Files
MoviePilot/app/modules/wechatclawbot/wechatclawbot.py

2217 lines
86 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.
import base64
import hashlib
import json
import mimetypes
import os
import random
import re
import threading
import time
import uuid
from dataclasses import dataclass, field
from pathlib import Path
from typing import Any, Dict, List, Optional, Tuple
from urllib.parse import quote
from Crypto.Cipher import AES
from Crypto.Util.Padding import pad
from app.core.cache import FileCache
from app.core.config import settings
from app.core.context import Context, MediaInfo
from app.core.metainfo import MetaInfo
from app.log import logger
from app.utils.http import RequestUtils
from app.utils.string import StringUtils
@dataclass
class ILinkIncomingMessage:
"""iLink 归一化后的入站消息。"""
user_id: str
text: Optional[str] = None
username: Optional[str] = None
message_id: Optional[str] = None
chat_id: Optional[str] = None
context_token: Optional[str] = None
images: List[Dict[str, Any]] = field(default_factory=list)
audio_refs: List[str] = field(default_factory=list)
files: List[Dict[str, Any]] = field(default_factory=list)
raw: Dict[str, Any] = field(default_factory=dict)
def to_message_payload(self) -> Dict[str, Any]:
payload = {
"__channel__": "wechatclawbot",
"userid": self.user_id,
"username": self.username or self.user_id,
"text": self.text or "",
"message_id": self.message_id,
"chat_id": self.chat_id,
"context_token": self.context_token,
"images": self.images or None,
"audio_refs": self.audio_refs or None,
"files": self.files or None,
}
return {key: value for key, value in payload.items() if value is not None}
class ILinkClient:
"""iLink HTTP 客户端,负责二维码登录、消息发送与长轮询。"""
channel_version = "1.0.2"
cdn_base_url = "https://novac2c.cdn.weixin.qq.com/c2c"
def __init__(
self,
base_url: str,
bot_token: Optional[str] = None,
account_id: Optional[str] = None,
sync_buf: Optional[str] = None,
timeout: int = 20,
):
"""保存 iLink 会话参数,供二维码登录、长轮询和消息发送复用。"""
self.base_url = (base_url or "https://ilinkai.weixin.qq.com").rstrip("/")
self.bot_token = bot_token
self.account_id = account_id
self.sync_buf = sync_buf
self.timeout = timeout
def set_credentials(
self,
bot_token: Optional[str],
account_id: Optional[str] = None,
sync_buf: Optional[str] = None,
) -> None:
"""更新 iLink 登录凭证与会话状态。"""
self.bot_token = bot_token
self.account_id = account_id
if sync_buf is not None:
self.sync_buf = sync_buf
def _headers(self, auth_required: bool = True) -> Dict[str, str]:
"""构建请求头,包含必要的身份验证信息。"""
headers = {
"Content-Type": "application/json",
"Accept": "application/json, text/plain, */*",
"User-Agent": "MoviePilot-WechatClawBot/1.0",
}
if auth_required and self.bot_token:
headers["AuthorizationType"] = "ilink_bot_token"
headers["Authorization"] = f"Bearer {self.bot_token}"
headers["X-WECHAT-UIN"] = self._build_wechat_uin()
return headers
@staticmethod
def _build_wechat_uin() -> str:
"""生成一个随机的微信 UIN用于标识请求来源。"""
random_u32 = random.getrandbits(32)
return base64.b64encode(str(random_u32).encode("utf-8")).decode("ascii")
def _with_base_info(self, body: Optional[Dict[str, Any]]) -> Dict[str, Any]:
"""向请求体中补充基础信息(如版本号)。"""
payload = dict(body or {})
base_info = payload.get("base_info")
if not isinstance(base_info, dict):
base_info = {}
base_info.setdefault("channel_version", self.channel_version)
payload["base_info"] = base_info
return payload
@staticmethod
def _json(resp: Any) -> Dict[str, Any]:
"""尝试将响应解析为 JSON处理异常情况。"""
if not resp:
return {}
try:
return resp.json() or {}
except Exception:
text = (getattr(resp, "text", "") or "").strip()
if not text:
return {}
try:
return json.loads(text)
except Exception:
return {}
@staticmethod
def _short_text(value: Any, max_len: int = 240) -> str:
"""将对象转换为简短的字符串描述,用于日志记录。"""
if value is None:
return ""
if isinstance(value, (dict, list)):
try:
text = json.dumps(value, ensure_ascii=False)
except Exception:
text = str(value)
else:
text = str(value)
text = text.replace("\n", " ").replace("\r", " ").strip()
if len(text) > max_len:
return f"{text[:max_len]}..."
