chore: normalize line endings to LF via .gitattributes
Root-cause fix for mixed CRLF/LF in the repo (no .gitattributes previously). Adds '* text=auto eol=lf' and explicit binary rules so all future checkouts are LF. Renormalizes all tracked text files (no logical content change). Backup tags: image qihuaiflow:backup-lineend-20260725-121632, git backup-lineend-pre-20260725-121632.
This commit is contained in:
QihuAI Design committed
1 parent
2a7bb875f9
commit
771c8456b8
39 files changed
+34810
-34777
No files matched your search
@@ -1,179 +1,179 @@
|
||||
"""
|
||||
AnySearch 联网检索客户端
|
||||
======================
|
||||
调用 AnySearch 网页/领域搜索 API (https://api.anysearch.com/v1/search)，
|
||||
为 AI 聊天注入实时联网检索结果，亦可被后端端点 / Agent 工具复用。
|
||||
|
||||
约束：平台容器存在智能路由代理，对外部 API 偶发重置；本客户端使用
|
||||
trust_env=False 直连 (等价于 NO_PROXY=api.anysearch.com)，规避代理重置。
|
||||
失败一律返回空/降级结果，保证聊天链路不中断。
|
||||
"""
|
||||
import json
|
||||
import logging
|
||||
from typing import List, Dict, Optional
|
||||
|
||||
logger = logging.getLogger("aiflow.anysearch")
|
||||
|
||||
DEFAULT_API_URL = "https://api.anysearch.com/v1/search"
|
||||
DEFAULT_ZONE = "cn"
|
||||
REQUEST_TIMEOUT = 8.0
|
||||
|
||||
|
||||
async def anysearch_search(
|
||||
api_key: str,
|
||||
api_url: str = DEFAULT_API_URL,
|
||||
query: str = "",
|
||||
top_k: int = 5,
|
||||
zone: str = DEFAULT_ZONE,
|
||||
freshness: Optional[str] = None,
|
||||
timeout: float = REQUEST_TIMEOUT,
|
||||
) -> List[Dict]:
|
||||
"""调用 AnySearch 搜索，返回结构化结果列表。
|
||||
|
||||
每个结果:
|
||||
{
|
||||
"content": snippet 摘要文本,
|
||||
"score": 0.9..0.5 递减,
|
||||
"source": "web_anysearch",
|
||||
"metadata": {"url", "title", "source_domain", "zone"}
|
||||
}
|
||||
失败时返回 [] (不抛异常)。
|
||||
"""
|
||||
if not api_key:
|
||||
return []
|
||||
try:
|
||||
import httpx
|
||||
except ImportError:
|
||||
logger.warning("httpx not available for AnySearch")
|
||||
return []
|
||||
|
||||
payload = {
|
||||
"query": query,
|
||||
"zone": zone,
|
||||
"top_k": max(1, min(int(top_k), 20)),
|
||||
}
|
||||
if freshness:
|
||||
payload["freshness"] = freshness
|
||||
|
||||
headers = {
|
||||
"Content-Type": "application/json",
|
||||
"Authorization": f"Bearer {api_key}",
|
||||
}
|
||||
try:
|
||||
# trust_env=False -> 忽略容器内 HTTP(S)_PROXY，直连 api.anysearch.com
|
||||
async with httpx.AsyncClient(trust_env=False, timeout=timeout) as client:
|
||||
resp = await client.post(api_url, json=payload, headers=headers)
|
||||
if resp.status_code != 200:
|
||||
logger.warning(f"AnySearch HTTP {resp.status_code}: {resp.text[:200]}")
|
||||
return []
|
||||
data = resp.json()
|
||||
if data.get("code") != 0:
|
||||
logger.warning(f"AnySearch biz error: {data.get('message')}")
|
||||
return []
|
||||
results = (data.get("data") or {}).get("results") or []
|
||||
hits = []
|
||||
for idx, r in enumerate(results[:top_k]):
|
||||
title = r.get("title", "") or ""
|
||||
url = r.get("url", "") or ""
|
||||
snippet = r.get("snippet") or r.get("content") or ""
|
||||
if not snippet and not title:
|
||||
continue
|
||||
try:
|
||||
from urllib.parse import urlparse
|
||||
domain = urlparse(url).netloc
|
||||
except Exception:
|
||||
domain = ""
|
||||
hits.append({
|
||||
"content": snippet,
|
||||
"score": max(0.4, 0.9 - idx * 0.08),
|
||||
"source": "web_anysearch",
|
||||
"metadata": {
|
||||
"url": url,
|
||||
"title": title,
|
||||
"source_domain": domain,
|
||||
"zone": zone,
|
||||
},
|
||||
})
|
||||
return hits
|
||||
except Exception as e:
|
||||
logger.warning(f"AnySearch search failed: {e}")
|
||||
return []
|
||||
|
||||
|
||||
async def anysearch_health(api_key: str, api_url: str = DEFAULT_API_URL, timeout: float = REQUEST_TIMEOUT) -> Dict:
|
||||
"""连通性探测：返回 {"status": "connected"/"disconnected", "detail": ...}。"""
|
||||
if not api_key:
|
||||
return {"status": "disconnected", "detail": "未配置 API Key", "connected": False, "usable": False}
|
||||
try:
|
||||
import httpx
|
||||
async with httpx.AsyncClient(trust_env=False, timeout=timeout) as client:
|
||||
resp = await client.post(
|
||||
api_url,
|
||||
json={"query": "ping", "zone": DEFAULT_ZONE, "top_k": 1},
|
||||
headers={"Content-Type": "application/json", "Authorization": f"Bearer {api_key}"},
|
||||
)
|
||||
if resp.status_code == 200:
|
||||
return {"status": "connected", "detail": "AnySearch 连接正常", "connected": True, "usable": True}
|
||||
return {"status": "error", "detail": f"HTTP {resp.status_code}", "connected": False, "usable": False}
|
||||
except Exception as e:
|
||||
return {"status": "error", "detail": str(e)[:200], "connected": False, "usable": False}
|
||||
|
||||
|
||||
def load_anysearch_config(db_session) -> Dict:
|
||||
"""从 site_settings 读取全局 AnySearch 配置（失败返回空字典）。"""
|
||||
if db_session is None:
|
||||
return {}
|
||||
try:
|
||||
from app.models import SiteSetting
|
||||
row = db_session.query(SiteSetting).filter(SiteSetting.key == "anysearch_config").first()
|
||||
if not row or not row.value:
|
||||
return {}
|
||||
cfg = json.loads(row.value)
|
||||
if not isinstance(cfg, dict):
|
||||
return {}
|
||||
return cfg
|
||||
except Exception as e:
|
||||
logger.warning(f"load_anysearch_config failed: {e}")
|
||||
return {}
|
||||
|
||||
|
||||
def save_anysearch_config(
|
||||
db_session,
|
||||
api_key: str,
|
||||
is_enabled: bool = True,
|
||||
api_url: str = DEFAULT_API_URL,
|
||||
zone: str = DEFAULT_ZONE,
|
||||
freshness: Optional[str] = None,
|
||||
top_k: int = 5,
|
||||
) -> bool:
|
||||
"""保存/更新全局 AnySearch 配置到 site_settings。"""
|
||||
if db_session is None:
|
||||
return False
|
||||
try:
|
||||
from app.models import SiteSetting
|
||||
from datetime import datetime
|
||||
cfg = {
|
||||
"api_key": api_key.strip(),
|
||||
"is_enabled": bool(is_enabled),
|
||||
"api_url": api_url.strip() or DEFAULT_API_URL,
|
||||
"zone": zone or DEFAULT_ZONE,
|
||||
"freshness": freshness,
|
||||
"top_k": max(1, min(int(top_k), 20)),
|
||||
}
|
||||
row = db_session.query(SiteSetting).filter(SiteSetting.key == "anysearch_config").first()
|
||||
if row:
|
||||
row.value = json.dumps(cfg, ensure_ascii=False)
|
||||
row.updated_at = datetime.now()
|
||||
else:
|
||||
from app.models import gen_uuid
|
||||
row = SiteSetting(id=gen_uuid(), key="anysearch_config", value=json.dumps(cfg, ensure_ascii=False))
|
||||
db_session.add(row)
|
||||
db_session.commit()
|
||||
return True
|
||||
except Exception as e:
|
||||
logger.warning(f"save_anysearch_config failed: {e}")
|
||||
try:
|
||||
db_session.rollback()
|
||||
except Exception:
|
||||
pass
|
||||
return False
|
||||
"""
|
||||
AnySearch 联网检索客户端
|
||||
======================
|
||||
调用 AnySearch 网页/领域搜索 API (https://api.anysearch.com/v1/search)，
|
||||
为 AI 聊天注入实时联网检索结果，亦可被后端端点 / Agent 工具复用。
|
||||
|
||||
约束：平台容器存在智能路由代理，对外部 API 偶发重置；本客户端使用
|
||||
trust_env=False 直连 (等价于 NO_PROXY=api.anysearch.com)，规避代理重置。
|
||||
失败一律返回空/降级结果，保证聊天链路不中断。
|
||||
"""
|
||||
import json
|
||||
import logging
|
||||
from typing import List, Dict, Optional
|
||||
|
||||
logger = logging.getLogger("aiflow.anysearch")
|
||||
|
||||
DEFAULT_API_URL = "https://api.anysearch.com/v1/search"
|
||||
DEFAULT_ZONE = "cn"
|
||||
REQUEST_TIMEOUT = 8.0
|
||||
|
||||
|
||||
async def anysearch_search(
|
||||
api_key: str,
|
||||
api_url: str = DEFAULT_API_URL,
|
||||
query: str = "",
|
||||
top_k: int = 5,
|
||||
zone: str = DEFAULT_ZONE,
|
||||
freshness: Optional[str] = None,
|
||||
timeout: float = REQUEST_TIMEOUT,
|
||||
) -> List[Dict]:
|
||||
"""调用 AnySearch 搜索，返回结构化结果列表。
|
||||
|
||||
每个结果:
|
||||
{
|
||||
"content": snippet 摘要文本,
|
||||
"score": 0.9..0.5 递减,
|
||||
"source": "web_anysearch",
|
||||
"metadata": {"url", "title", "source_domain", "zone"}
|
||||
}
|
||||
失败时返回 [] (不抛异常)。
|
||||
"""
|
||||
if not api_key:
|
||||
return []
|
||||
try:
|
||||
import httpx
|
||||
except ImportError:
|
||||
logger.warning("httpx not available for AnySearch")
|
||||
return []
|
||||
|
||||
payload = {
|
||||
"query": query,
|
||||
"zone": zone,
|
||||
"top_k": max(1, min(int(top_k), 20)),
|
||||
}
|
||||
if freshness:
|
||||
payload["freshness"] = freshness
|
||||
|
||||
headers = {
|
||||
"Content-Type": "application/json",
|
||||
"Authorization": f"Bearer {api_key}",
|
||||
}
|
||||
try:
|
||||
# trust_env=False -> 忽略容器内 HTTP(S)_PROXY，直连 api.anysearch.com
|
||||
async with httpx.AsyncClient(trust_env=False, timeout=timeout) as client:
|
||||
resp = await client.post(api_url, json=payload, headers=headers)
|
||||
if resp.status_code != 200:
|
||||
logger.warning(f"AnySearch HTTP {resp.status_code}: {resp.text[:200]}")
|
||||
return []
|
||||
data = resp.json()
|
||||
if data.get("code") != 0:
|
||||
logger.warning(f"AnySearch biz error: {data.get('message')}")
|
||||
return []
|
||||
results = (data.get("data") or {}).get("results") or []
|
||||
hits = []
|
||||
for idx, r in enumerate(results[:top_k]):
|
||||
title = r.get("title", "") or ""
|
||||
url = r.get("url", "") or ""
|
||||
snippet = r.get("snippet") or r.get("content") or ""
|
||||
if not snippet and not title:
|
||||
continue
|
||||
try:
|
||||
from urllib.parse import urlparse
|
||||
domain = urlparse(url).netloc
|
||||
except Exception:
|
||||
domain = ""
|
||||
hits.append({
|
||||
"content": snippet,
|
||||
"score": max(0.4, 0.9 - idx * 0.08),
|
||||
"source": "web_anysearch",
|
||||
"metadata": {
|
||||
"url": url,
|
||||
"title": title,
|
||||
"source_domain": domain,
|
||||
"zone": zone,
|
||||
},
|
||||
})
|
||||
return hits
|
||||
except Exception as e:
|
||||
logger.warning(f"AnySearch search failed: {e}")
|
||||
return []
|
||||
|
||||
|
||||
async def anysearch_health(api_key: str, api_url: str = DEFAULT_API_URL, timeout: float = REQUEST_TIMEOUT) -> Dict:
|
||||
"""连通性探测：返回 {"status": "connected"/"disconnected", "detail": ...}。"""
|
||||
if not api_key:
|
||||
return {"status": "disconnected", "detail": "未配置 API Key", "connected": False, "usable": False}
|
||||
try:
|
||||
import httpx
|
||||
async with httpx.AsyncClient(trust_env=False, timeout=timeout) as client:
|
||||
resp = await client.post(
|
||||
api_url,
|
||||
json={"query": "ping", "zone": DEFAULT_ZONE, "top_k": 1},
|
||||
headers={"Content-Type": "application/json", "Authorization": f"Bearer {api_key}"},
|
||||
)
|
||||
if resp.status_code == 200:
|
||||
return {"status": "connected", "detail": "AnySearch 连接正常", "connected": True, "usable": True}
|
||||
return {"status": "error", "detail": f"HTTP {resp.status_code}", "connected": False, "usable": False}
|
||||
except Exception as e:
|
||||
return {"status": "error", "detail": str(e)[:200], "connected": False, "usable": False}
|
||||
|
||||
|
||||
def load_anysearch_config(db_session) -> Dict:
|
||||
"""从 site_settings 读取全局 AnySearch 配置（失败返回空字典）。"""
|
||||
if db_session is None:
|
||||
return {}
|
||||
try:
|
||||
from app.models import SiteSetting
|
||||
row = db_session.query(SiteSetting).filter(SiteSetting.key == "anysearch_config").first()
|
||||
if not row or not row.value:
|
||||
return {}
|
||||
cfg = json.loads(row.value)
|
||||
if not isinstance(cfg, dict):
|
||||
return {}
|
||||
return cfg
|
||||
except Exception as e:
|
||||
logger.warning(f"load_anysearch_config failed: {e}")
|
||||
return {}
|
||||
|
||||
|
||||
def save_anysearch_config(
|
||||
db_session,
|
||||
api_key: str,
|
||||
is_enabled: bool = True,
|
||||
api_url: str = DEFAULT_API_URL,
|
||||
zone: str = DEFAULT_ZONE,
|
||||
freshness: Optional[str] = None,
|
||||
top_k: int = 5,
|
||||
) -> bool:
|
||||
"""保存/更新全局 AnySearch 配置到 site_settings。"""
|
||||
if db_session is None:
|
||||
return False
|
||||
try:
|
||||
from app.models import SiteSetting
|
||||
from datetime import datetime
|
||||
cfg = {
|
||||
"api_key": api_key.strip(),
|
||||
"is_enabled": bool(is_enabled),
|
||||
"api_url": api_url.strip() or DEFAULT_API_URL,
|
||||
"zone": zone or DEFAULT_ZONE,
|
||||
"freshness": freshness,
|
||||
"top_k": max(1, min(int(top_k), 20)),
|
||||
}
|
||||
row = db_session.query(SiteSetting).filter(SiteSetting.key == "anysearch_config").first()
|
||||
if row:
|
||||
row.value = json.dumps(cfg, ensure_ascii=False)
|
||||
row.updated_at = datetime.now()
|
||||
else:
|
||||
from app.models import gen_uuid
|
||||
row = SiteSetting(id=gen_uuid(), key="anysearch_config", value=json.dumps(cfg, ensure_ascii=False))
|
||||
db_session.add(row)
|
||||
db_session.commit()
|
||||
return True
|
||||
except Exception as e:
|
||||
logger.warning(f"save_anysearch_config failed: {e}")
|
||||
try:
|
||||
db_session.rollback()
|
||||
except Exception:
|
||||
pass
|
||||
return False
|
||||
File diff suppressed because it is too large.
Load diff
File diff suppressed because it is too large.
Load diff
File diff suppressed because it is too large.
Load diff
File diff suppressed because it is too large.
Load diff
File diff suppressed because it is too large.
Load diff
+2905
-2905
File diff suppressed because it is too large.
Load diff
@@ -1,319 +1,319 @@
|
||||
# Auto-extracted from chat.py during Phase 1 refactor (Strangler Fig).
|
||||
# Source of truth: scripts/phase1_extract.py. Do not hand-merge back.
|
||||
|
||||
from fastapi import APIRouter, Depends, HTTPException, Query, Response
|
||||
from sqlalchemy.orm import Session
|
||||
from sqlalchemy import and_, desc, asc, func
|
||||
from pydantic import BaseModel, Field
|
||||
from typing import Optional
|
||||
import uuid
|
||||
from datetime import datetime, timezone, timedelta
|
||||
import re
|
||||
import json
|
||||
import asyncio
|
||||
import math
|
||||
import httpx
|
||||
import os
|
||||
import logging
|
||||
from ..database import get_db, SessionLocal
|
||||
from ..models import (
|
||||
|
||||
|
||||
|
||||
User, Conversation, Message, AIModel, ModelProvider, TokenUsageLog,
|
||||
|
||||
|
||||
|
||||
KnowledgeBase, KnowledgeDocument, KnowledgeDocumentChunk, EvolutionSkill,
|
||||
|
||||
|
||||
|
||||
EvolutionTrajectory, MarketApp, AppPurchase, AppInstallation,
|
||||
|
||||
|
||||
|
||||
MCPServer,
|
||||
|
||||
|
||||
|
||||
gen_uuid
|
||||
|
||||
|
||||
|
||||
)
|
||||
from ..auth import get_current_user
|
||||
from ..platform_knowledge import get_platform_knowledge
|
||||
from ..mcp_client import list_mcp_tools, call_mcp_tool
|
||||
from ..middleware.log_service import LogService
|
||||
from ..capability_registry import resolve_auto_model
|
||||
from app.api.chat_utils import (cst_now)
|
||||
|
||||
def _build_knowledge_graph(db: Session, kb: KnowledgeBase) -> dict:
|
||||
|
||||
|
||||
|
||||
"""从知识库文档构建知识图谱节点和边，返回 graph_data。"""
|
||||
|
||||
|
||||
|
||||
documents = db.query(KnowledgeDocument).filter(
|
||||
|
||||
|
||||
|
||||
and_(
|
||||
|
||||
|
||||
|
||||
KnowledgeDocument.knowledge_base_id == kb.id,
|
||||
|
||||
|
||||
|
||||
KnowledgeDocument.status == "completed",
|
||||
|
||||
|
||||
|
||||
)
|
||||
|
||||
|
||||
|
||||
).all()
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
graph_nodes = []
|
||||
|
||||
|
||||
|
||||
graph_edges = []
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
doc_entity_types = [
|
||||
|
||||
|
||||
|
||||
{"type": "concept", "label": "核心概念"},
|
||||
|
||||
|
||||
|
||||
{"type": "entity", "label": "关键实体"},
|
||||
|
||||
|
||||
|
||||
{"type": "attribute", "label": "属性特征"},
|
||||
|
||||
|
||||
|
||||
{"type": "relation", "label": "关联关系"},
|
||||
|
||||
|
||||
|
||||
]
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
for idx, doc in enumerate(documents):
|
||||
|
||||
|
||||
|
||||
doc_node_id = f"doc-{doc.id[:8]}"
|
||||
|
||||
|
||||
|
||||
graph_nodes.append({
|
||||
|
||||
|
||||
|
||||
"id": doc_node_id, "label": doc.filename, "type": "document", "size": 18, "group": "document",
|
||||
|
||||
|
||||
|
||||
})
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
filename_base = doc.filename.rsplit(".", 1)[0]
|
||||
|
||||
|
||||
|
||||
words = filename_base.replace("_", " ").replace("-", " ").split()
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
for wi, word in enumerate(words):
|
||||
|
||||
|
||||
|
||||
if len(word) >= 2:
|
||||
|
||||
|
||||
|
||||
entity_type = doc_entity_types[wi % len(doc_entity_types)]
|
||||
|
||||
|
||||
|
||||
node_id = f"ent-{doc.id[:8]}-{wi}"
|
||||
|
||||
|
||||
|
||||
graph_nodes.append({
|
||||
|
||||
|
||||
|
||||
"id": node_id, "label": word, "type": entity_type["type"],
|
||||
|
||||
|
||||
|
||||
"size": 10 + (wi % 8), "group": entity_type["type"],
|
||||
|
||||
|
||||
|
||||
})
|
||||
|
||||
|
||||
|
||||
graph_edges.append({
|
||||
|
||||
|
||||
|
||||
"source": doc_node_id, "target": node_id,
|
||||
|
||||
|
||||
|
||||
"label": entity_type["label"], "weight": 0.5 + (0.5 / (wi + 1)),
|
||||
|
||||
|
||||
|
||||
})
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
for i in range(len(documents)):
|
||||
|
||||
|
||||
|
||||
for j in range(i + 1, len(documents)):
|
||||
|
||||
|
||||
|
||||
fi = documents[i].filename.lower().rsplit(".", 1)[0]
|
||||
|
||||
|
||||
|
||||
fj = documents[j].filename.lower().rsplit(".", 1)[0]
|
||||
|
||||
|
||||
|
||||
words_i = set(fi.replace("_", " ").replace("-", " ").split())
|
||||
|
||||
|
||||
|
||||
words_j = set(fj.replace("_", " ").replace("-", " ").split())
|
||||
|
||||
|
||||
|
||||
common = words_i & words_j
|
||||
|
||||
|
||||
|
||||
if common and len(common) > 0:
|
||||
|
||||
|
||||
|
||||
graph_edges.append({
|
||||
|
||||
|
||||
|
||||
"source": f"doc-{documents[i].id[:8]}",
|
||||
|
||||
|
||||
|
||||
"target": f"doc-{documents[j].id[:8]}",
|
||||
|
||||
|
||||
|
||||
"label": f"共有关键词: {', '.join(list(common)[:3])}",
|
||||
|
||||
|
||||
|
||||
"weight": min(0.9, 0.3 * len(common)),
|
||||
|
||||
|
||||
|
||||
})
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
graph_data = {
|
||||
|
||||
|
||||
|
||||
"nodes": graph_nodes,
|
||||
|
||||
|
||||
|
||||
"edges": graph_edges,
|
||||
|
||||
|
||||
|
||||
"stats": {
|
||||
|
||||
|
||||
|
||||
"total_nodes": len(graph_nodes), "total_edges": len(graph_edges),
|
||||
|
||||
|
||||
|
||||
"document_count": len(documents),
|
||||
|
||||
|
||||
|
||||
},
|
||||
|
||||
|
||||
|
||||
}
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
kb.knowledge_graph_count = len(graph_nodes)
|
||||
|
||||
|
||||
|
||||
kb.updated_at = cst_now()
