
-                "bitbucket_links": ["https://bitbucket.example/projects/PG/repos/service-a"],
-                "bitbucket": [],
-                "bitbucket_projects": [
-                    {
-                        "bitbucket_project_key": "pg",
-                        "repo_count": 1,
-                        "repos": [
-                            {
-                                "repo_storage_key": "pg-service-a",
-                                "bitbucket_repo_key": "service-a",
-                                "repo_slug": "service-a",
-                                "repo_url": "https://bitbucket.example/projects/PG/repos/service-a",
-                            }
-                        ],
-                    }
-                ],
-                "bitbucket_quarantine": {
-                    "total_entries": 2,
-                    "quarantined_entries": 1,
-                    "missing_bitbucket_project_key": 1,
-                    "missing_repo_locator": 1,
-                    "missing_repo_storage_key": 0,
-                    "samples": [
-                        {
-                            "reasons": ["missing_bitbucket_project_key"],
-                            "repo_storage_key": "orphan-repo",
-                        }
-                    ],
-                },
-                "sync_metadata": {"source": "extract-docs"},
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 711}
-        store.upsert_repo.return_value = {"id": 811}
-        mock_store_cls.return_value = store
-
-        result = record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    assert result["recorded"] is True
-    project_metadata = store.upsert_project.call_args.kwargs["metadata"]
-    assert project_metadata["bitbucket_projects"] == payload["projects"][0]["bitbucket_projects"]
-    assert project_metadata["bitbucket_quarantine"] == payload["projects"][0]["bitbucket_quarantine"]
-    artifact_payload = store.upsert_artifact.call_args.kwargs["payload"]
-    assert artifact_payload["bitbucket_projects"] == payload["projects"][0]["bitbucket_projects"]
-    assert artifact_payload["bitbucket_quarantine"] == payload["projects"][0]["bitbucket_quarantine"]
-    store.close.assert_called_once()
-
-
-def test_record_artifacts_repo_sync_complete_status_marks_repo_ready() -> None:
-    payload = {
-        "kind": "repo_sync",
-        "projects": [
-            {
-                "project_name": "Project Complete",
-                "project_storage_key": "project-complete",
-                "page_id": "808",
-                "bitbucket_links": ["https://bitbucket.example/projects/C/repos/service-complete"],
-                "bitbucket": [
-                    {
-                        "repo": "service-complete",
-                        "repo_url": "https://bitbucket.example/projects/C/repos/service-complete",
-                        "default_branch": "main",
-                        "status": "complete",
-                        "local_code_path": "/tmp/workspace/service-complete",
-                    }
-                ],
-                "sync_metadata": {
-                    "source": "sync-repos",
-                    "operation": "pull",
-                },
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 171}
-        store.upsert_repo.side_effect = [{"id": 181}, {"id": 182}]
-        mock_store_cls.return_value = store
-
-        result = record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    assert result["recorded"] is True
-    store.upsert_repo_preparation.assert_called_once()
-    repo_prep_kwargs = store.upsert_repo_preparation.call_args.kwargs
-    assert repo_prep_kwargs["status"] == "ready"
-    assert repo_prep_kwargs["sync_status"] == "complete"
-    assert repo_prep_kwargs["sync_decision"] == "not_modified"
-    assert repo_prep_kwargs["readiness"] == "ready"
-    store.close.assert_called_once()
-
-
-def test_record_artifacts_repo_sync_preserves_existing_project_documents_and_confluence_sync_artifacts() -> None:
-    payload = {
-        "kind": "repo_sync",
-        "projects": [
-            {
-                "project_name": "Project Merge",
-                "project_storage_key": "project-merge",
-                "page_id": "852",
-                "bitbucket_links": ["https://bitbucket.example/projects/MERGE/repos/service-merge"],