return text
@staticmethod
def _normalize_qrcode_url(value: Any) -> Optional[str]:
"""规范化二维码展示字段,兼容图片 URL、data URL 与裸 base64。"""
if value is None:
return None
raw = str(value).strip()
if not raw:
return None
lowered = raw.lower()
if lowered.startswith("data:image/"):
return raw
if lowered.startswith("//"):
return f"https:{raw}"
# 某些实现会直接返回二维码图片的 base64 内容,这里自动补齐 data URL 前缀
if len(raw) >= 128 and re.fullmatch(r"[A-Za-z0-9+/=_-]+", raw):
return f"data:image/png;base64,{raw}"
return raw
@staticmethod
def _pick_value(obj: Dict[str, Any], keys: List[str]) -> Optional[Any]:
"""从字典中尝试获取指定键中的第一个非空值。"""
for key in keys:
if key in obj and obj.get(key) not in (None, ""):
return obj.get(key)
return None
@staticmethod
def _pick_present_value(obj: Dict[str, Any], keys: List[str]) -> Optional[Any]:
"""
获取已存在的字段值,允许空字符串和 0 作为有效值。
sync_buf 这类游标字段可能会被服务端明确置为空串,表示游标已经回到初始状态。
这里不能复用默认的“忽略空字符串”逻辑,否则会错误沿用上一轮旧游标。
"""
for key in keys:
if key in obj and obj.get(key) is not None:
return obj.get(key)
return None
@classmethod
def _find_first_value(
cls, data: Any, keys: List[str], max_depth: int = 5
) -> Optional[Any]:
"""在嵌套结构中递归查找指定键中的第一个非空值。"""
if max_depth < 0 or data is None:
return None
if isinstance(data, dict):
direct = cls._pick_value(data, keys)
if direct not in (None, ""):
return direct
for value in data.values():
found = cls._find_first_value(value, keys, max_depth - 1)
if found not in (None, ""):
return found
elif isinstance(data, list):
for value in data:
found = cls._find_first_value(value, keys, max_depth - 1)
if found not in (None, ""):
return found
return None
@classmethod
def _find_first_list(
cls, data: Any, prefer_keys: List[str], max_depth: int = 5
) -> Optional[List[Any]]:
"""在嵌套结构中递归查找第一个符合条件的列表。"""
if max_depth < 0 or data is None:
return None
if isinstance(data, dict):
for key in prefer_keys:
value = data.get(key)
if isinstance(value, list):
return value
for value in data.values():
found = cls._find_first_list(value, prefer_keys, max_depth - 1)
if found is not None:
return found
elif isinstance(data, list):
if data and all(isinstance(item, dict) for item in data):
return data
for value in data:
found = cls._find_first_list(value, prefer_keys, max_depth - 1)
if found is not None:
return found
return None
@staticmethod
def _ok(payload: Dict[str, Any]) -> bool:
"""检查 API 响应是否表示成功。"""
if not payload:
return False
code = payload.get("errcode")
if code is None:
code = payload.get("code")
if code is None:
code = payload.get("ret")
if code is None:
err = payload.get("errmsg") or payload.get("error") or payload.get(
"error_msg"
)
if err and str(err).strip().lower() not in {"ok", "success", "succeed"}:
return False
state = payload.get("status") or payload.get("state")
if isinstance(state, str) and state.strip().lower() in {
"error",
"failed",
"fail",
}:
return False
return True
try:
return int(str(code)) == 0
except Exception:
return str(code).strip().lower() in {"0", "ok", "success", "succeed"}
def _is_send_success(self, payload: Dict[str, Any]) -> bool:
"""检查 API 响应中是否包含明确的发送成功信号。"""
if not payload:
return False
code = self._find_first_value(
payload, ["errcode", "code", "ret", "result_code", "status_code"]
)
if code is not None:
try:
return int(str(code)) == 0
except Exception:
return str(code).strip().lower() in {"0", "ok", "success", "succeed"}
success_flag = self._find_first_value(
payload, ["success", "ok", "is_success", "sent"]
)
if isinstance(success_flag, bool):
return success_flag
if success_flag is not None:
if str(success_flag).strip().lower() in {
"1",
"true",
"ok",
"success",
"succeed",
"sent",
}:
return True
state = self._find_first_value(payload, ["status", "state", "send_status"])
if state is not None:
if str(state).strip().lower() in {"ok", "success", "succeed", "sent", "done"}:
return True
if str(state).strip().lower() in {"failed", "error", "denied"}:
return False
err_text = self._find_first_value(payload, ["errmsg", "error", "error_msg", "detail"])
if err_text is not None and str(err_text).strip():
err_value = str(err_text).strip().lower()
if err_value not in {"ok", "success", "succeed", "sent"}:
return False
return False
def _is_send_explicit_failure(self, payload: Dict[str, Any]) -> bool:
"""检查 API 响应中是否包含明确的发送失败信号。"""
if not payload:
return False
code = self._find_first_value(
payload, ["errcode", "code", "ret", "result_code", "status_code"]
)
if code is not None:
try:
return int(str(code)) != 0
except Exception:
return str(code).strip().lower() not in {"0", "ok", "success", "succeed"}
success_flag = self._find_first_value(
payload, ["success", "ok", "is_success", "sent"]
)
if isinstance(success_flag, bool):
return not success_flag
if success_flag is not None:
if str(success_flag).strip().lower() in {
"0",
"false",
"fail",
"failed",
"error",
"denied",
}:
return True
state = self._find_first_value(payload, ["status", "state", "send_status"])
if state is not None:
if str(state).strip().lower() in {
"failed",
"error",
"denied",
"forbidden",
"blocked",
}:
return True
err_text = self._find_first_value(payload, ["errmsg", "error", "error_msg", "detail"])
if err_text is not None and str(err_text).strip():
if str(err_text).strip().lower() not in {"ok", "success", "succeed", "sent"}:
return True
return False
def _is_send_http_success(self, resp: Any, payload: Dict[str, Any]) -> bool:
"""根据 HTTP 状态码和解析后的响应判断发送是否成功。"""
if resp is None:
return False
status_code = getattr(resp, "status_code", None)
if status_code is None:
return False
try:
status_ok = 200 <= int(status_code) < 300
except Exception:
status_ok = False
if not status_ok:
return False
if not payload:
return True
return not self._is_send_explicit_failure(payload)
@staticmethod
def _build_user_candidates(to_user: str) -> List[str]:
"""为目标用户 ID 构建多种格式的候选列表以提高匹配成功率。"""
raw = str(to_user or "").strip()
if not raw:
return []
candidates = [raw]
if "@" in raw:
candidates.append(raw.split("@", 1)[0])
if raw.endswith("@im.wechat"):
candidates.append(raw[: -len("@im.wechat")])
else:
candidates.append(f"{raw}@im.wechat")
uniq: List[str] = []
for item in candidates:
value = str(item or "").strip()
if value and value not in uniq:
uniq.append(value)
return uniq
@staticmethod
def _build_text_payloads(user_id: str, text: str) -> List[Dict[str, Any]]:
"""构建多种格式的纯文本消息报文以兼容不同接口。"""
return [
{"to_user": user_id, "msg_type": "text", "text": {"content": text}},
{"to_user": user_id, "msg_type": "text", "text": text},
{"touser": user_id, "msgtype": "text", "text": {"content": text}},
{"touser": user_id, "msgtype": "text", "text": text},
{"to": user_id, "type": "text", "content": text},
{"to_user_id": user_id, "msg_type": "text", "content": text},
]
@staticmethod
def _aes_ecb_padded_size(plaintext_size: int) -> int:
"""计算 AES ECB 加密后的填充大小。"""
return ((int(plaintext_size) + 1 + 15) // 16) * 16
@staticmethod
def _encrypt_aes_ecb(plaintext: bytes, key: bytes) -> bytes:
"""使用 AES ECB 模式对明文进行加密。"""
cipher = AES.new(key, AES.MODE_ECB)
return cipher.encrypt(pad(plaintext, AES.block_size))
@staticmethod
def _encode_media_aes_key(aeskey: bytes) -> str:
"""将媒体 AES 密钥进行 Base64 编码。"""
return base64.b64encode(aeskey.hex().encode("ascii")).decode("ascii")
def _build_protocol_text_payload(
self, user_id: str, text: str, context_token: Optional[str]
) -> Dict[str, Any]:
"""构建协议层标准文本消息报文。"""
msg = {
"from_user_id": str(self.account_id or ""),
"to_user_id": user_id,
"client_id": f"mp-{uuid.uuid4()}",
"message_type": 2,
"message_state": 2,
"item_list": [{"type": 1, "text_item": {"text": text}}],
}
if context_token:
msg["context_token"] = context_token
return {"msg": msg}
def _build_protocol_image_payload(
self,
user_id: str,
context_token: Optional[str],
download_param: str,
aeskey_b64: str,
cipher_size: int,
) -> Dict[str, Any]:
"""构建协议层标准图片消息报文。"""
msg: Dict[str, Any] = {
"from_user_id": str(self.account_id or ""),
"to_user_id": user_id,
"client_id": f"mp-{uuid.uuid4()}",
"message_type": 2,
"message_state": 2,
"item_list": [
{
"type": 2,
"image_item": {
"media": {
"encrypt_query_param": download_param,
"aes_key": aeskey_b64,
"encrypt_type": 1,
},
"mid_size": int(cipher_size),
},
}
],
}
if context_token:
msg["context_token"] = context_token
return {"msg": msg}
def _build_protocol_file_payloads(
self,
user_id: str,
context_token: Optional[str],
download_param: str,
aeskey_b64: str,
cipher_size: int,
raw_size: int,
file_name: str,
mime_type: str,
file_md5: str,
) -> List[Dict[str, Any]]:
"""构建协议层标准文件消息报文(含多种格式候选)。"""
media = {
"encrypt_query_param": download_param,
"aes_key": aeskey_b64,
"encrypt_type": 1,
}
file_item = {
"name": file_name,
"file_name": file_name,
"filename": file_name,
"title": file_name,
"size": int(raw_size),
"file_size": int(raw_size),
"raw_size": int(raw_size),
"mid_size": int(cipher_size),
"mime_type": mime_type,
"content_type": mime_type,
"md5": file_md5,
"media": media,
}
msg_base: Dict[str, Any] = {
