YG-Soft/X-Financial
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(ai): recognize conversational document filters

Browse Source
This commit is contained in:
caoxiaozhu
2026-07-20 10:31:15 +08:00
parent 7a3755ac55
commit 2a9b3e3273
12 changed files with 250 additions and 54 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
server/src/app/services/ontology_detection.py
View File

@@ -83,6 +83,13 @@ ENGLISH_FINANCE_BUSINESS_KEYWORDS = (
class OntologyDetectionMixin:
@staticmethod
def _has_document_query_signal(compact_query: str) -> bool:
return "单据" in compact_query and (
any(keyword in compact_query for keyword in QUERY_KEYWORDS)
or any(keyword in compact_query for keyword in STATUS_KEYWORDS)
)
@staticmethod
def _is_expense_application_context(context_json: dict[str, Any]) -> bool:
document_type = str(context_json.get("document_type") or "").strip()
@@ -141,9 +148,7 @@ class OntologyDetectionMixin:
)
):
return True
if "单据" in compact_query and (
"状态" in compact_query or any(keyword in compact_query for keyword in STATUS_KEYWORDS)
):
if self._has_document_query_signal(compact_query):
return True
if any(keyword in compact_query for keyword in EXPENSE_NARRATIVE_KEYWORDS):
return True
@@ -224,9 +229,7 @@ class OntologyDetectionMixin:
best_scenario = "budget"
best_score = scores["budget"]
if best_score <= 0:
if "单据" in compact_query and any(
keyword in compact_query for keyword in STATUS_KEYWORDS
):
if self._has_document_query_signal(compact_query):
return "expense", 0.14
return "unknown", 0.0
@@ -539,7 +542,7 @@ class OntologyDetectionMixin:
"出现“客户”不等于应收，出现“供应商”不等于应付，必须结合动作词和业务目标判断。"
"预算编制、预算金额、成本中心、预算科目、预算预警、预算占用、"
"剩余预算、可用预算、超预算、预算不足等问题必须使用 budget 场景。"
"只有明确查询、统计、列出、多少、明细、对比时才优先使用 query 或 compare。"
"只有明确查询、筛选、过滤、统计、列出、多少、明细、对比时才优先使用 query 或 compare。"
"附件名称和 OCR 摘要只作为辅助证据，不能编造未出现的事实。"
"如果用户明确提到打车、的士票、出租车票、网约车、乘车费、车费等交通票据，"
"即使句子里出现“客户”，也必须优先识别为 transport，不要推断为 entertainment。"

4
server/src/app/services/ontology_rules.py
View File

@@ -36,6 +36,7 @@ SCENARIO_KEYWORDS = {
("单据报销", 0.18),
("报账", 0.20),
("差旅", 0.20),
("出差", 0.20),
("费用", 0.14),
("发票", 0.14),
("票据", 0.12),
@@ -99,6 +100,8 @@ QUERY_KEYWORDS = (
"查",
"查询",
"查看",
"筛选",
"过滤",
"列出",
"统计",
"汇总",
@@ -158,6 +161,7 @@ EXPENSE_TYPE_KEYWORDS = build_expense_type_keyword_map()
EXPENSE_NARRATIVE_KEYWORDS = (
"报销",
"报账",
"出差",
"招待",
"招待费",
"花销",

40
server/tests/test_ontology_service.py
View File

@@ -266,6 +266,46 @@ def test_semantic_ontology_service_extracts_entities_time_and_constraints() -> N
assert result.time_range.end_date == "2026-04-30"
def test_semantic_ontology_service_recognizes_colloquial_travel_city_filter(
monkeypatch,
) -> None:
session_factory = build_session_factory()
with session_factory() as db:
service = SemanticOntologyService(db)
monkeypatch.setattr(
service,
"_parse_with_model",
lambda **_kwargs: (None, [], "model_unavailable_for_rule_regression"),
)
result = service.parse(
OntologyParseRequest(
query="出差上海的单据请筛选一下",
user_id="pytest",
)
)
entity_map = {item.type: item.normalized_value for item in result.entities}
assert result.scenario == "expense"
assert result.intent == "query"
assert result.permission.level == "read"
assert entity_map["expense_type"] == "travel"
assert entity_map["location"] == "上海"
generic_result = service.parse(
OntologyParseRequest(
query="上海的单据请筛选一下",
user_id="pytest",
)
)
generic_entities = {
item.type: item.normalized_value for item in generic_result.entities
}
assert generic_result.scenario == "expense"
assert generic_result.intent == "query"
assert generic_entities["location"] == "上海"
def test_semantic_ontology_service_extracts_budget_query_fields() -> None:
session_factory = build_session_factory()
with session_factory() as db:

71
server/tests/test_orchestrator_review_flow.py
View File

@@ -562,6 +562,77 @@ def test_orchestrator_history_query_filters_location_time_and_returns_real_amoun
assert "321.45" in response.result["answer"]
def test_orchestrator_colloquial_travel_city_filter_returns_only_matching_documents(
monkeypatch,
) -> None:
monkeypatch.setattr(
"app.services.runtime_chat.RuntimeChatService.complete",
lambda *_args, **_kwargs: None,
)
session_factory = build_session_factory()
with session_factory() as db:
employee = Employee(
id="emp-colloquial-filter",
employee_no="E9021",
name="张三",
email="colloquial-filter@example.com",
)
db.add_all(
[
employee,
ExpenseClaim(
id="claim-colloquial-shanghai",
claim_no="EXP-SHANGHAI-001",
employee=employee,
employee_id=employee.id,
employee_name="张三",
department_name="交付部",
expense_type="travel",
reason="前往上海支持客户项目",
location="上海",
amount=Decimal("888.00"),
currency="CNY",
occurred_at=datetime(2026, 7, 8, 9, 0, tzinfo=UTC),
status="paid",
),
ExpenseClaim(
id="claim-colloquial-beijing",
claim_no="EXP-BEIJING-001",
employee=employee,
employee_id=employee.id,
employee_name="张三",
department_name="交付部",
expense_type="travel",
reason="前往北京支持客户项目",
location="北京",
amount=Decimal("666.00"),
currency="CNY",
occurred_at=datetime(2026, 7, 9, 9, 0, tzinfo=UTC),
status="paid",
),
]
)
db.commit()
response = run_for_fixture_tenant(
OrchestratorService(db),
OrchestratorRequest(
source="user_message",
user_id="colloquial-filter@example.com",
message="出差上海的单据请筛选一下",
),
)
query_payload = response.result["query_payload"]
assert response.status == "succeeded"
assert response.trace_summary.scenario == "expense"
assert response.trace_summary.intent == "query"
assert query_payload["record_count"] == 1
assert [item["claim_no"] for item in query_payload["records"]] == [
"EXP-SHANGHAI-001"
]
def test_orchestrator_archive_query_filters_archived_claims_and_limits_preview(
monkeypatch,
) -> None: