Files
ailine/backend/app/subgraphs/contact/graph.py
root b5c15ef445
All checks were successful
构建并部署 AI Agent 服务 / deploy (push) Successful in 12m9s
refactor: 单图方案重构 + 动态模型选择 + chat_services优化
## 核心改动

### 1. 单图方案重构
- 删除了多图(self.graphs),改为单图(self.graph)
- 新增 MainGraphState.current_model 字段用于运行时注入模型
- llm_call 节点改为动态选择模型(create_dynamic_llm_call_node)

### 2. chat_services 优化
- 添加 _cached_services 缓存,避免重复初始化
- 新增 get_cached_chat_services() 函数,用于单图注入
- 新增 _check_http_service_available() 统一HTTP探测逻辑
- 减少重复代码,LocalVLLMChatProvider和LocalSmallModelProvider共用探测方法

### 3. AIAgentService 重构
- initialize() 只构建一次图,传入 chat_services 字典
- 新增 _resolve_model() 模型回退逻辑
- 新增 _build_invocation() 统一构建调用参数
- process_message() 和 process_message_stream() 改为注入 current_model
- 流式处理代码拆分,增加可读性

### 4. 新增和删除文件
- 新增:backend/app/main_graph/main_graph_builder.py(图构建)
- 新增:backend/app/main_graph/subgraph_wrapper.py(子图封装)
- 新增:tools/test/test_tavily_search.py(测试)
- 删除:backend/app/main_graph/graph.py(旧图)
- 删除:backend/app/main_graph/utils/main_graph_builder.py(旧构建器)
- 删除:backend/app/main_graph/utils/__init__.py

### 5. 其他更新
- README.md:新增模型服务使用情况详解章节
- backend/app/model_services/__init__.py:新增 get_cached_chat_services 导出

## 方案优势

- 内存优化:N张图 → 1张图
- 灵活性:运行时动态选择模型,支持同会话不同模型
- 性能:模型服务缓存,初始化仅一次
- 可维护性:减少重复代码,统一HTTP探测逻辑
2026-05-05 17:30:55 +08:00

110 lines
3.3 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

"""
通讯录子图构建器
Contact Subgraph Builder
支持 API 注入的工厂模式
"""
from langgraph.graph import StateGraph, START, END
from .state import ContactState
from .nodes import create_contact_nodes
def build_contact_subgraph(contact_api=None):
"""
构建通讯录子图(工厂模式)
Args:
contact_api: 可选的 ContactAPIClient 实例(支持真实数据库或模拟模式)
不传入则使用默认模拟 API向后兼容
Returns:
配置好的 StateGraph
"""
# 创建节点(传入 API
nodes = create_contact_nodes(contact_api) if contact_api else None
# 如果没有传入 API使用向后兼容的导入
if nodes is None:
from .nodes import (
parse_intent,
list_contacts,
add_contact,
list_emails,
generate_email_draft,
human_review,
send_email,
sniff_contacts,
format_result,
should_continue
)
else:
parse_intent = nodes["parse_intent"]
list_contacts = nodes["list_contacts"]
add_contact = nodes["add_contact"]
list_emails = nodes["list_emails"]
generate_email_draft = nodes["generate_email_draft"]
human_review = nodes["human_review"]
send_email = nodes["send_email"]
sniff_contacts = nodes["sniff_contacts"]
format_result = nodes["format_result"]
should_continue = nodes["should_continue"]
# 创建图
graph = StateGraph(ContactState)
# 添加节点
graph.add_node("parse_intent", parse_intent)
graph.add_node("list_contacts", list_contacts)
graph.add_node("add_contact", add_contact)
graph.add_node("list_emails", list_emails)
graph.add_node("generate_email_draft", generate_email_draft)
graph.add_node("human_review", human_review)
graph.add_node("send_email", send_email)
graph.add_node("sniff_contacts", sniff_contacts)
graph.add_node("format_result", format_result)
# 添加边
# 从START开始
graph.add_edge(START, "parse_intent")
# 从parse_intent根据条件路由
graph.add_conditional_edges(
"parse_intent",
should_continue,
{
"list_contacts": "list_contacts",
"add_contact": "add_contact",
"list_emails": "list_emails",
"generate_email_draft": "generate_email_draft",
"sniff_contacts": "sniff_contacts",
}
)
# 从各个操作节点到format_result
graph.add_edge("list_contacts", "format_result")
graph.add_edge("add_contact", "format_result")
graph.add_edge("list_emails", "format_result")
graph.add_edge("sniff_contacts", "format_result")
# 邮件发送的特殊流程
graph.add_edge("generate_email_draft", "human_review")
# 从human_review根据条件路由
graph.add_conditional_edges(
"human_review",
should_continue,
{
"send_email": "send_email",
"format_result": "format_result",
}
)
# 发送邮件后到格式化
graph.add_edge("send_email", "format_result")
# 最终到END
graph.add_edge("format_result", END)
return graph