"from_user_id": str(self.account_id or ""),
"to_user_id": user_id,
"client_id": f"mp-{uuid.uuid4()}",
"message_type": 2,
"message_state": 2,
}
if context_token:
msg_base["context_token"] = context_token
protocol_candidates = []
for item_type in (6, 5, 3):
msg = dict(msg_base)
msg["item_list"] = [{"type": item_type, "file_item": dict(file_item)}]
protocol_candidates.append({"msg": msg})
simple_candidates = [
{
"to_user": user_id,
"msg_type": "file",
"file": dict(file_item),
},
{
"touser": user_id,
"msgtype": "file",
"file": dict(file_item),
},
{"to": user_id, "type": "file", "file": dict(file_item)},
]
return [*protocol_candidates, *simple_candidates]
def _request_upload_param(
self,
to_user: str,
plaintext: bytes,
media_types: Optional[List[int]] = None,
) -> Tuple[Optional[str], Optional[str], Optional[bytes], Optional[int], Optional[str]]:
"""
向 iLink 申请 CDN 上传参数。
iLink 对同一文件类型的兼容性并不稳定,这里会按候选 media_type 依次尝试,
成功后返回上传地址、AES 密钥和文件元信息给后续 CDN 上传流程复用。
"""
rawsize = len(plaintext)
rawfilemd5 = hashlib.md5(plaintext).hexdigest()
filesize = self._aes_ecb_padded_size(rawsize)
filekey = os.urandom(16).hex()
aeskey = os.urandom(16)
last_payload = None
media_type_candidates = media_types or [1]
for media_type in media_type_candidates:
body = self._with_base_info(
{
"filekey": filekey,
"media_type": media_type,
"to_user_id": to_user,
"rawsize": rawsize,
"rawfilemd5": rawfilemd5,
"filesize": filesize,
"no_need_thumb": True,
"aeskey": aeskey.hex(),
}
)
url = f"{self.base_url}/ilink/bot/getuploadurl"
resp = RequestUtils(
headers=self._headers(auth_required=True), timeout=self.timeout
).post(url, json=body)
payload = self._json(resp)
last_payload = payload or getattr(resp, "text", None)
upload_param = (
self._find_first_value(payload, ["upload_param", "uploadParam"])
if payload
else None
)
upload_full_url = (
self._find_first_value(payload, ["upload_full_url", "uploadFullUrl", "full_url"])
if payload
else None
)
if upload_param or upload_full_url:
return (
str(upload_param) if upload_param else None,
str(upload_full_url) if upload_full_url else None,
aeskey,
filesize,
filekey,
)
logger.warning(f"getuploadurl 失败: resp={self._short_text(last_payload)}")
return None, None, None, None, None
def _upload_encrypted_to_cdn(
self,
upload_param: Optional[str],
upload_full_url: Optional[str],
filekey: str,
plaintext: bytes,
aeskey: bytes,
) -> Tuple[Optional[str], Optional[int]]:
"""将文件按协议加密后上传到微信 CDN并返回后续发送消息所需的下载参数。"""
ciphertext = self._encrypt_aes_ecb(plaintext, aeskey)
if upload_full_url:
upload_url = str(upload_full_url).strip()
elif upload_param:
upload_url = (
f"{self.cdn_base_url}/upload?encrypted_query_param={quote(str(upload_param), safe='')}&"
f"filekey={quote(filekey, safe='')}"
)
else:
logger.warning("CDN 上传失败: 缺少 upload_url 参数")
return None, None
resp = RequestUtils(
headers={"Content-Type": "application/octet-stream"},
timeout=self.timeout,
).post(upload_url, data=ciphertext)
if getattr(resp, "status_code", None) != 200:
logger.warning(
f"CDN 上传失败: http={getattr(resp, 'status_code', None)}, "
f"err={self._short_text(getattr(resp, 'text', ''))}"
)
return None, None
download_param = None
if resp is not None and getattr(resp, "headers", None):
download_param = resp.headers.get("x-encrypted-param")
if not download_param:
logger.warning("CDN 上传成功但缺少 x-encrypted-param")
return None, None
return str(download_param), len(ciphertext)
def _send_payload_candidates(
self,
to_user: str,
payload_candidates: List[Dict[str, Any]],
url_candidates: Optional[List[str]] = None,
) -> bool:
"""
尝试一组候选发送报文,兼容 iLink 不同接口形态。
这里不会只依赖单一 URL 或单一报文结构,而是按“用户候选 -> 接口候选 -> 报文候选”
逐层回退,尽量提高不同账号和不同版本服务端的发送成功率。
"""
url_candidates = url_candidates or [
f"{self.base_url}/ilink/bot/sendmessage",
f"{self.base_url}/ilink/bot/sendmessage?bot_type=3",
]
last_error = ""
for user_id in self._build_user_candidates(to_user):
for url in url_candidates:
for index, body in enumerate(payload_candidates, start=1):
request_body = self._with_base_info(body)
resp = RequestUtils(
headers=self._headers(auth_required=True), timeout=self.timeout
).post(url, json=request_body)
payload = self._json(resp)
if self._is_send_success(payload) or self._is_send_http_success(resp, payload):
logger.info(f"发送消息成功: to_user={user_id}, variant={index}")
return True
http_code = getattr(resp, "status_code", None)
err_msg = (
self._find_first_value(payload, ["errmsg", "message", "error", "detail"])