-                "bitbucket": [
-                    {
-                        "repo": "service-merge",
-                        "repo_url": "https://bitbucket.example/projects/MERGE/repos/service-merge",
-                        "status": "success",
-                        "local_code_path": "/tmp/workspace/service-merge",
-                    }
-                ],
-                "sync_metadata": {
-                    "source": "sync-repos",
-                    "operation": "pull",
-                },
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {
-            "projects": [
-                {
-                    "id": 211,
-                    "project_key": "852",
-                    "metadata": {
-                        "page_id": "852",
-                        "project_name": "Project Merge",
-                        "documents": [{"title": "Architecture Spec"}],
-                        "confluence_links": ["https://wiki.example/pages/merge"],
-                        "sync_metadata": {
-                            "fetch_policy": "if-changed",
-                            "artifacts": [
-                                {
-                                    "resource_type": "confluence_page",
-                                    "resource_id": "merge-1",
-                                    "fetch_reason": "materialized",
-                                }
-                            ],
-                        },
-                    },
-                }
-            ],
-            "repos": [],
-            "artifacts": [],
-        }
-        store.upsert_project.return_value = {"id": 211}
-        store.upsert_repo.side_effect = [{"id": 301}, {"id": 302}]
-        mock_store_cls.return_value = store
-
-        result = record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    assert result["recorded"] is True
-    project_metadata = store.upsert_project.call_args.kwargs["metadata"]
-    assert project_metadata["documents"] == [{"title": "Architecture Spec"}]
-    assert project_metadata["confluence_links"] == ["https://wiki.example/pages/merge"]
-    assert project_metadata["sync_metadata"]["source"] == "sync-repos"
-    assert project_metadata["sync_metadata"]["operation"] == "pull"
-    assert project_metadata["sync_metadata"]["artifacts"] == [
-        {
-            "resource_type": "confluence_page",
-            "resource_id": "merge-1",
-            "fetch_reason": "materialized",
-        }
-    ]
-
-
-def test_record_artifacts_normalizes_and_dedupes_links() -> None:
-    payload = {
-        "kind": "repo_sync",
-        "projects": [
-            {
-                "project_name": "Project F",
-                "project_storage_key": "project-f",
-                "page_id": "505",
-                "page_url": "HTTPS://Wiki.Example:443/pages/505/?b=2&a=1#frag",
-                "confluence_links": ["https://wiki.example/pages/505?a=1&b=2"],
-                "bitbucket_links": [
-                    "HTTPS://bitbucket.example:443/projects/F/repos/service-f?a=2&a=1",
-                    "https://bitbucket.example/projects/F/repos/service-f?a=1&a=2#frag",
-                ],
-                "bitbucket": [
-                    {
-                        "repo": "service-f",
-                        "repo_url": "HTTPS://bitbucket.example:443/projects/F/repos/service-f?a=2&a=1#frag",
-                        "path": "HTTPS://bitbucket.example:443/projects/F/repos/service-f/browse/README.md?at=refs%2Fheads%2Fmain#L1",
-                        "status": "success",
-                        "local_code_path": "/tmp/workspace/service-f",
-                    }
-                ],
-                "sync_metadata": {"source": "sync-repos"},
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 171}
-        store.upsert_repo.side_effect = [{"id": 181}, {"id": 182}]
-        mock_store_cls.return_value = store
-
-        record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    project_normalized_urls = [call.kwargs["normalized_url"] for call in store.upsert_project_link.call_args_list]
-    assert project_normalized_urls == [
-        "https://wiki.example/pages/505?a=1&b=2",
-        "https://bitbucket.example/projects/F/repos/service-f?a=1&a=2",
-    ]
-    project_raw_urls = [call.kwargs["raw_url"] for call in store.upsert_project_link.call_args_list]
