Files
group_fqcd_jr/tests/unit/service/test_run_query_service.py
T
qyqy a7e2d3eac0 fix(customer-service): 统一客服热线来源;补 transfer_required 出参(docs/05 §6.3 未兑现的一半)
两个都是"代码里存在但没接对"的缺陷,都不是新功能。

## A1 客服热线在代码里有两个值(一个出口给假号码)

- `app/core/customer_service_rules.py:35` `CONTACT_PHONE = "15936583816"` ← 真号码,安全路由 6 处在用
- `app/service/agent/implementations/customer_service.py:155` `HOTLINE = "400-XXX-XXXX"` ← 占位符,兜底出口在用

后果:**同一个客服给客户两个不同的电话号码**。问"风险等级怎么划分"被安全路由处理时给真号码;
问一个知识库答不了的问题走兜底时给 `400-XXX-XXXX` —— 客户按这个号码永远打不通。

修法:`HOTLINE` / `SERVICE_HOURS` 改为**转发** `customer_service_rules` 的两个常量
(不是"改成相同的值",而是引用同一对象,避免日后再次漂移);工作时间也随之从
"每日 7:00-22:00" 统一为 "工作日 09:00-18:00"(与安全路由出口一致)。
新增守卫测试用 `is` 断言对象同一性 —— 值相等挡不住"两边各写一份恰好相同"的漂移。

## A2 `transfer_required` 既没落库也没出参

`docs/05` §6.3 一直规定 `GET /agent-runs/{run_id}` 的 `result` 里有
`transfer_required` / `transfer_reason`,但实现里两个都没有:前端判断"这轮要不要转人工"
只能靠**猜正文里有没有兜底话术的开头**(`docs/24` 自己把这称为权宜之计)。

- 写入侧:`conversation_message` **没有** `transfer_required` 列,加列要迁移且规则 4 禁止改既有
  字段定义 ⇒ 放进 `tool_calls` 这个现成 JSON 列,作为 `calls` 的兄弟键
  (`{"calls": [...], "transfer_required": bool, "transfer_reason": str|None}`)
- 读取侧:`RunQueryService.get` 取出来放进 `result`;**兼容历史行**(`calls` 裸列表 / None →
  按 False/None 处理,不抛异常、也不凭正文猜)

刻意**没做**的一半:`docs/05` §6.3 的 `result` 里还有 `degraded` / `degradation_reason`,
但 `CoreResult` 里根本没有这两个字段(降级信息目前只在工具出参里)—— 补它要改
`CoreResult` 并让各 Agent 传递降级状态,属另一个改动范围。**已在交付说明里注明这一半仍缺。**

## 真机验证

| 问题 | transfer_required | transfer_reason | 正文电话 |
|---|---|---|---|
「请介绍一下量子纠缠在基金估值中的应用」 | **True** | 置信度不足:score=0.571 gap=0.004 | 15936583816 ✅ |
「请帮我计算一下三体问题的数值解」 | **True** | 置信度不足:score=0.499 gap=0.011 | 15936583816 ✅ |
「基金申购后多久确认」(正常知识直返) | False | — | 无(正确) |
「你们公司明天会下雪吗」(闲聊出口) | False | — | 无(正确) |

(第一次我用"下雪"当兜底用例,结果它被闲聊出口正确接住了 —— 是我的期望值写错,不是代码问题。)

门禁:测试 1223 passed(新增 4 个用例)/ 3 failed(均为已知非代码缺陷)/ mypy 0 错。
2026-09-11 21:17:47 +08:00