if payload
else None
)
if not err_msg and resp is not None:
err_msg = self._short_text(getattr(resp, "text", ""))
last_error = f"http={http_code}, err={self._short_text(err_msg)}"
logger.debug(
f"发送候选失败: to_user={user_id}, variant={index}, "
f"{last_error}, req={self._short_text(request_body)}, "
f"resp={self._short_text(payload)}"
)
logger.warning(f"发送消息失败: to_user={to_user}, {last_error}")
return False
def get_qrcode(self) -> Dict[str, Any]:
"""向服务端请求并解析二维码登录信息。"""
url = f"{self.base_url}/ilink/bot/get_bot_qrcode?bot_type=3"
logger.debug(f"请求二维码: {url}")
resp = RequestUtils(
headers=self._headers(auth_required=False), timeout=self.timeout
).get_res(url)
payload = self._json(resp)
if not payload:
return {"success": False, "message": "获取二维码失败"}
data = payload.get("data") or payload.get("result") or payload
qrcode = data.get("qrcode") or data.get("qr_code") or data.get(
"qrcode_id"
) or data.get("ticket")
qrcode_url = (
data.get("qrcode_url")
or data.get("url")
or data.get("qrcodeUrl")
or data.get("qr_url")
or data.get("qrcode_img_content")
or data.get("qrcode_img_url")
or data.get("qr_img")
)
qrcode_url = self._normalize_qrcode_url(qrcode_url)
if not qrcode_url and qrcode:
qrcode_url = f"https://liteapp.weixin.qq.com/q/7GiQu1?qrcode={qrcode}&bot_type=3"
return {
"success": self._ok(payload) and bool(qrcode or qrcode_url),
"qrcode": qrcode,
"qrcode_url": qrcode_url,
"raw": payload,
"message": payload.get("errmsg") or payload.get("message"),
}
def get_qrcode_status(self, qrcode: str) -> Dict[str, Any]:
"""查询二维码扫描状态并获取登录凭证。"""
url = f"{self.base_url}/ilink/bot/get_qrcode_status"
resp = RequestUtils(
headers=self._headers(auth_required=False), timeout=self.timeout
).get_res(url, params={"qrcode": qrcode})
payload = self._json(resp)
if not payload:
retry_resp = RequestUtils(
headers=self._headers(auth_required=False), timeout=self.timeout
).get_res(f"{url}?qrcode={qrcode}")
payload = self._json(retry_resp)
if not payload:
return {
"success": False,
"status": "waiting",
"token": None,
"account_id": None,
"raw": {},
"message": "二维码状态接口返回空响应",
}
data = payload.get("data") or payload.get("result") or payload
token = (
data.get("bot_token")
or data.get("token")
or data.get("access_token")
or self._find_first_value(
data, ["bot_token", "access_token", "token", "jwt", "auth_token"]
)
)
account_id = (
data.get("account_id")
or data.get("ilink_bot_id")
or data.get("wxid")
or data.get("uid")
or data.get("user_id")
or self._find_first_value(
data,
[
"account_id",
"ilink_bot_id",
"wxid",
"uid",
"user_id",
"from_user",
"from_uid",
],
)
)
base_url = (
data.get("baseurl")
or data.get("base_url")
or payload.get("baseurl")
or payload.get("base_url")
)
if token:
self.bot_token = token
if account_id:
self.account_id = str(account_id)
state = (
data.get("status")
or data.get("state")
or payload.get("status")
or payload.get("state")
or self._find_first_value(data, ["status", "state", "scan_status"])
or "waiting"
)
return {
"success": self._ok(payload),
"status": str(state).lower(),
"token": token,
"account_id": account_id,
"base_url": base_url,
"qrcode_url": self._normalize_qrcode_url(
data.get("qrcode_url")
or data.get("url")
or data.get("qrcodeUrl")
or data.get("qr_url")
or data.get("qrcode_img_content")
or data.get("qrcode_img_url")
or data.get("qr_img")
),
"raw": payload,
"message": payload.get("errmsg") or payload.get("message"),
}
def send_text(self, to_user: str, text: str, context_token: Optional[str] = None) -> bool:
"""发送纯文本消息。"""
if not self.bot_token:
logger.warning("发送消息失败bot token 未配置")
return False
if not to_user or not text:
logger.warning("发送消息失败to_user 或 text 为空")
return False
payload_candidates = [
self._build_protocol_text_payload(
user_id=str(to_user), text=text, context_token=context_token
),
*self._build_text_payloads(str(to_user), text),
]
return self._send_payload_candidates(to_user=to_user, payload_candidates=payload_candidates)
def send_image_text_png(
self,
to_user: str,
image_bytes: bytes,
text: str,
context_token: Optional[str] = None,
) -> bool:
"""发送包含图片和文本的消息。"""
if not self.bot_token:
logger.warning("发送图文失败bot token 未配置")
return False
if not to_user or not image_bytes or not text:
logger.warning("发送图文失败to_user 或 image_bytes 或 text 为空")
return False
for user_id in self._build_user_candidates(to_user):
upload_param, upload_full_url, aeskey, _, filekey = self._request_upload_param(
user_id, image_bytes, media_types=[1]
