Eric Wang
e72d9c9bc0
- merge main (b0ff2e0..c7ca781): workflow/20261022 moves dynamic context out of fixed prompts - collection/extraction prompts carry no inline vars; reference time and current form travel in the user message <turn_context> block; photo/confirm data kept at prompt tails - prompts: all 8 files regenerated from 20261022 (byte-identity tests point at the new export) - collect: user message = <turn_context>(本轮参考时间) + <user_input>; extraction message adds 当前表单; single trusted clock per turn threads through reply/extraction/repair - photo: raw input only (20261022 photo nodes use system_text, no wrapper) - workflow fake llm: unwraps <user_input> before event parsing - fields: accident_time normalized to extraction format (YYYY年M月D日H点m分, strip zero padding) - fixes form assertion mismatch with padded output - evals/run.py: trust_env=False - bypass Windows system proxy (registry proxy at 192.168.0.4:7897 was intercepting localhost and returning 503) - verified: 172 unit tests; ALL golden suites pass on real model (46/46): collection_skip 12/12, photo_skip 4/4, no_false 2/2, button_guidance 12/12, non_motor_vehicle 14/14, time_confirmation 2/2
119 lines
5.0 KiB
Python
119 lines
5.0 KiB
Python
"""20261020 工作流的提示词移植与模型决策回归。"""
|
||
|
||
from datetime import datetime, timezone
|
||
import json
|
||
from pathlib import Path
|
||
|
||
import pytest
|
||
|
||
from src.agent.collect import CollectReplyError, run_collect
|
||
from src.agent.llm import FakeLLMService
|
||
from src.agent.prompts import COLLECTION, load_prompt, render_prompt
|
||
|
||
WORKFLOW = Path(__file__).resolve().parents[2] / "workflow/20261022/事故信息采集2025118.json"
|
||
NODES = {
|
||
"collection": "事故信息收集",
|
||
"single_photo": "单车拍照",
|
||
"dual_photo": "双车拍照",
|
||
"single_photo_addon": "单车拍照#6",
|
||
"dual_photo_addon": "双车拍照#7",
|
||
"single_confirm": "单车信息确认",
|
||
"dual_confirm": "双车信息确认",
|
||
"extraction": "文本内容提取事故信息",
|
||
}
|
||
_INPUT_KEYS = {"extraction": "description"}
|
||
|
||
|
||
@pytest.mark.parametrize(("prompt", "node_name"), NODES.items())
|
||
def test_archived_prompt_matches_workflow(prompt, node_name):
|
||
workflow = json.loads(WORKFLOW.read_text(encoding="utf-8-sig"))
|
||
node = next(n for n in workflow["nodes"] if n["name"] == node_name)
|
||
input_key = _INPUT_KEYS.get(prompt, "systemPrompt")
|
||
original = next(i["value"] for i in node["inputs"] if i["key"] == input_key)
|
||
archived = load_prompt(prompt).split("--- 原文开始 ---", 1)[1].split("--- 原文结束 ---", 1)[0]
|
||
assert archived.strip() == original.strip()
|
||
|
||
|
||
def test_render_replaces_time_and_person_variables():
|
||
# 20261022：collection/extraction 提示词本体不含变量（动态上下文走
|
||
# user 消息的 <turn_context>）；拍照/确认提示词尾部仍渲染人员变量。
|
||
rendered = render_prompt(COLLECTION, {})
|
||
assert "{{$" not in rendered
|
||
rendered = render_prompt("dual_confirm", {
|
||
"person_fields": {"hphm1": "测试A", "hphm2": "测试B", "sfzmwh1": "1234"}
|
||
})
|
||
assert "测试A" in rendered and "测试B" in rendered
|
||
assert "1234" in rendered
|
||
assert "{{$" not in rendered
|
||
assert "未提供" in rendered
|
||
|
||
|
||
def test_user_message_carries_turn_context():
|
||
from src.agent.collect import _build_user_message
|
||
from datetime import datetime, timezone
|
||
now = datetime(2026, 10, 22, 8, 0, tzinfo=timezone.utc)
|
||
message = _build_user_message({"input_text": "两辆"}, now=now)
|
||
assert "<turn_context>" in message
|
||
assert "本轮参考时间：2026-10-22 08:00:00 +0000" in message
|
||
assert "<user_input>" in message
|
||
assert "两辆" in message
|
||
|
||
|
||
@pytest.mark.asyncio
|
||
async def test_short_answer_does_not_override_model_routing():
|
||
# “在”可能是明确回答；运行时不得按字数一律拒绝分流。
|
||
llm = FakeLLMService(lambda s, u: "<state>2000</state>信息已齐，请拍摄车辆前方。")
|
||
result = await run_collect({
|
||
"stage_code": "1002", "input_text": "在", "clarify_count": 0,
|
||
"dialogue": [{"role": "assistant", "content": "您还在事故现场吗？"}],
|
||
}, llm)
|
||
assert result["stage_code"] == "2000"
|
||
assert result["clarify_count"] == 0
|
||
assert len(llm.calls) == 2 # 对话 + 提取
|
||
|
||
|
||
@pytest.mark.asyncio
|
||
async def test_format_and_stage_errors_share_one_repair_budget():
|
||
replies = iter(["缺少前缀", "<state>9999</state>非法阶段"])
|
||
llm = FakeLLMService(lambda s, u: next(replies))
|
||
with pytest.raises(CollectReplyError):
|
||
await run_collect({"stage_code": "1002", "input_text": "追尾了"}, llm)
|
||
assert len(llm.calls) == 2
|
||
|
||
|
||
@pytest.mark.asyncio
|
||
async def test_collect_sends_history_as_messages():
|
||
history = [
|
||
{"role": "user", "content": "没有人受伤"},
|
||
{"role": "assistant", "content": "<state>1002</state>请问有几辆车？"},
|
||
]
|
||
llm = FakeLLMService(lambda s, u: "<state>1002</state>请问您还在现场吗？")
|
||
await run_collect({
|
||
"stage_code": "1002", "input_text": "两辆", "dialogue": history,
|
||
}, llm)
|
||
assert llm.calls[0]["history"] == history
|
||
# 20261022 格式：<turn_context> + <user_input>
|
||
assert "<turn_context>" in llm.calls[0]["user_message"]
|
||
assert "<user_input>\n两辆\n</user_input>" in llm.calls[0]["user_message"]
|
||
assert "没有人受伤" not in llm.calls[0]["user_message"]
|
||
|
||
|
||
@pytest.mark.asyncio
|
||
async def test_openai_adapter_preserves_message_roles():
|
||
from types import SimpleNamespace
|
||
from unittest.mock import AsyncMock
|
||
from src.agent.llm import OpenAILLMService
|
||
|
||
create = AsyncMock(return_value=SimpleNamespace(choices=[
|
||
SimpleNamespace(message=SimpleNamespace(content="<state>1002</state>收到。"))
|
||
]))
|
||
llm = OpenAILLMService(model="test", api_key="test", max_retries=0)
|
||
llm._client = SimpleNamespace(chat=SimpleNamespace(completions=SimpleNamespace(create=create)))
|
||
history = [{"role": "assistant", "content": "之前的问题"}]
|
||
await llm.complete("系统提示词", "本轮输入", history=history)
|
||
assert create.call_args.kwargs["messages"] == [
|
||
{"role": "system", "content": "系统提示词"},
|
||
{"role": "assistant", "content": "之前的问题"},
|
||
{"role": "user", "content": "本轮输入"},
|
||
]
|