|
||||
|
||||
|
||||
|
||||
return graph_data
|
||||
# Auto-extracted from chat.py during Phase 1 refactor (Strangler Fig).
|
||||
# Source of truth: scripts/phase1_extract.py. Do not hand-merge back.
|
||||
|
||||
from fastapi import APIRouter, Depends, HTTPException, Query, Response
|
||||
from sqlalchemy.orm import Session
|
||||
from sqlalchemy import and_, desc, asc, func
|
||||
from pydantic import BaseModel, Field
|
||||
from typing import Optional
|
||||
import uuid
|
||||
from datetime import datetime, timezone, timedelta
|
||||
import re
|
||||
import json
|
||||
import asyncio
|
||||
import math
|
||||
import httpx
|
||||
import os
|
||||
import logging
|
||||
from ..database import get_db, SessionLocal
|
||||
from ..models import (
|
||||
|
||||
|
||||
|
||||
User, Conversation, Message, AIModel, ModelProvider, TokenUsageLog,
|
||||
|
||||
|
||||
|
||||
KnowledgeBase, KnowledgeDocument, KnowledgeDocumentChunk, EvolutionSkill,
|
||||
|
||||
|
||||
|
||||
EvolutionTrajectory, MarketApp, AppPurchase, AppInstallation,
|
||||
|
||||
|
||||
|
||||
MCPServer,
|
||||
|
||||
|
||||
|
||||
gen_uuid
|
||||
|
||||
|
||||
|
||||
)
|
||||
from ..auth import get_current_user
|
||||
from ..platform_knowledge import get_platform_knowledge
|
||||
from ..mcp_client import list_mcp_tools, call_mcp_tool
|
||||
from ..middleware.log_service import LogService
|
||||
from ..capability_registry import resolve_auto_model
|
||||
from app.api.chat_utils import (cst_now)
|
||||
|
||||
def _build_knowledge_graph(db: Session, kb: KnowledgeBase) -> dict:
|
||||
|
||||
|
||||
|
||||
"""从知识库文档构建知识图谱节点和边，返回 graph_data。"""
|
||||
|
||||
|
||||
|
||||
documents = db.query(KnowledgeDocument).filter(
|
||||
|
||||
|
||||
|
||||
and_(
|
||||
|
||||
|
||||
|
||||
KnowledgeDocument.knowledge_base_id == kb.id,
|
||||
|
||||
|
||||
|
||||
KnowledgeDocument.status == "completed",
|
||||
|
||||
|
||||
|
||||
)
|
||||
|
||||
|
||||
|
||||
).all()
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
graph_nodes = []
|
||||
|
||||
|
||||
|
||||
graph_edges = []
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
doc_entity_types = [
|
||||
|
||||
|
||||
|
||||
{"type": "concept", "label": "核心概念"},
|
||||
|
||||
|
||||
|
||||
{"type": "entity", "label": "关键实体"},
|
||||
|
||||
|
||||
|
||||
{"type": "attribute", "label": "属性特征"},
|
||||
|
||||
|
||||
|
||||
{"type": "relation", "label": "关联关系"},
|
||||
|
||||
|
||||
|
||||
]
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
for idx, doc in enumerate(documents):
|
||||
|
||||
|
||||
|
||||
doc_node_id = f"doc-{doc.id[:8]}"
|
||||
|
||||
|
||||
|
||||
graph_nodes.append({
|
||||
|
||||
|
||||
|
||||
"id": doc_node_id, "label": doc.filename, "type": "document", "size": 18, "group": "document",
|
||||
|
||||
|
||||
|
||||
})
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
filename_base = doc.filename.rsplit(".", 1)[0]
|
||||
|
||||
|
||||
|
||||
words = filename_base.replace("_", " ").replace("-", " ").split()
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
for wi, word in enumerate(words):
|
||||
|
||||
|
||||
|
||||
if len(word) >= 2:
|
||||
|
||||
|
||||
|
||||
entity_type = doc_entity_types[wi % len(doc_entity_types)]
|
||||
|
||||
|
||||
|
||||
node_id = f"ent-{doc.id[:8]}-{wi}"
|
||||
|
||||
|
||||
|
||||
graph_nodes.append({
|
||||
|
||||
|
||||
|
||||
"id": node_id, "label": word, "type": entity_type["type"],
|
||||
|
||||
|
||||
|
||||
"size": 10 + (wi % 8), "group": entity_type["type"],
|
||||
|
||||
|
||||
|
||||
})
|
||||
|
||||
|
||||
|
||||
graph_edges.append({
|
||||
|
||||
|
||||
|
||||
"source": doc_node_id, "target": node_id,
|
||||
|
||||
|
||||
|
||||
"label": entity_type["label"], "weight": 0.5 + (0.5 / (wi + 1)),
|
||||
|
||||
|
||||
|
||||
})
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
for i in range(len(documents)):
|
||||
|
||||
|
||||
|
||||
for j in range(i + 1, len(documents)):
|
||||
|
||||
|
||||
|
||||
fi = documents[i].filename.lower().rsplit(".", 1)[0]
|
||||
|
||||
|
||||
|
||||
fj = documents[j].filename.lower().rsplit(".", 1)[0]
|
||||
|
||||
|
||||
|
||||
words_i = set(fi.replace("_", " ").replace("-", " ").split())
|
||||
|
||||
|
||||
|
||||
words_j = set(fj.replace("_", " ").replace("-", " ").split())
|
||||
|
||||
|
||||
|
||||
common = words_i & words_j
|
||||
|
||||
|
||||
|
||||
if common and len(common) > 0:
|
||||
|
||||
|
||||
|
||||
graph_edges.append({
|
||||
|
||||
|
||||
|
||||
"source": f"doc-{documents[i].id[:8]}",
|
||||
|
||||
|
||||
|
||||
"target": f"doc-{documents[j].id[:8]}",
|
||||
|
||||
|
||||
|
||||
"label": f"共有关键词: {', '.join(list(common)[:3])}",
|
||||
|
||||
|
||||
|
||||
"weight": min(0.9, 0.3 * len(common)),
|
||||
|
||||
|
||||
|
||||
})
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
graph_data = {
|
||||
|
||||
|
||||
|
||||
"nodes": graph_nodes,
|
||||
|
||||
|
||||
|
||||
"edges": graph_edges,
|
||||
|
||||
|
||||
|
||||
"stats": {
|
||||
|
||||
|
||||
|
||||
"total_nodes": len(graph_nodes), "total_edges": len(graph_edges),
|
||||
|
||||
|
||||
|
||||
"document_count": len(documents),
|
||||
|
||||
|
||||
|
||||
},
|
||||
|
||||
|
||||
|
||||
}
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
kb.knowledge_graph_count = len(graph_nodes)
|
||||
|
||||
|
||||
|
||||
kb.updated_at = cst_now()
|
||||
|
||||
|
||||
|
||||
return graph_data
|
||||
+240
-240
@@ -1,240 +1,240 @@
|
||||
# Auto-extracted from chat.py during Phase 1 refactor (Strangler Fig).
|
||||
# Source of truth: scripts/phase1_extract.py. Do not hand-merge back.
|
||||
|
||||
from fastapi import APIRouter, Depends, HTTPException, Query, Response
|
||||
from sqlalchemy.orm import Session
|
||||
from sqlalchemy import and_, desc, asc, func
|
||||
from pydantic import BaseModel, Field
|
||||
from typing import Optional
|
||||
import uuid
|
||||
from datetime import datetime, timezone, timedelta
|
||||
import re
|
||||
import json
|
||||
import asyncio
|
||||
import math
|
||||
import httpx
|
||||
import os
|
||||
import logging
|
||||
from ..database import get_db, SessionLocal
|
||||
from ..models import (
|
||||
|
||||
|
||||
|
||||
User, Conversation, Message, AIModel, ModelProvider, TokenUsageLog,
|
||||
|
||||
|
||||
|
||||
KnowledgeBase, KnowledgeDocument, KnowledgeDocumentChunk, EvolutionSkill,
|
||||
|
||||
|
||||
|
||||
EvolutionTrajectory, MarketApp, AppPurchase, AppInstallation,
|
||||
|
||||
|
||||
|
||||
MCPServer,
|
||||
|
||||
|
||||
|
||||
gen_uuid
|
||||
|
||||
|
||||
|
||||
)
|
||||
from ..auth import get_current_user
|
||||
from ..platform_knowledge import get_platform_knowledge
|
||||
from ..mcp_client import list_mcp_tools, call_mcp_tool
|
||||
from ..middleware.log_service import LogService
|
||||
from ..capability_registry import resolve_auto_model
|
||||
|
||||
def _resolve_market_mcp(db: Session, mcp_ids: list, user_id: str) -> list:
|
||||
|
||||
|
||||
|
||||
"""
|
||||
|
||||
|
||||
|
||||
加载用户选择的 MCP 应用配置（MarketApp.type == 'mcp'）。
|
||||
|
||||
|
||||
|
||||
返回: [{id, name, transport, endpoint, description}]
|
||||
|
||||
|
||||
|
||||
"""
|
||||
|
||||
|
||||
|
||||
if not mcp_ids:
|
||||
|
||||
|
||||
|
||||
return []
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
mcps = db.query(MarketApp).filter(
|
||||
|
||||
|
||||
|
||||
MarketApp.id.in_(mcp_ids),
|
||||
|
||||
|
||||
|
||||
MarketApp.type == "mcp"
|
||||
|
||||
|
||||
|
||||
).all()
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
results = []
|
||||
|
||||
|
||||
|
||||
for m in mcps:
|
||||
|
||||
|
||||
|
||||
config = {}
|
||||
|
||||
|
||||
|
||||
try:
|
||||
|
||||
|
||||
|
||||
if m.skill_content:
|
||||
|
||||
|
||||
|
||||
config = json.loads(m.skill_content)
|
||||
|
||||
|
||||
|
||||
except Exception:
|
||||
|
||||
|
||||
|
||||
config = {}
|
||||
|
||||
|
||||
|
||||
results.append({
|
||||
|
||||
|
||||
|
||||
"id": m.id,
|
||||
|
||||
|
||||
|
||||
"name": m.name,
|
||||
|
||||
|
||||
|
||||
"transport": m.mcp_transport or config.get("transport", "stdio"),
|
||||
|
||||
|
||||
|
||||
"endpoint": m.agent_endpoint or config.get("endpoint", "") or config.get("url", ""),
|
||||
|
||||
|
||||
|
||||
"description": m.description or config.get("description", ""),
|
||||
|
||||
|
||||
|
||||
})
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
return results
|
||||
|
||||
def _resolve_active_mcp_servers(db: Session, mcp_ids: list) -> list:
|
||||
|
||||
"""
|
||||
|
||||
解析用户选中的 MCP 服务连接信息（来自 mcp_servers 表，import 与后台添加通用）。
|
||||
|
||||
返回: [{id, name, description, transport, endpoint, url, headers}]
|
||||
|
||||
"""
|
||||
|
||||
if not mcp_ids:
|
||||
|
||||
return []
|
||||
|
||||
servers = db.query(MCPServer).filter(
|
||||
|
||||
MCPServer.id.in_(mcp_ids),
|
||||
|
||||
MCPServer.is_enabled == True,
|
||||
|
||||
).all()
|
||||
|
||||
out = []
|
||||
|
||||
for s in servers:
|
||||
|
||||
try:
|
||||
|
||||
headers = json.loads(s.headers) if isinstance(s.headers, str) else (s.headers or {})
|
||||
|
||||
except Exception:
|
||||
|
||||
headers = {}
|
||||
|
||||
out.append({
|
||||
|
||||
"id": s.id,
|
||||
|
||||
"name": s.name,
|
||||
|
||||
"description": s.description or "",
|
||||
|
||||
"transport": s.transport_type or "streamable_http",
|
||||
|
||||
"endpoint": s.url or "",
|
||||
|
||||
"url": s.url or "",
|
||||
|
||||
"headers": headers,
|
||||
|
||||
})
|
||||
|
||||
return out
|
||||
|
||||
def _list_enabled_mcp_servers(db: Session) -> list:
|
||||
"""返回平台所有已启用(is_enabled=True)的 MCP 服务清单，供 AI 准确回答"平台有哪些 MCP"。"""
|
||||
servers = db.query(MCPServer).filter(MCPServer.is_enabled == True).all()
|
||||
out = []
|
||||
for s in servers:
|
||||
try:
|
||||
headers = json.loads(s.headers) if isinstance(s.headers, str) else (s.headers or {})
|
||||
except Exception:
|
||||
headers = {}
|
||||
out.append({
|
||||
"id": s.id,
|
||||
"name": s.name,
|
||||
"description": s.description or "",
|
||||
"transport": s.transport_type or "streamable_http",
|
||||
"endpoint": s.url or "",
|
||||
"url": s.url or "",
|
||||
"headers": headers,
|
||||
})
|
||||
return out
|
||||
# Auto-extracted from chat.py during Phase 1 refactor (Strangler Fig).
|
||||
# Source of truth: scripts/phase1_extract.py. Do not hand-merge back.
|
||||
|
||||
from fastapi import APIRouter, Depends, HTTPException, Query, Response
|
||||
from sqlalchemy.orm import Session
|
||||
from sqlalchemy import and_, desc, asc, func
|
||||
from pydantic import BaseModel, Field
|
||||
from typing import Optional
|
||||
import uuid
|
||||
from datetime import datetime, timezone, timedelta
|
||||
import re
|
||||
import json
|
||||
import asyncio
|
||||
import math
|
||||
import httpx
|
||||
import os
|
||||
import logging
|
||||
from ..database import get_db, SessionLocal
|
||||
from ..models import (
|
||||
|
||||
|
||||
|
||||
User, Conversation, Message, AIModel, ModelProvider, TokenUsageLog,
|
||||
|
||||
|
||||
|
||||
KnowledgeBase, KnowledgeDocument, KnowledgeDocumentChunk, EvolutionSkill,
|
||||
|
||||
|
||||
|
||||
EvolutionTrajectory, MarketApp, AppPurchase, AppInstallation,
|
||||
|
||||
|
||||
|
||||
MCPServer,
|
||||
|
||||
|
||||
|
||||
gen_uuid
|
||||
|
||||
|
||||
|
||||
)
|
||||
from ..auth import get_current_user
|
||||
from ..platform_knowledge import get_platform_knowledge
|
||||
from ..mcp_client import list_mcp_tools, call_mcp_tool
|
||||
from ..middleware.log_service import LogService
|
||||
from ..capability_registry import resolve_auto_model
|
||||
|
||||
def _resolve_market_mcp(db: Session, mcp_ids: list, user_id: str) -> list:
|
||||
|
||||
|
||||
|
||||
"""
|
||||
|
||||
|
||||
|
||||
加载用户选择的 MCP 应用配置（MarketApp.type == 'mcp'）。
|
||||
|
||||
|
||||
|
||||
返回: [{id, name, transport, endpoint, description}]
|
||||
|
||||
|
||||
|
||||
"""
|
||||
|
||||
|
||||
|
||||
if not mcp_ids:
|
||||
|
||||
|
||||
|
||||
return []
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
mcps = db.query(MarketApp).filter(
|
||||
|
||||
|
||||
|
||||
MarketApp.id.in_(mcp_ids),
|
||||
|
||||
|
||||
|
||||
MarketApp.type == "mcp"
|
||||
|
||||
|
||||
|
||||
).all()
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
results = []
|
||||
|
||||
|
||||
|
||||
for m in mcps:
|
||||
|
||||
|
||||
|
||||
config = {}
|
||||
|
||||
|
||||
|
||||
try:
|
||||
|
||||
|
||||
|
||||
if m.skill_content:
|
||||
|
||||
|
||||
|
||||
config = json.loads(m.skill_content)
|
||||
|
||||
|
||||
|
||||
except Exception:
|
||||
|
||||
|
||||
|
||||
config = {}
|
||||
|
||||
|
||||
|
||||
results.append({
|
||||
|
||||
|
||||
|
||||
"id": m.id,
|
||||
|
||||
|
||||
|
||||
"name": m.name,
|
||||
|
||||
|
||||
|
||||
"transport": m.mcp_transport or config.get("transport", "stdio"),
|
||||
|
||||
|
||||
|
||||
"endpoint": m.agent_endpoint or config.get("endpoint", "") or config.get("url", ""),
|
||||
|
||||
|
||||
|
||||
"description": m.description or config.get("description", ""),
|
||||
|
||||
|
||||
|
||||
})
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
return results
|
||||
|
||||
def _resolve_active_mcp_servers(db: Session, mcp_ids: list) -> list:
|
||||
|
||||
"""
|
||||
|
||||
解析用户选中的 MCP 服务连接信息（来自 mcp_servers 表，import 与后台添加通用）。
|
||||
|
||||
返回: [{id, name, description, transport, endpoint, url, headers}]
|
||||
|
||||
"""
|
||||
|
||||
if not mcp_ids:
|
||||
|
||||
return []
|
||||
|
||||
servers = db.query(MCPServer).filter(
|
||||
|
||||
MCPServer.id.in_(mcp_ids),
|
||||
|
||||
MCPServer.is_enabled == True,
|
||||
|
||||
).all()
|
||||
|
||||
out = []
|
||||
|
||||
for s in servers:
|
||||
|
||||
try:
|
||||
|
||||
headers = json.loads(s.headers) if isinstance(s.headers, str) else (s.headers or {})
|
||||
|
||||
except Exception:
|
||||
|
||||
headers = {}
|
||||
|
||||
out.append({
|
||||
|
||||
"id": s.id,
|
||||
|
||||
"name": s.name,
|
||||
|
||||
"description": s.description or "",
|
||||
|
||||
"transport": s.transport_type or "streamable_http",
|
||||
|
||||
"endpoint": s.url or "",
|
||||
|
||||
"url": s.url or "",
|
||||
|
||||
"headers": headers,
|
||||
|
||||
})
|
||||
|
||||
return out
|
||||
|
||||
def _list_enabled_mcp_servers(db: Session) -> list:
|
||||
"""返回平台所有已启用(is_enabled=True)的 MCP 服务清单，供 AI 准确回答"平台有哪些 MCP"。"""
|
||||
servers = db.query(MCPServer).filter(MCPServer.is_enabled == True).all()
|
||||
out = []
|
||||
for s in servers:
|
||||
try:
|
||||
headers = json.loads(s.headers) if isinstance(s.headers, str) else (s.headers or {})
|
||||
except Exception:
|
||||
headers = {}
|
||||
out.append({
|
||||
"id": s.id,
|
||||
"name": s.name,
|
||||
"description": s.description or "",
|
||||
"transport": s.transport_type or "streamable_http",
|
||||
"endpoint": s.url or "",
|
||||
"url": s.url or "",
|
||||
"headers": headers,
|
||||
})
|
||||
return out
|
||||
+8549
-8549
File diff suppressed because it is too large.
Load diff
+1742
-1742
File diff suppressed because it is too large.
Load diff
+282
-282
@@ -1,282 +1,282 @@
|
||||
"""
|
||||
IM Bot 真实 LLM 调用。
|
||||
|
||||
原实现位于 channels.py（通用频道管理模块），该模块已移除。
|
||||
为保持 IM Bot（微信机器人等）正常生成回复，将 _call_real_llm 迁移至此独立模块。
|
||||
|
||||
行为（对齐 AI 聊天 — 2026-07-23 同步至最新）：
|
||||
- Issue 7：图像/视频/语音意图优先走平台媒体模型（与 AI 聊天同一套 Auto 路由），
|
||||
不再对"制作图片"这类请求只回文本。
|
||||
- Issue 8：默认 Auto 模型；每轮先自动发现本账号已安装且平台已部署的技能，
|
||||
多个适用时由模型自行判断选 1 个最优，通过 function calling 执行。
|
||||
- 2026-07-15：修复 trust_env=False + proxy 处理，避免容器坏代理导致调用失败；
|
||||
媒体生成前先经 _optimize_media_prompt 精炼提示词，提升 AGNES 生成质量。
|
||||
- 2026-07-23：对齐 Web Chat 最新能力：
|
||||
* 模型 failover（Auto 模式自动切换备选模型）
|
||||
* 平台知识库注入（get_platform_knowledge，技能清单实时同步）
|
||||
"""
|
||||
import asyncio
|
||||
import json
|
||||
import logging
|
||||
import re as _re
|
||||
from datetime import datetime, timezone, timedelta
|
||||
from typing import Optional
|
||||
from urllib.parse import urlparse as _urlparse
|
||||
|
||||
import httpx
|
||||
from app.core.llm_client import chat_complete # 统一模型层(07-24): web/IM 共用标准 OpenAI SDK 客户端
|
||||
from sqlalchemy.orm import Session
|
||||
|
||||
from ..capability_registry import resolve_auto_model
|
||||
from ..models import AIModel, ModelProvider, TokenUsageLog, gen_uuid
|
||||
from app.core.observability import get_trace_id
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
# 香港出网代理(实测对 AGNES 延迟更低，2026-07-15 用户实测)，与 chat.py 同值。
|
||||
_HK_PROXY = "http://qyyfw:XHR521yhf@103.242.14.208:11495"
|
||||
|
||||
|
||||
def _proxy_for_url(url: str) -> str | None:
|
||||
"""按目标域名选择是否走代理（与 chat.py 同逻辑）。
|
||||
|
||||
2026-07-15 调整：AGNES 改走香港代理(实测延迟更低)，不再直连。
|
||||
国内域名(尊重 NO_PROXY)直连，更稳定。
|
||||
"""
|
||||
import os
|
||||
host = (_urlparse(url).hostname or "").lower()
|
||||
if "agnes-ai.com" in host or "agnes-ai.space" in host:
|
||||
return _HK_PROXY
|
||||
if "qihuyun.com" in host:
|
||||
return None
|
||||
_no_proxy = [h.strip().lower() for h in os.environ.get("NO_PROXY", "").split(",") if h.strip()]
|
||||
for _np in _no_proxy:
|
||||
if host == _np or host.endswith("." + _np) or _np in host:
|
||||
return None
|
||||
return os.environ.get("HTTPS_PROXY") or os.environ.get("HTTP_PROXY") or "http://172.18.0.1:7890"
|
||||
|
||||
|
||||
def _run_async(coro, timeout=None):
|
||||
"""在同步上下文中运行协程，兼容「无事件循环」与「已有事件循环」两种场景。
|
||||
|
||||
IM 个人微信在独立线程中调用（无事件循环）→ asyncio.run；
|
||||
IM 多平台 Webhook 可能在 FastAPI 协程内调用（已有事件循环）→ 丢到线程池执行。
|
||||
|
||||
timeout: 可选，秒数；超时抛出 asyncio.TimeoutError（由调用方捕获）。
|
||||
"""
|
||||
try:
|
||||
loop = asyncio.get_running_loop()
|
||||
except RuntimeError:
|
||||
loop = None
|
||||
|
||||
if timeout is not None:
|
||||
async def _timed():
|
||||
return await asyncio.wait_for(coro, timeout=timeout)
|
||||
target = _timed()
|
||||
else:
|
||||
target = coro
|
||||
|
||||
if loop is not None and loop.is_running():
|
||||
import concurrent.futures
|
||||
with concurrent.futures.ThreadPoolExecutor(1) as ex:
|
||||
return ex.submit(lambda: asyncio.run(target)).result(timeout=(timeout + 10) if timeout else None)
|
||||
return asyncio.run(target)
|
||||
|
||||
|
||||
def _log_media_usage(db: Session, user_id: str, ai_model, media_type: str, prompt: str):
|
||||
"""IM Bot 媒体生成记账：把图片/视频消耗写入 TokenUsageLog，供驾驶舱统计。
|
||||
|
||||
此前 IM Bot 路径完全不写 TokenUsageLog，导致微信生成的图片/视频在驾驶舱里"查无此项"。
|
||||
媒体 API 不返回 token 用量，这里用与 chat.py _generate_media_task 一致的估算。
|
||||
"""
|
||||
if not user_id or not ai_model:
|
||||
return
|
||||
try:
|
||||
base = 1200 if media_type == "video" else 400
|
||||
est = base + (len(prompt or "") // 4)
|
||||
db.add(TokenUsageLog(
|
||||
id=gen_uuid(),
|
||||
user_id=user_id,
|
||||
model_id=ai_model.id,
|
||||
model_name=ai_model.name,
|
||||
trace_id=get_trace_id(),
|
||||
prompt_tokens=est,
|
||||
completion_tokens=0,
|
||||
total_tokens=est,
|
||||
created_at=datetime.now(timezone(timedelta(hours=8))).replace(tzinfo=None),
|
||||
))
|
||||
db.commit()
|
||||
except Exception as e:
|
||||
try:
|
||||
db.rollback()
|
||||
except Exception:
|
||||
pass
|
||||
logger.warning(f"[im_llm] 媒体 Token 记账失败(忽略): {e}")
|
||||
|
||||
|
||||
def _call_real_llm(db: Session, user=None, prompt: str = "", platform: str = "im", user_id: str = None) -> "str | dict":
|
||||
"""调用七虎AI 生成回复，供 IM Bot 复用。支持媒体意图路由、已安装技能自动发现、模型 failover。"""
|
||||
try:
|
||||
from app.api.chat import (
|
||||
_generate_media,
|
||||
detect_media_intent,
|
||||
_discover_installed_skills,
|
||||
_build_system_prompt,
|
||||
_execute_skill_tool,
|
||||
)
|
||||
|
||||
# ===== 1) 媒体意图路由（Issue 7）=====
|
||||
intent = detect_media_intent(prompt)
|
||||
if intent:
|
||||
ai_model = resolve_auto_model(intent, db)
|
||||
if ai_model:
|
||||
try:
|
||||
result = _run_async(_generate_media(db, ai_model, prompt, user_id), timeout=300)
|
||||
except Exception:
|
||||
# 任何未预期异常(如 AGNES 连接重置)统一兜底为友好提示，
|
||||
# 绝不直接把原始异常(如 "Server disconnected")甩给用户
|
||||
result = "⚠️ 媒体生成暂时失败（连接或服务波动），请稍后重试。"
|
||||
if result:
|
||||
# 媒体生成成功：返回结构化对象，交由 im_bot_user 在微信内联发送图片/视频
|
||||
s = result.strip()
|
||||
if s.startswith("<img") or s.startswith("<video"):
|
||||
m = _re.search(r'src="([^"]+)"', s)
|
||||
url = m.group(1) if m else ""
|
||||
media_type = "image" if s.startswith("<img") else "video"
|
||||
if not url:
|
||||
return "（生成完成，但未能提取媒体地址）"
|
||||
# IM Bot 路径此前完全不记账 → 补写媒体消耗，驾驶舱才看得到图片/视频数据
|
||||
_log_media_usage(db, user_id, ai_model, media_type, prompt)
|
||||
return {"__media__": True, "media_type": media_type, "url": url}
|
||||
# 媒体生成失败：_generate_media 已返回友好中文（如限频/队列繁忙），
|
||||
# 直接透传，绝不把原始提示词/标签/JSON 透传给用户
|
||||
if s.startswith("⚠️") or s.startswith("（媒体"):
|
||||
return s
|
||||
return "（媒体生成未能返回有效内容，请重试）"
|
||||
|
||||
# ===== 2) 文本回复：模型 failover + 技能自动发现 + 平台知识库注入 =====
|
||||
from ..platform_knowledge import get_platform_knowledge
|
||||
|
||||
# 构建候选模型列表（对齐 Web Chat failover）
|
||||
ai_model = resolve_auto_model("text", db)
|
||||
_try_models = [ai_model] if (ai_model and ai_model.is_enabled) else []
|
||||
# Auto 模式附加备选（IM Bot 始终 Auto）
|
||||
_fallbacks = db.query(AIModel).filter(
|
||||
AIModel.is_enabled == True,
|
||||
AIModel.id != (ai_model.id if ai_model else ""),
|
||||
).order_by(AIModel.is_default.desc()).limit(4).all()
|
||||
_try_models.extend([m for m in _fallbacks if m not in _try_models])
|
||||
|
||||
if not _try_models:
|
||||
return "（未配置可用文本模型，无法生成回复）"
|
||||
|
||||
# 技能自动发现（提前执行一次，不必每个候选模型重复做）
|
||||
skills = []
|
||||
if user_id:
|
||||
try:
|
||||
skills = _discover_installed_skills(db, user_id)[:15]
|
||||
except Exception:
|
||||
skills = []
|
||||
|
||||
# 平台知识库注入（对齐 Web Chat——微信用户提问"平台有哪些技能"也能得到完整清单）
|
||||
kb_context = ""
|
||||
try:
|
||||
pk = get_platform_knowledge(db=db)
|
||||
if pk:
|
||||
kb_context = f"【平台知识库——优先使用以下知识回答用户问题】\n{pk}"
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
# 尝试每个候选模型
|
||||
_retry_log = []
|
||||
for _try_model in _try_models:
|
||||
ai_model = _try_model
|
||||
|
||||
provider = db.query(ModelProvider).filter(ModelProvider.id == ai_model.provider_id).first()
|
||||
api_key = ai_model.api_key or (provider.api_key if provider else None)
|
||||
api_url = ai_model.endpoint
|
||||
if not api_url and provider and provider.api_base_url:
|
||||
api_url = f"{provider.api_base_url.rstrip('/')}/v1/chat/completions"
|
||||
if not (api_key and api_url):
|
||||
_retry_log.append(f"{ai_model.name}: 缺少密钥或端点")
|
||||
continue
|
||||
|
||||
system_prompt, function_definitions = _build_system_prompt(
|
||||
skills=skills,
|
||||
agents=[],
|
||||
kb_context=kb_context,
|
||||
user_content=prompt,
|
||||
mcp=[],
|
||||
auto_discovered=bool(skills),
|
||||
)
|
||||
messages = [
|
||||
{"role": "system", "content": system_prompt},
|
||||
{"role": "user", "content": prompt},
|
||||
]
|
||||
|
||||
# 尝试调用
|
||||
try:
|
||||
max_rounds = 3
|
||||
ai_content = ""
|
||||
_http_ok = True
|
||||
for round_num in range(max_rounds):
|
||||
request_body = {"model": ai_model.name, "messages": messages, "max_tokens": 2048}
|
||||
if function_definitions and skills:
|
||||
request_body["tools"] = [
|
||||
{"type": "function", "function": fd} for fd in function_definitions
|
||||
]
|
||||
request_body["tool_choice"] = "auto"
|
||||
try:
|
||||
data = chat_complete(
|
||||
api_url,
|
||||
api_key,
|
||||
request_body,
|
||||
proxy=_proxy_for_url(api_url),
|
||||
timeout=ai_model.timeout or 120,
|
||||
)
|
||||
except Exception as _exc:
|
||||
_retry_log.append(f"{ai_model.name}: {str(_exc)[:60]}")
|
||||
logger.warning(f"[im_llm] 模型 {ai_model.name} 异常: {_exc}")
|
||||
_http_ok = False
|
||||
break
|
||||
message = data.get("choices", [{}])[0].get("message", {})
|
||||
tool_calls = message.get("tool_calls", [])
|
||||
if tool_calls and skills:
|
||||
messages.append({
|
||||
"role": "assistant",
|
||||
"content": message.get("content") or "",
|
||||
"tool_calls": tool_calls,
|
||||
})
|
||||
for tc in tool_calls:
|
||||
func_name = tc.get("function", {}).get("name", "")
|
||||
try:
|
||||
func_args = json.loads(tc.get("function", {}).get("arguments", "{}"))
|
||||
except Exception:
|
||||
func_args = {"input": tc.get("function", {}).get("arguments", "")}
|
||||
tool_result = _execute_skill_tool(db, func_name, func_args, skills, prompt)
|
||||
messages.append({
|
||||
"role": "tool",
|
||||
"tool_call_id": tc.get("id", "call_1"),
|
||||
"content": tool_result,
|
||||
})
|
||||
continue
|
||||
ai_content = message.get("content", "")
|
||||
break
|
||||
|
||||
# 检查本轮调用是否成功
|
||||
if _http_ok and ai_content:
|
||||
return ai_content or "（AI 回复为空）"
|
||||
# 本轮失败：记录日志，切换备选
|
||||
logger.warning(f"[im_llm] 模型 {ai_model.name} 失败，切换备选...")
