Hermes CI Fix
242f3cf5ec
R3 期间口径（仅 YYYY-MM）延伸到取数层：新增 alert_entity_scope.month_period_clause()（跨库可移植 SQL 等价式），应用到 15 个文件 26 处「按 period 倒序取最新值/最近N期」位点；dashboard 原 MySQL-only REGEXP 正则收口到同单一事实源。 GET /api/cma/subjects 响应增补 entity_id（NULL=全局标准准则科目 / N=账套私有明细），既有 10 字段语义与名称不变。 证据：pending-tasks/_reports/yanxue-20260924-cma-fetch-caliber-fixes-fullstack.md（改前/改后 A/B：F_NET_PROFIT 2026-H2/-55.0 → 2026-08/-10.46）。新增 7 例回归测试；全量 1 failed(预存在) 1473 passed，基线同项。
360 lines
13 KiB
Python
360 lines
13 KiB
Python
"""知识摘要服务 — 管理会计OS持久记忆
|
||
|
||
仿 OpenCode SummarizeProvider 的持久记忆机制，做三处改进:
|
||
1. 分层压缩：日→周→月→全部，不是一次性全量压缩
|
||
2. 结构化存储：MySQL 关系表，不是 SQLite JSON 消息
|
||
3. 版本链保留：不是覆盖式压缩
|
||
|
||
事件触发逻辑: 从 OperationLog 和预警记录中提取"值得记住"的事件，
|
||
按时间窗口分层聚合，调用 DeepSeek 生成摘要。
|
||
"""
|
||
|
||
import logging
|
||
import os
|
||
import json
|
||
import httpx
|
||
from datetime import datetime, date, timedelta
|
||
from typing import Optional
|
||
|
||
from sqlalchemy.orm import Session
|
||
from sqlalchemy import func, and_
|
||
|
||
from app.models import KnowledgeEvent, KnowledgeSummary, OperationLog, KPIAlert, KPIDefinition, KPIValue, UserEntity
|
||
from app.utils.alert_entity_scope import month_period_clause
|
||
|
||
logger = logging.getLogger("cma.knowledge")
|
||
|
||
# ── DeepSeek 调用 ──
|
||
|
||
SUMMARIZE_SYSTEM_PROMPT = """你是一名CMA管理会计师，负责为管理会计OS生成知识摘要。
|
||
你的工作是：审核一组经营事件记录，提炼出"必须记住"的核心信息。
|
||
|
||
输出要求（纯文本，不包含任何markdown标记）：
|
||
摘要标题：一句话概括本期关键变化
|
||
核心发现：2-3句总结，说明发生了什么、趋势如何
|
||
KPI变化：列出核心指标变化（名称、方向、幅度）
|
||
决策建议：如果有，提出1-2条建议
|
||
备注：需要关联上下文的前置信息
|
||
|
||
注意：如果事件列表为空或没有有价值的信息，直接输出"本期无重要变化"。"""
|
||
|
||
|
||
async def _call_deepseek(prompt: str, timeout: int = 30) -> str:
|
||
"""调用DeepSeek API生成摘要"""
|
||
api_key = os.getenv("DEEPSEEK_API_KEY", "sk-8e2...c2e8")
|
||
try:
|
||
async with httpx.AsyncClient(timeout=timeout) as client:
|
||
resp = await client.post(
|
||
"https://api.deepseek.com/v1/chat/completions",
|
||
headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
|
||
json={
|
||
"model": "deepseek-chat",
|
||
"messages": [
|
||
{"role": "system", "content": SUMMARIZE_SYSTEM_PROMPT},
|
||
{"role": "user", "content": prompt},
|
||
],
|
||
"stream": False,
|
||
"temperature": 0.2,
|
||
"max_tokens": 1024,
|
||
},
|
||
)
|
||
data = resp.json()
|
||
return data.get("choices", [{}])[0].get("message", {}).get("content", "")
|
||
except Exception as e:
|
||
logger.error(f"DeepSeek摘要调用失败: {e}")
|
||
return ""
|
||
|
||
|
||
# ── 事件抽取 ──
|
||
|
||
def extract_events(db: Session, since: datetime, until: Optional[datetime] = None, entity_id: int = 1) -> list[dict]:
|
||
"""从 OperationLog + KPIAlert 中抽取关键事件（按企业抽屉隔离）
|
||
返回 dict 列表，用于喂给摘要 prompt
|
||
|
||
隔离说明（2026-09-03 多租户分区）：
|
||
- KPIAlert 带 entity_id 列 → 直接按 entity_id 过滤
|
||
- OperationLog 无 entity_id 列（数据债，P2 follow-up）→ 经 user_id→user_entities
|
||
归属当前企业；无授权归属的日志（user_id NULL/未授权）不进入本企业记忆
|
||
"""
|
||
now = until or datetime.utcnow()
|
||