242 lines
9.0 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
"""`RunQueryService` 契约测试(全部使用替身,不连数据库)。
关注三条对客户端与安全都重要的语义:
1. 运行不存在与越权访问**返回同一个 404**——不能靠错误码区分"存在但无权";
2. 未成功的运行**不返回任何结果内容**(`result` 为 None),避免未提交结果外泄;
3. 快照里的时间统一以 `...Z` 结尾(UTC),以及 `watch` 的心跳与终态退出行为。
"""
from dataclasses import dataclass, replace
from datetime import datetime
from decimal import Decimal
from types import SimpleNamespace
from typing import Any
import pytest
from app.core.contracts import RequestContext
from app.core.errors import RunNotFoundError
from app.service.run_query_service import RunQueryService, RunSnapshot
CONTEXT = RequestContext(user_id="9001", trace_id="trace-1")
NOW = datetime(2026, 9, 10, 12, 0, 0)
@dataclass
class FakeRun:
run_id: str = "run-1"
trace_id: str = "trace-1"
status: str = "running"
agent_type: str = "customer_service"
session_id: str = "session-1"
error_code: str | None = None
created_at: datetime = NOW
completed_at: datetime | None = None
@dataclass
class FakeMessage:
content: str = "稳健型"
tool_calls: Any = None
intent: str | None = "general"
confidence: Any = None
source_references: Any = None
class FakeRepository:
def __init__(self, rows: tuple[Any, Any] | None) -> None:
self.rows = rows
async def run_result(self, run_id: str, user_id: int) -> tuple[Any, Any] | None:
del run_id, user_id
return self.rows
class FakeSession:
async def __aenter__(self) -> "FakeSession":
return self
async def __aexit__(self, *exc: object) -> bool:
return False
def patch_repository(monkeypatch: pytest.MonkeyPatch, rows: tuple[Any, Any] | None) -> None:
monkeypatch.setattr("app.service.run_query_service.SessionFactory", FakeSession)
monkeypatch.setattr(
"app.service.run_query_service.ConversationRepository",
lambda _session: FakeRepository(rows),
)
async def test_unknown_run_raises_not_found(monkeypatch: pytest.MonkeyPatch) -> None:
"""不存在与越权必须是同一个错误,否则可以据此探测资源是否存在。"""
patch_repository(monkeypatch, None)
with pytest.raises(RunNotFoundError) as excinfo:
await RunQueryService().get("run-x", CONTEXT)
assert excinfo.value.code == "RUN_NOT_FOUND"
assert excinfo.value.status_code == 404
async def test_running_run_does_not_expose_result(monkeypatch: pytest.MonkeyPatch) -> None:
patch_repository(monkeypatch, (FakeRun(status="running"), None))
snapshot = await RunQueryService().get("run-1", CONTEXT)
assert snapshot.status == "running"
assert snapshot.result is None
assert snapshot.completed_at is None
assert snapshot.created_at.endswith("Z")
async def test_succeeded_run_exposes_result_with_string_confidence(
monkeypatch: pytest.MonkeyPatch,
) -> None:
message = FakeMessage(
content="稳健型",
tool_calls={"calls": []},
confidence=Decimal("0.9000"),
source_references=[{"source_type": "memory"}],
)
patch_repository(monkeypatch, (FakeRun(status="succeeded", completed_at=NOW), message))
snapshot = await RunQueryService().get("run-1", CONTEXT)
assert snapshot.result is not None
assert snapshot.result["content"] == "稳健型"
# DECIMAL(5,4) 经驱动回来是 Decimal,接口层统一转字符串避免浮点精度歧义。
assert snapshot.result["confidence"] == "0.9000"
assert snapshot.result["source_references"] == [{"source_type": "memory"}]
assert snapshot.completed_at is not None and snapshot.completed_at.endswith("Z")
async def test_failed_run_does_not_expose_result(monkeypatch: pytest.MonkeyPatch) -> None:
"""即使库里有结果消息,失败运行也不得把它当作成功结果返回。"""
patch_repository(
monkeypatch, (FakeRun(status="failed", error_code="AGENT_INTERNAL_ERROR"), FakeMessage())
)
snapshot = await RunQueryService().get("run-1", CONTEXT)
assert snapshot.result is None
assert snapshot.error_code == "AGENT_INTERNAL_ERROR"
# --- `docs/05` §6.3 要求的 `transfer_required` / `transfer_reason` 出参 -----------------
async def test_result_exposes_transfer_marker_when_transferred(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""兜底/转人工分支必须把标记**出参**给客户端。