)
if (not upload_param and not upload_full_url) or not aeskey or not filekey:
continue
download_param, cipher_size = self._upload_encrypted_to_cdn(
upload_param=upload_param,
upload_full_url=upload_full_url,
filekey=filekey,
plaintext=image_bytes,
aeskey=aeskey,
)
if not download_param or not cipher_size:
continue
aeskey_b64 = self._encode_media_aes_key(aeskey)
message_items = [
{"type": 1, "text_item": {"text": text}},
{
"type": 2,
"image_item": {
"media": {
"encrypt_query_param": download_param,
"aes_key": aeskey_b64,
"encrypt_type": 1,
},
"mid_size": int(cipher_size),
},
},
]
sent_all = True
for item in message_items:
msg = {
"from_user_id": str(self.account_id or ""),
"to_user_id": user_id,
"client_id": f"mp-{uuid.uuid4()}",
"message_type": 2,
"message_state": 2,
"item_list": [item],
}
if context_token:
msg["context_token"] = context_token
if not self._send_payload_candidates(
to_user=user_id,
payload_candidates=[{"msg": msg}],
):
sent_all = False
break
if sent_all:
return True
return False
def send_image_png(
self, to_user: str, image_bytes: bytes, context_token: Optional[str] = None
) -> bool:
"""发送图片消息。"""
if not self.bot_token:
logger.warning("发送图片失败bot token 未配置")
return False
if not to_user or not image_bytes:
logger.warning("发送图片失败to_user 或 image_bytes 为空")
return False
for user_id in self._build_user_candidates(to_user):
upload_param, upload_full_url, aeskey, _, filekey = self._request_upload_param(
user_id, image_bytes, media_types=[1]
)
if (not upload_param and not upload_full_url) or not aeskey or not filekey:
continue
download_param, cipher_size = self._upload_encrypted_to_cdn(
upload_param=upload_param,
upload_full_url=upload_full_url,
filekey=filekey,
plaintext=image_bytes,
aeskey=aeskey,
)
if not download_param or not cipher_size:
continue
aeskey_b64 = self._encode_media_aes_key(aeskey)
body = self._build_protocol_image_payload(
user_id=user_id,
context_token=context_token,
download_param=download_param,
aeskey_b64=aeskey_b64,
cipher_size=cipher_size,
)
if self._send_payload_candidates(
to_user=user_id,
payload_candidates=[body],
):
return True
return False
def send_file_bytes(
self,
to_user: str,
file_bytes: bytes,
file_name: str,
mime_type: str,
context_token: Optional[str] = None,
) -> bool:
"""发送文件消息。"""
if not self.bot_token:
logger.warning("发送文件失败bot token 未配置")
return False
if not to_user or not file_bytes:
logger.warning("发送文件失败to_user 或 file_bytes 为空")
return False
file_name = file_name or "attachment"
mime_type = mime_type or "application/octet-stream"
file_md5 = hashlib.md5(file_bytes).hexdigest()
for user_id in self._build_user_candidates(to_user):
upload_param, upload_full_url, aeskey, _, filekey = self._request_upload_param(
user_id, file_bytes, media_types=[2, 6, 1]
)
if (not upload_param and not upload_full_url) or not aeskey or not filekey:
continue
download_param, cipher_size = self._upload_encrypted_to_cdn(
upload_param=upload_param,
upload_full_url=upload_full_url,
filekey=filekey,
plaintext=file_bytes,
aeskey=aeskey,
)
if not download_param or not cipher_size:
continue
aeskey_b64 = self._encode_media_aes_key(aeskey)
payload_candidates = self._build_protocol_file_payloads(
user_id=user_id,
context_token=context_token,
download_param=download_param,
aeskey_b64=aeskey_b64,
cipher_size=cipher_size,
raw_size=len(file_bytes),
file_name=file_name,
mime_type=mime_type,
file_md5=file_md5,
)
if self._send_payload_candidates(
to_user=user_id,
payload_candidates=payload_candidates,
):
return True
return False
@classmethod
def _encode_ref_payload(cls, kind: str, payload: Dict[str, Any]) -> str:
"""将附件元信息编码为 wxclaw:// 协议链接。"""
encoded = base64.urlsafe_b64encode(
json.dumps(payload, ensure_ascii=False).encode("utf-8")
).decode("ascii").rstrip("=")
return f"wxclaw://{kind}/{encoded}"
def _build_attachment_ref(
self,
kind: str,
attachment: Dict[str, Any],
default_name: Optional[str] = None,
) -> Optional[Tuple[str, Dict[str, Any]]]:
"""从附件信息构建 wxclaw 协议引用。"""
if not isinstance(attachment, dict):
return None
download_url = (
attachment.get("download_url")
or attachment.get("url")
or attachment.get("cdnurl")
or attachment.get("cdn_url")
or attachment.get("file_url")
)
if not download_url:
return None
payload = {
"url": download_url,
"aeskey": attachment.get("aeskey")
or attachment.get("encoding_aes_key")
or attachment.get("encrypt_key"),
"encrypt_type": attachment.get("encrypt_type"),
