feat: 足球 LLM 预测服务初始提交
Profeto — 给 LLM 提供数据,让 LLM 预测足球比分。 核心模块: - FastAPI 后端 + PostgreSQL (SQLAlchemy async) - 多 Agent LLM 预测 (5 专家 + 终裁) - 数据采集 (bzzoiro / understat / injuries) - React 前端 (Vite + Tailwind) 包含: - 数据源抽象 (DataSource 协议 + 注册表) - Alembic 数据库迁移 - Prompt 模板 (单/多 Agent) - 核心路径单元测试
This commit is contained in:
shangfangjian
@@ -0,0 +1,213 @@
|
||||
"""多 agent 架构测试。"""
|
||||
import pytest
|
||||
from src.llm.agents.base import AgentReport, load_agent_prompt, _is_no_data
|
||||
from src.llm.context_builder import MatchHeader
|
||||
|
||||
|
||||
class TestNoDataGate:
|
||||
"""no_data 门控: 切片无数据 → 跳过 LLM。"""
|
||||
|
||||
def test_is_no_data_all_lines(self):
|
||||
assert _is_no_data("── 伤停 ──\n 无数据") is True
|
||||
|
||||
def test_is_no_data_with_content(self):
|
||||
assert _is_no_data("── 交锋 ──\n 2026-03: A 2-1 B") is False
|
||||
|
||||
def test_is_no_data_empty(self):
|
||||
assert _is_no_data("") is True
|
||||
|
||||
def test_is_no_data_mixed(self):
|
||||
# 部分有数据部分无 → 不是 no_data
|
||||
assert _is_no_data("── xG ──\n A: xG 1.5 vs B 0.8\n B: 无 xG 数据") is False
|
||||
|
||||
def test_stub_no_data_report(self):
|
||||
from src.llm.agents.base import _stub_no_data
|
||||
r = _stub_no_data("injuries")
|
||||
assert r.status == "no_data"
|
||||
assert r.data_sufficiency == "none"
|
||||
assert r.home_edge is None
|
||||
|
||||
|
||||
class TestPromptLoading:
|
||||
"""agent prompt 模板加载。"""
|
||||
|
||||
@pytest.mark.parametrize("name", ["form", "stats", "home_away", "injuries", "h2h", "aggregator"])
|
||||
def test_all_prompts_exist(self, name):
|
||||
tpl = load_agent_prompt(name, "v1")
|
||||
assert "{{context}}" in tpl or "{{agent_reports}}" in tpl
|
||||
|
||||
def test_prompt_not_found(self):
|
||||
with pytest.raises(FileNotFoundError):
|
||||
load_agent_prompt("nonexistent", "v1")
|
||||
|
||||
|
||||
class TestReportParsing:
|
||||
"""LLM JSON 输出 → AgentReport 解析(宽容处理)。"""
|
||||
|
||||
def _make_header(self) -> MatchHeader:
|
||||
return MatchHeader(
|
||||
match_id=1, home_name="A", away_name="B", league_name="PL",
|
||||
season="2026-2027", match_date="2026-09-15", match_dt=None,
|
||||
stage=None, home_team_id=10, away_team_id=20, league_id=1,
|
||||
)
|
||||
|
||||
def test_parse_full_report(self):
|
||||
from src.llm.agents.base import _parse_report
|
||||
from src.llm.provider import LLMResponse
|
||||
|
||||
parsed = {
|
||||
"data_sufficiency": "high",
|
||||
"analysis": "主队交锋占优",
|
||||
"home_edge": 0.6,
|
||||
"confidence": 0.8,
|
||||
"key_evidence": ["近5次交锋主队4胜", "主场交锋3连胜"],
|
||||
}
|
||||
resp = LLMResponse(content="{}", parsed=parsed, prompt_tokens=100, completion_tokens=50, latency_ms=500)
|
||||
r = _parse_report("h2h", parsed, resp, "gpt-4o-mini")
|
||||
assert r.status == "ok"
|
||||
assert r.home_edge == 0.6
|
||||
assert r.confidence == 0.8
|
||||
assert len(r.key_evidence) == 2
|
||||
assert r.data_sufficiency == "high"
|
||||
|
||||
def test_parse_xg_report_with_score(self):
|
||||