|
||||
continue
|
||||
except Exception as exc:
|
||||
_retry_log.append(f"{ai_model.name}: {str(exc)[:60]}")
|
||||
logger.warning(f"[im_llm] 模型 {ai_model.name} 异常: {exc}")
|
||||
continue
|
||||
|
||||
# 全部失败
|
||||
_summary = "; ".join(_retry_log[-3:])
|
||||
return f"（已尝试 {len(_try_models)} 个模型均调用失败。最后错误: {_summary}。请稍后重试。）"
|
||||
except Exception as e:
|
||||
return f"（AI 回复生成失败：{e}）"
|
||||
"""
|
||||
IM Bot 真实 LLM 调用。
|
||||
|
||||
原实现位于 channels.py（通用频道管理模块），该模块已移除。
|
||||
为保持 IM Bot（微信机器人等）正常生成回复，将 _call_real_llm 迁移至此独立模块。
|
||||
|
||||
行为（对齐 AI 聊天 — 2026-07-23 同步至最新）：
|
||||
- Issue 7：图像/视频/语音意图优先走平台媒体模型（与 AI 聊天同一套 Auto 路由），
|
||||
不再对"制作图片"这类请求只回文本。
|
||||
- Issue 8：默认 Auto 模型；每轮先自动发现本账号已安装且平台已部署的技能，
|
||||
多个适用时由模型自行判断选 1 个最优，通过 function calling 执行。
|
||||
- 2026-07-15：修复 trust_env=False + proxy 处理，避免容器坏代理导致调用失败；
|
||||
媒体生成前先经 _optimize_media_prompt 精炼提示词，提升 AGNES 生成质量。
|
||||
- 2026-07-23：对齐 Web Chat 最新能力：
|
||||
* 模型 failover（Auto 模式自动切换备选模型）
|
||||
* 平台知识库注入（get_platform_knowledge，技能清单实时同步）
|
||||
"""
|
||||
import asyncio
|
||||
import json
|
||||
import logging
|
||||
import re as _re
|
||||
from datetime import datetime, timezone, timedelta
|
||||
from typing import Optional
|
||||
from urllib.parse import urlparse as _urlparse
|
||||
|
||||
import httpx
|
||||
from app.core.llm_client import chat_complete # 统一模型层(07-24): web/IM 共用标准 OpenAI SDK 客户端
|
||||
from sqlalchemy.orm import Session
|
||||
|
||||
from ..capability_registry import resolve_auto_model
|
||||
from ..models import AIModel, ModelProvider, TokenUsageLog, gen_uuid
|
||||
from app.core.observability import get_trace_id
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
||||
# 香港出网代理(实测对 AGNES 延迟更低，2026-07-15 用户实测)，与 chat.py 同值。
|
||||
_HK_PROXY = "http://qyyfw:XHR521yhf@103.242.14.208:11495"
|
||||
|
||||
|
||||
def _proxy_for_url(url: str) -> str | None:
|
||||
"""按目标域名选择是否走代理（与 chat.py 同逻辑）。
|
||||
|
||||
2026-07-15 调整：AGNES 改走香港代理(实测延迟更低)，不再直连。
|
||||
国内域名(尊重 NO_PROXY)直连，更稳定。
|
||||
"""
|
||||
import os
|
||||
host = (_urlparse(url).hostname or "").lower()
|
||||
if "agnes-ai.com" in host or "agnes-ai.space" in host:
|
||||
return _HK_PROXY
|
||||
if "qihuyun.com" in host:
|
||||
return None
|
||||
_no_proxy = [h.strip().lower() for h in os.environ.get("NO_PROXY", "").split(",") if h.strip()]
|
||||
for _np in _no_proxy:
|
||||
if host == _np or host.endswith("." + _np) or _np in host:
|
||||
return None
|
||||
return os.environ.get("HTTPS_PROXY") or os.environ.get("HTTP_PROXY") or "http://172.18.0.1:7890"
|
||||
|
||||
|
||||
def _run_async(coro, timeout=None):
|
||||
"""在同步上下文中运行协程，兼容「无事件循环」与「已有事件循环」两种场景。
|
||||
|
||||
IM 个人微信在独立线程中调用（无事件循环）→ asyncio.run；
|
||||
IM 多平台 Webhook 可能在 FastAPI 协程内调用（已有事件循环）→ 丢到线程池执行。
|
||||
|
||||
timeout: 可选，秒数；超时抛出 asyncio.TimeoutError（由调用方捕获）。
|
||||
"""
|
||||
try:
|
||||
loop = asyncio.get_running_loop()
|
||||
except RuntimeError:
|
||||
loop = None
|
||||
|
||||
if timeout is not None:
|
||||
async def _timed():
|
||||
return await asyncio.wait_for(coro, timeout=timeout)
|
||||
target = _timed()
|
||||
else:
|
||||
target = coro
|
||||
|
||||
if loop is not None and loop.is_running():
|
||||
import concurrent.futures
|
||||
with concurrent.futures.ThreadPoolExecutor(1) as ex:
|
||||
return ex.submit(lambda: asyncio.run(target)).result(timeout=(timeout + 10) if timeout else None)
|
||||
return asyncio.run(target)
|
||||
|
||||
|
||||
def _log_media_usage(db: Session, user_id: str, ai_model, media_type: str, prompt: str):
|
||||
"""IM Bot 媒体生成记账：把图片/视频消耗写入 TokenUsageLog，供驾驶舱统计。
|
||||
|
||||
此前 IM Bot 路径完全不写 TokenUsageLog，导致微信生成的图片/视频在驾驶舱里"查无此项"。
|
||||
媒体 API 不返回 token 用量，这里用与 chat.py _generate_media_task 一致的估算。
|
||||
"""
|
||||
if not user_id or not ai_model:
|
||||
return
|
||||
try:
|
||||
base = 1200 if media_type == "video" else 400
|
||||
est = base + (len(prompt or "") // 4)
|
||||
db.add(TokenUsageLog(
|
||||
id=gen_uuid(),
|
||||
user_id=user_id,
|
||||
model_id=ai_model.id,
|
||||
model_name=ai_model.name,
|
||||
trace_id=get_trace_id(),
|
||||
prompt_tokens=est,
|
||||
completion_tokens=0,
|
||||
total_tokens=est,
|
||||
created_at=datetime.now(timezone(timedelta(hours=8))).replace(tzinfo=None),
|
||||
))
|
||||
db.commit()
|
||||
except Exception as e:
|
||||
try:
|
||||
db.rollback()
|
||||
except Exception:
|
||||
pass
|
||||
logger.warning(f"[im_llm] 媒体 Token 记账失败(忽略): {e}")
|
||||
|
||||
|
||||
def _call_real_llm(db: Session, user=None, prompt: str = "", platform: str = "im", user_id: str = None) -> "str | dict":
|
||||
"""调用七虎AI 生成回复，供 IM Bot 复用。支持媒体意图路由、已安装技能自动发现、模型 failover。"""
|
||||
try:
|
||||
from app.api.chat import (
|
||||
_generate_media,
|
||||
detect_media_intent,
|
||||
_discover_installed_skills,
|
||||
_build_system_prompt,
|
||||
_execute_skill_tool,
|
||||
)
|
||||
|
||||
# ===== 1) 媒体意图路由（Issue 7）=====
|
||||
intent = detect_media_intent(prompt)
|
||||
if intent:
|
||||
ai_model = resolve_auto_model(intent, db)
|
||||
if ai_model:
|
||||
try:
|
||||
result = _run_async(_generate_media(db, ai_model, prompt, user_id), timeout=300)
|
||||
except Exception:
|
||||
# 任何未预期异常(如 AGNES 连接重置)统一兜底为友好提示，
|
||||
# 绝不直接把原始异常(如 "Server disconnected")甩给用户
|
||||
result = "⚠️ 媒体生成暂时失败（连接或服务波动），请稍后重试。"
|
||||
if result:
|
||||
# 媒体生成成功：返回结构化对象，交由 im_bot_user 在微信内联发送图片/视频
|
||||
s = result.strip()
|
||||
if s.startswith("<img") or s.startswith("<video"):
|
||||
m = _re.search(r'src="([^"]+)"', s)
|
||||
url = m.group(1) if m else ""
|
||||
media_type = "image" if s.startswith("<img") else "video"
|
||||
if not url:
|
||||
return "（生成完成，但未能提取媒体地址）"
|
||||
# IM Bot 路径此前完全不记账 → 补写媒体消耗，驾驶舱才看得到图片/视频数据
|
||||
_log_media_usage(db, user_id, ai_model, media_type, prompt)
|
||||
return {"__media__": True, "media_type": media_type, "url": url}
|
||||
# 媒体生成失败：_generate_media 已返回友好中文（如限频/队列繁忙），
|
||||
# 直接透传，绝不把原始提示词/标签/JSON 透传给用户
|
||||
if s.startswith("⚠️") or s.startswith("（媒体"):
|
||||
return s
|
||||
return "（媒体生成未能返回有效内容，请重试）"
|
||||
|
||||
# ===== 2) 文本回复：模型 failover + 技能自动发现 + 平台知识库注入 =====
|
||||
from ..platform_knowledge import get_platform_knowledge
|
||||
|
||||
# 构建候选模型列表（对齐 Web Chat failover）
|
||||
ai_model = resolve_auto_model("text", db)
|
||||
_try_models = [ai_model] if (ai_model and ai_model.is_enabled) else []
|
||||
# Auto 模式附加备选（IM Bot 始终 Auto）
|
||||
_fallbacks = db.query(AIModel).filter(
|
||||
AIModel.is_enabled == True,
|
||||
AIModel.id != (ai_model.id if ai_model else ""),
|
||||
).order_by(AIModel.is_default.desc()).limit(4).all()
|
||||
_try_models.extend([m for m in _fallbacks if m not in _try_models])
|
||||
|
||||
if not _try_models:
|
||||
return "（未配置可用文本模型，无法生成回复）"
|
||||
|
||||
# 技能自动发现（提前执行一次，不必每个候选模型重复做）
|
||||
skills = []
|
||||
if user_id:
|
||||
try:
|
||||
skills = _discover_installed_skills(db, user_id)[:15]
|
||||
except Exception:
|
||||
skills = []
|
||||
|
||||
# 平台知识库注入（对齐 Web Chat——微信用户提问"平台有哪些技能"也能得到完整清单）
|
||||
kb_context = ""
|
||||
try:
|
||||
pk = get_platform_knowledge(db=db)
|
||||
if pk:
|
||||
kb_context = f"【平台知识库——优先使用以下知识回答用户问题】\n{pk}"
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
# 尝试每个候选模型
|
||||
_retry_log = []
|
||||
for _try_model in _try_models:
|
||||
ai_model = _try_model
|
||||
|
||||
provider = db.query(ModelProvider).filter(ModelProvider.id == ai_model.provider_id).first()
|
||||
api_key = ai_model.api_key or (provider.api_key if provider else None)
|
||||
api_url = ai_model.endpoint
|
||||
if not api_url and provider and provider.api_base_url:
|
||||
api_url = f"{provider.api_base_url.rstrip('/')}/v1/chat/completions"
|
||||
if not (api_key and api_url):
|
||||
_retry_log.append(f"{ai_model.name}: 缺少密钥或端点")
|
||||
continue
|
||||
|
||||
system_prompt, function_definitions = _build_system_prompt(
|
||||
skills=skills,
|
||||
agents=[],
|
||||
kb_context=kb_context,
|
||||
user_content=prompt,
|
||||
mcp=[],
|
||||
auto_discovered=bool(skills),
|
||||
)
|
||||
messages = [
|
||||
{"role": "system", "content": system_prompt},
|
||||
{"role": "user", "content": prompt},
|
||||
]
|
||||
|
||||
# 尝试调用
|
||||
try:
|
||||
max_rounds = 3
|
||||
ai_content = ""
|
||||
_http_ok = True
|
||||
for round_num in range(max_rounds):
|
||||
request_body = {"model": ai_model.name, "messages": messages, "max_tokens": 2048}
|
||||
if function_definitions and skills:
|
||||
request_body["tools"] = [
|
||||
{"type": "function", "function": fd} for fd in function_definitions
|
||||
]
|
||||
request_body["tool_choice"] = "auto"
|
||||
try:
|
||||
data = chat_complete(
|
||||
api_url,
|
||||
api_key,
|
||||
request_body,
|
||||
proxy=_proxy_for_url(api_url),
|
||||
timeout=ai_model.timeout or 120,
|
||||
)
|
||||
except Exception as _exc:
|
||||
_retry_log.append(f"{ai_model.name}: {str(_exc)[:60]}")
|
||||
logger.warning(f"[im_llm] 模型 {ai_model.name} 异常: {_exc}")
|
||||
_http_ok = False
|
||||
break
|
||||
message = data.get("choices", [{}])[0].get("message", {})
|
||||
tool_calls = message.get("tool_calls", [])
|
||||
if tool_calls and skills:
|
||||
messages.append({
|
||||
"role": "assistant",
|
||||
"content": message.get("content") or "",
|
||||
"tool_calls": tool_calls,
|
||||
})
|
||||
for tc in tool_calls:
|
||||
func_name = tc.get("function", {}).get("name", "")
|
||||
try:
|
||||
func_args = json.loads(tc.get("function", {}).get("arguments", "{}"))
|
||||
except Exception:
|
||||
func_args = {"input": tc.get("function", {}).get("arguments", "")}
|
||||
tool_result = _execute_skill_tool(db, func_name, func_args, skills, prompt)
|
||||
messages.append({
|
||||
"role": "tool",
|
||||
"tool_call_id": tc.get("id", "call_1"),
|
||||
"content": tool_result,
|
||||
})
|
||||
continue
|
||||
ai_content = message.get("content", "")
|
||||
break
|
||||
|
||||
# 检查本轮调用是否成功
|
||||
if _http_ok and ai_content:
|
||||
return ai_content or "（AI 回复为空）"
|
||||
# 本轮失败：记录日志，切换备选
|
||||
logger.warning(f"[im_llm] 模型 {ai_model.name} 失败，切换备选...")
|
||||
continue
|
||||
except Exception as exc:
|
||||
_retry_log.append(f"{ai_model.name}: {str(exc)[:60]}")
|
||||
logger.warning(f"[im_llm] 模型 {ai_model.name} 异常: {exc}")
|
||||
continue
|
||||
|
||||
# 全部失败
|
||||
_summary = "; ".join(_retry_log[-3:])
|
||||
return f"（已尝试 {len(_try_models)} 个模型均调用失败。最后错误: {_summary}。请稍后重试。）"
|
||||
except Exception as e:
|
||||
return f"（AI 回复生成失败：{e}）"
|
||||
+1341
-1341
File diff suppressed because it is too large.
