Files
zhenxun_bot/zhenxun/utils/utils.py
T
80fc5b86a7 ✨ feat!(llm): 重构并升级大语言模型服务为全新 AI 智能体框架 (#2146)
* ✨ feat!(llm): 重构并升级大语言模型服务为全新 AI 智能体框架

- 【重构】将原 services/llm 重构并迁移至全新的 services/ai 架构,提供向下兼容垫片
- 【新增】引入 Agent、Team、Workflow 三大智能体与工作流编排范式
- 【新增】引入基于 RAG 的长期向量记忆与中期槽位记忆系统
- 【新增】引入基于 Docker 的安全代码执行沙箱环境
- 【新增】支持 MCP 协议,允许动态管理和调用 MCP 服务
- 【新增】引入输入输出安全合规护栏与自愈反思机制
- 【优化】重构并优化多厂商 API 适配器 (Gemini, OpenAI, DeepSeek, GLM 等)
- 【优化】优化日志脱敏与 Token 预估机制
- 【移除】移除旧版 llm default 和 llm reset-key 命令,新增 llm mcp 管理命令

* 🔧 chore(deps): 更新项目依赖与配置

- 添加 mcp、jieba 和 aiodocker 依赖到配置文件及 requirements.txt
- 在 pyright 配置中设置 reportMissingImports 为 none
- 调整 .gitignore 中 resources 目录的忽略规则

* ♻️ refactor(tools): 重构工具终止机制并清理知识库日志输出

- 统一使用 `context.state["__end_run__"]` 替代 `EndRunResult` 控制任务结束
- 移除文件系统和向量知识库检索工具中 `ToolResult` 的 `.with_log` 调用
- 调整指令处理器(Directive)的返回值为 `tool_res.output`
- 修复部分类型检查警告并优化联合类型判断语法

* ♻️ refactor(tools): 重构工具副作用指令与控制流熔断机制

- 引入 `DirectivePayload` 及 `ToolResult` 的子类以结构化表达工具副作用
- 移除通过 `context.state` 传递魔术变量的隐式控制流设计
- 重构 `DirectiveManager` 处理器接口,直接在处理器中修改 `AgentState` 并构建 `AgentRunResult`
- 在 `StandardAgentExecutor` 中统一通过 `directive_manager` 调度工具返回的副作用指令
- 补全 `MessageBuilder` 中部分核心方法的文档注释

* 🐛 fix(sandbox): 修复 Docker 沙箱容器状态检测与会话清理逻辑

-【修复】修正 `is_alive` 中直接读取私有属性的问题,改用 `show()` 返回值
-【修复】解决 `execute_code` 中缓存的执行器与当前会话不一致的问题
-【优化】在清理工作区前增加容器存活检测,避免向已死容器发送请求
-【优化】创建容器时增加运行状态校验,若已停止则自动从缓存中移除并重建
-【优化】优化容器销毁和清理逻辑,静默处理容器不存在 (404) 的异常

* 📝 docs(core): 补充核心模块初始化方法的文档注释

* 🚨 auto fix by pre-commit hooks

---------

Co-authored-by: webjoin111 <455457521@qq.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2026-07-03 08:53:56 +08:00