from src.llm.agents.base import _parse_report
|
||||
from src.llm.provider import LLMResponse
|
||||
|
||||
parsed = {
|
||||
"data_sufficiency": "medium",
|
||||
"analysis": "主队火力更强",
|
||||
"home_edge": 0.4,
|
||||
"confidence": 0.7,
|
||||
"key_evidence": ["场均xG 2.1"],
|
||||
"exp_home_goals": 2.1,
|
||||
"exp_away_goals": 1.2,
|
||||
"probable_score": {"home": 2, "away": 1, "prob": 0.14},
|
||||
}
|
||||
resp = LLMResponse(content="{}", parsed=parsed)
|
||||
r = _parse_report("xg", parsed, resp, "gpt-4o-mini")
|
||||
assert r.exp_home_goals == 2.1
|
||||
assert r.probable_score == "2-1"
|
||||
|
||||
def test_parse_bad_values_forgiving(self):
|
||||
"""非法数值/字段宽容降级,不抛异常。"""
|
||||
from src.llm.agents.base import _parse_report
|
||||
from src.llm.provider import LLMResponse
|
||||
|
||||
parsed = {
|
||||
"data_sufficiency": "bogus", # 非法 → medium
|
||||
"home_edge": "very strong", # 非法 → None
|
||||
"confidence": None,
|
||||
"key_evidence": "单字符串", # → [str]
|
||||
}
|
||||
resp = LLMResponse(content="{}", parsed=parsed)
|
||||
r = _parse_report("form", parsed, resp, "m")
|
||||
assert r.data_sufficiency == "medium"
|
||||
assert r.home_edge is None
|
||||
assert r.key_evidence == ["单字符串"]
|
||||
|
||||
|
||||
class TestRunAgent:
|
||||
"""run_agent 执行器: 门控 + fail-open。"""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_no_data_skips_llm(self):
|
||||
"""切片无数据 → 不调 LLM,直接 stub。"""
|
||||
from src.llm.agents.base import AgentSpec, run_agent
|
||||
|
||||
async def empty_slice(header, before=None):
|
||||
return "── 伤停 ──\n 无数据"
|
||||
|
||||
spec = AgentSpec(name="injuries", system_prompt="s", slice_fn=empty_slice)
|
||||
header = self._make_header()
|
||||
|
||||
class ExplodingProvider:
|
||||
async def chat(self, *a, **kw):
|
||||
raise AssertionError("LLM 不应被调用")
|
||||
|
||||
r = await run_agent(spec, header, ExplodingProvider(), version="v1")
|
||||
assert r.status == "no_data"
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_llm_error_fail_open(self):
|
||||
"""LLM 调用失败 → status=error,不抛异常。"""
|
||||
from src.llm.agents.base import AgentSpec, run_agent
|
||||
|
||||
async def good_slice(header, before=None):
|
||||
return "── 交锋 ──\n 2026-03: A 2-1 B"
|
||||
|
||||
spec = AgentSpec(name="h2h", system_prompt="s", slice_fn=good_slice)
|
||||
header = self._make_header()
|
||||
|
||||
class FailProvider:
|
||||
async def chat(self, *a, **kw):
|
||||
from src.llm.provider import LLMResponse
|
||||
return LLMResponse(content="", error="timeout")
|
||||
|
||||
r = await run_agent(spec, header, FailProvider(), version="v1")
|
||||
assert r.status == "error"
|
||||
assert "LLM 调用失败" in r.analysis
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_success_path(self):
|
||||
"""正常路径: 切片 → LLM → 解析。"""
|
||||
from src.llm.agents.base import AgentSpec, run_agent
|
||||
from src.llm.provider import LLMResponse
|
||||
|
||||
async def good_slice(header, before=None):
|
||||
return "── 交锋 ──\n 2026-03: A 2-1 B"
|
||||
|
||||