-    assert project_raw_urls == [
-        "HTTPS://Wiki.Example:443/pages/505/?b=2&a=1#frag",
-        "HTTPS://bitbucket.example:443/projects/F/repos/service-f?a=2&a=1",
-    ]
-    repo_normalized_urls = [call.kwargs["normalized_url"] for call in store.upsert_repo_link.call_args_list]
-    assert repo_normalized_urls == [
-        "https://bitbucket.example/projects/F/repos/service-f?a=1&a=2",
-        "https://bitbucket.example/projects/F/repos/service-f/browse/README.md?at=refs%2Fheads%2Fmain",
-    ]
-    repo_raw_urls = [call.kwargs["raw_url"] for call in store.upsert_repo_link.call_args_list]
-    assert repo_raw_urls == [
-        "HTTPS://bitbucket.example:443/projects/F/repos/service-f?a=2&a=1#frag",
-        "HTTPS://bitbucket.example:443/projects/F/repos/service-f/browse/README.md?at=refs%2Fheads%2Fmain#L1",
-    ]
-    for call in store.upsert_repo_link.call_args_list:
-        assert call.kwargs["metadata"]["source"] == "record_artifacts"
-
-
-def test_record_artifacts_classifies_internal_confluence_links_correctly() -> None:
-    payload = {
-        "kind": "repo_sync",
-        "projects": [
-            {
-                "project_name": "CEP",
-                "project_storage_key": "88718940",
-                "page_id": "88718940",
-                "page_url": "http://10.254.136.35:8090/pages/viewpage.action?pageId=88718940",
-                "confluence_links": [
-                    "http://10.254.136.35:8090/display/TTCN24/CEP+Project",
-                    "http://10.254.136.35:8090/pages/viewpage.action?pageId=92179009",
-                ],
-                "bitbucket_links": [
-                    "http://bitbucket.digital.vn/projects/CPM/repos/campaign-service/browse",
-                ],
-                "bitbucket": [
-                    {
-                        "repo": "campaign-service",
-                        "repo_url": "http://bitbucket.digital.vn/projects/CPM/repos/campaign-service",
-                        "status": "success",
-                        "local_code_path": "/tmp/workspace/campaign-service",
-                    }
-                ],
-                "sync_metadata": {"source": "sync-repos"},
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 401}
-        store.upsert_repo.side_effect = [{"id": 501}, {"id": 502}]
-        mock_store_cls.return_value = store
-
-        record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    project_links = {
-        call.kwargs["normalized_url"]: call.kwargs["link_type"] for call in store.upsert_project_link.call_args_list
-    }
-    assert project_links["http://10.254.136.35:8090/pages/viewpage.action?pageId=88718940"] == "project_page"
-    assert project_links["http://10.254.136.35:8090/display/TTCN24/CEP%2BProject"] == "confluence"
-    assert project_links["http://10.254.136.35:8090/pages/viewpage.action?pageId=92179009"] == "confluence"
-    assert project_links["http://bitbucket.digital.vn/projects/CPM/repos/campaign-service/browse"] == "bitbucket"
-    store.delete_project_links_by_type_and_urls.assert_called_once()
-    cleanup_kwargs = store.delete_project_links_by_type_and_urls.call_args.kwargs
-    assert cleanup_kwargs["project_id"] == 401
-    assert cleanup_kwargs["link_type"] == "bitbucket"
-    assert cleanup_kwargs["normalized_urls"] == [
-        "http://10.254.136.35:8090/display/TTCN24/CEP%2BProject",
-        "http://10.254.136.35:8090/pages/viewpage.action?pageId=88718940",
-        "http://10.254.136.35:8090/pages/viewpage.action?pageId=92179009",
-    ]
-    store.cleanup_project_confluence_links_misclassified_as_bitbucket.assert_called_once_with(project_id=401)
-
-
-def test_record_artifacts_skips_local_project_placeholder_repo_urls() -> None:
-    payload = {
-        "kind": "repo_sync",
-        "projects": [
-            {
-                "project_name": "Mini App",
-                "page_id": "983",
-                "bitbucket_links": ["http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core/browse"],
-                "bitbucket": [
-                    {
-                        "project": "local",
-                        "repo": "merchant-profile-service",
-                        "status": "success",
-                        "local_code_path": "/tmp/workspace/merchant-profile-service",
-                    },
-                    {