|
||
# 0. 当前企业授权用户集合（OperationLog 归属过滤用）
|
||
auth_user_ids = [
|
||
uid for (uid,) in db.query(UserEntity.user_id)
|
||
.filter(UserEntity.entity_id == entity_id).all()
|
||
]
|
||
|
||
# 1. 操作日志 → 事件（本企业授权用户的操作）
|
||
logs = (
|
||
db.query(OperationLog)
|
||
.filter(
|
||
OperationLog.created_at >= since,
|
||
OperationLog.created_at <= now,
|
||
OperationLog.user_id.in_(auth_user_ids), # 空集合 → IN (NULL) 不匹配任何行
|
||
)
|
||
.order_by(OperationLog.created_at.asc())
|
||
.all()
|
||
)
|
||
|
||
events = []
|
||
for log in logs:
|
||
detail_str = ""
|
||
if log.detail:
|
||
# 裁短 detail 避免 token 浪费
|
||
d = json.dumps(log.detail, ensure_ascii=False)
|
||
detail_str = d[:300] + ("..." if len(d) > 300 else "")
|
||
|
||
events.append({
|
||
"type": "action",
|
||
"time": log.created_at.isoformat() if log.created_at else "",
|
||
"action": log.action,
|
||
"target": f"{log.target_type}#{log.target_id}",
|
||
"detail": detail_str,
|
||
})
|
||
|
||
# 2. 预警记录 → 事件（KPIAlert 直接按企业过滤）
|
||
alerts = (
|
||
db.query(KPIAlert, KPIDefinition.kpi_name, KPIDefinition.kpi_code)
|
||
.join(KPIDefinition, KPIAlert.kpi_id == KPIDefinition.id)
|
||
.filter(
|
||
KPIAlert.created_at >= since,
|
||
KPIAlert.created_at <= now,
|
||
KPIAlert.entity_id == entity_id,
|
||
)
|
||
.order_by(KPIAlert.created_at.asc())
|
||
.all()
|
||
)
|
||
|
||
for alert, kpi_name, kpi_code in alerts:
|
||
events.append({
|
||
"type": "alert",
|
||
"time": alert.created_at.isoformat() if alert.created_at else "",
|
||
"level": alert.alert_level,
|
||
"kpi": f"{kpi_code} ({kpi_name})",
|
||
"message": (alert.alert_message or "")[:200],
|
||
"status": alert.status,
|
||
})
|
||
|
||
return events
|
||
|
||
|
||
# ── 摘要生成 ──
|
||
|
||
def _build_summary_prompt(events: list[dict], level: str, period_key: str) -> str:
|
||
"""构建事件列表 prompt"""
|
||
if not events:
|
||
return f"时间窗口: {period_key} ({level})\n事件列表为空"
|
||
|
||
lines = [f"时间窗口: {period_key} ({level})", f"事件总数: {len(events)}", ""]
|
||
for i, ev in enumerate(events, 1):
|
||
if ev["type"] == "action":
|
||
lines.append(f"{i}. [操作] {ev['time']} {ev['action']} on {ev['target']} | {ev['detail']}")
|
||
elif ev["type"] == "alert":
|
||
lines.append(f"{i}. [预警] {ev['time']} [{ev['level']}] {ev['kpi']} | {ev['message']} (状态: {ev['status']})")
|
||
else:
|
||
lines.append(f"{i}. [其他] {ev['time']} {ev}")
|
||
return "\n".join(lines)
|
||
|
||
|
||
async def _kpi_snapshot(db: Session, entity_id: int = 1) -> list[dict]:
|
||
"""当前KPI快照 — 本企业每个活跃KPI的最新实际值（按账套隔离）"""
|
||
kpis = db.query(KPIDefinition).filter(
|
||
KPIDefinition.status == "active",
|
||
KPIDefinition.entity_id == entity_id,
|
||
).all()
|
||
snapshot = []
|
||
for k in kpis:
|
||
latest = (
|
||
db.query(KPIValue)
|
||
.filter(KPIValue.kpi_id == k.id, month_period_clause(KPIValue.period))
|
||
.order_by(KPIValue.period.desc())
|
||
.first()
|
||
)
|
||
if latest:
|
||
snapshot.append({
|
||
"code": k.kpi_code,
|
||
"name": k.kpi_name,