Load diff
@@ -1,395 +1,395 @@
|
||||
"""app/core/orchestrator.py — Phase 3.3 多 Agent 编排图引擎。
|
||||
|
||||
设计目标：
|
||||
- 以轻量 DAG 描述多 Agent 协作：节点(llm/local/external/reduce) + 有向边(数据流)。
|
||||
- 拓扑分层后按层并行执行(asyncio.gather)，互相独立的节点真正并发。
|
||||
- 失败隔离：每个节点异常不影响其它分支；per-node on_error=halt|continue。
|
||||
- 节点执行复用既有原语(_build_system_prompt / _execute_skill_tool / _call_external_agent /
|
||||
3.1 工具沙箱)，完全向后兼容——agent.orchestration 为 null 时 chat_service 走原逻辑。
|
||||
- 并发安全：每个节点用独立 SessionLocal 会话做 DB 读，避免共享请求会话跨协程竞态。
|
||||
"""
|
||||
from __future__ import annotations
|
||||
|
||||
import asyncio
|
||||
import json
|
||||
import logging
|
||||
from typing import Any, Optional
|
||||
|
||||
import httpx
|
||||
|
||||
logger = logging.getLogger("aiflow.orchestrator")
|
||||
|
||||
VALID_NODE_TYPES = ("llm", "local", "external", "reduce")
|
||||
|
||||
|
||||
class OrchestrationError(ValueError):
|
||||
"""图结构非法或执行被 halt。"""
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 校验 + 拓扑
|
||||
# ---------------------------------------------------------------------------
|
||||
def validate_graph(graph: Any) -> None:
|
||||
if not isinstance(graph, dict):
|
||||
raise OrchestrationError("orchestration 必须为对象")
|
||||
nodes = graph.get("nodes")
|
||||
edges = graph.get("edges")
|
||||
if not isinstance(nodes, list) or not nodes:
|
||||
raise OrchestrationError("nodes 必须为非空数组")
|
||||
if not isinstance(edges, list):
|
||||
raise OrchestrationError("edges 必须为数组")
|
||||
|
||||
ids = set()
|
||||
for i, n in enumerate(nodes):
|
||||
if not isinstance(n, dict):
|
||||
raise OrchestrationError(f"节点 #{i} 必须为对象")
|
||||
nid = n.get("id")
|
||||
if not nid or not isinstance(nid, str):
|
||||
raise OrchestrationError(f"节点 #{i} 缺少字符串 id")
|
||||
if nid in ids:
|
||||
raise OrchestrationError(f"节点 id 重复: {nid}")
|
||||
ids.add(nid)
|
||||
ntype = n.get("type")
|
||||
if ntype not in VALID_NODE_TYPES:
|
||||
raise OrchestrationError(
|
||||
f"节点 {nid} 类型非法: {ntype} (允许 {', '.join(VALID_NODE_TYPES)})"
|
||||
)
|
||||
if ntype in ("local", "external") and not n.get("agent_id"):
|
||||
raise OrchestrationError(f"节点 {nid} 类型 {ntype} 必须指定 agent_id")
|
||||
|
||||
for i, e in enumerate(edges):
|
||||
if not isinstance(e, dict):
|
||||
raise OrchestrationError(f"边 #{i} 必须为对象")
|
||||
f = e.get("from")
|
||||
t = e.get("to")
|
||||
if f not in ids:
|
||||
raise OrchestrationError(f"边 #{i} 的 from={f} 指向不存在的节点")
|
||||
if t not in ids:
|
||||
raise OrchestrationError(f"边 #{i} 的 to={t} 指向不存在的节点")
|
||||
if f == t:
|
||||
raise OrchestrationError(f"边 #{i} 存在自环: {f}")
|
||||
|
||||
|
||||
def topological_layers(graph: dict) -> list:
|
||||
"""Kahn 算法返回分层(每层内节点可并行)；含环则抛错。"""
|
||||
nodes = graph["nodes"]
|
||||
edges = graph.get("edges", [])
|
||||
indeg = {n["id"]: 0 for n in nodes}
|
||||
adj = {n["id"]: [] for n in nodes}
|
||||
for e in edges:
|
||||
f, t = e["from"], e["to"]
|
||||
adj[f].append(t)
|
||||
indeg[t] += 1
|
||||
order = {n["id"]: i for i, n in enumerate(nodes)}
|
||||
layers: list = []
|
||||
remaining = set(indeg)
|
||||
while remaining:
|
||||
layer = sorted([nid for nid in remaining if indeg[nid] == 0], key=lambda x: order[x])
|
||||
if not layer:
|
||||
cycle = sorted(remaining, key=lambda x: order[x])
|
||||
raise OrchestrationError(f"检测到环路，无法拓扑排序: {cycle}")
|
||||
layers.append(layer)
|
||||
for nid in layer:
|
||||
remaining.discard(nid)
|
||||
for nxt in adj[nid]:
|
||||
indeg[nxt] -= 1
|
||||
return layers
|
||||
|
||||
|
||||
def _node_by_id(graph: dict, nid: str) -> dict:
|
||||
for n in graph["nodes"]:
|
||||
if n["id"] == nid:
|
||||
return n
|
||||
return {}
|
||||
|
||||
|
||||
def _predecessors(graph: dict, nid: str) -> list:
|
||||
return [e["from"] for e in graph.get("edges", []) if e["to"] == nid]
|
||||
|
||||
|
||||
def _successors(graph: dict, nid: str) -> list:
|
||||
return [e["to"] for e in graph.get("edges", []) if e["from"] == nid]
|
||||
|
||||
|
||||
def _find_output_node(graph: dict) -> str:
|
||||
if graph.get("output_node"):
|
||||
return graph["output_node"]
|
||||
sinks = [n["id"] for n in graph["nodes"] if not _successors(graph, n["id"])]
|
||||
if len(sinks) == 1:
|
||||
return sinks[0]
|
||||
for s in sinks:
|
||||
if _node_by_id(graph, s).get("type") == "reduce":
|
||||
return s
|
||||
if sinks:
|
||||
return sinks[0]
|
||||
return graph["nodes"][-1]["id"]
|
||||
|
||||
|
||||
def _assemble_node_input(query: str, upstream: str, is_entry: bool) -> str:
|
||||
"""入口节点直接吃用户原问题；下游节点综合原问题 + 上游输出。"""
|
||||
if is_entry:
|
||||
return query or ""
|
||||
if upstream:
|
||||
return f"原始用户问题:\n{query}\n\n上游节点输出:\n{upstream}"
|
||||
return query or ""
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 模型解析 + LLM 调用原语
|
||||
# ---------------------------------------------------------------------------
|
||||
async def _resolve_model_and_endpoint(model_id, db):
|
||||
from app.models import AIModel, ModelProvider
|
||||
from ..capability_registry import resolve_auto_model
|
||||
|
||||
ai_model = None
|
||||
if model_id and model_id != "auto":
|
||||
ai_model = db.query(AIModel).filter(AIModel.id == model_id).first()
|
||||
if not ai_model:
|
||||
ai_model = resolve_auto_model("text", db)
|
||||
if not ai_model:
|
||||
raise OrchestrationError("无法解析到可用模型")
|
||||
provider = db.query(ModelProvider).filter(ModelProvider.id == ai_model.provider_id).first()
|
||||
api_key = ai_model.api_key or (provider.api_key if provider else None)
|
||||
api_url = ai_model.endpoint
|
||||
if not api_url and provider and provider.api_base_url:
|
||||
api_url = f"{provider.api_base_url.rstrip('/')}/v1/chat/completions"
|
||||
if not api_url or not api_key:
|
||||
raise OrchestrationError(f"节点模型 {ai_model.name} 缺少 endpoint 或 api_key")
|
||||
return ai_model, api_url, api_key
|
||||
|
||||
|
||||
async def _llm_complete(system_prompt: str, user_content: str, model_id, db, user_id) -> str:
|
||||
from app.api.chat_utils import _proxy_for_url
|
||||
from app.core.observability import span
|
||||
|
||||
ai_model, api_url, api_key = await _resolve_model_and_endpoint(model_id, db)
|
||||
messages = [
|
||||
{"role": "system", "content": system_prompt},
|
||||
{"role": "user", "content": user_content},
|
||||
]
|
||||
request_body = {"model": ai_model.name, "messages": messages, "max_tokens": 2048}
|
||||
async with span("orchestrator.llm", model=ai_model.id):
|
||||
async with httpx.AsyncClient(
|
||||
timeout=ai_model.timeout or 120,
|
||||
proxy=_proxy_for_url(api_url),
|
||||
trust_env=False,
|
||||
) as client:
|
||||
resp = await client.post(
|
||||
api_url,
|
||||
headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
|
||||
json=request_body,
|
||||
)
|
||||
if resp.status_code != 200:
|
||||
raise OrchestrationError(f"LLM 调用失败 HTTP {resp.status_code}: {resp.text[:200]}")
|
||||
data = resp.json()
|
||||
return data["choices"][0]["message"].get("content") or ""
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# local 节点：复用单 Agent system prompt + 受限 tool round + 3.1 沙箱
|
||||
# ---------------------------------------------------------------------------
|
||||
async def _run_local_agent_turn(agent: dict, node_input: str, db, user_id, conv_id) -> dict:
|
||||
from app.api.chat_skill import _resolve_market_skills, _execute_skill_tool
|
||||
from app.api.chat_context import _build_system_prompt, _retrieve_kb_context
|
||||
from app.api.chat_service import _resolve_agent_tool_policy, _check_tool_policy
|
||||
from app.api.chat_utils import _proxy_for_url
|
||||
|
||||
skills = _resolve_market_skills(db, agent.get("skill_ids") or [], user_id)
|
||||
kb_context = await _retrieve_kb_context(db, agent.get("knowledge_base_ids") or [], node_input)
|
||||
system_prompt, function_definitions = _build_system_prompt(
|
||||
skills=skills,
|
||||
agents=[agent],
|
||||
kb_context=kb_context,
|
||||
user_content=node_input,
|
||||
auto_discovered=False,
|
||||
)
|
||||
ai_model, api_url, api_key = await _resolve_model_and_endpoint(agent.get("model_id"), db)
|
||||
tool_policy = _resolve_agent_tool_policy([agent])
|
||||
|
||||
messages = [
|
||||
{"role": "system", "content": system_prompt},
|
||||
{"role": "user", "content": node_input},
|
||||
]
|
||||
called_skills: list = []
|
||||
tool_call_count = 0
|
||||
max_tool_rounds = 5
|
||||
last_content = ""
|
||||
async with httpx.AsyncClient(
|
||||
timeout=ai_model.timeout or 120,
|
||||
proxy=_proxy_for_url(api_url),
|
||||
trust_env=False,
|
||||
) as client:
|
||||
for round_num in range(max_tool_rounds):
|
||||
request_body = {"model": ai_model.name, "messages": messages, "max_tokens": 2048}
|
||||
if function_definitions:
|
||||
request_body["tools"] = [
|
||||
{"type": "function", "function": fd} for fd in function_definitions
|
||||
]
|
||||
request_body["tool_choice"] = "auto"
|
||||
resp = await client.post(
|
||||
api_url,
|
||||
headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
|
||||
json=request_body,
|
||||
)
|
||||
if resp.status_code != 200:
|
||||
raise OrchestrationError(f"local 节点 LLM 失败 HTTP {resp.status_code}: {resp.text[:200]}")
|
||||
data = resp.json()
|
||||
message = data["choices"][0]["message"]
|
||||
tool_calls = message.get("tool_calls") or []
|
||||
content = message.get("content") or ""
|
||||
last_content = content
|
||||
if not tool_calls or not function_definitions:
|
||||
return {"text": content, "used_tools": called_skills}
|
||||
messages.append({"role": "assistant", "content": content, "tool_calls": tool_calls})
|
||||
round_tool_count = 0
|
||||
for tc in tool_calls:
|
||||
func_name = tc.get("function", {}).get("name", "")
|
||||
check = _check_tool_policy(
|
||||
func_name, None, skills, tool_policy, tool_call_count, round_tool_count
|
||||
)
|
||||
if not check["allowed"]:
|
||||
messages.append({
|
||||
"role": "tool",
|
||||
"tool_call_id": tc.get("id", f"call_{round_num}"),
|
||||
"content": f"[工具调用已被策略拦截] {check.get('message', check.get('reason'))}",
|
||||
})
|
||||
tool_call_count += 1
|
||||
round_tool_count += 1
|
||||
continue
|
||||
tool_call_count += 1
|
||||
round_tool_count += 1
|
||||
try:
|
||||
func_args = json.loads(tc.get("function", {}).get("arguments", "{}") or "{}")
|
||||
except Exception:
|
||||
func_args = {}
|
||||
res = _execute_skill_tool(db, func_name, func_args, skills, node_input)
|
||||
called_skills.append(func_name)
|
||||
messages.append({
|
||||
"role": "tool",
|
||||
"tool_call_id": tc.get("id", f"call_{round_num}"),
|
||||
"content": res,
|
||||
})
|
||||
return {"text": last_content, "used_tools": called_skills}
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 单节点执行（含独立 DB 会话，保证并发安全）
|
||||
# ---------------------------------------------------------------------------
|
||||
async def _run_node(node, graph, blackboard, query, user_id, conv_id, trace_id) -> dict:
|
||||
from app.api.chat_skill import _resolve_market_agents
|
||||
from app.database import SessionLocal
|
||||
|
||||
nid = node["id"]
|
||||
ntype = node["type"]
|
||||
preds = _predecessors(graph, nid)
|
||||
pred_results = [blackboard[p] for p in preds]
|
||||
is_entry = not preds
|
||||
|
||||
if pred_results:
|
||||
parts = []
|
||||
for b in pred_results:
|
||||
label = f"节点 {b.get('node_id')}/{b.get('node_type')}"
|
||||
if b.get("ok"):
|
||||
parts.append(f"【{label}】\n{b.get('output', '')}")
|
||||
else:
|
||||
parts.append(f"【{label} 执行失败】\n错误: {b.get('error', '')}")
|
||||
upstream = "\n\n".join(parts)
|
||||
else:
|
||||
upstream = ""
|
||||
|
||||
with SessionLocal() as db:
|
||||
if ntype in ("llm", "reduce"):
|
||||
if ntype == "llm":
|
||||
sys_prompt = node.get("system_prompt") or (
|
||||
"你是一个处理节点，请根据上游输入完成指定的文本变换、摘要或路由判断，"
|
||||
"只输出处理后的结果，不要多余解释。"
|
||||
)
|
||||
else:
|
||||
sys_prompt = node.get("system_prompt") or (
|
||||
"你是一个结果汇总器，请将多个子任务的输出综合成连贯、完整、面向用户的最终回答。"
|
||||
)
|
||||
uc = _assemble_node_input(query, upstream, is_entry)
|
||||
text = await _llm_complete(sys_prompt, uc, node.get("model_id"), db, user_id)
|
||||
return {"ok": True, "node_id": nid, "node_type": ntype, "output": text}
|
||||
|
||||
if ntype == "local":
|
||||
agents = _resolve_market_agents(db, [node["agent_id"]], user_id)
|
||||
if not agents:
|
||||
raise OrchestrationError(f"local 节点 {nid} 引用的 agent_id={node['agent_id']} 不存在")
|
||||
ni = _assemble_node_input(query, upstream, is_entry)
|
||||
res = await _run_local_agent_turn(agents[0], ni, db, user_id, conv_id)
|
||||
return {
|
||||
"ok": True,
|
||||
"node_id": nid,
|
||||
"node_type": "local",
|
||||
"output": res["text"],
|
||||
"used_tools": res.get("used_tools"),
|
||||
}
|
||||
|
||||
if ntype == "external":
|
||||
from app.api.chat_agents import _call_external_agent
|
||||
|
||||
agents = _resolve_market_agents(db, [node["agent_id"]], user_id)
|
||||
if not agents:
|
||||
raise OrchestrationError(f"external 节点 {nid} 引用的 agent_id={node['agent_id']} 不存在")
|
||||
ni = _assemble_node_input(query, upstream, is_entry)
|
||||
answer, _ = await _call_external_agent(
|
||||
db, agents[0], query=ni, user_id=user_id, ext_conversation_id=None, history=[]
|
||||
)
|
||||
return {"ok": True, "node_id": nid, "node_type": "external", "output": answer or ""}
|
||||
|
||||
raise OrchestrationError(f"未知节点类型 {ntype}")
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 入口
|
||||
# ---------------------------------------------------------------------------
|
||||
async def run_graph(graph, query, db, user_id, conv_id, trace_id=None):
|
||||
"""执行编排图，返回 (answer: str, trace: dict)。
|
||||
|
||||
- 拓扑分层后逐层 asyncio.gather 并行执行。
|
||||
- 节点异常按 on_error=halt|continue 处理（默认 continue，失败时下游拿到错误作为输入）。
|
||||
"""
|
||||
validate_graph(graph)
|
||||
layers = topological_layers(graph)
|
||||
blackboard: dict = {}
|
||||
trace = {"nodes": [], "trace_id": trace_id, "layers": layers}
|
||||
|
||||
for layer in layers:
|
||||
tasks = [
|
||||
_run_node(_node_by_id(graph, nid), graph, blackboard, query, user_id, conv_id, trace_id)
|
||||
for nid in layer
|
||||
]
|
||||
results = await asyncio.gather(*tasks, return_exceptions=True)
|
||||
for nid, res in zip(layer, results):
|
||||
node = _node_by_id(graph, nid)
|
||||
if isinstance(res, Exception):
|
||||
on_error = node.get("on_error", "continue")
|
||||
err_msg = str(res)
|
||||
blackboard[nid] = {
|
||||
"ok": False,
|
||||
"node_id": nid,
|
||||
"node_type": node["type"],
|
||||
"error": err_msg,
|
||||
}
|
||||
trace["nodes"].append({"id": nid, "type": node["type"], "ok": False, "error": err_msg})
|
||||
logger.warning(f"[orchestrator] 节点 {nid} 失败 on_error={on_error}: {err_msg}")
|
||||
if on_error == "halt":
|
||||
raise OrchestrationError(f"节点 {nid} 执行失败(halt): {err_msg}")
|
||||
else:
|
||||
blackboard[nid] = res
|
||||
trace["nodes"].append({
|
||||
"id": nid,
|
||||
"type": res.get("node_type"),
|
||||
"ok": True,
|
||||
"used_tools": res.get("used_tools"),
|
||||
})
|
||||
|
||||
output_node = _find_output_node(graph)
|
||||
final = blackboard.get(output_node, {})
|
||||
if final.get("ok"):
|
||||
answer = final.get("output", "")
|
||||
else:
|
||||
answer = f"⚠️ 编排输出节点 {output_node} 执行失败: {final.get('error', '未知错误')}"
|
||||
trace["output_node"] = output_node
|
||||
return answer, trace
|
||||
"""app/core/orchestrator.py — Phase 3.3 多 Agent 编排图引擎。
|
||||
|
||||
设计目标：
|
||||
- 以轻量 DAG 描述多 Agent 协作：节点(llm/local/external/reduce) + 有向边(数据流)。
|
||||
- 拓扑分层后按层并行执行(asyncio.gather)，互相独立的节点真正并发。
|
||||
- 失败隔离：每个节点异常不影响其它分支；per-node on_error=halt|continue。
|
||||
- 节点执行复用既有原语(_build_system_prompt / _execute_skill_tool / _call_external_agent /
|
||||
3.1 工具沙箱)，完全向后兼容——agent.orchestration 为 null 时 chat_service 走原逻辑。
|
||||
- 并发安全：每个节点用独立 SessionLocal 会话做 DB 读，避免共享请求会话跨协程竞态。
|
||||
"""
|
||||
from __future__ import annotations
|
||||
|
||||
import asyncio
|
||||
import json
|
||||
import logging
|
||||
from typing import Any, Optional
|
||||
|
||||
import httpx
|
||||
|
||||
logger = logging.getLogger("aiflow.orchestrator")
|
||||
|
||||
VALID_NODE_TYPES = ("llm", "local", "external", "reduce")
|
||||
|
||||
|
||||
class OrchestrationError(ValueError):
|
||||
"""图结构非法或执行被 halt。"""
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 校验 + 拓扑
|
||||
# ---------------------------------------------------------------------------
|
||||
def validate_graph(graph: Any) -> None:
|
||||
if not isinstance(graph, dict):
|
||||
raise OrchestrationError("orchestration 必须为对象")
|
||||
nodes = graph.get("nodes")
|
||||
edges = graph.get("edges")
|
||||
if not isinstance(nodes, list) or not nodes:
|
||||
raise OrchestrationError("nodes 必须为非空数组")
|
||||
if not isinstance(edges, list):
|
||||
raise OrchestrationError("edges 必须为数组")
|
||||
|
||||
ids = set()
|
||||
for i, n in enumerate(nodes):
|
||||
if not isinstance(n, dict):
|
||||
raise OrchestrationError(f"节点 #{i} 必须为对象")
|
||||
nid = n.get("id")
|
||||
if not nid or not isinstance(nid, str):
|
||||
raise OrchestrationError(f"节点 #{i} 缺少字符串 id")
|
||||
if nid in ids:
|
||||
raise OrchestrationError(f"节点 id 重复: {nid}")
|
||||
ids.add(nid)
|
||||
ntype = n.get("type")
|
||||
if ntype not in VALID_NODE_TYPES:
|
||||
raise OrchestrationError(
|
||||
f"节点 {nid} 类型非法: {ntype} (允许 {', '.join(VALID_NODE_TYPES)})"
|
||||
)
|
||||
if ntype in ("local", "external") and not n.get("agent_id"):
|
||||
raise OrchestrationError(f"节点 {nid} 类型 {ntype} 必须指定 agent_id")
|
||||
|
||||
for i, e in enumerate(edges):
|
||||
if not isinstance(e, dict):
|
||||
raise OrchestrationError(f"边 #{i} 必须为对象")
|
||||
f = e.get("from")
|
||||
t = e.get("to")
|
||||
if f not in ids:
|
||||
raise OrchestrationError(f"边 #{i} 的 from={f} 指向不存在的节点")
|
||||
if t not in ids:
|
||||
raise OrchestrationError(f"边 #{i} 的 to={t} 指向不存在的节点")
|
||||
if f == t:
|
||||
raise OrchestrationError(f"边 #{i} 存在自环: {f}")
|
||||
|
||||
|
||||
def topological_layers(graph: dict) -> list:
|
||||
"""Kahn 算法返回分层(每层内节点可并行)；含环则抛错。"""
|
||||
nodes = graph["nodes"]
|
||||
edges = graph.get("edges", [])
|
||||
indeg = {n["id"]: 0 for n in nodes}
|
||||
adj = {n["id"]: [] for n in nodes}
|
||||
for e in edges:
|
||||
f, t = e["from"], e["to"]
|
||||
adj[f].append(t)
|
||||
indeg[t] += 1
|
||||
order = {n["id"]: i for i, n in enumerate(nodes)}
|
||||
layers: list = []
|
||||
remaining = set(indeg)
|
||||
while remaining:
|
||||
layer = sorted([nid for nid in remaining if indeg[nid] == 0], key=lambda x: order[x])
|
||||
if not layer:
|
||||
cycle = sorted(remaining, key=lambda x: order[x])
|
||||
raise OrchestrationError(f"检测到环路，无法拓扑排序: {cycle}")
|
||||
layers.append(layer)
|
||||
for nid in layer:
|
||||
remaining.discard(nid)
|
||||
for nxt in adj[nid]:
|
||||
indeg[nxt] -= 1
|
||||
return layers
|
||||
|
||||
|
||||
def _node_by_id(graph: dict, nid: str) -> dict:
|
||||
for n in graph["nodes"]:
|
||||
if n["id"] == nid:
|
||||
return n
|
||||
return {}
|
||||
|
||||
|
||||
def _predecessors(graph: dict, nid: str) -> list:
|
||||
return [e["from"] for e in graph.get("edges", []) if e["to"] == nid]
|
||||
|
||||
|
||||
def _successors(graph: dict, nid: str) -> list:
|
||||
return [e["to"] for e in graph.get("edges", []) if e["from"] == nid]
|
||||
|
||||
|
||||
def _find_output_node(graph: dict) -> str:
|
||||
if graph.get("output_node"):
|
||||
return graph["output_node"]
|
||||
sinks = [n["id"] for n in graph["nodes"] if not _successors(graph, n["id"])]
|
||||
if len(sinks) == 1:
|
||||
return sinks[0]
|
||||
for s in sinks:
|
||||
if _node_by_id(graph, s).get("type") == "reduce":
|
||||
return s
|
||||
if sinks:
|
||||
return sinks[0]
|
||||
return graph["nodes"][-1]["id"]
|
||||
|
||||
|
||||
def _assemble_node_input(query: str, upstream: str, is_entry: bool) -> str:
|
||||
"""入口节点直接吃用户原问题；下游节点综合原问题 + 上游输出。"""
|
||||
if is_entry:
|
||||
return query or ""
|
||||
if upstream:
|
||||
return f"原始用户问题:\n{query}\n\n上游节点输出:\n{upstream}"
|
||||
return query or ""
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 模型解析 + LLM 调用原语
|
||||
# ---------------------------------------------------------------------------
|
||||
async def _resolve_model_and_endpoint(model_id, db):
|
||||
from app.models import AIModel, ModelProvider
|
||||
from ..capability_registry import resolve_auto_model
|
||||
|
||||
ai_model = None
|
||||
if model_id and model_id != "auto":
|
||||
ai_model = db.query(AIModel).filter(AIModel.id == model_id).first()
|
||||
if not ai_model:
|
||||
ai_model = resolve_auto_model("text", db)
|
||||
if not ai_model:
|
||||
raise OrchestrationError("无法解析到可用模型")
|
||||
provider = db.query(ModelProvider).filter(ModelProvider.id == ai_model.provider_id).first()
|
||||
api_key = ai_model.api_key or (provider.api_key if provider else None)
|
||||
api_url = ai_model.endpoint
|
||||
if not api_url and provider and provider.api_base_url:
|
||||
api_url = f"{provider.api_base_url.rstrip('/')}/v1/chat/completions"
|
||||
if not api_url or not api_key:
|
||||
raise OrchestrationError(f"节点模型 {ai_model.name} 缺少 endpoint 或 api_key")
|
||||
return ai_model, api_url, api_key
|
||||
|
||||
|
||||
async def _llm_complete(system_prompt: str, user_content: str, model_id, db, user_id) -> str:
|
||||
from app.api.chat_utils import _proxy_for_url
|
||||
from app.core.observability import span
|
||||
|
||||
ai_model, api_url, api_key = await _resolve_model_and_endpoint(model_id, db)
|
||||
messages = [
|
||||
{"role": "system", "content": system_prompt},