"mime_type": attachment.get("mime_type")
or attachment.get("content_type"),
"name": attachment.get("name")
or attachment.get("filename")
or default_name,
"size": attachment.get("size") or attachment.get("file_size"),
}
ref = self._encode_ref_payload(kind=kind, payload=payload)
return ref, payload
@staticmethod
def _as_scalar(value: Any) -> Optional[Any]:
"""将值转换为标量,过滤掉字典和列表。"""
if value in (None, ""):
return None
if isinstance(value, (dict, list, tuple, set)):
return None
return value
def _build_poll_result(
self,
success: bool,
payload: Optional[Dict[str, Any]] = None,
message: Optional[str] = None,
item_count: int = 0,
parsed_count: int = 0,
) -> Dict[str, Any]:
"""构建轮询结果摘要。"""
payload = payload or {}
resolved_message = message or self._find_first_value(
payload, ["errmsg", "message", "error", "error_msg", "detail"]
)
return {
"success": success,
"raw": payload,
"message": self._short_text(resolved_message) if resolved_message else None,
"item_count": item_count,
"parsed_count": parsed_count,
}
def _parse_incoming(self, item: Dict[str, Any]) -> Optional[ILinkIncomingMessage]:
"""
将 getupdates 返回的原始事件归一化为 MoviePilot 可消费的入站消息。
iLink 返回结构存在多种嵌套和字段别名,这里集中做兼容处理,统一提取
用户、消息ID、文本、图片、语音和文件附件。
"""
if not isinstance(item, dict):
return None
message = item
for key in ["message", "msg", "event", "payload", "data", "body"]:
child = item.get(key)
if isinstance(child, dict):
message = child
break
sender = (
message.get("from")
if isinstance(message.get("from"), dict)
else message.get("sender")
if isinstance(message.get("sender"), dict)
else message.get("user")
if isinstance(message.get("user"), dict)
else message.get("from_user")
if isinstance(message.get("from_user"), dict)
else {}
)
user_id = (
self._pick_value(sender, ["user_id", "id", "wxid", "uid"])
or self._pick_value(
message,
[
"from_user",
"from_user_id",
"user_id",
"uid",
"wxid",
"from_uid",
"fromUser",
"fromUserId",
"openid",
],
)
or self._pick_value(
item,
[
"from_user",
"from_user_id",
"user_id",
"uid",
"wxid",
"from_uid",
"fromUser",
"fromUserId",
"openid",
],
)
or self._find_first_value(
message,
[
"from_user",
"from_user_id",
"user_id",
"sender_id",
"uid",
"wxid",
"from_uid",
"fromUserId",
"openid",
],
)
)
user_id = self._as_scalar(user_id)
if not user_id:
return None
username = (
self._pick_value(sender, ["name", "nickname", "username", "remark"])
or self._pick_value(
message, ["username", "nickname", "from_name", "fromNick", "sender_name"]
)
or str(user_id)
)
message_id = self._pick_value(
message, ["message_id", "msg_id", "id", "client_msg_id", "msgId", "seq"]
) or self._pick_value(
item, ["message_id", "msg_id", "id", "client_msg_id", "msgId", "seq"]
)
chat_id = self._pick_value(
message,
["chat_id", "conversation_id", "room_id", "chatId", "conversationId", "roomId"],
) or self._pick_value(
item,
["chat_id", "conversation_id", "room_id", "chatId", "conversationId", "roomId"],
)
context_token = self._pick_value(message, ["context_token", "contextToken"]) or self._pick_value(
item, ["context_token", "contextToken"]
)
text_parts: List[str] = []
images: List[Dict[str, Any]] = []
audio_refs: List[str] = []
files: List[Dict[str, Any]] = []
def append_text(value: Any) -> None:
if value in (None, ""):
return
if isinstance(value, dict):
value = self._pick_value(value, ["content", "text", "value", "message"])
if value in (None, ""):
return
text_value = str(value).strip()
if text_value and text_value not in text_parts:
text_parts.append(text_value)
msgtype = str(
message.get("msgtype") or message.get("msg_type") or message.get("type") or ""
).lower()
if msgtype == "text":
append_text((message.get("text") or {}).get("content") if isinstance(message.get("text"), dict) else None)
append_text(message.get("content"))
elif msgtype == "image":
image_data = message.get("image") or {}
image_ref = self._build_attachment_ref("image", image_data)
if image_ref:
ref, payload = image_ref
images.append(
{
"ref": ref,
"name": payload.get("name"),
"mime_type": payload.get("mime_type"),
"size": payload.get("size"),
}
)
elif msgtype == "file":
file_data = message.get("file") or {}
file_ref = self._build_attachment_ref("file", file_data)
if file_ref:
ref, payload = file_ref
files.append(
{
"ref": ref,
"name": payload.get("name"),
"mime_type": payload.get("mime_type"),
"size": payload.get("size"),
}
)
elif msgtype == "voice":
voice_data = message.get("voice") or {}
append_text((voice_data or {}).get("content"))