315 lines
8.1 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.
from collections.abc import Callable
from dataclasses import dataclass
from datetime import datetime
import os
from pathlib import Path
import stat
import time
from types import TracebackType
from typing import Any, ClassVar
import httpx
from nonebot_plugin_uninfo import Uninfo
import pypinyin
from zhenxun.configs.config import Config
from zhenxun.services.log import logger
from .limiters import CountLimiter, FreqLimiter, UserBlockLimiter # noqa: F401
@dataclass
class EntityIDs:
user_id: str
"""用户id"""
group_id: str | None
"""群组id"""
channel_id: str | None
"""频道id"""
class ResourceDirManager:
"""
临时文件管理器
"""
temp_path: ClassVar[set[Path]] = set()
@classmethod
def __tree_append(cls, path: Path, deep: int = 1, current: int = 0):
"""递归添加文件夹"""
if current >= deep and deep != -1:
return
path = path.resolve() # 标准化路径
for f in os.listdir(path):
file = (path / f).resolve() # 标准化子路径
if file.is_dir():
if file not in cls.temp_path:
cls.temp_path.add(file)
logger.debug(f"添加临时文件夹: {file}")
cls.__tree_append(file, deep, current + 1)
@classmethod
def add_temp_dir(cls, path: str | Path, tree: bool = False, deep: int = 1):
"""添加临时清理文件夹,这些文件夹会被自动清理
参数:
path: 文件夹路径
tree: 是否递归添加文件夹
deep: 深度, -1 为无限深度
"""
if isinstance(path, str):
path = Path(path)
if path not in cls.temp_path:
cls.temp_path.add(path)
logger.debug(f"添加临时文件夹: {path}")
if tree:
cls.__tree_append(path, deep)
def is_binary_file(file_path: str) -> bool:
"""判断是否为二进制文件
参数:
file_path: 文件路径
返回:
bool: 是否为二进制文件
"""
# fmt: off
# 精简但包含图片和字体的二进制文件扩展名集合
BINARY_EXTENSIONS = frozenset({
# 图片文件
"jpg", "jpeg", "png", "gif", "bmp", "ico", "webp", "tiff", "tif", "svg",
# 字体文件
"ttf", "otf", "woff", "woff2", "eot",
# 压缩文件
"zip", "rar", "7z", "tar", "gz", "bz2", "xz",
# 可执行文件和库
"exe", "dll", "so", "dylib",
# 文档文件
"pdf", "doc", "docx", "xls", "xlsx", "ppt", "pptx",
# 多媒体文件
"mp3", "mp4", "avi", "mov", "wmv", "flv",
# 其他常见二进制文件
"bin", "dat", "db", "class", "pyc"
})
# 使用os.path.splitext高效提取扩展名
_, ext = os.path.splitext(file_path)
# 去除点号并转换为小写
ext_clean = ext.lstrip(".").lower()
return ext_clean in BINARY_EXTENSIONS
def cn2py(word: str) -> str:
"""将字符串转化为拼音
参数:
word: 文本
"""
return "".join("".join(i) for i in pypinyin.pinyin(word, style=pypinyin.NORMAL))
async def get_user_avatar(uid: int | str) -> bytes | None:
"""快捷获取用户头像
参数:
uid: 用户id
"""
url = f"http://q1.qlogo.cn/g?b=qq&nk={uid}&s=160"
async with httpx.AsyncClient() as client:
for _ in range(3):
try:
return (await client.get(url)).content
except Exception:
logger.error("获取用户头像错误", "Util", target=uid)
return None
async def get_group_avatar(gid: int | str) -> bytes | None:
"""快捷获取用群头像
参数:
gid: 群号
"""
url = f"http://p.qlogo.cn/gh/{gid}/{gid}/640/"
async with httpx.AsyncClient() as client:
for _ in range(3):
try:
return (await client.get(url)).content
except Exception:
logger.error("获取群头像错误", "Util", target=gid)
return None
def change_pixiv_image_links(
url: str, size: str | None = None, nginx_url: str | None = None
) -> str:
"""根据配置改变图片大小和反代链接
参数:
url: 图片原图链接
size: 模式
nginx_url: 反代
返回:
str: url
"""
if size == "master":
img_sp = url.rsplit(".", maxsplit=1)
url = img_sp[0]
img_type = img_sp[1]
url = url.replace("original", "master") + f"_master1200.{img_type}"
if not nginx_url:
nginx_url = Config.get_config("pixiv", "PIXIV_NGINX_URL")
if nginx_url:
url = (
url.replace("i.pximg.net", nginx_url)
.replace("i.pixiv.cat", nginx_url)
.replace("i.pixiv.re", nginx_url)
.replace("_webp", "")
)
return url
def change_img_md5(path_file: str | Path) -> bool:
"""改变图片MD5
参数:
path_file: 图片路径
返还:
bool: 是否修改成功
"""
try:
with open(path_file, "a", encoding="utf-8") as f:
f.write(str(int(time.time() * 1000)))
return True
except Exception as e:
logger.warning(f"改变图片MD5错误 Path:{path_file}", e=e)
return False
def is_valid_date(date_text: str, separator: str = "-") -> bool:
"""日期是否合法
参数:
date_text: 日期
separator: 分隔符
返回:
bool: 日期是否合法
"""
try:
datetime.strptime(date_text, f"%Y{separator}%m{separator}%d")
return True
except ValueError:
return False
def get_entity_ids(session: Uninfo) -> EntityIDs:
"""获取用户id,群组id,频道id
参数:
session: Uninfo
返回:
EntityIDs: 用户id,群组id,频道id
"""
user_id = session.user.id
group_id = None
channel_id = None
if session.group:
if session.group.parent:
group_id = session.group.parent.id
channel_id = session.group.id
else:
group_id = session.group.id
return EntityIDs(user_id=user_id, group_id=group_id, channel_id=channel_id)
def is_number(text: str) -> bool:
"""是否为数字
参数:
text: 文本
返回:
bool: 是否为数字
"""
try:
float(text)
return True
except ValueError:
return False
def win_on_rm_error(
func: Callable[[str], Any],
path: str,
_exc_info: tuple[type[BaseException], BaseException, TracebackType],
) -> None:
"""Windows下删除只读文件/目录时的回调。
去除只读属性后重试删除,避免 WinError 5。
"""
try:
os.chmod(path, stat.S_IWRITE)
except Exception:
# 即使去除权限失败也继续尝试
pass
try:
func(path)
except Exception:
# 仍失败则记录调试日志并忽略,交由上层继续处理
logger.debug(f"删除失败重试仍失败: {path}")
def infer_plugin_namespace(
default: str = "global",
) -> str:
"""
智能推断调用者所在的插件命名空间。
"""
import inspect
from nonebot.plugin import get_plugin_by_module_name
from nonebot.plugin.manager import _current_plugin
plugin = _current_plugin.get()
if plugin:
return plugin.name
try:
stack = inspect.stack()
for frame_info in stack[1:]:
module = inspect.getmodule(frame_info.frame)
if not module:
continue
m_name = module.__name__
if m_name.startswith("zhenxun.services.") or m_name.startswith(
"zhenxun.utils."
):
continue
plugin = get_plugin_by_module_name(m_name)
if plugin:
return plugin.name
parts = m_name.split(".")
for keyword in ("plugins", "builtin_plugins"):
if keyword in parts:
idx = parts.index(keyword)
if len(parts) > idx + 1:
return parts[idx + 1]
continue
except Exception:
pass
return default