|
||||
{"role": "user", "content": user_content},
|
||||
]
|
||||
request_body = {"model": ai_model.name, "messages": messages, "max_tokens": 2048}
|
||||
async with span("orchestrator.llm", model=ai_model.id):
|
||||
async with httpx.AsyncClient(
|
||||
timeout=ai_model.timeout or 120,
|
||||
proxy=_proxy_for_url(api_url),
|
||||
trust_env=False,
|
||||
) as client:
|
||||
resp = await client.post(
|
||||
api_url,
|
||||
headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
|
||||
json=request_body,
|
||||
)
|
||||
if resp.status_code != 200:
|
||||
raise OrchestrationError(f"LLM 调用失败 HTTP {resp.status_code}: {resp.text[:200]}")
|
||||
data = resp.json()
|
||||
return data["choices"][0]["message"].get("content") or ""
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# local 节点：复用单 Agent system prompt + 受限 tool round + 3.1 沙箱
|
||||
# ---------------------------------------------------------------------------
|
||||
async def _run_local_agent_turn(agent: dict, node_input: str, db, user_id, conv_id) -> dict:
|
||||
from app.api.chat_skill import _resolve_market_skills, _execute_skill_tool
|
||||
from app.api.chat_context import _build_system_prompt, _retrieve_kb_context
|
||||
from app.api.chat_service import _resolve_agent_tool_policy, _check_tool_policy
|
||||
from app.api.chat_utils import _proxy_for_url
|
||||
|
||||
skills = _resolve_market_skills(db, agent.get("skill_ids") or [], user_id)
|
||||
kb_context = await _retrieve_kb_context(db, agent.get("knowledge_base_ids") or [], node_input)
|
||||
system_prompt, function_definitions = _build_system_prompt(
|
||||
skills=skills,
|
||||
agents=[agent],
|
||||
kb_context=kb_context,
|
||||
user_content=node_input,
|
||||
auto_discovered=False,
|
||||
)
|
||||
ai_model, api_url, api_key = await _resolve_model_and_endpoint(agent.get("model_id"), db)
|
||||
tool_policy = _resolve_agent_tool_policy([agent])
|
||||
|
||||
messages = [
|
||||
{"role": "system", "content": system_prompt},
|
||||
{"role": "user", "content": node_input},
|
||||
]
|
||||
called_skills: list = []
|
||||
tool_call_count = 0
|
||||
max_tool_rounds = 5
|
||||
last_content = ""
|
||||
async with httpx.AsyncClient(
|
||||
timeout=ai_model.timeout or 120,
|
||||
proxy=_proxy_for_url(api_url),
|
||||
trust_env=False,
|
||||
) as client:
|
||||
for round_num in range(max_tool_rounds):
|
||||
request_body = {"model": ai_model.name, "messages": messages, "max_tokens": 2048}
|
||||
if function_definitions:
|
||||
request_body["tools"] = [
|
||||
{"type": "function", "function": fd} for fd in function_definitions
|
||||
]
|
||||
request_body["tool_choice"] = "auto"
|
||||
resp = await client.post(
|
||||
api_url,
|
||||
headers={"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"},
|
||||
json=request_body,
|
||||
)
|
||||
if resp.status_code != 200:
|
||||
raise OrchestrationError(f"local 节点 LLM 失败 HTTP {resp.status_code}: {resp.text[:200]}")
|
||||
data = resp.json()
|
||||
message = data["choices"][0]["message"]
|
||||
tool_calls = message.get("tool_calls") or []
|
||||
content = message.get("content") or ""
|
||||
last_content = content
|
||||
if not tool_calls or not function_definitions:
|
||||
return {"text": content, "used_tools": called_skills}
|
||||
messages.append({"role": "assistant", "content": content, "tool_calls": tool_calls})
|
||||
round_tool_count = 0
|
||||
for tc in tool_calls:
|
||||
func_name = tc.get("function", {}).get("name", "")
|
||||
check = _check_tool_policy(
|
||||
func_name, None, skills, tool_policy, tool_call_count, round_tool_count
|
||||
)
|
||||
if not check["allowed"]:
|
||||
messages.append({
|
||||
"role": "tool",
|
||||
"tool_call_id": tc.get("id", f"call_{round_num}"),
|
||||
"content": f"[工具调用已被策略拦截] {check.get('message', check.get('reason'))}",
|
||||
})
|
||||
tool_call_count += 1
|
||||
round_tool_count += 1
|
||||
continue
|
||||
tool_call_count += 1
|
||||
round_tool_count += 1
|
||||
try:
|
||||
func_args = json.loads(tc.get("function", {}).get("arguments", "{}") or "{}")
|
||||
except Exception:
|
||||
func_args = {}
|
||||
res = _execute_skill_tool(db, func_name, func_args, skills, node_input)
|
||||
called_skills.append(func_name)
|
||||
messages.append({
|
||||
"role": "tool",
|
||||
"tool_call_id": tc.get("id", f"call_{round_num}"),
|
||||
"content": res,
|
||||
})
|
||||
return {"text": last_content, "used_tools": called_skills}
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 单节点执行（含独立 DB 会话，保证并发安全）
|
||||
# ---------------------------------------------------------------------------
|
||||
async def _run_node(node, graph, blackboard, query, user_id, conv_id, trace_id) -> dict:
|
||||
from app.api.chat_skill import _resolve_market_agents
|
||||
from app.database import SessionLocal
|
||||
|
||||
nid = node["id"]
|
||||
ntype = node["type"]
|
||||
preds = _predecessors(graph, nid)
|
||||
pred_results = [blackboard[p] for p in preds]
|
||||
is_entry = not preds
|
||||
|
||||
if pred_results:
|
||||
parts = []
|
||||
for b in pred_results:
|
||||
label = f"节点 {b.get('node_id')}/{b.get('node_type')}"
|
||||
if b.get("ok"):
|
||||
parts.append(f"【{label}】\n{b.get('output', '')}")
|
||||
else:
|
||||
parts.append(f"【{label} 执行失败】\n错误: {b.get('error', '')}")
|
||||
upstream = "\n\n".join(parts)
|
||||
else:
|
||||
upstream = ""
|
||||
|
||||
with SessionLocal() as db:
|
||||
if ntype in ("llm", "reduce"):
|
||||
if ntype == "llm":
|
||||
sys_prompt = node.get("system_prompt") or (
|
||||
"你是一个处理节点，请根据上游输入完成指定的文本变换、摘要或路由判断，"
|
||||
"只输出处理后的结果，不要多余解释。"
|
||||
)
|
||||
else:
|
||||
sys_prompt = node.get("system_prompt") or (
|
||||
"你是一个结果汇总器，请将多个子任务的输出综合成连贯、完整、面向用户的最终回答。"
|
||||
)
|
||||
uc = _assemble_node_input(query, upstream, is_entry)
|
||||
text = await _llm_complete(sys_prompt, uc, node.get("model_id"), db, user_id)
|
||||
return {"ok": True, "node_id": nid, "node_type": ntype, "output": text}
|
||||
|
||||
if ntype == "local":
|
||||
agents = _resolve_market_agents(db, [node["agent_id"]], user_id)
|
||||
if not agents:
|
||||
raise OrchestrationError(f"local 节点 {nid} 引用的 agent_id={node['agent_id']} 不存在")
|
||||
ni = _assemble_node_input(query, upstream, is_entry)
|
||||
res = await _run_local_agent_turn(agents[0], ni, db, user_id, conv_id)
|
||||
return {
|
||||
"ok": True,
|
||||
"node_id": nid,
|
||||
"node_type": "local",
|
||||
"output": res["text"],
|
||||
"used_tools": res.get("used_tools"),
|
||||
}
|
||||
|
||||
if ntype == "external":
|
||||
from app.api.chat_agents import _call_external_agent
|
||||
|
||||
agents = _resolve_market_agents(db, [node["agent_id"]], user_id)
|
||||
if not agents:
|
||||
raise OrchestrationError(f"external 节点 {nid} 引用的 agent_id={node['agent_id']} 不存在")
|
||||
ni = _assemble_node_input(query, upstream, is_entry)
|
||||
answer, _ = await _call_external_agent(
|
||||
db, agents[0], query=ni, user_id=user_id, ext_conversation_id=None, history=[]
|
||||
)
|
||||
return {"ok": True, "node_id": nid, "node_type": "external", "output": answer or ""}
|
||||
|
||||
raise OrchestrationError(f"未知节点类型 {ntype}")
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# 入口
|
||||
# ---------------------------------------------------------------------------
|
||||
async def run_graph(graph, query, db, user_id, conv_id, trace_id=None):
|
||||
"""执行编排图，返回 (answer: str, trace: dict)。
|
||||
|
||||
- 拓扑分层后逐层 asyncio.gather 并行执行。
|
||||
- 节点异常按 on_error=halt|continue 处理（默认 continue，失败时下游拿到错误作为输入）。
|
||||
"""
|
||||
validate_graph(graph)
|
||||
layers = topological_layers(graph)
|
||||
blackboard: dict = {}
|
||||
trace = {"nodes": [], "trace_id": trace_id, "layers": layers}
|
||||
|
||||
for layer in layers:
|
||||
tasks = [
|
||||
_run_node(_node_by_id(graph, nid), graph, blackboard, query, user_id, conv_id, trace_id)
|
||||
for nid in layer
|
||||
]
|
||||
results = await asyncio.gather(*tasks, return_exceptions=True)
|
||||
for nid, res in zip(layer, results):
|
||||
node = _node_by_id(graph, nid)
|
||||
if isinstance(res, Exception):
|
||||
on_error = node.get("on_error", "continue")
|
||||
err_msg = str(res)
|
||||
blackboard[nid] = {
|
||||
"ok": False,
|
||||
"node_id": nid,
|
||||
"node_type": node["type"],
|
||||
"error": err_msg,
|
||||
}
|
||||
trace["nodes"].append({"id": nid, "type": node["type"], "ok": False, "error": err_msg})
|
||||
logger.warning(f"[orchestrator] 节点 {nid} 失败 on_error={on_error}: {err_msg}")
|
||||
if on_error == "halt":
|
||||
raise OrchestrationError(f"节点 {nid} 执行失败(halt): {err_msg}")
|
||||
else:
|
||||
blackboard[nid] = res
|
||||
trace["nodes"].append({
|
||||
"id": nid,
|
||||
"type": res.get("node_type"),
|
||||
"ok": True,
|
||||
"used_tools": res.get("used_tools"),
|
||||
})
|
||||
|
||||
output_node = _find_output_node(graph)
|
||||
final = blackboard.get(output_node, {})
|
||||
if final.get("ok"):
|
||||
answer = final.get("output", "")
|
||||
else:
|
||||
answer = f"⚠️ 编排输出节点 {output_node} 执行失败: {final.get('error', '未知错误')}"
|
||||
trace["output_node"] = output_node
|
||||
return answer, trace
|
||||
+131
-131
@@ -1,131 +1,131 @@
|
||||
"""
|
||||
OpenTelemetry 桥接 (Phase 1.I 标准化)
|
||||
=====================================
|
||||
设计基准（标准对照表见 STANDARDS_AUDIT_2026-07-24.md §I）：
|
||||
- 标准：OpenTelemetry SDK + OTLP gRPC exporter —— 云原生可观测事实标准。
|
||||
- traces：span 经 observability.span() 嵌套导出（见 observability.py）。
|
||||
- metrics：MeterProvider 经 OTLP 导出 span 耗时直方图 + 调用计数，供 Grafana 出图。
|
||||
- 失败安全（生产红线）：若 OTLP 端点不可达 / 初始化异常，自动降级为 no-op，
|
||||
绝不阻断业务请求、不产生未捕获异常。
|
||||
- 门控：OTEL_ENABLED=1 默认开启；置 0 则零开销、不走 OTel。
|
||||
- 发射目标：OTEL_EXPORTER_OTLP_ENDPOINT，默认 http://qihu-aiflow-alloy:4317
|
||||
（经 Grafana Alloy 桥接 OTLP→Prometheus，再由 Prometheus 喂给 Grafana 出图）。
|
||||
|
||||
对外 API：
|
||||
init_otel() 首次调用时构建 TracerProvider + MeterProvider（幂等）
|
||||
get_tracer() 返回 tracer（未启用/失败则 None）
|
||||
get_meter() 返回 meter（未启用/失败则 None）
|
||||
current_span_cm(name) 返回一个 OTel span 的上下文管理器（供 observability 嵌套）
|
||||
record_span_metrics(name, elapsed_ms, status) 记录 span 指标（耗时直方图 + 计数）
|
||||
"""
|
||||
|
||||
import os
|
||||
import logging
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
_ENABLED = os.getenv("OTEL_ENABLED", "1") not in ("0", "false", "False", "")
|
||||
_ENDPOINT = os.getenv("OTEL_EXPORTER_OTLP_ENDPOINT", "http://qihu-aiflow-alloy:4317").rstrip("/")
|
||||
if not _ENDPOINT.startswith("http://") and not _ENDPOINT.startswith("https://"):
|
||||
_ENDPOINT = "http://" + _ENDPOINT
|
||||
|
||||
_tracer = None
|
||||
_meter = None
|
||||
_initialized = False
|
||||
_duration_hist = None
|
||||
_span_counter = None
|
||||
|
||||
|
||||
def init_otel() -> None:
|
||||
"""构建 TracerProvider + MeterProvider（幂等）。失败则降级为 no-op。"""
|
||||
global _tracer, _meter, _initialized
|
||||
if _initialized:
|
||||
return
|
||||
_initialized = True
|
||||
if not _ENABLED:
|
||||
logger.info("[otel] disabled (OTEL_ENABLED=0)")
|
||||
return
|
||||
try:
|
||||
from opentelemetry import trace, metrics
|
||||
from opentelemetry.sdk.resources import Resource
|
||||
from opentelemetry.semconv.resource import ResourceAttributes
|
||||
from opentelemetry.sdk.trace import TracerProvider
|
||||
from opentelemetry.sdk.trace.export import BatchSpanProcessor
|
||||
from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import (
|
||||
OTLPSpanExporter,
|
||||
)
|
||||
from opentelemetry.sdk.metrics import MeterProvider
|
||||
from opentelemetry.sdk.metrics.export import PeriodicExportingMetricReader
|
||||
from opentelemetry.exporter.otlp.proto.grpc.metric_exporter import (
|
||||
OTLPMetricExporter,
|
||||
)
|
||||
|
||||
resource = Resource.create({ResourceAttributes.SERVICE_NAME: "qihu-aiflow"})
|
||||
|
||||
# --- traces ---
|
||||
trace_provider = TracerProvider(resource=resource)
|
||||
trace_provider.add_span_processor(
|
||||
BatchSpanProcessor(OTLPSpanExporter(endpoint=_ENDPOINT, insecure=True))
|
||||
)
|
||||
trace.set_tracer_provider(trace_provider)
|
||||
_tracer = trace.get_tracer("qihu-aiflow")
|
||||
|
||||
# --- metrics ---
|
||||
metric_reader = PeriodicExportingMetricReader(
|
||||
OTLPMetricExporter(endpoint=_ENDPOINT, insecure=True)
|
||||
)
|
||||
meter_provider = MeterProvider(resource=resource, metric_readers=[metric_reader])
|
||||
metrics.set_meter_provider(meter_provider)
|
||||
_meter = metrics.get_meter("qihu-aiflow")
|
||||
|
||||
logger.info(f"[otel] tracer+meter ready -> {_ENDPOINT}")
|
||||
except Exception as e: # pragma: no cover - 降级保护
|
||||
logger.warning(f"[otel] init failed, degraded to no-op: {e}")
|
||||
_tracer = None
|
||||
_meter = None
|
||||
|
||||
|
||||
def get_tracer():
|
||||
"""返回 tracer；首次调用触发 init。未启用/失败返回 None。"""
|
||||
if _tracer is None and _ENABLED:
|
||||
init_otel()
|
||||
return _tracer
|
||||
|
||||
|
||||
def get_meter():
|
||||
"""返回 meter；首次调用触发 init。未启用/失败返回 None。"""
|
||||
if _meter is None and _ENABLED:
|
||||
init_otel()
|
||||
return _meter
|
||||
|
||||
|
||||
def current_span_cm(name: str, attributes: dict | None = None):
|
||||
"""返回一个 OTel span 的上下文管理器（同步/异步通用）。失败时返回 None。"""
|
||||
tr = get_tracer()
|
||||
if tr is None:
|
||||
return None
|
||||
try:
|
||||
return tr.start_as_current_span(name, attributes=attributes or {})
|
||||
except Exception as e: # pragma: no cover
|
||||
logger.warning(f"[otel] start_span failed: {e}")
|
||||
return None
|
||||
|
||||
|
||||
def record_span_metrics(name: str, elapsed_ms: float, status: str) -> None:
|
||||
"""记录 span 指标：耗时直方图 + 调用计数（供 Grafana 出图）。失败安全。"""
|
||||
m = get_meter()
|
||||
if m is None:
|
||||
return
|
||||
global _duration_hist, _span_counter
|
||||
try:
|
||||
if _duration_hist is None:
|
||||
_duration_hist = m.create_histogram(
|
||||
"aiflow_span_duration_ms", "Span duration in milliseconds"
|
||||
)
|
||||
_span_counter = m.create_counter("aiflow_span_count", "Span invocation count")
|
||||
attrs = {"span": name, "status": status}
|
||||
_duration_hist.record(elapsed_ms, attrs)
|
||||
_span_counter.add(1, attrs)
|
||||
except Exception: # pragma: no cover
|
||||
pass
|
||||
|
||||
"""
|
||||
OpenTelemetry 桥接 (Phase 1.I 标准化)
|
||||
=====================================
|
||||
设计基准（标准对照表见 STANDARDS_AUDIT_2026-07-24.md §I）：
|
||||
- 标准：OpenTelemetry SDK + OTLP gRPC exporter —— 云原生可观测事实标准。
|
||||
- traces：span 经 observability.span() 嵌套导出（见 observability.py）。
|
||||
- metrics：MeterProvider 经 OTLP 导出 span 耗时直方图 + 调用计数，供 Grafana 出图。
|
||||
- 失败安全（生产红线）：若 OTLP 端点不可达 / 初始化异常，自动降级为 no-op，
|
||||
绝不阻断业务请求、不产生未捕获异常。
|
||||
- 门控：OTEL_ENABLED=1 默认开启；置 0 则零开销、不走 OTel。
|
||||
- 发射目标：OTEL_EXPORTER_OTLP_ENDPOINT，默认 http://qihu-aiflow-alloy:4317
|
||||
（经 Grafana Alloy 桥接 OTLP→Prometheus，再由 Prometheus 喂给 Grafana 出图）。
|
||||
|
||||
对外 API：
|
||||
init_otel() 首次调用时构建 TracerProvider + MeterProvider（幂等）
|
||||
get_tracer() 返回 tracer（未启用/失败则 None）
|
||||
get_meter() 返回 meter（未启用/失败则 None）
|
||||
current_span_cm(name) 返回一个 OTel span 的上下文管理器（供 observability 嵌套）
|
||||
record_span_metrics(name, elapsed_ms, status) 记录 span 指标（耗时直方图 + 计数）
|
||||
"""
|
||||
|
||||
import os
|
||||
import logging
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
_ENABLED = os.getenv("OTEL_ENABLED", "1") not in ("0", "false", "False", "")
|
||||
_ENDPOINT = os.getenv("OTEL_EXPORTER_OTLP_ENDPOINT", "http://qihu-aiflow-alloy:4317").rstrip("/")
|
||||
if not _ENDPOINT.startswith("http://") and not _ENDPOINT.startswith("https://"):
|
||||
_ENDPOINT = "http://" + _ENDPOINT
|
||||
|
||||
_tracer = None
|
||||
_meter = None
|
||||
_initialized = False
|
||||
_duration_hist = None
|
||||
_span_counter = None
|
||||
|
||||
|
||||
def init_otel() -> None:
|
||||
"""构建 TracerProvider + MeterProvider（幂等）。失败则降级为 no-op。"""
|
||||
global _tracer, _meter, _initialized
|
||||
if _initialized:
|
||||
return
|
||||
_initialized = True
|
||||
if not _ENABLED:
|
||||
logger.info("[otel] disabled (OTEL_ENABLED=0)")
|
||||
return
|
||||
try:
|
||||
from opentelemetry import trace, metrics
|
||||
from opentelemetry.sdk.resources import Resource
|
||||
from opentelemetry.semconv.resource import ResourceAttributes
|
||||
from opentelemetry.sdk.trace import TracerProvider
|
||||
from opentelemetry.sdk.trace.export import BatchSpanProcessor
|
||||
from opentelemetry.exporter.otlp.proto.grpc.trace_exporter import (
|
||||
OTLPSpanExporter,
|
||||
)
|
||||
from opentelemetry.sdk.metrics import MeterProvider
|
||||
from opentelemetry.sdk.metrics.export import PeriodicExportingMetricReader
|
||||
from opentelemetry.exporter.otlp.proto.grpc.metric_exporter import (
|
||||
OTLPMetricExporter,
|
||||
)
|
||||
|
||||
resource = Resource.create({ResourceAttributes.SERVICE_NAME: "qihu-aiflow"})
|
||||
|
||||
# --- traces ---
|
||||
trace_provider = TracerProvider(resource=resource)
|
||||
trace_provider.add_span_processor(
|
||||
BatchSpanProcessor(OTLPSpanExporter(endpoint=_ENDPOINT, insecure=True))
|
||||
)
|
||||
trace.set_tracer_provider(trace_provider)
|
||||
_tracer = trace.get_tracer("qihu-aiflow")
|
||||
|
||||
# --- metrics ---
|
||||
metric_reader = PeriodicExportingMetricReader(
|
||||
OTLPMetricExporter(endpoint=_ENDPOINT, insecure=True)
|
||||
)
|
||||
meter_provider = MeterProvider(resource=resource, metric_readers=[metric_reader])
|
||||
metrics.set_meter_provider(meter_provider)
|
||||
_meter = metrics.get_meter("qihu-aiflow")
|
||||
|
||||
logger.info(f"[otel] tracer+meter ready -> {_ENDPOINT}")
|
||||
except Exception as e: # pragma: no cover - 降级保护
|
||||
logger.warning(f"[otel] init failed, degraded to no-op: {e}")
|
||||
_tracer = None
|
||||
_meter = None
|
||||
|
||||
|
||||
def get_tracer():
|
||||
"""返回 tracer；首次调用触发 init。未启用/失败返回 None。"""
|
||||
if _tracer is None and _ENABLED:
|
||||
init_otel()
|
||||
return _tracer
|
||||
|
||||
|
||||
def get_meter():
|
||||
"""返回 meter；首次调用触发 init。未启用/失败返回 None。"""
|
||||
if _meter is None and _ENABLED:
|
||||
init_otel()
|
||||
return _meter
|
||||
|
||||
|
||||
def current_span_cm(name: str, attributes: dict | None = None):
|
||||
"""返回一个 OTel span 的上下文管理器（同步/异步通用）。失败时返回 None。"""
|
||||
tr = get_tracer()
|
||||
if tr is None:
|
||||
return None
|
||||
try:
|
||||
return tr.start_as_current_span(name, attributes=attributes or {})
|
||||
except Exception as e: # pragma: no cover
|
||||
logger.warning(f"[otel] start_span failed: {e}")
|
||||
return None
|
||||
|
||||
|
||||
def record_span_metrics(name: str, elapsed_ms: float, status: str) -> None:
|
||||
"""记录 span 指标：耗时直方图 + 调用计数（供 Grafana 出图）。失败安全。"""
|
||||
m = get_meter()
|
||||
if m is None:
|
||||
return
|
||||
global _duration_hist, _span_counter
|
||||
try:
|
||||
if _duration_hist is None:
|
||||
_duration_hist = m.create_histogram(
|
||||
"aiflow_span_duration_ms", "Span duration in milliseconds"
|
||||
)
|
||||
_span_counter = m.create_counter("aiflow_span_count", "Span invocation count")
|
||||
attrs = {"span": name, "status": status}
|
||||
_duration_hist.record(elapsed_ms, attrs)
|
||||
_span_counter.add(1, attrs)
|
||||
except Exception: # pragma: no cover
|
||||
pass
|
||||
|
||||
+271
-271
@@ -1,271 +1,271 @@
|
||||
"""
|
||||
知识库路由层
|
||||
============
|
||||
将检索请求智能路由到本地BM25、向量检索、或外部KB服务(Dify)。
|
||||
根据知识库配置自动选择最优检索策略。
|
||||
|
||||
使用方式:
|
||||
from app.kb_router import kb_router_search
|
||||
results = await kb_router_search(query, kb_id, top_k=5)
|
||||
"""
|
||||
|
||||
import logging
|
||||
from typing import List, Dict, Optional
|
||||
|
||||
logger = logging.getLogger("aiflow.kb_router")
|
||||
|
||||
|
||||
async def _get_external_services(db_session, enabled_only: bool = True) -> List[Dict]:
|
||||
"""获取已配置的外部KB服务（仅 dify）"""
|
||||
from sqlalchemy import text
|
||||
try:
|
||||
query = "SELECT * FROM external_kb_services"
|
||||
if enabled_only:
|
||||
query += " WHERE is_enabled = 1"
|
||||
query += " ORDER BY created_at DESC"
|
||||
result = db_session.execute(text(query))
|
||||
columns = result.keys()
|
||||
return [dict(zip(columns, row)) for row in result.fetchall()]
|
||||
except Exception as e:
|
||||
logger.error(f"Failed to get external services: {e}")
|
||||
return []
|
||||
|
||||
|
||||
async def _search_dify(service: Dict, query: str, top_k: int = 5) -> List[Dict]:
|
||||
"""检索 Dify 知识库（走 app.dify_kb 客户端）"""
|
||||
try:
|
||||
from app.dify_kb import dify_retrieve
|
||||
api_url = service.get("api_url", "")
|
||||
api_key = service.get("api_key", "")
|
||||
dataset_name = None
|
||||
cfg = service.get("config")
|
||||
if isinstance(cfg, str):
|
||||
try:
|
||||
import json
|
||||
cfg = json.loads(cfg)
|
||||
except Exception:
|
||||
cfg = None
|
||||
if isinstance(cfg, dict):
|
||||
dataset_name = cfg.get("dataset_name")
|
||||
hits = dify_retrieve(api_url, api_key, query, top_k=top_k, dataset_name=dataset_name)
|
||||
return hits
|
||||
except Exception as e:
|
||||
logger.error(f"Dify search error: {e}")
|
||||
return []
|
||||
|
||||
|
||||
async def _search_external(service: Dict, query: str, top_k: int = 5) -> List[Dict]:
|
||||
"""路由到对应的外部KB服务（当前仅支持 dify）"""
|
||||
service_type = service.get("service_type", "")
|
||||
if service_type == "dify":
|
||||
return await _search_dify(service, query, top_k)
|
||||
else:
|
||||
logger.warning(f"Unknown external KB type: {service_type}")
|
||||
return []
|
||||
|
||||
async def _get_web_services(db_session, enabled_only: bool = True) -> List[Dict]:
|