spec = AgentSpec(name="h2h", system_prompt="s", slice_fn=good_slice)
|
||||
header = self._make_header()
|
||||
|
||||
class OkProvider:
|
||||
model = "test-model"
|
||||
async def chat(self, system, user, **kw):
|
||||
assert "{{context}}" not in user # 模板已渲染
|
||||
assert "A 2-1 B" in user
|
||||
return LLMResponse(
|
||||
content="{}",
|
||||
parsed={"data_sufficiency": "high", "analysis": "ok", "home_edge": 0.5, "confidence": 0.9},
|
||||
prompt_tokens=10, completion_tokens=5, latency_ms=100,
|
||||
)
|
||||
|
||||
r = await run_agent(spec, header, OkProvider(), version="v1")
|
||||
assert r.status == "ok"
|
||||
assert r.home_edge == 0.5
|
||||
assert r.model == "test-model"
|
||||
|
||||
def _make_header(self) -> MatchHeader:
|
||||
return MatchHeader(
|
||||
match_id=1, home_name="A", away_name="B", league_name="PL",
|
||||
season="2026-2027", match_date="2026-09-15", match_dt=None,
|
||||
stage=None, home_team_id=10, away_team_id=20, league_id=1,
|
||||
)
|
||||
|
||||
|
||||
class TestOrchestratorAggregation:
|
||||
"""终裁输入拼装逻辑。"""
|
||||
|
||||
def test_reports_to_json(self):
|
||||
from src.llm.agents.orchestrator import _reports_to_json
|
||||
import json
|
||||
|
||||
reports = [
|
||||
AgentReport(agent="h2h", status="ok", home_edge=0.5, confidence=0.8, analysis="a"),
|
||||
AgentReport(agent="injuries", status="no_data", data_sufficiency="none"),
|
||||
]
|
||||
text = _reports_to_json(reports)
|
||||
data = json.loads(text)
|
||||
assert len(data) == 2
|
||||
assert data[0]["agent"] == "h2h"
|
||||
assert data[1]["status"] == "no_data"
|
||||
|
||||
def test_aggregator_prompt_renders(self):
|
||||
"""终裁 prompt 模板两占位符都能渲染。"""
|
||||
tpl = load_agent_prompt("aggregator", "v1")
|
||||
rendered = (
|
||||
tpl
|
||||
.replace("{{match_header}}", "对阵: A vs B")
|
||||
.replace("{{agent_reports}}", '[{"agent": "h2h"}]')
|
||||
)
|
||||
assert "{{match_header}}" not in rendered
|
||||
assert "{{agent_reports}}" not in rendered
|
||||
@@ -0,0 +1,112 @@
|
||||
"""测试核心路径。"""
|
||||
import pytest
|
||||
from datetime import datetime, timezone
|
||||
from src.data.normalize import NormalizedMatch, normalize_bzzoiro, _parse_date, _to_int, _to_float, derive_season_label
|
||||
|
||||
|
||||
class TestNormalize:
|
||||
def test_parse_date_iso(self):
|
||||
assert _parse_date("2026-09-15T15:00:00Z") is not None
|
||||
|
||||
def test_parse_date_bare(self):
|
||||
assert _parse_date("2026-09-15") is not None
|
||||
|
||||
def test_parse_date_none(self):
|
||||
assert _parse_date(None) is None
|
||||
assert _parse_date("") is None
|
||||
|
||||
def test_to_int_strict(self):
|
||||
assert _to_int("2") == 2
|
||||
assert _to_int("2.0") == 2
|
||||
assert _to_int("2.8") is None # 拒绝非整数值
|
||||
assert _to_int(None) is None
|
||||
assert _to_int("-") is None
|
||||
|
||||
def test_to_float(self):
|
||||
assert _to_float("1.5") == 1.5
|
||||
assert _to_float(None) is None
|
||||
|
||||
def test_normalize_bzzoiro_finished(self):
|
||||
raw = {
|
||||
"event_date": "2026-09-15T15:00:00Z",