|
||
"value": latest.actual_value,
|
||
"period": latest.period,
|
||
"unit": k.unit,
|
||
})
|
||
return snapshot
|
||
|
||
|
||
async def generate_summary(
|
||
db: Session,
|
||
level: str,
|
||
period_key: str,
|
||
since: datetime,
|
||
until: Optional[datetime] = None,
|
||
prev_summary: Optional[KnowledgeSummary] = None,
|
||
entity_id: int = 1,
|
||
) -> KnowledgeSummary:
|
||
"""生成一层摘要（daily/weekly/monthly/cumulative）
|
||
|
||
Args:
|
||
level: daily / weekly / monthly / cumulative
|
||
period_key: 期间标识，如 "2026-06-12" / "2026-W24" / "2026-06" / "cumulative"
|
||
since: 事件开始时间
|
||
until: 事件结束时间
|
||
prev_summary: 前一层摘要（用于累积摘要继承）
|
||
entity_id: 企业账套（企业记忆分区 2026-09-03）
|
||
"""
|
||
now = until or datetime.utcnow()
|
||
|
||
# 1. 抽取事件（本企业）
|
||
events = extract_events(db, since, now, entity_id=entity_id)
|
||
|
||
# 2. 构建 prompt
|
||
prompt = _build_summary_prompt(events, level, period_key)
|
||
|
||
# 3. 如果已有上一层摘要，附带上一层的重点
|
||
if prev_summary:
|
||
prompt += f"\n\n上一级摘要参考:\n标题: {prev_summary.title}\n内容: {prev_summary.content[:500]}\n"
|
||
|
||
# 4. 调用 DeepSeek
|
||
result_text = await _call_deepseek(prompt)
|
||
|
||
# 5. 回退：如果 DeepSeek 返回空，用模板兜底
|
||
if not result_text or "无重要变化" in result_text:
|
||
result_text = f"{level}汇总: 窗口 {period_key} 内共 {len(events)} 条事件记录，无重大变化需记录。"
|
||
|
||
# 6. 提取 KPI 快照（本企业）
|
||
kpi_snapshot = await _kpi_snapshot(db, entity_id=entity_id)
|
||
|
||
# 7. 存入数据库
|
||
summary = KnowledgeSummary(
|
||
entity_id=entity_id,
|
||
level=level,
|
||
period_key=period_key,
|
||
title=f"{level.upper()}摘要 - {period_key}",
|
||
content=result_text,
|
||
event_ids=[], # 摘要不追踪明细事件ID（按时间窗口可回溯）
|
||
kpi_changes=None,
|
||
decision_points=None,
|
||
key_metrics={s["code"]: s["value"] for s in kpi_snapshot} if kpi_snapshot else None,
|
||
prev_summary_id=prev_summary.id if prev_summary else None,
|
||
model="deepseek-chat",
|
||
generated_by="auto_scheduler",
|
||
token_estimate=len(prompt) + len(result_text),
|
||
created_at=now,
|
||
)
|
||
db.add(summary)
|
||
db.commit()
|
||
db.refresh(summary)
|
||
|
||
logger.info(f"知识摘要已生成: level={level} period={period_key} id={summary.id}")
|
||
return summary
|
||
|
||
|
||
# ── 分层调度 ──
|
||
|
||
def get_last_summary(db: Session, level: str, entity_id: int = 1) -> Optional[KnowledgeSummary]:
|
||
"""获取该层最新（最新创建）的摘要（按企业隔离）"""
|
||
return (
|
||
db.query(KnowledgeSummary)
|
||
.filter(
|
||
KnowledgeSummary.level == level,
|
||
KnowledgeSummary.is_stale == 0,
|
||
KnowledgeSummary.entity_id == entity_id,
|
||
)
|
||
.order_by(KnowledgeSummary.id.desc())
|
||
.first()
|
||
)
|
||
|
||
|
||
async def run_daily_summary(db: Session, entity_id: int = 1) -> KnowledgeSummary:
|
||
"""运行每日摘要（企业记忆分区）"""
|
||
today = date.today()
|
||
period_key = today.isoformat()
|
||
since = datetime(today.year, today.month, today.day)
|
||
|
||
# 获取前一天摘要作为 prev
|
||
yesterday = today - timedelta(days=1)
|
||
prev = get_last_summary(db, "daily", entity_id=entity_id)