||||
"""获取全局 AnySearch 联网检索配置（从 site_settings 读取）。"""
|
||||
from app.anysearch_client import load_anysearch_config
|
||||
cfg = load_anysearch_config(db_session)
|
||||
if not cfg:
|
||||
return []
|
||||
if enabled_only and not cfg.get("is_enabled"):
|
||||
return []
|
||||
api_key = cfg.get("api_key", "")
|
||||
if not api_key:
|
||||
return []
|
||||
return [{
|
||||
"id": "anysearch_global",
|
||||
"service_type": "anysearch",
|
||||
"api_key": api_key,
|
||||
"api_url": cfg.get("api_url") or "https://api.anysearch.com/v1/search",
|
||||
"config": {
|
||||
"zone": cfg.get("zone", "cn"),
|
||||
"freshness": cfg.get("freshness"),
|
||||
"top_k": cfg.get("top_k", 5),
|
||||
},
|
||||
"is_enabled": cfg.get("is_enabled", False),
|
||||
}]
|
||||
|
||||
|
||||
async def _search_anysearch(service: Dict, query: str, top_k: int = 5) -> List[Dict]:
|
||||
"""调用 AnySearch 联网检索（走 app.anysearch_client）。"""
|
||||
try:
|
||||
from app.anysearch_client import anysearch_search
|
||||
api_key = service.get("api_key", "")
|
||||
api_url = service.get("api_url") or "https://api.anysearch.com/v1/search"
|
||||
cfg = service.get("config")
|
||||
if isinstance(cfg, str):
|
||||
try:
|
||||
import json
|
||||
cfg = json.loads(cfg)
|
||||
except Exception:
|
||||
cfg = None
|
||||
cfg = cfg or {}
|
||||
zone = cfg.get("zone", "cn")
|
||||
freshness = cfg.get("freshness")
|
||||
k = int(cfg.get("top_k", top_k))
|
||||
return await anysearch_search(api_key, api_url, query, top_k=k, zone=zone, freshness=freshness)
|
||||
except Exception as e:
|
||||
logger.error(f"AnySearch search error: {e}")
|
||||
return []
|
||||
|
||||
|
||||
async def kb_router_search(
|
||||
query: str,
|
||||
kb_id: Optional[str] = None,
|
||||
kb_ids: Optional[List[str]] = None,
|
||||
top_k: int = 5,
|
||||
bm25_results: Optional[List[Dict]] = None,
|
||||
db_session=None,
|
||||
use_vector: bool = True,
|
||||
use_external: bool = True,
|
||||
use_tree: bool = True,
|
||||
use_web: bool = False,
|
||||
) -> List[Dict]:
|
||||
"""
|
||||
知识库路由层 — 混合检索
|
||||
|
||||
检索策略:
|
||||
1. 本地 BM25 检索 (MySQL, 已有)
|
||||
2. 向量语义检索 (PostgreSQL+pgvector, 新增)
|
||||
3. 结构化分片段检索 (TreeSearch, 新增: 知识库目录树 + chunk 级精准命中)
|
||||
4. 外部KB检索 (Dify, 按配置)
|
||||
5. 联网检索 (AnySearch, 可选 use_web)
|
||||
|
||||
融合方式: Reciprocal Rank Fusion (RRF)
|
||||
|
||||
Args:
|
||||
query: 查询文本
|
||||
kb_id: 知识库ID
|
||||
top_k: 返回结果数
|
||||
bm25_results: BM25检索结果
|
||||
db_session: 数据库会话(用于获取外部KB配置与分片段检索)
|
||||
use_vector: 是否使用向量检索
|
||||
use_external: 是否使用外部KB检索
|
||||
use_tree: 是否使用 TreeSearch 结构化分片段检索
|
||||
use_web: 是否使用 AnySearch 联网检索
|
||||
|
||||
Returns:
|
||||
融合后的检索结果列表
|
||||
"""
|
||||
all_results = []
|
||||
|
||||
# 1. BM25 结果
|
||||
if bm25_results:
|
||||
for rank, item in enumerate(bm25_results):
|
||||
all_results.append({
|
||||
"content": item.get("content", item.get("text", "")),
|
||||
"score": item.get("score", 1.0 / (60 + rank + 1)),
|
||||
"source": "local_bm25",
|
||||
"rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
})
|
||||
|
||||
# 2. 向量检索
|
||||
if use_vector:
|
||||
try:
|
||||
from app.pgvector_search import vector_search
|
||||
kb_targets = kb_ids if kb_ids else ([kb_id] if kb_id else [None])
|
||||
vec_results = []
|
||||
for _kid in kb_targets:
|
||||
try:
|
||||
vec_results.extend(await vector_search(query, _kid, top_k))
|
||||
except Exception as e:
|
||||
logger.warning(f"Vector search failed for kb={_kid}: {e}")
|
||||
for rank, item in enumerate(vec_results):
|
||||
all_results.append({
|
||||
"content": item.get("content", ""),
|
||||
"score": item.get("score", 0.5),
|
||||
"source": "vector_pg",
|
||||
"rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
})
|
||||
except ImportError:
|
||||
logger.debug("pgvector_search not available, skipping vector search")
|
||||
except Exception as e:
|
||||
logger.warning(f"Vector search failed: {e}")
|
||||
|
||||
# 2.5 结构化分片段检索 (TreeSearch, Level 2)
|
||||
if use_tree and db_session is not None:
|
||||
try:
|
||||
from app.tree_search import tree_search
|
||||
kb_targets = kb_ids if kb_ids else ([kb_id] if kb_id else None)
|
||||
tree_hits = await tree_search(
|
||||
db_session, query, kb_ids=kb_targets, top_k=top_k, mode="hybrid",
|
||||
)
|
||||
for rank, item in enumerate(tree_hits):
|
||||
all_results.append({
|
||||
"content": item.get("content", ""),
|
||||
"score": item.get("score", 0.5),
|
||||
"source": "tree_chunk",
|
||||
"rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
})
|
||||
except Exception as e:
|
||||
logger.warning(f"TreeSearch failed: {e}")
|
||||
|
||||
# 3. 外部KB检索 (Dify)
|
||||
if use_external and db_session:
|
||||
services = await _get_external_services(db_session)
|
||||
for service in services:
|
||||
if service.get("status") == "connected":
|
||||
ext_results = await _search_external(service, query, top_k)
|
||||
for rank, item in enumerate(ext_results):
|
||||
all_results.append({
|
||||
"content": item.get("content", ""),
|
||||
"score": item.get("score", 0.5),
|
||||
"source": item.get("source", "external_dify"),
|
||||
"rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
})
|
||||
|
||||
# 4. 联网检索 (AnySearch, 可选)
|
||||
if use_web and db_session is not None:
|
||||
try:
|
||||
web_services = await _get_web_services(db_session)
|
||||
for svc in web_services:
|
||||
web_hits = await _search_anysearch(svc, query, top_k)
|
||||
for rank, item in enumerate(web_hits):
|
||||
all_results.append({
|
||||
"content": item.get("content", ""),
|
||||
"score": item.get("score", 0.6),
|
||||
"source": item.get("source", "web_anysearch"),
|
||||
"rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
})
|
||||
except Exception as e:
|
||||
logger.warning(f"AnySearch web search failed: {e}")
|
||||
|
||||
if not all_results:
|
||||
return []
|
||||
|
||||
# RRF 融合
|
||||
rrf_k = 60
|
||||
score_map = {}
|
||||
for item in all_results:
|
||||
content_key = item["content"][:200] # 用内容前200字符作为去重键
|
||||
rrf_score = 1.0 / (rrf_k + item.get("rank", 1))
|
||||
|
||||
if content_key in score_map:
|
||||
score_map[content_key]["score"] += rrf_score
|
||||
score_map[content_key]["sources"].append(item["source"])
|
||||
else:
|
||||
score_map[content_key] = {
|
||||
"content": item["content"],
|
||||
"score": rrf_score,
|
||||
"sources": [item["source"]],
|
||||
"metadata": item.get("metadata", {}),
|
||||
}
|
||||
|
||||
# 按融合分数排序
|
||||
merged = sorted(score_map.values(), key=lambda x: x["score"], reverse=True)
|
||||
|
||||
# 标准化分数到 0-1
|
||||
if merged:
|
||||
max_score = merged[0]["score"]
|
||||
for item in merged:
|
||||
item["score"] = item["score"] / max_score if max_score > 0 else 0
|
||||
item["sources"] = list(set(item["sources"]))
|
||||
|
||||
return merged[:top_k]
|
||||
"""
|
||||
知识库路由层
|
||||
============
|
||||
将检索请求智能路由到本地BM25、向量检索、或外部KB服务(Dify)。
|
||||
根据知识库配置自动选择最优检索策略。
|
||||
|
||||
使用方式:
|
||||
from app.kb_router import kb_router_search
|
||||
results = await kb_router_search(query, kb_id, top_k=5)
|
||||
"""
|
||||
|
||||
import logging
|
||||
from typing import List, Dict, Optional
|
||||
|
||||
logger = logging.getLogger("aiflow.kb_router")
|
||||
|
||||
|
||||
async def _get_external_services(db_session, enabled_only: bool = True) -> List[Dict]:
|
||||
"""获取已配置的外部KB服务（仅 dify）"""
|
||||
from sqlalchemy import text
|
||||
try:
|
||||
query = "SELECT * FROM external_kb_services"
|
||||
if enabled_only:
|
||||
query += " WHERE is_enabled = 1"
|
||||
query += " ORDER BY created_at DESC"
|
||||
result = db_session.execute(text(query))
|
||||
columns = result.keys()
|
||||
return [dict(zip(columns, row)) for row in result.fetchall()]
|
||||
except Exception as e:
|
||||
logger.error(f"Failed to get external services: {e}")
|
||||
return []
|
||||
|
||||
|
||||
async def _search_dify(service: Dict, query: str, top_k: int = 5) -> List[Dict]:
|
||||
"""检索 Dify 知识库（走 app.dify_kb 客户端）"""
|
||||
try:
|
||||
from app.dify_kb import dify_retrieve
|
||||
api_url = service.get("api_url", "")
|
||||
api_key = service.get("api_key", "")
|
||||
dataset_name = None
|
||||
cfg = service.get("config")
|
||||
if isinstance(cfg, str):
|
||||
try:
|
||||
import json
|
||||
cfg = json.loads(cfg)
|
||||
except Exception:
|
||||
cfg = None
|
||||
if isinstance(cfg, dict):
|
||||
dataset_name = cfg.get("dataset_name")
|
||||
hits = dify_retrieve(api_url, api_key, query, top_k=top_k, dataset_name=dataset_name)
|
||||
return hits
|
||||
except Exception as e:
|
||||
logger.error(f"Dify search error: {e}")
|
||||
return []
|
||||
|
||||
|
||||
async def _search_external(service: Dict, query: str, top_k: int = 5) -> List[Dict]:
|
||||
"""路由到对应的外部KB服务（当前仅支持 dify）"""
|
||||
service_type = service.get("service_type", "")
|
||||
if service_type == "dify":
|
||||
return await _search_dify(service, query, top_k)
|
||||
else:
|
||||
logger.warning(f"Unknown external KB type: {service_type}")
|
||||
return []
|
||||
|
||||
async def _get_web_services(db_session, enabled_only: bool = True) -> List[Dict]:
|
||||
"""获取全局 AnySearch 联网检索配置（从 site_settings 读取）。"""
|
||||
from app.anysearch_client import load_anysearch_config
|
||||
cfg = load_anysearch_config(db_session)
|
||||
if not cfg:
|
||||
return []
|
||||
if enabled_only and not cfg.get("is_enabled"):
|
||||
return []
|
||||
api_key = cfg.get("api_key", "")
|
||||
if not api_key:
|
||||
return []
|
||||
return [{
|
||||
"id": "anysearch_global",
|
||||
"service_type": "anysearch",
|
||||
"api_key": api_key,
|
||||
"api_url": cfg.get("api_url") or "https://api.anysearch.com/v1/search",
|
||||
"config": {
|
||||
"zone": cfg.get("zone", "cn"),
|
||||
"freshness": cfg.get("freshness"),
|
||||
"top_k": cfg.get("top_k", 5),
|
||||
},
|
||||
"is_enabled": cfg.get("is_enabled", False),
|
||||
}]
|
||||
|
||||
|
||||
async def _search_anysearch(service: Dict, query: str, top_k: int = 5) -> List[Dict]:
|
||||
"""调用 AnySearch 联网检索（走 app.anysearch_client）。"""
|
||||
try:
|
||||
from app.anysearch_client import anysearch_search
|
||||
api_key = service.get("api_key", "")
|
||||
api_url = service.get("api_url") or "https://api.anysearch.com/v1/search"
|
||||
cfg = service.get("config")
|
||||
if isinstance(cfg, str):
|
||||
try:
|
||||
import json
|
||||
cfg = json.loads(cfg)
|
||||
except Exception:
|
||||
cfg = None
|
||||
cfg = cfg or {}
|
||||
zone = cfg.get("zone", "cn")
|
||||
freshness = cfg.get("freshness")
|
||||
k = int(cfg.get("top_k", top_k))
|
||||
return await anysearch_search(api_key, api_url, query, top_k=k, zone=zone, freshness=freshness)
|
||||
except Exception as e:
|
||||
logger.error(f"AnySearch search error: {e}")
|
||||
return []
|
||||
|
||||
|
||||
async def kb_router_search(
|
||||
query: str,
|
||||
kb_id: Optional[str] = None,
|
||||
kb_ids: Optional[List[str]] = None,
|
||||
top_k: int = 5,
|
||||
bm25_results: Optional[List[Dict]] = None,
|
||||
db_session=None,
|
||||
use_vector: bool = True,
|
||||
use_external: bool = True,
|
||||
use_tree: bool = True,
|
||||
use_web: bool = False,
|
||||
) -> List[Dict]:
|
||||
"""
|
||||
知识库路由层 — 混合检索
|
||||
|
||||
检索策略:
|
||||
1. 本地 BM25 检索 (MySQL, 已有)
|
||||
2. 向量语义检索 (PostgreSQL+pgvector, 新增)
|
||||
3. 结构化分片段检索 (TreeSearch, 新增: 知识库目录树 + chunk 级精准命中)
|
||||
4. 外部KB检索 (Dify, 按配置)
|
||||
5. 联网检索 (AnySearch, 可选 use_web)
|
||||
|
||||
融合方式: Reciprocal Rank Fusion (RRF)
|
||||
|
||||
Args:
|
||||
query: 查询文本
|
||||
kb_id: 知识库ID
|
||||
top_k: 返回结果数
|
||||
bm25_results: BM25检索结果
|
||||
db_session: 数据库会话(用于获取外部KB配置与分片段检索)
|
||||
use_vector: 是否使用向量检索
|
||||
use_external: 是否使用外部KB检索
|
||||
use_tree: 是否使用 TreeSearch 结构化分片段检索
|
||||
use_web: 是否使用 AnySearch 联网检索
|
||||
|
||||
Returns:
|
||||
融合后的检索结果列表
|
||||
"""
|
||||
all_results = []
|
||||
|
||||
# 1. BM25 结果
|
||||
if bm25_results:
|
||||
for rank, item in enumerate(bm25_results):
|
||||
all_results.append({
|
||||
"content": item.get("content", item.get("text", "")),
|
||||
"score": item.get("score", 1.0 / (60 + rank + 1)),
|
||||
"source": "local_bm25",
|
||||
"rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
})
|
||||
|
||||
# 2. 向量检索
|
||||
if use_vector:
|
||||
try:
|
||||
from app.pgvector_search import vector_search
|
||||
kb_targets = kb_ids if kb_ids else ([kb_id] if kb_id else [None])
|
||||
vec_results = []
|
||||
for _kid in kb_targets:
|
||||
try:
|
||||
vec_results.extend(await vector_search(query, _kid, top_k))
|
||||
except Exception as e:
|
||||
logger.warning(f"Vector search failed for kb={_kid}: {e}")
|
||||
for rank, item in enumerate(vec_results):
|
||||
all_results.append({
|
||||
"content": item.get("content", ""),
|
||||
"score": item.get("score", 0.5),
|
||||
"source": "vector_pg",
|
||||
"rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
})
|
||||
except ImportError:
|
||||
logger.debug("pgvector_search not available, skipping vector search")
|
||||
except Exception as e:
|
||||
logger.warning(f"Vector search failed: {e}")
|
||||
|
||||
# 2.5 结构化分片段检索 (TreeSearch, Level 2)
|
||||
if use_tree and db_session is not None:
|
||||
try:
|
||||
from app.tree_search import tree_search
|
||||
kb_targets = kb_ids if kb_ids else ([kb_id] if kb_id else None)
|
||||
tree_hits = await tree_search(
|
||||
db_session, query, kb_ids=kb_targets, top_k=top_k, mode="hybrid",
|
||||
)
|
||||
for rank, item in enumerate(tree_hits):
|
||||
all_results.append({
|
||||
"content": item.get("content", ""),
|
||||
"score": item.get("score", 0.5),
|
||||
"source": "tree_chunk",
|
||||
"rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
})
|
||||
except Exception as e:
|
||||
logger.warning(f"TreeSearch failed: {e}")
|
||||
|
||||
# 3. 外部KB检索 (Dify)
|
||||
if use_external and db_session:
|
||||
services = await _get_external_services(db_session)
|
||||
for service in services:
|
||||
if service.get("status") == "connected":
|
||||
ext_results = await _search_external(service, query, top_k)
|
||||
for rank, item in enumerate(ext_results):
|
||||
all_results.append({
|
||||
"content": item.get("content", ""),
|
||||
"score": item.get("score", 0.5),
|
||||
"source": item.get("source", "external_dify"),
|
||||
"rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
})
|
||||
|
||||
# 4. 联网检索 (AnySearch, 可选)
|
||||
if use_web and db_session is not None:
|
||||
try:
|
||||
web_services = await _get_web_services(db_session)
|
||||
for svc in web_services:
|
||||
web_hits = await _search_anysearch(svc, query, top_k)
|
||||
for rank, item in enumerate(web_hits):
|
||||
all_results.append({
|
||||
"content": item.get("content", ""),
|
||||
"score": item.get("score", 0.6),
|
||||
"source": item.get("source", "web_anysearch"),
|
||||
"rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
})
|
||||
except Exception as e:
|
||||
logger.warning(f"AnySearch web search failed: {e}")
|
||||
|
||||
if not all_results:
|
||||
return []
|
||||
|
||||
# RRF 融合
|
||||
rrf_k = 60
|
||||
score_map = {}
|
||||
for item in all_results:
|
||||
content_key = item["content"][:200] # 用内容前200字符作为去重键
|
||||
rrf_score = 1.0 / (rrf_k + item.get("rank", 1))
|
||||
|
||||
if content_key in score_map:
|
||||
score_map[content_key]["score"] += rrf_score
|
||||
score_map[content_key]["sources"].append(item["source"])
|
||||
else:
|
||||
score_map[content_key] = {
|
||||
"content": item["content"],
|
||||
"score": rrf_score,
|
||||
"sources": [item["source"]],
|
||||
"metadata": item.get("metadata", {}),
|
||||
}
|
||||
|
||||
# 按融合分数排序
|
||||
merged = sorted(score_map.values(), key=lambda x: x["score"], reverse=True)
|
||||
|
||||
# 标准化分数到 0-1
|
||||
if merged:
|
||||
max_score = merged[0]["score"]
|
||||
for item in merged:
|
||||
item["score"] = item["score"] / max_score if max_score > 0 else 0
|
||||
item["sources"] = list(set(item["sources"]))
|
||||
|
||||
return merged[:top_k]
|
||||
@@ -1,379 +1,379 @@
|
||||
"""
|
||||
PGVector 向量检索模块
|
||||
====================
|
||||
使用 PostgreSQL + pgvector 实现语义向量检索，
|
||||
与现有 MySQL BM25 检索配合，实现混合检索。
|
||||
|
||||
使用方式:
|
||||
from app.pgvector_search import vector_search
|
||||
results = await vector_search(query, kb_id, top_k=5)
|
||||
"""
|
||||
|
||||
import os
|
||||
import logging
|
||||
import asyncio
|
||||
from typing import List, Dict, Optional
|
||||
from datetime import datetime
|
||||
|
||||
logger = logging.getLogger("aiflow.pgvector")
|
||||
|
||||
# PostgreSQL 连接配置
|
||||
PG_HOST = os.environ.get("POSTGRES_HOST", "qihu-aiflow-pgvector")
|
||||
PG_PORT = int(os.environ.get("POSTGRES_PORT", "5432"))
|
||||
PG_USER = os.environ.get("POSTGRES_USER", "aiflow")
|
||||
PG_PASSWORD = os.environ.get("POSTGRES_PASSWORD", "pwd2026")
|
||||
PG_DB = os.environ.get("POSTGRES_DB", "aiflow_vectors")
|
||||
|
||||
# 嵌入模型配置
|
||||
EMBEDDING_API_URL = os.environ.get("SILICONFLOW_BASE_URL", "https://api.siliconflow.cn")
|
||||
EMBEDDING_API_KEY = os.environ.get("SILICONFLOW_API_KEY", "")
|
||||
EMBEDDING_MODEL = os.environ.get("EMBEDDING_MODEL", "BAAI/bge-large-zh-v1.5")
|
||||
|
||||
# ---- 本地嵌入缓存（SQLite 持久化）：破解"每次嵌入消耗外部 API 额度"的隐性成本障碍 ----
|
||||
import sqlite3, hashlib, json as _json
|
||||
_EMBED_CACHE_DB = os.environ.get("EMBED_CACHE_DB", "/app/backend/python/.embed_cache.db")
|
||||
def _embed_cache_get(text, model):
|
||||
try:
|
||||
key = hashlib.sha256(f"{model}::{text}".encode("utf-8")).hexdigest()
|
||||
conn = sqlite3.connect(_EMBED_CACHE_DB, timeout=5)
|
||||
row = conn.execute("SELECT vec FROM embed_cache WHERE k=?", (key,)).fetchone()
|
||||
conn.close()
|
||||
if row:
|
||||
return _json.loads(row[0])
|
||||
except Exception:
|
||||
return None
|
||||
return None
|
||||
def _embed_cache_put(text, model, vec):
|
||||
try:
|
||||
key = hashlib.sha256(f"{model}::{text}".encode("utf-8")).hexdigest()
|
||||
conn = sqlite3.connect(_EMBED_CACHE_DB, timeout=5)
|
||||
conn.execute("CREATE TABLE IF NOT EXISTS embed_cache (k TEXT PRIMARY KEY, vec TEXT, ts TIMESTAMP DEFAULT CURRENT_TIMESTAMP)")
|
||||
conn.execute("INSERT OR REPLACE INTO embed_cache (k, vec) VALUES (?, ?)", (key, _json.dumps(vec)))
|
||||
conn.commit()
|
||||
conn.close()
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
|
||||
# 连接池
|
||||
_pg_pool = None
|
||||
_pool_lock = asyncio.Lock()
|
||||
|
||||
|
||||
async def get_pg_pool():
|
||||
"""获取 PostgreSQL 异步连接池"""
|
||||
global _pg_pool
|
||||
if _pg_pool is not None:
|
||||
return _pg_pool
|
||||
|
||||
async with _pool_lock:
|
||||
if _pg_pool is not None:
|
||||
return _pg_pool
|
||||
|
||||
try:
|
||||
import asyncpg
|
||||
except ImportError:
|
||||
logger.warning("asyncpg not installed, vector search disabled")
|
||||
return None
|
||||
|
||||
try:
|
||||
_pg_pool = await asyncpg.create_pool(
|
||||
host=PG_HOST,
|
||||
port=PG_PORT,
|
||||
user=PG_USER,
|
||||
password=PG_PASSWORD,
|
||||
database=PG_DB,
|
||||
min_size=2,
|
||||
max_size=10,
|
||||
command_timeout=30,
|
||||
)
|
||||
logger.info(f"PostgreSQL connection pool created: {PG_HOST}:{PG_PORT}/{PG_DB}")
|
||||
|
||||
# 初始化表结构
|
||||
async with _pg_pool.acquire() as conn:
|
||||
await conn.execute("""
|
||||
CREATE TABLE IF NOT EXISTS document_embeddings (
|
||||
id SERIAL PRIMARY KEY,
|
||||
doc_id VARCHAR(64) NOT NULL,
|
||||
kb_id VARCHAR(64) NOT NULL,
|
||||
chunk_id VARCHAR(64),
|
||||
content TEXT NOT NULL,
|
||||
embedding vector(1024),
|
||||
metadata JSONB DEFAULT '{}',
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
""")
|
||||
await conn.execute("""
|
||||
CREATE INDEX IF NOT EXISTS idx_embeddings_kb
|
||||
ON document_embeddings(kb_id);
|
||||
""")
|
||||
await conn.execute("""
|
||||
CREATE INDEX IF NOT EXISTS idx_embeddings_doc
|
||||
ON document_embeddings(doc_id);
|
||||
""")
|
||||
await conn.execute("""
|
||||
CREATE INDEX IF NOT EXISTS idx_embeddings_vector
|
||||
ON document_embeddings USING ivfflat (embedding vector_cosine_ops)
|
||||
WITH (lists = 100);
|
||||
""")
|
||||
logger.info("PGVector tables initialized")
|
||||
|
||||
return _pg_pool
|
||||
except Exception as e:
|
||||
logger.error(f"Failed to create PostgreSQL pool: {e}")
|
||||
_pg_pool = None
|
||||
return None
|
||||
|
||||
|
||||
async def get_embedding(text: str) -> Optional[List[float]]:
|
||||
"""调用嵌入API获取文本向量"""
|
||||
if not EMBEDDING_API_KEY:
|
||||
logger.warning("No embedding API key configured")
|
||||
return None
|
||||
|
||||
cached = _embed_cache_get(text, EMBEDDING_MODEL)
|
||||
if cached:
|
||||
return cached
|
||||
try:
|
||||
# 收敛到共享标准模型层（app.core.llm_client.aembed），与聊天共用代理/超时/failover 语义
|
||||
from app.core.llm_client import aembed
|
||||
embedding = await aembed(
|
||||
EMBEDDING_API_URL, EMBEDDING_API_KEY, text[:8000], EMBEDDING_MODEL, timeout=30
|
||||
)
|
||||
if embedding:
|
||||
_embed_cache_put(text, EMBEDDING_MODEL, embedding)
|
||||
return embedding
|
||||
logger.error("Embedding API returned empty")
|
||||
return None
|
||||
except Exception as e:
|
||||
logger.error(f"Embedding generation failed: {e}")
|
||||
return None
|
||||
|
||||
|
||||
async def vector_search(
|
||||
query: str,
|
||||
kb_id: Optional[str] = None,
|
||||
top_k: int = 5,
|
||||
threshold: float = 0.5,
|
||||
) -> List[Dict]:
|
||||
"""
|
||||
向量语义检索
|
||||
|
||||
Args:
|
||||
query: 查询文本
|
||||
kb_id: 知识库ID (可选，不指定则搜索全部)
|
||||
top_k: 返回结果数
|
||||
threshold: 相似度阈值 (0-1)
|
||||
|
||||
Returns:
|
||||
[{"doc_id": ..., "content": ..., "score": ..., "metadata": ...}, ...]