|
||||
"status": "finished",
|
||||
"home_team": "Man City",
|
||||
"away_team": "Man United",
|
||||
"home_score": 2,
|
||||
"away_score": 1,
|
||||
"round_number": 5,
|
||||
}
|
||||
m = normalize_bzzoiro(raw, "E0")
|
||||
assert m is not None
|
||||
assert m.home_team == "Manchester City"
|
||||
assert m.away_team == "Manchester United"
|
||||
assert m.home_goals == 2
|
||||
assert m.away_goals == 1
|
||||
assert m.match_status == "finished"
|
||||
assert m.match_stage == "第 5 轮"
|
||||
|
||||
def test_normalize_bzzoiro_unknown_status(self):
|
||||
raw = {"event_date": "2026-09-15", "status": "weird", "home_team": "A", "away_team": "B"}
|
||||
assert normalize_bzzoiro(raw, "E0") is None
|
||||
|
||||
def test_normalized_match_validate_finished_no_score(self):
|
||||
m = NormalizedMatch(
|
||||
league_type="E0", date=_parse_date("2026-09-15"),
|
||||
home_team="A", away_team="B", match_status="finished",
|
||||
)
|
||||
with pytest.raises(ValueError, match="must have score"):
|
||||
m.validate()
|
||||
|
||||
def test_normalized_match_validate_goals_range(self):
|
||||
m = NormalizedMatch(
|
||||
league_type="E0", date=_parse_date("2026-09-15"),
|
||||
home_team="A", away_team="B", match_status="finished",
|
||||
home_goals=50, away_goals=0,
|
||||
)
|
||||
with pytest.raises(ValueError, match="out of range"):
|
||||
m.validate()
|
||||
|
||||
|
||||
class TestSeasonLabel:
|
||||
def test_august_is_new_season(self):
|
||||
# 8 月属于新赛季
|
||||
d = datetime(2026, 8, 15, tzinfo=timezone.utc)
|
||||
assert derive_season_label(d) == "2026-2027"
|
||||
|
||||
def test_july_is_old_season(self):
|
||||
# 7 月属于上一赛季
|
||||
d = datetime(2026, 7, 15, tzinfo=timezone.utc)
|
||||
assert derive_season_label(d) == "2025-2026"
|
||||
|
||||
def test_january_is_old_season(self):
|
||||
d = datetime(2026, 1, 15, tzinfo=timezone.utc)
|
||||
assert derive_season_label(d) == "2025-2026"
|
||||
|
||||
|
||||
class TestProviderMock:
|
||||
"""用 mock 测试 LLM provider 的解析逻辑。"""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_provider_parses_json(self, monkeypatch):
|
||||
from src.llm.provider import LLMProvider
|
||||
|
||||
async def fake_post(*args, **kwargs):
|
||||
class FakeResp:
|
||||
status_code = 200
|
||||
def raise_for_status(self): pass
|
||||
def json(self):
|
||||
return {
|
||||
"choices": [{"message": {"content": '{"pred_home_goals": 1.5, "pred_away_goals": 1.0, "1x2": "1", "confidence": 0.7, "reasoning": "test"}'}}],
|
||||
"usage": {"prompt_tokens": 100, "completion_tokens": 50},
|
||||
}
|
||||
return FakeResp()
|
||||
|
||||
import httpx
|
||||
monkeypatch.setattr(httpx.AsyncClient, "post", fake_post)
|
||||
|
||||
p = LLMProvider(api_key="test", model="gpt-4o")
|
||||
resp = await p.chat("sys", "user", json_mode=True)
|
||||
assert resp.error is None
|
||||
assert resp.parsed is not None
|
||||
assert resp.parsed["pred_home_goals"] == 1.5
|
||||
assert resp.parsed["1x2"] == "1"
|
||||
Reference in New Issue
Block a user