-                        "project": "WEBAPPV2",
-                        "repo": "miniapp-core",
-                        "status": "success",
-                        "local_code_path": "/tmp/workspace/miniapp-core",
-                    },
-                ],
-                "sync_metadata": {"source": "sync-repos"},
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 211}
-        store.upsert_repo.side_effect = [{"id": 311}, {"id": 312}, {"id": 313}]
-        mock_store_cls.return_value = store
-
-        record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    project_metadata = store.upsert_project.call_args.kwargs["metadata"]
-    assert project_metadata["bitbucket_links"] == [
-        "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core/browse"
-    ]
-
-    repo_url_by_key = {
-        str(call.kwargs["repo_key"]): str(call.kwargs["repo_url"] or "") for call in store.upsert_repo.call_args_list
-    }
-    assert repo_url_by_key["local-merchant-profile-service"] == ""
-    assert (
-        repo_url_by_key["webappv2-miniapp-core"] == "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core"
-    )
-    assert "https://bitbucket.local/projects/local/repos/merchant-profile-service" not in repo_url_by_key.values()
-
-    repo_link_urls = [call.kwargs["normalized_url"] for call in store.upsert_repo_link.call_args_list]
-    assert "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core" in repo_link_urls
-    assert "https://bitbucket.local/projects/local/repos/merchant-profile-service" not in repo_link_urls
-
-
-def test_record_artifacts_skips_project_only_bitbucket_repo_entries_without_repo_identity() -> None:
-    payload = {
-        "kind": "extract_docs",
-        "projects": [
-            {
-                "project_name": "Core Transfer",
-                "project_storage_key": "88718975",
-                "page_id": "88718975",
-                "bitbucket_links": ["http://bitbucket.digital.vn/projects/SWIT"],
-                "bitbucket": [
-                    {
-                        "project": "SWIT",
-                        "repo_url": "http://bitbucket.digital.vn/projects/SWIT",
-                        "status": "error",
-                        "sync_decision": "failed",
-                    }
-                ],
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 701}
-        mock_store_cls.return_value = store
-
-        record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    repo_keys = [str(call.kwargs["repo_key"]) for call in store.upsert_repo.call_args_list]
-    assert repo_keys == ["88718975:project"]
-    hierarchy_repo_keys = [str(call.kwargs["repo_key"]) for call in store.link_repo_hierarchy.call_args_list]
-    assert hierarchy_repo_keys == ["88718975:project"]
-    store.upsert_repo_link.assert_not_called()
-    project_metadata = store.upsert_project.call_args.kwargs["metadata"]
-    assert project_metadata["bitbucket_links"] == ["http://bitbucket.digital.vn/projects/SWIT"]
-
-
-def test_record_artifacts_filters_non_bitbucket_links_from_project_bitbucket_links() -> None:
-    payload = {
-        "kind": "extract_docs",
-        "projects": [
-            {
-                "project_name": "Core Lien ket nap tien",
-                "project_storage_key": "88719105",
-                "page_id": "88719105",
-                "bitbucket_links": [
-                    "http://sonar-10.digital.vn/dashboard?id=vm-link-service",
-                    "http://bitbucket.digital.vn/projects/PAY/repos/vm-link-service/browse",
-                ],
-                "bitbucket": [],
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = {"projects": [], "repos": [], "artifacts": []}
-        store.upsert_project.return_value = {"id": 801}
-        mock_store_cls.return_value = store
-
-        record_artifacts(state_dsn="postgresql://state", payload=payload)
-
-    project_metadata = store.upsert_project.call_args.kwargs["metadata"]
-    assert project_metadata["bitbucket_links"] == [
-        "http://bitbucket.digital.vn/projects/PAY/repos/vm-link-service/browse"
-    ]
-
-
-def test_load_manifest_payload_drops_local_placeholder_links_from_repo_rows() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 401,
-                "project_key": "3-2-1-4-mini-app",
-                "display_name": "3.2.1.4 Mini app",
-                "metadata": {