|
||||
"""
|
||||
pool = await get_pg_pool()
|
||||
if pool is None:
|
||||
return []
|
||||
|
||||
# 获取查询向量
|
||||
query_embedding = await get_embedding(query)
|
||||
if query_embedding is None:
|
||||
return []
|
||||
|
||||
try:
|
||||
async with pool.acquire() as conn:
|
||||
# 将向量转换为 PostgreSQL 格式
|
||||
embedding_str = f"[{','.join(str(x) for x in query_embedding)}]"
|
||||
|
||||
if kb_id:
|
||||
rows = await conn.fetch(
|
||||
"""
|
||||
SELECT doc_id, chunk_id, content, metadata,
|
||||
1 - (embedding <=> $1::vector) as score
|
||||
FROM document_embeddings
|
||||
WHERE kb_id = $2
|
||||
ORDER BY embedding <=> $1::vector
|
||||
LIMIT $3;
|
||||
""",
|
||||
embedding_str, kb_id, top_k,
|
||||
)
|
||||
else:
|
||||
rows = await conn.fetch(
|
||||
"""
|
||||
SELECT doc_id, chunk_id, content, metadata,
|
||||
1 - (embedding <=> $1::vector) as score
|
||||
FROM document_embeddings
|
||||
ORDER BY embedding <=> $1::vector
|
||||
LIMIT $2;
|
||||
""",
|
||||
embedding_str, top_k,
|
||||
)
|
||||
|
||||
results = []
|
||||
for row in rows:
|
||||
score = float(row["score"])
|
||||
if score >= threshold:
|
||||
results.append({
|
||||
"doc_id": row["doc_id"],
|
||||
"chunk_id": row["chunk_id"],
|
||||
"content": row["content"],
|
||||
"score": score,
|
||||
"metadata": row["metadata"] if isinstance(row["metadata"], dict) else {},
|
||||
})
|
||||
|
||||
return results
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"Vector search failed: {e}")
|
||||
return []
|
||||
|
||||
|
||||
async def index_document(
|
||||
doc_id: str,
|
||||
kb_id: str,
|
||||
content: str,
|
||||
chunk_id: Optional[str] = None,
|
||||
metadata: Optional[Dict] = None,
|
||||
) -> bool:
|
||||
"""
|
||||
将文档内容索引到 PGVector
|
||||
|
||||
Args:
|
||||
doc_id: 文档ID
|
||||
kb_id: 知识库ID
|
||||
content: 文档内容
|
||||
chunk_id: 分块ID (可选)
|
||||
metadata: 元数据
|
||||
|
||||
Returns:
|
||||
是否成功
|
||||
"""
|
||||
pool = await get_pg_pool()
|
||||
if pool is None:
|
||||
return False
|
||||
|
||||
# 分块处理（简单按长度分块）
|
||||
chunks = []
|
||||
chunk_size = 500
|
||||
overlap = 50
|
||||
if len(content) <= chunk_size:
|
||||
chunks = [(content, chunk_id or f"{doc_id}_0")]
|
||||
else:
|
||||
for i in range(0, len(content), chunk_size - overlap):
|
||||
chunk = content[i:i + chunk_size]
|
||||
if len(chunk.strip()) > 10:
|
||||
chunks.append((chunk, f"{doc_id}_{i // (chunk_size - overlap)}"))
|
||||
|
||||
success_count = 0
|
||||
for chunk_text, cid in chunks:
|
||||
embedding = await get_embedding(chunk_text)
|
||||
if embedding is None:
|
||||
continue
|
||||
|
||||
try:
|
||||
async with pool.acquire() as conn:
|
||||
embedding_str = f"[{','.join(str(x) for x in embedding)}]"
|
||||
import json
|
||||
await conn.execute(
|
||||
"""
|
||||
INSERT INTO document_embeddings (doc_id, kb_id, chunk_id, content, embedding, metadata)
|
||||
VALUES ($1, $2, $3, $4, $5::vector, $6)
|
||||
ON CONFLICT DO NOTHING;
|
||||
""",
|
||||
doc_id, kb_id, cid, chunk_text, embedding_str,
|
||||
json.dumps(metadata or {}),
|
||||
)
|
||||
success_count += 1
|
||||
except Exception as e:
|
||||
logger.error(f"Failed to index chunk {cid}: {e}")
|
||||
|
||||
logger.info(f"Indexed document {doc_id}: {success_count}/{len(chunks)} chunks")
|
||||
return success_count > 0
|
||||
|
||||
|
||||
async def hybrid_search(
|
||||
query: str,
|
||||
kb_id: Optional[str] = None,
|
||||
top_k: int = 5,
|
||||
bm25_results: Optional[List[Dict]] = None,
|
||||
) -> List[Dict]:
|
||||
"""
|
||||
混合检索：BM25 + 向量检索融合
|
||||
|
||||
Args:
|
||||
query: 查询文本
|
||||
kb_id: 知识库ID
|
||||
top_k: 返回结果数
|
||||
bm25_results: BM25 检索结果（来自MySQL）
|
||||
|
||||
Returns:
|
||||
融合后的检索结果
|
||||
"""
|
||||
# 获取向量检索结果
|
||||
vec_results = await vector_search(query, kb_id, top_k)
|
||||
|
||||
if not vec_results and not bm25_results:
|
||||
return []
|
||||
|
||||
if not vec_results:
|
||||
return bm25_results or []
|
||||
|
||||
if not bm25_results:
|
||||
return vec_results
|
||||
|
||||
# 融合：Reciprocal Rank Fusion (RRF)
|
||||
rrf_k = 60
|
||||
score_map = {}
|
||||
|
||||
for rank, item in enumerate(bm25_results):
|
||||
doc_id = item.get("doc_id", item.get("id", ""))
|
||||
score_map[doc_id] = {
|
||||
"content": item.get("content", item.get("text", "")),
|
||||
"score": 1.0 / (rrf_k + rank + 1),
|
||||
"bm25_rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
}
|
||||
|
||||
for rank, item in enumerate(vec_results):
|
||||
doc_id = item["doc_id"]
|
||||
if doc_id in score_map:
|
||||
score_map[doc_id]["score"] += 1.0 / (rrf_k + rank + 1)
|
||||
score_map[doc_id]["vec_score"] = item["score"]
|
||||
score_map[doc_id]["vec_rank"] = rank + 1
|
||||
else:
|
||||
score_map[doc_id] = {
|
||||
"content": item["content"],
|
||||
"score": 1.0 / (rrf_k + rank + 1),
|
||||
"vec_score": item["score"],
|
||||
"vec_rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
}
|
||||
|
||||
# 按融合分数排序
|
||||
merged = sorted(score_map.values(), key=lambda x: x["score"], reverse=True)
|
||||
return merged[:top_k]
|
||||
|
||||
|
||||
async def delete_document_embeddings(doc_id: str) -> bool:
|
||||
"""删除文档的所有向量索引"""
|
||||
pool = await get_pg_pool()
|
||||
if pool is None:
|
||||
return False
|
||||
|
||||
try:
|
||||
async with pool.acquire() as conn:
|
||||
await conn.execute(
|
||||
"DELETE FROM document_embeddings WHERE doc_id = $1;",
|
||||
doc_id,
|
||||
)
|
||||
logger.info(f"Deleted embeddings for document {doc_id}")
|
||||
return True
|
||||
except Exception as e:
|
||||
logger.error(f"Failed to delete embeddings: {e}")
|
||||
return False
|
||||
|
||||
|
||||
async def close_pool():
|
||||
"""关闭连接池"""
|
||||
global _pg_pool
|
||||
if _pg_pool is not None:
|
||||
await _pg_pool.close()
|
||||
_pg_pool = None
|
||||
logger.info("PostgreSQL connection pool closed")
|
||||
"""
|
||||
PGVector 向量检索模块
|
||||
====================
|
||||
使用 PostgreSQL + pgvector 实现语义向量检索，
|
||||
与现有 MySQL BM25 检索配合，实现混合检索。
|
||||
|
||||
使用方式:
|
||||
from app.pgvector_search import vector_search
|
||||
results = await vector_search(query, kb_id, top_k=5)
|
||||
"""
|
||||
|
||||
import os
|
||||
import logging
|
||||
import asyncio
|
||||
from typing import List, Dict, Optional
|
||||
from datetime import datetime
|
||||
|
||||
logger = logging.getLogger("aiflow.pgvector")
|
||||
|
||||
# PostgreSQL 连接配置
|
||||
PG_HOST = os.environ.get("POSTGRES_HOST", "qihu-aiflow-pgvector")
|
||||
PG_PORT = int(os.environ.get("POSTGRES_PORT", "5432"))
|
||||
PG_USER = os.environ.get("POSTGRES_USER", "aiflow")
|
||||
PG_PASSWORD = os.environ.get("POSTGRES_PASSWORD", "pwd2026")
|
||||
PG_DB = os.environ.get("POSTGRES_DB", "aiflow_vectors")
|
||||
|
||||
# 嵌入模型配置
|
||||
EMBEDDING_API_URL = os.environ.get("SILICONFLOW_BASE_URL", "https://api.siliconflow.cn")
|
||||
EMBEDDING_API_KEY = os.environ.get("SILICONFLOW_API_KEY", "")
|
||||
EMBEDDING_MODEL = os.environ.get("EMBEDDING_MODEL", "BAAI/bge-large-zh-v1.5")
|
||||
|
||||
# ---- 本地嵌入缓存（SQLite 持久化）：破解"每次嵌入消耗外部 API 额度"的隐性成本障碍 ----
|
||||
import sqlite3, hashlib, json as _json
|
||||
_EMBED_CACHE_DB = os.environ.get("EMBED_CACHE_DB", "/app/backend/python/.embed_cache.db")
|
||||
def _embed_cache_get(text, model):
|
||||
try:
|
||||
key = hashlib.sha256(f"{model}::{text}".encode("utf-8")).hexdigest()
|
||||
conn = sqlite3.connect(_EMBED_CACHE_DB, timeout=5)
|
||||
row = conn.execute("SELECT vec FROM embed_cache WHERE k=?", (key,)).fetchone()
|
||||
conn.close()
|
||||
if row:
|
||||
return _json.loads(row[0])
|
||||
except Exception:
|
||||
return None
|
||||
return None
|
||||
def _embed_cache_put(text, model, vec):
|
||||
try:
|
||||
key = hashlib.sha256(f"{model}::{text}".encode("utf-8")).hexdigest()
|
||||
conn = sqlite3.connect(_EMBED_CACHE_DB, timeout=5)
|
||||
conn.execute("CREATE TABLE IF NOT EXISTS embed_cache (k TEXT PRIMARY KEY, vec TEXT, ts TIMESTAMP DEFAULT CURRENT_TIMESTAMP)")
|
||||
conn.execute("INSERT OR REPLACE INTO embed_cache (k, vec) VALUES (?, ?)", (key, _json.dumps(vec)))
|
||||
conn.commit()
|
||||
conn.close()
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
|
||||
# 连接池
|
||||
_pg_pool = None
|
||||
_pool_lock = asyncio.Lock()
|
||||
|
||||
|
||||
async def get_pg_pool():
|
||||
"""获取 PostgreSQL 异步连接池"""
|
||||
global _pg_pool
|
||||
if _pg_pool is not None:
|
||||
return _pg_pool
|
||||
|
||||
async with _pool_lock:
|
||||
if _pg_pool is not None:
|
||||
return _pg_pool
|
||||
|
||||
try:
|
||||
import asyncpg
|
||||
except ImportError:
|
||||
logger.warning("asyncpg not installed, vector search disabled")
|
||||
return None
|
||||
|
||||
try:
|
||||
_pg_pool = await asyncpg.create_pool(
|
||||
host=PG_HOST,
|
||||
port=PG_PORT,
|
||||
user=PG_USER,
|
||||
password=PG_PASSWORD,
|
||||
database=PG_DB,
|
||||
min_size=2,
|
||||
max_size=10,
|
||||
command_timeout=30,
|
||||
)
|
||||
logger.info(f"PostgreSQL connection pool created: {PG_HOST}:{PG_PORT}/{PG_DB}")
|
||||
|
||||
# 初始化表结构
|
||||
async with _pg_pool.acquire() as conn:
|
||||
await conn.execute("""
|
||||
CREATE TABLE IF NOT EXISTS document_embeddings (
|
||||
id SERIAL PRIMARY KEY,
|
||||
doc_id VARCHAR(64) NOT NULL,
|
||||
kb_id VARCHAR(64) NOT NULL,
|
||||
chunk_id VARCHAR(64),
|
||||
content TEXT NOT NULL,
|
||||
embedding vector(1024),
|
||||
metadata JSONB DEFAULT '{}',
|
||||
created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
|
||||
);
|
||||
""")
|
||||
await conn.execute("""
|
||||
CREATE INDEX IF NOT EXISTS idx_embeddings_kb
|
||||
ON document_embeddings(kb_id);
|
||||
""")
|
||||
await conn.execute("""
|
||||
CREATE INDEX IF NOT EXISTS idx_embeddings_doc
|
||||
ON document_embeddings(doc_id);
|
||||
""")
|
||||
await conn.execute("""
|
||||
CREATE INDEX IF NOT EXISTS idx_embeddings_vector
|
||||
ON document_embeddings USING ivfflat (embedding vector_cosine_ops)
|
||||
WITH (lists = 100);
|
||||
""")
|
||||
logger.info("PGVector tables initialized")
|
||||
|
||||
return _pg_pool
|
||||
except Exception as e:
|
||||
logger.error(f"Failed to create PostgreSQL pool: {e}")
|
||||
_pg_pool = None
|
||||
return None
|
||||
|
||||
|
||||
async def get_embedding(text: str) -> Optional[List[float]]:
|
||||
"""调用嵌入API获取文本向量"""
|
||||
if not EMBEDDING_API_KEY:
|
||||
logger.warning("No embedding API key configured")
|
||||
return None
|
||||
|
||||
cached = _embed_cache_get(text, EMBEDDING_MODEL)
|
||||
if cached:
|
||||
return cached
|
||||
try:
|
||||
# 收敛到共享标准模型层（app.core.llm_client.aembed），与聊天共用代理/超时/failover 语义
|
||||
from app.core.llm_client import aembed
|
||||
embedding = await aembed(
|
||||
EMBEDDING_API_URL, EMBEDDING_API_KEY, text[:8000], EMBEDDING_MODEL, timeout=30
|
||||
)
|
||||
if embedding:
|
||||
_embed_cache_put(text, EMBEDDING_MODEL, embedding)
|
||||
return embedding
|
||||
logger.error("Embedding API returned empty")
|
||||
return None
|
||||
except Exception as e:
|
||||
logger.error(f"Embedding generation failed: {e}")
|
||||
return None
|
||||
|
||||
|
||||
async def vector_search(
|
||||
query: str,
|
||||
kb_id: Optional[str] = None,
|
||||
top_k: int = 5,
|
||||
threshold: float = 0.5,
|
||||
) -> List[Dict]:
|
||||
"""
|
||||
向量语义检索
|
||||
|
||||
Args:
|
||||
query: 查询文本
|
||||
kb_id: 知识库ID (可选，不指定则搜索全部)
|
||||
top_k: 返回结果数
|
||||
threshold: 相似度阈值 (0-1)
|
||||
|
||||
Returns:
|
||||
[{"doc_id": ..., "content": ..., "score": ..., "metadata": ...}, ...]
|
||||
"""
|
||||
pool = await get_pg_pool()
|
||||
if pool is None:
|
||||
return []
|
||||
|
||||
# 获取查询向量
|
||||
query_embedding = await get_embedding(query)
|
||||
if query_embedding is None:
|
||||
return []
|
||||
|
||||
try:
|
||||
async with pool.acquire() as conn:
|
||||
# 将向量转换为 PostgreSQL 格式
|
||||
embedding_str = f"[{','.join(str(x) for x in query_embedding)}]"
|
||||
|
||||
if kb_id:
|
||||
rows = await conn.fetch(
|
||||
"""
|
||||
SELECT doc_id, chunk_id, content, metadata,
|
||||
1 - (embedding <=> $1::vector) as score
|
||||
FROM document_embeddings
|
||||
WHERE kb_id = $2
|
||||
ORDER BY embedding <=> $1::vector
|
||||
LIMIT $3;
|
||||
""",
|
||||
embedding_str, kb_id, top_k,
|
||||
)
|
||||
else:
|
||||
rows = await conn.fetch(
|
||||
"""
|
||||
SELECT doc_id, chunk_id, content, metadata,
|
||||
1 - (embedding <=> $1::vector) as score
|
||||
FROM document_embeddings
|
||||
ORDER BY embedding <=> $1::vector
|
||||
LIMIT $2;
|
||||
""",
|
||||
embedding_str, top_k,
|
||||
)
|
||||
|
||||
results = []
|
||||
for row in rows:
|
||||
score = float(row["score"])
|
||||
if score >= threshold:
|
||||
results.append({
|
||||
"doc_id": row["doc_id"],
|
||||
"chunk_id": row["chunk_id"],
|
||||
"content": row["content"],
|
||||
"score": score,
|
||||
"metadata": row["metadata"] if isinstance(row["metadata"], dict) else {},
|
||||
})
|
||||
|
||||
return results
|
||||
|
||||
except Exception as e:
|
||||
logger.error(f"Vector search failed: {e}")
|
||||
return []
|
||||
|
||||
|
||||
async def index_document(
|
||||
doc_id: str,
|
||||
kb_id: str,
|
||||
content: str,
|
||||
chunk_id: Optional[str] = None,
|
||||
metadata: Optional[Dict] = None,
|
||||
) -> bool:
|
||||
"""
|
||||
将文档内容索引到 PGVector
|
||||
|
||||
Args:
|
||||
doc_id: 文档ID
|
||||
kb_id: 知识库ID
|
||||
content: 文档内容
|
||||
chunk_id: 分块ID (可选)
|
||||
metadata: 元数据
|
||||
|
||||
Returns:
|
||||
是否成功
|
||||
"""
|
||||
pool = await get_pg_pool()
|
||||
if pool is None:
|
||||
return False
|
||||
|
||||
# 分块处理（简单按长度分块）
|
||||
chunks = []
|
||||
chunk_size = 500
|
||||
overlap = 50
|
||||
if len(content) <= chunk_size:
|
||||
chunks = [(content, chunk_id or f"{doc_id}_0")]
|
||||
else:
|
||||
for i in range(0, len(content), chunk_size - overlap):
|
||||
chunk = content[i:i + chunk_size]
|
||||
if len(chunk.strip()) > 10:
|
||||
chunks.append((chunk, f"{doc_id}_{i // (chunk_size - overlap)}"))
|
||||
|
||||
success_count = 0
|
||||
for chunk_text, cid in chunks:
|
||||
embedding = await get_embedding(chunk_text)
|
||||
if embedding is None:
|
||||
continue
|
||||
|
||||
try:
|
||||
async with pool.acquire() as conn:
|
||||
embedding_str = f"[{','.join(str(x) for x in embedding)}]"
|
||||
import json
|
||||
await conn.execute(
|
||||
"""
|
||||
INSERT INTO document_embeddings (doc_id, kb_id, chunk_id, content, embedding, metadata)
|
||||
VALUES ($1, $2, $3, $4, $5::vector, $6)
|
||||
ON CONFLICT DO NOTHING;
|
||||
""",
|
||||
doc_id, kb_id, cid, chunk_text, embedding_str,
|
||||
json.dumps(metadata or {}),
|
||||
)
|
||||
success_count += 1
|
||||
except Exception as e:
|
||||
logger.error(f"Failed to index chunk {cid}: {e}")
|
||||
|
||||
logger.info(f"Indexed document {doc_id}: {success_count}/{len(chunks)} chunks")
|
||||
return success_count > 0
|
||||
|
||||
|
||||
async def hybrid_search(
|
||||
query: str,
|
||||
kb_id: Optional[str] = None,
|
||||
top_k: int = 5,
|
||||
bm25_results: Optional[List[Dict]] = None,
|
||||
) -> List[Dict]:
|
||||
"""
|
||||
混合检索：BM25 + 向量检索融合
|
||||
|
||||
Args:
|
||||
query: 查询文本
|
||||
kb_id: 知识库ID
|
||||
top_k: 返回结果数
|
||||
bm25_results: BM25 检索结果（来自MySQL）
|
||||
|
||||
Returns:
|
||||
融合后的检索结果
|
||||
"""
|
||||
# 获取向量检索结果
|
||||
vec_results = await vector_search(query, kb_id, top_k)
|
||||
|
||||
if not vec_results and not bm25_results:
|
||||
return []
|
||||
|
||||
if not vec_results:
|
||||
return bm25_results or []
|
||||
|
||||
if not bm25_results:
|
||||
return vec_results
|
||||
|
||||
# 融合：Reciprocal Rank Fusion (RRF)
|
||||
rrf_k = 60
|
||||
score_map = {}
|
||||
|
||||
for rank, item in enumerate(bm25_results):
|
||||
doc_id = item.get("doc_id", item.get("id", ""))
|
||||
score_map[doc_id] = {
|
||||
"content": item.get("content", item.get("text", "")),
|
||||
"score": 1.0 / (rrf_k + rank + 1),
|
||||
"bm25_rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
}
|
||||
|
||||
for rank, item in enumerate(vec_results):
|
||||
doc_id = item["doc_id"]
|
||||
if doc_id in score_map:
|
||||
score_map[doc_id]["score"] += 1.0 / (rrf_k + rank + 1)
|
||||
score_map[doc_id]["vec_score"] = item["score"]
|
||||
score_map[doc_id]["vec_rank"] = rank + 1
|
||||
else:
|
||||
score_map[doc_id] = {
|
||||
"content": item["content"],
|
||||
"score": 1.0 / (rrf_k + rank + 1),
|
||||
"vec_score": item["score"],
|
||||
"vec_rank": rank + 1,
|
||||
"metadata": item.get("metadata", {}),
|
||||
}
|
||||
|
||||
# 按融合分数排序
|
||||
merged = sorted(score_map.values(), key=lambda x: x["score"], reverse=True)
|
||||
return merged[:top_k]
|
||||
|
||||
|
||||
async def delete_document_embeddings(doc_id: str) -> bool:
|
||||
"""删除文档的所有向量索引"""
|
||||
pool = await get_pg_pool()
|
||||
if pool is None:
|
||||
return False
|
||||
|
||||
try:
|
||||
async with pool.acquire() as conn:
|
||||
await conn.execute(
|
||||
"DELETE FROM document_embeddings WHERE doc_id = $1;",
|
||||
doc_id,
|
||||
)
|
||||
logger.info(f"Deleted embeddings for document {doc_id}")
|
||||
return True
|
||||
except Exception as e:
|
||||
logger.error(f"Failed to delete embeddings: {e}")
|
||||
return False
|
||||
|
||||
|
||||
async def close_pool():
|
||||
"""关闭连接池"""
|
||||
global _pg_pool
|
||||
if _pg_pool is not None:
|
||||
await _pg_pool.close()
|
||||
_pg_pool = None
|
||||
logger.info("PostgreSQL connection pool closed")
|
||||
+249
-249
@@ -1,250 +1,250 @@
|
||||
"""
|
||||
Qihu-AIFlow FastAPI 主应用
|
||||
完整后端API，监听18080端口，所有接口以 /app 为前缀
|
||||
"""
|
||||
import sys
|
||||
import os
|
||||
import time
|
||||
import uuid
|
||||
import logging
|
||||
from contextlib import asynccontextmanager
|
||||
from fastapi import FastAPI, Request
|
||||
from fastapi.middleware.cors import CORSMiddleware
|
||||
from starlette.middleware.base import BaseHTTPMiddleware
|
||||
from datetime import datetime
|
||||
|
||||
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
||||
|
||||
from app.api import auth, chat, knowledge, market, token_usage, payment, profile, notifications, assets, evolution
|
||||
from app.casdoor_auth import router as casdoor_router
|
||||
from app.api.admin import dashboard, users, models, settings, payment_admin, services, membership_admin, ads, market_admin, logs, skills_admin, agents_admin, knowledge_admin, workflow_admin, storage_admin, automation_admin, mcp_admin, developer_admin, apps_admin, distro_admin, code_exec_admin, privacy_admin, im_bot_admin, promotion_admin, achievement_admin, tenant_admin, billing_admin, analytics_admin, observability_admin
|
||||
from app.api.im_bot_user import router as im_bot_user_router
|
||||
from app.api.im_bot_platform import router as im_bot_platform_router
|
||||
from app.api.portal import user_portal
|
||||
from app.api.admin.developer_tools_admin import router as developer_tools_router
|
||||
from app.api.admin.mcp_market_admin import router as mcp_market_router
|
||||
from app.api.admin import asset_admin, achievement_admin
|
||||
from app.config import setup_file_logging, SLOW_REQUEST_THRESHOLD
|
||||
from app.core.observability import set_trace_id
|
||||
from app.database import SessionLocal
|
||||
from app.models import OperationLog as OperationLogModel, TokenUsageLog as TokenUsageLogModel
|
||||
|
||||
# 导入平台适配器触发注册（AstrBot借鉴优化方案 P1）
|
||||
import app.platform_adapters # noqa: F401 - 触发所有平台适配器装饰器注册
|
||||
|
||||
# 初始化文件日志（按天轮转）
|
||||
access_logger = setup_file_logging()
|
||||
logger = logging.getLogger("aiflow")
|
||||
|
||||
|
||||
class APITrackingMiddleware(BaseHTTPMiddleware):
|
||||
"""API调用统计中间件 - 记录请求耗时，标记慢请求"""
|
||||
|
||||
async def dispatch(self, request: Request, call_next):
|
||||
request_id = str(uuid.uuid4())
|
||||
set_trace_id(request_id)
|
||||
start_time = time.time()
|
||||
path = request.url.path
|
||||
method = request.method
|
||||
|
||||
# 跳过健康检查和静态资源
|
||||
skip_paths = ["/app/api/health", "/favicon.ico"]
|
||||
should_skip = any(path.startswith(p) for p in skip_paths)
|
||||
|
||||
try:
|
||||
response = await call_next(request)
|
||||
elapsed = time.time() - start_time
|
||||
status_code = response.status_code
|
||||
response.headers["X-Trace-Id"] = request_id
|
||||
|
||||
if not should_skip:
|
||||
access_logger.info(
|
||||
f"[{request_id}] {method} {path} -> {status_code} ({elapsed:.3f}s)"
|
||||
)
|
||||
|
||||
# 异步记录API调用统计
|
||||
try:
|
||||
db = SessionLocal()
|
||||
try:
|
||||
_record_api_call(db, method, path, status_code, elapsed)
|
||||
finally:
|
||||
db.close()
|
||||
except Exception as e:
|
||||
logger.warning(f"API统计记录失败: {e}")
|
||||
|
||||
return response
|
||||
except Exception as e:
|
||||
elapsed = time.time() - start_time
|
||||
if not should_skip:
|
||||
access_logger.info(
|
||||
f"[{request_id}] {method} {path} -> ERROR ({elapsed:.3f}s): {str(e)}"
|
||||
)
|
||||
raise
|
||||
|
||||
async def __call__(self, scope, receive, send):
|
||||
if scope["type"] != "http":
|
||||
await self.app(scope, receive, send)
|
||||
return
|
||||
await super().__call__(scope, receive, send)
|
||||
|
||||
|
||||
def _record_api_call(db, method: str, path: str, status_code: int, elapsed: float):
|
||||
"""记录API调用到数据库统计表"""
|
||||
# 简化路径以减少基数（去掉动态ID）
|
||||
parts = path.split("/")
|
||||
simplified_parts = []
|
||||
for part in parts:
|
||||
# 将UUID样式的路径段替换为{id}
|
||||
if len(part) == 36 and part.count("-") == 4:
|
||||
simplified_parts.append("{id}")
|
||||
# 将纯数字路径段替换为{id}
|
||||
elif part.isdigit():
|
||||
simplified_parts.append("{id}")
|
||||
else:
|
||||
simplified_parts.append(part)
|
||||
simplified_path = "/".join(simplified_parts)
|
||||
|
||||
# 统计记录通过数据库实现，简化版记录到日志
|
||||
is_error = status_code >= 400
|
||||
is_slow = elapsed >= SLOW_REQUEST_THRESHOLD
|
||||
|
||||
if is_slow:
|
||||
logger.warning(
|
||||
f"慢请求: {method} {simplified_path} -> {status_code} ({elapsed:.3f}s, 阈值: {SLOW_REQUEST_THRESHOLD}s)"
|
||||
)
|
||||
|
||||
|
||||
@asynccontextmanager
|
||||
async def lifespan(app: FastAPI):
|
||||
logger.info("Qihu-AIFlow v2.0.0 启动中...")