为什么这条重要:前端判断"这轮要不要转人工"此前只能靠**猜正文里有没有兜底话术开头**
(`docs/24` 自称权宜之计)。`docs/05` §6.3 一直规定 `result` 里有这两个字段,
但此前既没落库也没出参 —— 这个用例把"契约已兑现"钉住。
"""
message = FakeMessage(
content="抱歉,这个问题我暂时无法给出准确答复…",
tool_calls={"calls": [], "transfer_required": True,
"transfer_reason": "置信度不足:score=0.571 gap=0.004"},
)
patch_repository(monkeypatch, (FakeRun(status="succeeded", completed_at=NOW), message))
snapshot = await RunQueryService().get("run-1", CONTEXT)
assert snapshot.result is not None
assert snapshot.result["transfer_required"] is True
assert snapshot.result["transfer_reason"] == "置信度不足:score=0.571 gap=0.004"
async def test_result_transfer_marker_defaults_to_false(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""正常回答:标记为 False、原因为 None(不能因为缺键就返回 None 让客户端混淆)。"""
message = FakeMessage(content="交易日 15:00 前提交…",
tool_calls={"calls": [], "transfer_required": False,
"transfer_reason": None})
patch_repository(monkeypatch, (FakeRun(status="succeeded", completed_at=NOW), message))
snapshot = await RunQueryService().get("run-1", CONTEXT)
assert snapshot.result is not None
assert snapshot.result["transfer_required"] is False
assert snapshot.result["transfer_reason"] is None
async def test_result_tolerates_legacy_tool_calls_shape(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""**兼容历史行**:本字段上线前落库的 `tool_calls` 里没有这两个键。
那种行的 `tool_calls` 可能就是裸列表,甚至为 None。读取时必须按 False/None 处理,
**不得抛异常、也不得凭正文内容去猜**——猜错方向会让"不需要转人工"的答复被标成转人工。
"""
for legacy in ({"calls": []}, [], None):
message = FakeMessage(content="稳健型", tool_calls=legacy)
patch_repository(monkeypatch, (FakeRun(status="succeeded", completed_at=NOW), message))
snapshot = await RunQueryService().get("run-1", CONTEXT)
assert snapshot.result is not None, legacy
assert snapshot.result["transfer_required"] is False, legacy
assert snapshot.result["transfer_reason"] is None, legacy
def terminal_snapshot(status: str = "succeeded") -> RunSnapshot:
return RunSnapshot(
run_id="run-1",
trace_id="trace-1",
status=status,
agent_type="customer_service",
session_id="session-1",
result={"content": "ok"},
error_code=None,
created_at="2026-09-10T12:00:00Z",
completed_at="2026-09-10T12:00:01Z",
)
async def test_watch_stops_immediately_on_terminal_snapshot(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""终态快照只发一次且不再轮询——断流重连也走这条路径。"""
calls = {"count": 0}
async def never_called(self: RunQueryService, run_id: str, context: RequestContext) -> Any:
calls["count"] += 1
raise AssertionError("终态不应继续轮询数据库")
monkeypatch.setattr(RunQueryService, "get", never_called)
initial = terminal_snapshot()
events = [snapshot async for snapshot in RunQueryService().watch(initial, CONTEXT)]
assert events == [initial]
assert calls["count"] == 0
async def test_watch_emits_heartbeat_between_polls(monkeypatch: pytest.MonkeyPatch) -> None:
monkeypatch.setattr(
"app.service.run_query_service.get_settings",
lambda: SimpleNamespace(sse_max_connection_seconds=30, sse_heartbeat_seconds=0),
)
polls = {"count": 0}
async def fake_get(self: RunQueryService, run_id: str, context: RequestContext) -> RunSnapshot:
del self, run_id, context
polls["count"] += 1
return replace(terminal_snapshot(), status="succeeded")
monkeypatch.setattr(RunQueryService, "get", fake_get)
initial = replace(terminal_snapshot(), status="running", result=None, completed_at=None)
events = [snapshot async for snapshot in RunQueryService().watch(initial, CONTEXT)]
# 序列:非终态快照 → 心跳(None) → 轮询得到的终态快照
assert [event is None for event in events] == [False, True, False]
assert events[0] is initial
assert events[2] is not None and events[2].status == "succeeded"
assert polls["count"] == 1