mirror of
https://github.com/kharonsec/br-acc
synced 2026-04-25 17:15:02 +02:00
Fix 9 query-schema alignment bugs + 3 AI-review findings
Entity ID alignment: add 6 missing ID fields (cnes_code, finance_id, embargo_id, school_id, convenio_id, stats_id) to entity_by_id and all 6 investigation Cypher queries (WHERE + coalesce chains). entity_by_element_id: add missing PublicOffice label. pattern_self_dealing: fix Amendment field reads with dual-source coalesce fallbacks (TransfereGov + Transparencia). init.cypher + schema_init.cypher: replace dead indexes (amendment_object→function, amendment_date→value_committed, convenio_date→date_published), expand fulltext index to 9 node types with 11 search fields including n.function. seed-dev.cypher: fix all property names (id→contract_id/sanction_id, value→valor, PublicOffice id→cpf), add Amendment node, fix AUTOR_EMENDA target to Amendment. search.py: add name extraction for Contract/Amendment/Convenio/Embargo /PublicOffice types in search results. 21 new tests, 570 total green. Triple-AI validated (Claude + Codex). Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
@@ -14,6 +14,8 @@ EXPECTED_LABELS = {
|
||||
# Expected entity ID property fields in lookup/coalesce chains.
|
||||
EXPECTED_ID_FIELDS = {
|
||||
"cpf", "cnpj", "contract_id", "sanction_id", "amendment_id",
|
||||
"cnes_code", "finance_id", "embargo_id", "school_id", "convenio_id",
|
||||
"stats_id",
|
||||
}
|
||||
|
||||
|
||||
@@ -109,3 +111,110 @@ async def test_connections_rejects_invalid_depth(client: AsyncClient) -> None:
|
||||
async def test_connections_rejects_zero_depth(client: AsyncClient) -> None:
|
||||
response = await client.get("/api/v1/entity/test-id/connections?depth=0")
|
||||
assert response.status_code == 422
|
||||
|
||||
|
||||
# ── entity_by_id: all 11 ID fields are present in WHERE clause ──────────
|
||||
|
||||
|
||||
def test_entity_by_id_has_all_11_id_fields() -> None:
|
||||
"""entity_by_id.cypher must resolve all 11 entity ID property fields."""
|
||||
cypher = _load_cypher("entity_by_id")
|
||||
all_fields = {
|
||||
"cpf", "cnpj", "contract_id", "sanction_id", "amendment_id",
|
||||
"cnes_code", "finance_id", "embargo_id", "school_id",
|
||||
"convenio_id", "stats_id",
|
||||
}
|
||||
for field in all_fields:
|
||||
assert f"e.{field}" in cypher, (
|
||||
f"entity_by_id.cypher missing ID field: e.{field}"
|
||||
)
|
||||
|
||||
|
||||
# ── entity_by_element_id: PublicOffice label ────────────────────────────
|
||||
|
||||
|
||||
def test_entity_by_element_id_has_public_office_label() -> None:
|
||||
"""entity_by_element_id.cypher must include PublicOffice in label allowlist."""
|
||||
cypher = _load_cypher("entity_by_element_id")
|
||||
assert "e:PublicOffice" in cypher, (
|
||||
"entity_by_element_id.cypher missing PublicOffice label"
|
||||
)
|
||||
|
||||
|
||||
def test_entity_by_element_id_has_all_labels() -> None:
|
||||
"""entity_by_element_id.cypher must include all 13 entity labels."""
|
||||
cypher = _load_cypher("entity_by_element_id")
|
||||
for label in EXPECTED_LABELS:
|
||||
assert f"e:{label}" in cypher, (
|
||||
f"entity_by_element_id.cypher missing label: {label}"
|
||||
)
|
||||
|
||||
|
||||
# ── Investigation coalesce chains: all 6 queries include all 11 ID fields ──
|
||||
|
||||
|
||||
INVESTIGATION_COALESCE_QUERIES = [
|
||||
"investigation_get",
|
||||
"investigation_list",
|
||||
"investigation_update",
|
||||
"investigation_by_token",
|
||||
"investigation_add_entity",
|
||||
]
|
||||
|
||||
ALL_ID_FIELDS = [
|
||||
"e.cpf", "e.cnpj", "e.contract_id", "e.sanction_id", "e.amendment_id",
|
||||
"e.cnes_code", "e.finance_id", "e.embargo_id", "e.school_id",
|
||||
"e.convenio_id", "e.stats_id",
|
||||
]
|
||||
|
||||
|
||||
@pytest.mark.parametrize("query_name", INVESTIGATION_COALESCE_QUERIES)
|
||||
def test_investigation_coalesce_has_all_id_fields(query_name: str) -> None:
|
||||
"""Every investigation coalesce chain must include all 11 entity ID fields."""
|
||||
cypher = _load_cypher(query_name)
|
||||
for field in ALL_ID_FIELDS:
|
||||
assert field in cypher, (
|
||||
f"{query_name}.cypher coalesce chain missing {field}"
|
||||
)
|
||||
|
||||
|
||||
INVESTIGATION_WHERE_QUERIES = [
|
||||
"investigation_add_entity",
|
||||
"investigation_remove_entity",
|
||||
]
|
||||
|
||||
|
||||
@pytest.mark.parametrize("query_name", INVESTIGATION_WHERE_QUERIES)
|
||||
def test_investigation_where_has_all_id_fields(query_name: str) -> None:
|
||||
"""Investigation add/remove WHERE clauses must look up all 11 entity ID fields."""
|
||||
cypher = _load_cypher(query_name)
|
||||
for field in ALL_ID_FIELDS:
|
||||
assert field in cypher, (
|
||||
f"{query_name}.cypher WHERE clause missing {field}"
|
||||
)
|
||||
|
||||
|
||||
# ── pattern_self_dealing: uses correct Amendment field names ─────────
|
||||
|
||||
|
||||
def test_self_dealing_uses_value_committed_or_value_paid() -> None:
|
||||
"""pattern_self_dealing.cypher must read value_committed/value_paid with a.value fallback."""
|
||||
cypher = _load_cypher("pattern_self_dealing")
|
||||
# Must use coalesce with both TransfereGov fields AND Transparencia fallback
|
||||
assert "a.value_committed" in cypher, (
|
||||
"pattern_self_dealing.cypher missing a.value_committed (TransfereGov)"
|
||||
)
|
||||
assert "a.value_paid" in cypher, (
|
||||
"pattern_self_dealing.cypher missing a.value_paid (TransfereGov)"
|
||||
)
|
||||
# Transparencia Amendments use a.value and a.object — must be in coalesce fallback
|
||||
lines = cypher.splitlines()
|
||||
for line in lines:
|
||||
if "amendment_value" in line:
|
||||
assert "a.value" in line, (
|
||||
f"amendment_value line missing a.value fallback for Transparencia: {line.strip()}"
|
||||
)
|
||||
if "amendment_object" in line:
|
||||
assert "a.function" in line and "a.object" in line, (
|
||||
f"amendment_object line missing dual-source fallback: {line.strip()}"
|
||||
)
|
||||
|
||||
Reference in New Issue
Block a user