-                    "project_storage_key": "3-2-1-4-mini-app",
-                    "project_name": "3.2.1.4 Mini app",
-                    "page_id": "88718882",
-                    "page_url": "http://confluence.digital.vn/pages/viewpage.action?pageId=88718882",
-                    "bitbucket_links": [
-                        "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core/browse",
-                        "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-lib/browse",
-                    ],
-                    "explicit_bitbucket_links": [],
-                    "project": {
-                        "project_name": "3.2.1.4 Mini app",
-                        "page_id": "88718882",
-                        "page_url": "http://confluence.digital.vn/pages/viewpage.action?pageId=88718882",
-                        "documents": [
-                            {"title": "BRD"},
-                            {
-                                "source": "bitbucket",
-                                "url": "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core/browse",
-                            },
-                        ],
-                        "confluence_links": ["http://confluence.digital.vn/display/TTCN24/3.2.1.4+Mini+app"],
-                        "bitbucket_links": [
-                            "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core/browse",
-                            "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-lib/browse",
-                        ],
-                        "explicit_bitbucket_links": [],
-                        "child_pages": [],
-                    },
-                },
-            }
-        ],
-        "repos": [
-            {
-                "id": 501,
-                "project_id": 401,
-                "repo_key": "local-merchant-profile-service",
-                "repo_url": "https://bitbucket.local/projects/local/repos/merchant-profile-service",
-                "default_branch": "main",
-                "metadata": {"repo_slug": "merchant-profile-service", "source": "extract-docs"},
-            },
-            {
-                "id": 502,
-                "project_id": 401,
-                "repo_key": "webappv2-miniapp-core",
-                "repo_url": "https://bitbucket.local/projects/WEBAPPV2/repos/miniapp-core",
-                "default_branch": "main",
-                "metadata": {"repo_slug": "miniapp-core", "source": "extract-docs"},
-            },
-            {
-                "id": 503,
-                "project_id": 401,
-                "repo_key": "webappv2-miniapp-lib",
-                "repo_url": "https://bitbucket.local/projects/WEBAPPV2/repos/miniapp-lib",
-                "default_branch": "main",
-                "metadata": {"repo_slug": "miniapp-lib", "source": "extract-docs"},
-            },
-        ],
-        "artifacts": [],
-        "project_links": [
-            {
-                "id": 601,
-                "project_id": 401,
-                "link_type": "bitbucket",
-                "link_role": "",
-                "normalized_url": "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core/browse",
-                "raw_url": "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core/browse",
-                "source_page_id": "88718882",
-                "source_page_url": "http://confluence.digital.vn/pages/viewpage.action?pageId=88718882",
-                "source_context": "record_discovery",
-                "last_seen_at": None,
-                "metadata": {},
-                "created_at": None,
-                "updated_at": None,
-            }
-        ],
-        "repo_links": [],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        payload = load_manifest_payload(state_dsn="postgresql://state")
-
-    project = payload["projects"][0]
-    assert "https://bitbucket.local/projects/local/repos/merchant-profile-service" not in project["bitbucket_links"]
-    assert "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core" in project["bitbucket_links"]
-    assert "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-lib" in project["bitbucket_links"]
-    assert project["explicit_bitbucket_links"] == [
-        "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core/browse"
-    ]
-
-    repo_entries = {entry["repo_storage_key"]: entry for entry in project["bitbucket"]}