|
||
|
||
return await generate_summary(
|
||
db, "daily", period_key, since, prev_summary=prev, entity_id=entity_id,
|
||
)
|
||
|
||
|
||
async def run_weekly_summary(db: Session, entity_id: int = 1) -> KnowledgeSummary:
|
||
"""运行每周摘要（周日执行）"""
|
||
today = date.today()
|
||
# ISO 周算法: 本周一到今天
|
||
iso_week = today.isocalendar()
|
||
period_key = f"{iso_week[0]}-W{iso_week[1]:02d}"
|
||
since = today - timedelta(days=today.weekday()) # 本周一
|
||
since_dt = datetime(since.year, since.month, since.day)
|
||
|
||
prev = get_last_summary(db, "weekly", entity_id=entity_id)
|
||
return await generate_summary(
|
||
db, "weekly", period_key, since_dt, prev_summary=prev, entity_id=entity_id,
|
||
)
|
||
|
||
|
||
async def run_monthly_summary(db: Session, entity_id: int = 1) -> KnowledgeSummary:
|
||
"""运行月度摘要"""
|
||
today = date.today()
|
||
period_key = today.strftime("%Y-%m")
|
||
since = datetime(today.year, today.month, 1)
|
||
|
||
# 附属前一个月的 daily 和 weekly 摘要
|
||
prev_month = today.replace(day=1) - timedelta(days=1)
|
||
prev = get_last_summary(db, "monthly", entity_id=entity_id)
|
||
|
||
return await generate_summary(
|
||
db, "monthly", period_key, since, prev_summary=prev, entity_id=entity_id,
|
||
)
|
||
|
||
|
||
# ── 对外接口（同步包装，供手动触发使用） ──
|
||
|
||
import asyncio
|
||
|
||
|
||
def generate_summary_sync(
|
||
db: Session,
|
||
level: str,
|
||
period_key: str,
|
||
since: datetime,
|
||
until: Optional[datetime] = None,
|
||
entity_id: int = 1,
|
||
) -> dict:
|
||
"""同步包装，用于 API 手动触发"""
|
||
loop = asyncio.new_event_loop()
|
||
try:
|
||
summary = loop.run_until_complete(
|
||
generate_summary(db, level, period_key, since, until, entity_id=entity_id)
|
||
)
|
||
return {"id": summary.id, "entity_id": summary.entity_id, "level": summary.level, "period_key": summary.period_key}
|
||
finally:
|
||
loop.close()
|
||
|
||
|
||
def generate_daily_sync(db: Session, entity_id: int = 1) -> dict:
|
||
loop = asyncio.new_event_loop()
|
||
try:
|
||
summary = loop.run_until_complete(run_daily_summary(db, entity_id))
|
||
return {"id": summary.id, "entity_id": summary.entity_id, "level": summary.level, "period_key": summary.period_key}
|
||
finally:
|
||
loop.close()
|
||
|
||
|
||
def generate_weekly_sync(db: Session, entity_id: int = 1) -> dict:
|
||
loop = asyncio.new_event_loop()
|
||
try:
|
||
summary = loop.run_until_complete(run_weekly_summary(db, entity_id))
|
||
return {"id": summary.id, "entity_id": summary.entity_id, "level": summary.level, "period_key": summary.period_key}
|
||
finally:
|
||
loop.close()
|
||
|
||
|
||
def generate_monthly_sync(db: Session, entity_id: int = 1) -> dict:
|
||
loop = asyncio.new_event_loop()
|
||
try:
|
||
summary = loop.run_until_complete(run_monthly_summary(db, entity_id))
|
||
return {"id": summary.id, "entity_id": summary.entity_id, "level": summary.level, "period_key": summary.period_key}
|
||
finally:
|
||
loop.close()
|