|
||||
|
||||
# 启动后台 Evolution Pipeline（知识库图谱已存在的用户配置自动恢复运行）
|
||||
try:
|
||||
from app.services.evolution_service import EvolutionWatcherManager
|
||||
from app.models import UserEvolutionConfig
|
||||
db = SessionLocal()
|
||||
try:
|
||||
configs = db.query(UserEvolutionConfig).filter(
|
||||
UserEvolutionConfig.watcher_status == "running"
|
||||
).all()
|
||||
if configs:
|
||||
watcher = EvolutionWatcherManager()
|
||||
for cfg in configs:
|
||||
try:
|
||||
watcher.start_user_evolution(cfg.user_id, lambda: SessionLocal())
|
||||
logger.info(f"[Evolution] 恢复用户 {cfg.user_id} 的进化Pipeline")
|
||||
except Exception as e:
|
||||
logger.warning(f"[Evolution] 恢复用户 {cfg.user_id} 失败: {e}")
|
||||
logger.info(f"[Evolution] 已恢复 {len(configs)} 个用户的进化Pipeline")
|
||||
else:
|
||||
logger.info("[Evolution] 无待恢复的进化配置")
|
||||
finally:
|
||||
db.close()
|
||||
except Exception as e:
|
||||
logger.warning(f"[Evolution] Pipeline自动启动异常(不影响正常服务): {e}")
|
||||
|
||||
# 启动Redis事件总线
|
||||
try:
|
||||
from app.event_bus import get_event_bus
|
||||
event_bus = get_event_bus()
|
||||
event_bus.start(num_consumers=2)
|
||||
logger.info("[EventBus] Redis事件总线已启动")
|
||||
except Exception as e:
|
||||
logger.warning(f"[EventBus] 启动失败(不影响正常服务): {e}")
|
||||
|
||||
logger.info("Qihu-AIFlow v2.0.0 启动完成")
|
||||
|
||||
yield
|
||||
|
||||
# 停止Redis事件总线
|
||||
try:
|
||||
from app.event_bus import get_event_bus
|
||||
get_event_bus().stop()
|
||||
logger.info("[EventBus] Redis事件总线已停止")
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
# 优雅关闭：停止所有 Evolution Pipeline
|
||||
try:
|
||||
from app.services.evolution_service import EvolutionWatcherManager
|
||||
EvolutionWatcherManager().shutdown_all()
|
||||
logger.info("[Evolution] 所有Pipeline已停止")
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
logger.info("Qihu-AIFlow 正在关闭...")
|
||||
|
||||
|
||||
app = FastAPI(
|
||||
title="Qihu-AIFlow - 智能自动化PaaS平台",
|
||||
version="1.0.0",
|
||||
lifespan=lifespan,
|
||||
)
|
||||
|
||||
# API调用追踪中间件
|
||||
app.add_middleware(APITrackingMiddleware)
|
||||
|
||||
# CORS配置
|
||||
app.add_middleware(
|
||||
CORSMiddleware,
|
||||
allow_origins=["*"],
|
||||
allow_credentials=True,
|
||||
allow_methods=["*"],
|
||||
allow_headers=["*"],
|
||||
)
|
||||
|
||||
# 前台 API
|
||||
app.include_router(auth.router, prefix="/app/api/auth", tags=["认证"])
|
||||
app.include_router(chat.router, prefix="/app/api/chat", tags=["AI聊天"])
|
||||
app.include_router(knowledge.router, prefix="/app/api/knowledge", tags=["知识库"])
|
||||
app.include_router(market.router, prefix="/app/api/market", tags=["应用市场"])
|
||||
app.include_router(token_usage.router, prefix="/app/api/token", tags=["Token驾驶舱"])
|
||||
app.include_router(payment.router, prefix="/app/api/payment", tags=["付费中心"])
|
||||
app.include_router(profile.router, prefix="/app/api/profile", tags=["个人中心"])
|
||||
app.include_router(notifications.router, prefix="/app/api/notifications", tags=["消息通知"])
|
||||
app.include_router(assets.router, prefix="/app/api/assets", tags=["资产管理"])
|
||||
app.include_router(evolution.router, prefix="/app/api/evolution", tags=["AI自进化"])
|
||||
|
||||
# 后台 API
|
||||
app.include_router(dashboard.router, prefix="/app/api/admin/dashboard", tags=["后台-仪表盘"])
|
||||
app.include_router(users.router, prefix="/app/api/admin/users", tags=["后台-用户管理"])
|
||||
app.include_router(models.router, prefix="/app/api/admin/models", tags=["后台-模型管理"])
|
||||
app.include_router(settings.router, prefix="/app/api/admin/settings", tags=["后台-网站设置"])
|
||||
app.include_router(payment_admin.router, prefix="/app/api/admin/payment", tags=["后台-支付配置"])
|
||||
app.include_router(services.router, prefix="/app/api/admin/services", tags=["后台-增值服务"])
|
||||
app.include_router(membership_admin.router, prefix="/app/api/admin/membership", tags=["后台-会员管理"])
|
||||
app.include_router(ads.router, prefix="/app/api/admin/ads", tags=["后台-广告管理"])
|
||||
app.include_router(market_admin.router, prefix="/app/api/admin/market", tags=["后台-应用审核"])
|
||||
app.include_router(logs.router, prefix="/app/api/admin/logs", tags=["后台-日志风控"])
|
||||
|
||||
app.include_router(skills_admin.router, prefix="/app/api/admin/skills", tags=["后台-技能管理"])
|
||||
app.include_router(agents_admin.router, prefix="/app/api/admin/agents", tags=["后台-Agent管理"])
|
||||
app.include_router(knowledge_admin.router, prefix="/app/api/admin/knowledge", tags=["后台-知识库管理"])
|
||||
app.include_router(workflow_admin.router, prefix="/app/api/admin/workflows", tags=["后台-工作流编排"])
|
||||
app.include_router(asset_admin.router, prefix="/app/api/admin/assets", tags=["后台-资产管理"])
|
||||
app.include_router(storage_admin.router, prefix="/app/api/admin/storage", tags=["后台-存储配置"])
|
||||
app.include_router(automation_admin.router, prefix="/app/api/admin/automation", tags=["后台-自动化任务"])
|
||||
app.include_router(mcp_admin.router, prefix="/app/api/admin/mcp", tags=["后台-MCP服务"])
|
||||
app.include_router(developer_admin.router, prefix="/app/api/admin/developer", tags=["后台-开发者API"])
|
||||
app.include_router(developer_tools_router, prefix="/app/api/admin/developer-tools", tags=["dev-tools"])
|
||||
app.include_router(apps_admin.router, prefix="/app/api/admin/apps", tags=["admin-ai-apps"])
|
||||
app.include_router(mcp_market_router, prefix="/app/api/admin/mcp-market", tags=["mcp-market"])
|
||||
app.include_router(code_exec_admin.router, prefix="/app/api/admin/code-exec", tags=["code-exec"])
|
||||
app.include_router(distro_admin.router, prefix="/app/api/admin/distributions", tags=["distributions"])
|
||||
app.include_router(privacy_admin.router, prefix="/app/api/admin/privacy", tags=["privacy"])
|
||||
app.include_router(im_bot_admin.router, prefix="/app/api/admin/im-bot", tags=["im-bot"])
|
||||
app.include_router(im_bot_user_router, prefix="/app/api/im/wechat", tags=["im-wechat-user"])
|
||||
app.include_router(im_bot_platform_router, prefix="/app/api/im/platform", tags=["im-platform-user"])
|
||||
app.include_router(promotion_admin.router, prefix="/app/api/admin/promotion", tags=["promotion"])
|
||||
app.include_router(tenant_admin.router, prefix="/app/api/admin/tenant", tags=["tenant"])
|
||||
app.include_router(billing_admin.router, prefix="/app/api/admin/billing", tags=["billing"])
|
||||
app.include_router(analytics_admin.router, prefix="/app/api/admin/analytics", tags=["analytics"])
|
||||
app.include_router(observability_admin.router, prefix="/app/api/admin/observability", tags=["后台-调用可观测"])
|
||||
app.include_router(user_portal.router, prefix="/app/api/portal", tags=["portal"])
|
||||
app.include_router(achievement_admin.router, prefix="/app/api/admin/achievement", tags=["achievement"])
|
||||
app.include_router(casdoor_router, prefix="/app/api/auth", tags=["Casdoor SSO"])
|
||||
|
||||
|
||||
@app.get("/app/api/health")
|
||||
async def health_check():
|
||||
"""
|
||||
Qihu-AIFlow FastAPI 主应用
|
||||
完整后端API，监听18080端口，所有接口以 /app 为前缀
|
||||
"""
|
||||
import sys
|
||||
import os
|
||||
import time
|
||||
import uuid
|
||||
import logging
|
||||
from contextlib import asynccontextmanager
|
||||
from fastapi import FastAPI, Request
|
||||
from fastapi.middleware.cors import CORSMiddleware
|
||||
from starlette.middleware.base import BaseHTTPMiddleware
|
||||
from datetime import datetime
|
||||
|
||||
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
||||
|
||||
from app.api import auth, chat, knowledge, market, token_usage, payment, profile, notifications, assets, evolution
|
||||
from app.casdoor_auth import router as casdoor_router
|
||||
from app.api.admin import dashboard, users, models, settings, payment_admin, services, membership_admin, ads, market_admin, logs, skills_admin, agents_admin, knowledge_admin, workflow_admin, storage_admin, automation_admin, mcp_admin, developer_admin, apps_admin, distro_admin, code_exec_admin, privacy_admin, im_bot_admin, promotion_admin, achievement_admin, tenant_admin, billing_admin, analytics_admin, observability_admin
|
||||
from app.api.im_bot_user import router as im_bot_user_router
|
||||
from app.api.im_bot_platform import router as im_bot_platform_router
|
||||
from app.api.portal import user_portal
|
||||
from app.api.admin.developer_tools_admin import router as developer_tools_router
|
||||
from app.api.admin.mcp_market_admin import router as mcp_market_router
|
||||
from app.api.admin import asset_admin, achievement_admin
|
||||
from app.config import setup_file_logging, SLOW_REQUEST_THRESHOLD
|
||||
from app.core.observability import set_trace_id
|
||||
from app.database import SessionLocal
|
||||
from app.models import OperationLog as OperationLogModel, TokenUsageLog as TokenUsageLogModel
|
||||
|
||||
# 导入平台适配器触发注册（AstrBot借鉴优化方案 P1）
|
||||
import app.platform_adapters # noqa: F401 - 触发所有平台适配器装饰器注册
|
||||
|
||||
# 初始化文件日志（按天轮转）
|
||||
access_logger = setup_file_logging()
|
||||
logger = logging.getLogger("aiflow")
|
||||
|
||||
|
||||
class APITrackingMiddleware(BaseHTTPMiddleware):
|
||||
"""API调用统计中间件 - 记录请求耗时，标记慢请求"""
|
||||
|
||||
async def dispatch(self, request: Request, call_next):
|
||||
request_id = str(uuid.uuid4())
|
||||
set_trace_id(request_id)
|
||||
start_time = time.time()
|
||||
path = request.url.path
|
||||
method = request.method
|
||||
|
||||
# 跳过健康检查和静态资源
|
||||
skip_paths = ["/app/api/health", "/favicon.ico"]
|
||||
should_skip = any(path.startswith(p) for p in skip_paths)
|
||||
|
||||
try:
|
||||
response = await call_next(request)
|
||||
elapsed = time.time() - start_time
|
||||
status_code = response.status_code
|
||||
response.headers["X-Trace-Id"] = request_id
|
||||
|
||||
if not should_skip:
|
||||
access_logger.info(
|
||||
f"[{request_id}] {method} {path} -> {status_code} ({elapsed:.3f}s)"
|
||||
)
|
||||
|
||||
# 异步记录API调用统计
|
||||
try:
|
||||
db = SessionLocal()
|
||||
try:
|
||||
_record_api_call(db, method, path, status_code, elapsed)
|
||||
finally:
|
||||
db.close()
|
||||
except Exception as e:
|
||||
logger.warning(f"API统计记录失败: {e}")
|
||||
|
||||
return response
|
||||
except Exception as e:
|
||||
elapsed = time.time() - start_time
|
||||
if not should_skip:
|
||||
access_logger.info(
|
||||
f"[{request_id}] {method} {path} -> ERROR ({elapsed:.3f}s): {str(e)}"
|
||||
)
|
||||
raise
|
||||
|
||||
async def __call__(self, scope, receive, send):
|
||||
if scope["type"] != "http":
|
||||
await self.app(scope, receive, send)
|
||||
return
|
||||
await super().__call__(scope, receive, send)
|
||||
|
||||
|
||||
def _record_api_call(db, method: str, path: str, status_code: int, elapsed: float):
|
||||
"""记录API调用到数据库统计表"""
|
||||
# 简化路径以减少基数（去掉动态ID）
|
||||
parts = path.split("/")
|
||||
simplified_parts = []
|
||||
for part in parts:
|
||||
# 将UUID样式的路径段替换为{id}
|
||||
if len(part) == 36 and part.count("-") == 4:
|
||||
simplified_parts.append("{id}")
|
||||
# 将纯数字路径段替换为{id}
|
||||
elif part.isdigit():
|
||||
simplified_parts.append("{id}")
|
||||
else:
|
||||
simplified_parts.append(part)
|
||||
simplified_path = "/".join(simplified_parts)
|
||||
|
||||
# 统计记录通过数据库实现，简化版记录到日志
|
||||
is_error = status_code >= 400
|
||||
is_slow = elapsed >= SLOW_REQUEST_THRESHOLD
|
||||
|
||||
if is_slow:
|
||||
logger.warning(
|
||||
f"慢请求: {method} {simplified_path} -> {status_code} ({elapsed:.3f}s, 阈值: {SLOW_REQUEST_THRESHOLD}s)"
|
||||
)
|
||||
|
||||
|
||||
@asynccontextmanager
|
||||
async def lifespan(app: FastAPI):
|
||||
logger.info("Qihu-AIFlow v2.0.0 启动中...")
|
||||
|
||||
# 启动后台 Evolution Pipeline（知识库图谱已存在的用户配置自动恢复运行）
|
||||
try:
|
||||
from app.services.evolution_service import EvolutionWatcherManager
|
||||
from app.models import UserEvolutionConfig
|
||||
db = SessionLocal()
|
||||
try:
|
||||
configs = db.query(UserEvolutionConfig).filter(
|
||||
UserEvolutionConfig.watcher_status == "running"
|
||||
).all()
|
||||
if configs:
|
||||
watcher = EvolutionWatcherManager()
|
||||
for cfg in configs:
|
||||
try:
|
||||
watcher.start_user_evolution(cfg.user_id, lambda: SessionLocal())
|
||||
logger.info(f"[Evolution] 恢复用户 {cfg.user_id} 的进化Pipeline")
|
||||
except Exception as e:
|
||||
logger.warning(f"[Evolution] 恢复用户 {cfg.user_id} 失败: {e}")
|
||||
logger.info(f"[Evolution] 已恢复 {len(configs)} 个用户的进化Pipeline")
|
||||
else:
|
||||
logger.info("[Evolution] 无待恢复的进化配置")
|
||||
finally:
|
||||
db.close()
|
||||
except Exception as e:
|
||||
logger.warning(f"[Evolution] Pipeline自动启动异常(不影响正常服务): {e}")
|
||||
|
||||
# 启动Redis事件总线
|
||||
try:
|
||||
from app.event_bus import get_event_bus
|
||||
event_bus = get_event_bus()
|
||||
event_bus.start(num_consumers=2)
|
||||
logger.info("[EventBus] Redis事件总线已启动")
|
||||
except Exception as e:
|
||||
logger.warning(f"[EventBus] 启动失败(不影响正常服务): {e}")
|
||||
|
||||
logger.info("Qihu-AIFlow v2.0.0 启动完成")
|
||||
|
||||
yield
|
||||
|
||||
# 停止Redis事件总线
|
||||
try:
|
||||
from app.event_bus import get_event_bus
|
||||
get_event_bus().stop()
|
||||
logger.info("[EventBus] Redis事件总线已停止")
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
# 优雅关闭：停止所有 Evolution Pipeline
|
||||
try:
|
||||
from app.services.evolution_service import EvolutionWatcherManager
|
||||
EvolutionWatcherManager().shutdown_all()
|
||||
logger.info("[Evolution] 所有Pipeline已停止")
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
logger.info("Qihu-AIFlow 正在关闭...")
|
||||
|
||||
|
||||
app = FastAPI(
|
||||
title="Qihu-AIFlow - 智能自动化PaaS平台",
|
||||
version="1.0.0",
|
||||
lifespan=lifespan,
|
||||
)
|
||||
|
||||
# API调用追踪中间件
|
||||
app.add_middleware(APITrackingMiddleware)
|
||||
|
||||
# CORS配置
|
||||
app.add_middleware(
|
||||
CORSMiddleware,
|
||||
allow_origins=["*"],
|
||||
allow_credentials=True,
|
||||
allow_methods=["*"],
|
||||
allow_headers=["*"],
|
||||
)
|
||||
|
||||
# 前台 API
|
||||
app.include_router(auth.router, prefix="/app/api/auth", tags=["认证"])
|
||||
app.include_router(chat.router, prefix="/app/api/chat", tags=["AI聊天"])
|
||||
app.include_router(knowledge.router, prefix="/app/api/knowledge", tags=["知识库"])
|
||||
app.include_router(market.router, prefix="/app/api/market", tags=["应用市场"])
|
||||
app.include_router(token_usage.router, prefix="/app/api/token", tags=["Token驾驶舱"])
|
||||
app.include_router(payment.router, prefix="/app/api/payment", tags=["付费中心"])
|
||||
app.include_router(profile.router, prefix="/app/api/profile", tags=["个人中心"])
|
||||
app.include_router(notifications.router, prefix="/app/api/notifications", tags=["消息通知"])
|
||||
app.include_router(assets.router, prefix="/app/api/assets", tags=["资产管理"])
|
||||
app.include_router(evolution.router, prefix="/app/api/evolution", tags=["AI自进化"])
|
||||
|
||||
# 后台 API
|
||||
app.include_router(dashboard.router, prefix="/app/api/admin/dashboard", tags=["后台-仪表盘"])
|
||||
app.include_router(users.router, prefix="/app/api/admin/users", tags=["后台-用户管理"])
|
||||
app.include_router(models.router, prefix="/app/api/admin/models", tags=["后台-模型管理"])
|
||||
app.include_router(settings.router, prefix="/app/api/admin/settings", tags=["后台-网站设置"])
|
||||
app.include_router(payment_admin.router, prefix="/app/api/admin/payment", tags=["后台-支付配置"])
|
||||
app.include_router(services.router, prefix="/app/api/admin/services", tags=["后台-增值服务"])
|
||||
app.include_router(membership_admin.router, prefix="/app/api/admin/membership", tags=["后台-会员管理"])
|
||||
app.include_router(ads.router, prefix="/app/api/admin/ads", tags=["后台-广告管理"])
|
||||
app.include_router(market_admin.router, prefix="/app/api/admin/market", tags=["后台-应用审核"])
|
||||
app.include_router(logs.router, prefix="/app/api/admin/logs", tags=["后台-日志风控"])
|
||||
|
||||
app.include_router(skills_admin.router, prefix="/app/api/admin/skills", tags=["后台-技能管理"])
|
||||
app.include_router(agents_admin.router, prefix="/app/api/admin/agents", tags=["后台-Agent管理"])
|
||||
app.include_router(knowledge_admin.router, prefix="/app/api/admin/knowledge", tags=["后台-知识库管理"])
|
||||
app.include_router(workflow_admin.router, prefix="/app/api/admin/workflows", tags=["后台-工作流编排"])
|
||||
app.include_router(asset_admin.router, prefix="/app/api/admin/assets", tags=["后台-资产管理"])
|
||||
app.include_router(storage_admin.router, prefix="/app/api/admin/storage", tags=["后台-存储配置"])
|
||||
app.include_router(automation_admin.router, prefix="/app/api/admin/automation", tags=["后台-自动化任务"])
|
||||
app.include_router(mcp_admin.router, prefix="/app/api/admin/mcp", tags=["后台-MCP服务"])
|
||||
app.include_router(developer_admin.router, prefix="/app/api/admin/developer", tags=["后台-开发者API"])
|
||||
app.include_router(developer_tools_router, prefix="/app/api/admin/developer-tools", tags=["dev-tools"])
|
||||
app.include_router(apps_admin.router, prefix="/app/api/admin/apps", tags=["admin-ai-apps"])
|
||||
app.include_router(mcp_market_router, prefix="/app/api/admin/mcp-market", tags=["mcp-market"])
|
||||
app.include_router(code_exec_admin.router, prefix="/app/api/admin/code-exec", tags=["code-exec"])
|
||||
app.include_router(distro_admin.router, prefix="/app/api/admin/distributions", tags=["distributions"])
|
||||
app.include_router(privacy_admin.router, prefix="/app/api/admin/privacy", tags=["privacy"])
|
||||
app.include_router(im_bot_admin.router, prefix="/app/api/admin/im-bot", tags=["im-bot"])
|
||||
app.include_router(im_bot_user_router, prefix="/app/api/im/wechat", tags=["im-wechat-user"])
|
||||
app.include_router(im_bot_platform_router, prefix="/app/api/im/platform", tags=["im-platform-user"])
|
||||
app.include_router(promotion_admin.router, prefix="/app/api/admin/promotion", tags=["promotion"])
|
||||
app.include_router(tenant_admin.router, prefix="/app/api/admin/tenant", tags=["tenant"])
|
||||
app.include_router(billing_admin.router, prefix="/app/api/admin/billing", tags=["billing"])
|
||||
app.include_router(analytics_admin.router, prefix="/app/api/admin/analytics", tags=["analytics"])
|
||||
app.include_router(observability_admin.router, prefix="/app/api/admin/observability", tags=["后台-调用可观测"])
|
||||
app.include_router(user_portal.router, prefix="/app/api/portal", tags=["portal"])
|
||||
app.include_router(achievement_admin.router, prefix="/app/api/admin/achievement", tags=["achievement"])
|
||||
app.include_router(casdoor_router, prefix="/app/api/auth", tags=["Casdoor SSO"])
|
||||
|
||||
|
||||
@app.get("/app/api/health")
|
||||
async def health_check():
|
||||
return {"status": "ok", "service": "Qihu-AIFlow"}
|
||||
@@ -1,30 +1,30 @@
|
||||
# AIFlow 后端依赖
|
||||
fastapi==0.115.6
|
||||
uvicorn==0.34.0
|
||||
sqlalchemy==2.0.36
|
||||
pymysql==1.1.1
|
||||
cryptography==44.0.0
|
||||
passlib[bcrypt]==1.7.4
|
||||
python-jose[cryptography]==3.3.0
|
||||
python-multipart==0.0.18
|
||||
redis==5.2.1
|
||||
pydantic==2.10.3
|
||||
pydantic-settings==2.7.0
|
||||
python-dotenv==1.0.1
|
||||
httpx==0.28.1
|
||||
aiofiles==24.1.0
|
||||
celery==5.4.0
|
||||
qrcode==8.2
|
||||
pillow==12.2.0
|
||||
xskill==0.5.2
|
||||
|
||||
# ---- pinned celery ecosystem (known-good, extracted from running container) ----
|
||||
# 避免 pip 浮动解析到与 celery 5.4.0 冲突的新版传递依赖(如 click-repl 0.2.0 的 prompt-toolkit 冲突)
|
||||
amqp==5.3.1
|
||||
kombu==5.6.2
|
||||
billiard==4.2.4
|
||||
vine==5.1.0
|
||||
click==8.4.2
|
||||
click-didyoumean==0.3.1
|
||||
click-plugins==1.1.1.2
|
||||
click-repl==0.3.0
|
||||
# AIFlow 后端依赖
|
||||
fastapi==0.115.6
|
||||
uvicorn==0.34.0
|
||||
sqlalchemy==2.0.36
|
||||
pymysql==1.1.1
|
||||
cryptography==44.0.0
|
||||
passlib[bcrypt]==1.7.4
|
||||
python-jose[cryptography]==3.3.0
|
||||
python-multipart==0.0.18
|
||||
redis==5.2.1
|
||||
pydantic==2.10.3
|
||||
pydantic-settings==2.7.0
|
||||
python-dotenv==1.0.1
|
||||
httpx==0.28.1
|
||||
aiofiles==24.1.0
|
||||
celery==5.4.0
|
||||
qrcode==8.2
|
||||
pillow==12.2.0
|
||||
xskill==0.5.2
|
||||
|
||||
# ---- pinned celery ecosystem (known-good, extracted from running container) ----
|
||||
# 避免 pip 浮动解析到与 celery 5.4.0 冲突的新版传递依赖(如 click-repl 0.2.0 的 prompt-toolkit 冲突)
|
||||
amqp==5.3.1
|
||||
kombu==5.6.2
|
||||
billiard==4.2.4
|
||||
vine==5.1.0
|
||||
click==8.4.2
|
||||
click-didyoumean==0.3.1
|
||||
click-plugins==1.1.1.2
|
||||
click-repl==0.3.0
|
||||
Reference in new issue
Block a user