-    assert list(repo_entries) == ["webappv2-miniapp-core"]
-    assert (
-        repo_entries["webappv2-miniapp-core"]["repo_url"]
-        == "http://bitbucket.digital.vn/projects/WEBAPPV2/repos/miniapp-core"
-    )
-    store.close.assert_called_once()
-
-
-def test_doctor_reports_integrity_issues() -> None:
-    snapshot = {
-        "projects": [
-            {
-                "id": 1,
-                "project_key": "project-a",
-                "display_name": "Project A",
-                "metadata": {
-                    "confluence_links": ["not-a-url"],
-                    "bitbucket_links": ["https://bitbucket.example/projects/A/repos/r1"],
-                },
-            }
-        ],
-        "repos": [
-            {"id": 2, "project_id": 999, "repo_key": "repo-a"},
-            {
-                "id": 4,
-                "project_id": 1,
-                "repo_key": "repo-b",
-                "metadata": {"project_storage_key": "other-project"},
-            },
-        ],
-        "artifacts": [
-            {
-                "id": 3,
-                "repo_id": 888,
-                "artifact_key": "artifact-a",
-                "payload": {"corpus_path": str(Path("/tmp/definitely-missing-corpus.jsonl"))},
-            }
-        ],
-    }
-
-    with patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls:
-        store = MagicMock()
-        store.validate_schema.return_value = True
-        store.export_snapshot.return_value = snapshot
-        mock_store_cls.return_value = store
-
-        result = doctor(state_dsn="postgresql://state")
-
-    checks = result["checks"]
-    assert result["schema_valid"] is True
-    assert result["integrity_valid"] is False
-    assert result["valid"] is False
-    assert result["integrity_issue_count"] == 6
-    assert checks["counts"] == {"projects": 1, "repos": 2, "artifacts": 1}
-    assert checks["orphan_repo_ids"] == [2]
-    assert checks["orphan_artifact_ids"] == [3]
-    assert checks["missing_project_identity"] == ["project-a"]
-    assert checks["invalid_urls"][0]["field"] == "confluence_links"
-    assert checks["missing_artifact_paths"][0]["artifact_key"] == "artifact-a"
-    assert checks["ownership_conflicts"][0]["repo_key"] == "repo-b"
-
-
-def test_repair_returns_doctor_report() -> None:
-    snapshot = {"projects": [], "repos": [], "artifacts": []}
-    refresh_summary = {
-        "projects_total": 0,
-        "projects_rekeyed": 0,
-        "projects_merged": 0,
-        "projects_already_canonical": 0,
-        "projects_missing_page_id": 0,
-        "projects_invalid_page_id": 0,
-        "project_metadata_backfilled": 0,
-        "repo_metadata_normalized": 0,
-        "project_reference_rows_rewritten": 0,
-        "project_merge_rows_touched": 0,
-        "project_reference_updates": [],
-        "unresolved_projects": [],
-    }
-
-    with (
-        patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls,
-        patch("vds_audit_orchestrator.state._alembic_upgrade_audit_state") as mock_alembic_upgrade,
-    ):
-        store = MagicMock()
-        store.validate_schema.return_value = True
-        store.export_snapshot.return_value = snapshot
-        store.refresh_project_identifiers.return_value = refresh_summary
-        store.cleanup_polluted_registry_project_rows.return_value = {"projects_removed": 0}
-        store.cleanup_missing_artifact_paths.return_value = {"artifacts_updated": 0, "fields_cleared": 0}
-        mock_store_cls.return_value = store
-
-        result = repair(state_dsn="postgresql://state")
-
-    assert result["repaired"] is True
-    assert result["fully_valid"] is True
-    assert result["project_identifier_refresh"] == refresh_summary
-    assert result["metadata_updates"] == 0
-    assert result["polluted_registry_rows_removed"] == 0
-    assert result["missing_artifact_path_cleanup"] == {"artifacts_updated": 0, "fields_cleared": 0}
-    assert result["report"]["schema_valid"] is True
-    assert result["report"]["integrity_valid"] is True
-    assert result["report"]["valid"] is True
-    store.refresh_project_identifiers.assert_called_once()
-    store.cleanup_polluted_registry_project_rows.assert_called_once_with(registry_page_id="88716675")
-    # P4 / TSK-4.1.8: repair() now invokes alembic upgrade head via init().
-    mock_alembic_upgrade.assert_called_once_with("postgresql://state")
-
-
-def test_repair_reports_corpus_identity_backfill_counts() -> None:
-    snapshot = {"projects": [], "repos": [], "artifacts": []}
-    refresh_summary = {
-        "projects_total": 0,
-        "projects_rekeyed": 0,
-        "projects_merged": 0,
-        "projects_already_canonical": 0,
-        "projects_missing_page_id": 0,
-        "projects_invalid_page_id": 0,
-        "project_metadata_backfilled": 0,
-        "repo_metadata_normalized": 0,
-        "project_reference_rows_rewritten": 0,
-        "project_merge_rows_touched": 0,
-        "project_reference_updates": [],
-        "unresolved_projects": [],
-    }
-    corpus_identity_summary = {
-        "documents_project_id_backfilled": 2,
-        "confluence_page_chunk_page_id_repaired": 3,
-        "orphans": 0,
-    }
-
-    with (
-        patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls,
-        patch("vds_audit_orchestrator.state._alembic_upgrade_audit_state"),
-    ):
-        store = MagicMock()
-        store.validate_schema.return_value = True
-        store.export_snapshot.return_value = snapshot
-        store.refresh_project_identifiers.return_value = refresh_summary
-        mock_store_cls.return_value = store
-
-        with patch.object(
-            state_module,
-            "_refresh_corpus_identity",
-            return_value=corpus_identity_summary,
-            create=True,
-        ) as mock_refresh_corpus_identity:
-            result = repair(state_dsn="postgresql://state")
-
-    assert result["repaired"] is True
-    assert "corpus_identity_backfill" not in result
-    mock_refresh_corpus_identity.assert_not_called()
-
-
-def test_repair_sanitizes_non_http_project_links() -> None:
-    snapshot_before = {
-        "projects": [
-            {
-                "id": 1,
-                "project_key": "local",
-                "display_name": "Local",
-                "metadata": {
-                    "confluence_links": [],
-                    "bitbucket_links": ["file:///tmp/local-repo"],
-                },
-            }
-        ],
-        "repos": [],
-        "artifacts": [],
-    }
-    snapshot_after = {
-        "projects": [
-            {
-                "id": 1,
-                "project_key": "local",
-                "display_name": "Local",
-                "metadata": {
-                    "confluence_links": [],
-                    "bitbucket_links": [],
-                },
-            }
-        ],
-        "repos": [],
-        "artifacts": [],
-    }
-
-    with (
-        patch("vds_audit_orchestrator.state.PostgresStateStore") as mock_store_cls,
-        patch("vds_audit_orchestrator.state._alembic_upgrade_audit_state"),
-    ):
-        store = MagicMock()
-        store.validate_schema.return_value = True
-        store.export_snapshot.side_effect = [snapshot_before, snapshot_after]
-        store.refresh_project_identifiers.return_value = {
-            "projects_total": 1,
-            "projects_rekeyed": 0,
-            "projects_merged": 0,
-            "projects_already_canonical": 1,
-            "projects_missing_page_id": 0,
-            "projects_invalid_page_id": 0,
-            "project_metadata_backfilled": 0,
-            "repo_metadata_normalized": 0,
-            "project_reference_rows_rewritten": 0,
-            "project_merge_rows_touched": 0,
-            "project_reference_updates": [],
-            "unresolved_projects": [],
-        }
-        mock_store_cls.return_value = store
-
-        result = repair(state_dsn="postgresql://state")
-
-    assert result["repaired"] is True
-    assert result["metadata_updates"] == 1
-    assert result["fully_valid"] is True
-    assert result["project_identifier_refresh"]["projects_already_canonical"] == 1
-    store.upsert_project.assert_called_once()
-    call = store.upsert_project.call_args
-    assert call.kwargs["project_key"] == "local"
-    assert call.kwargs["metadata"]["bitbucket_links"] == []
-
-
-def test_project_storage_key_from_payload_uses_page_id_only_even_with_bitbucket_links() -> None:
-    project_payload = {
-        "page_id": "88718856",
-        "project_name": "Mini App",
-        "bitbucket_links": ["http://bitbucket.digital.vn/projects/PAR"],
-    }
-
-    assert state_module._project_storage_key_from_payload(project_payload) == "88718856"
-
-
-def test_project_storage_key_from_payload_raises_without_numeric_page_id() -> None:
-    project_payload = {
-        "page_id": "mini-app",
-        "project_name_en": "Mini App",
-        "project_name": "Mini app (VN)",
-        "bitbucket_links": [],
-    }
-
-    with pytest.raises(ValueError, match="requires numeric Confluence page_id"):
-        state_module._project_storage_key_from_payload(project_payload)