voice_ref = self._build_attachment_ref("voice", voice_data, default_name="voice.amr")
if voice_ref:
ref, _ = voice_ref
audio_refs.append(ref)
elif msgtype == "mixed":
for msg_item in ((message.get("mixed") or {}).get("msg_item") or []):
item_type = str(msg_item.get("msgtype") or msg_item.get("type") or "").lower()
if item_type == "text":
append_text((msg_item.get("text") or {}).get("content"))
elif item_type == "image":
image_ref = self._build_attachment_ref("image", msg_item.get("image") or {})
if image_ref:
ref, payload = image_ref
images.append(
{
"ref": ref,
"name": payload.get("name"),
"mime_type": payload.get("mime_type"),
"size": payload.get("size"),
}
)
elif item_type == "file":
file_ref = self._build_attachment_ref("file", msg_item.get("file") or {})
if file_ref:
ref, payload = file_ref
files.append(
{
"ref": ref,
"name": payload.get("name"),
"mime_type": payload.get("mime_type"),
"size": payload.get("size"),
}
)
elif item_type == "voice":
append_text(((msg_item.get("voice") or {}).get("content") or "").strip())
voice_ref = self._build_attachment_ref(
"voice",
msg_item.get("voice") or {},
default_name="voice.amr",
)
if voice_ref:
ref, _ = voice_ref
audio_refs.append(ref)
item_list = message.get("item_list") if isinstance(message.get("item_list"), list) else []
for one in item_list:
if not isinstance(one, dict):
continue
item_type = one.get("type")
if item_type == 1 and isinstance(one.get("text_item"), dict):
append_text((one.get("text_item") or {}).get("text"))
elif item_type == 2 and isinstance(one.get("image_item"), dict):
image_ref = self._build_attachment_ref(
"image", (one.get("image_item") or {}).get("media") or {}
)
if image_ref:
ref, payload = image_ref
images.append(
{
"ref": ref,
"name": payload.get("name"),
"mime_type": payload.get("mime_type"),
"size": payload.get("size"),
}
)
elif item_type in {5, 6}:
file_ref = self._build_attachment_ref(
"file", (one.get("file_item") or {}).get("media") or {}
)
if file_ref:
ref, payload = file_ref
files.append(
{
"ref": ref,
"name": (one.get("file_item") or {}).get("name")
or (one.get("file_item") or {}).get("file_name")
or payload.get("name"),
"mime_type": (one.get("file_item") or {}).get("mime_type")
or payload.get("mime_type"),
"size": (one.get("file_item") or {}).get("size")
or payload.get("size"),
}
)
if not text_parts:
append_text(
self._pick_value(message, ["content", "message", "msg", "text", "body", "msg_content", "msgContent"])
)
append_text(
self._pick_value(item, ["content", "message", "msg", "text", "body", "msg_content", "msgContent"])
)
append_text(self._find_first_value(message, ["content", "text", "message", "msg", "body", "cmd"]))
text = "\n".join(part for part in text_parts if part).strip() or None
return ILinkIncomingMessage(
user_id=str(user_id),
text=text,
username=str(username) if username else None,
message_id=str(message_id) if message_id else None,
chat_id=str(chat_id) if chat_id else None,
context_token=str(context_token) if context_token else None,
images=images,
audio_refs=audio_refs,
files=files,
raw=item,
)
def _extract_updates(
self, payload: Dict[str, Any]
) -> Tuple[List[Dict[str, Any]], Optional[str]]:
"""
提取轮询结果中的消息列表与游标。
线上存在两种等价字段命名:较新的实现返回 `get_updates_buf`
部分实例仍然返回 `sync_buf`。两者都表示下一轮轮询应携带的游标。
"""
sync_buf = self._pick_present_value(
payload, ["get_updates_buf", "sync_buf", "syncBuf"]
)
items = payload.get("msgs")
if isinstance(items, list):
return items, sync_buf
return [], sync_buf
@staticmethod
def _has_canonical_poll_shape(payload: Dict[str, Any]) -> bool:
"""官方响应至少应包含顶层 msgs 列表。"""
return isinstance(payload.get("msgs"), list)
def _resolve_poll_success(self, payload: Dict[str, Any]) -> Optional[bool]:
"""
判断 getupdates 是否给出了明确的成功/失败信号。
轮询接口不能沿用“只要没有明显报错就算成功”的宽松策略,否则服务端返回旧消息列表、
但状态码其实失败时,会被误判为可消费响应,导致旧消息再次进入业务链路。
返回 `None` 表示响应里没有显式状态,需要交给协议结构继续判断。
"""
if not payload:
return False
code = self._find_first_value(
payload, ["errcode", "code", "ret", "result_code", "status_code"]
)
if code is not None:
try:
return int(str(code)) == 0
except Exception:
return str(code).strip().lower() in {"0", "ok", "success", "succeed"}
success_flag = self._find_first_value(
payload, ["success", "ok", "is_success"]
)
if isinstance(success_flag, bool):
return success_flag
if success_flag is not None:
return str(success_flag).strip().lower() in {
"1",
"true",
"ok",
"success",
"succeed",
}
state = self._find_first_value(payload, ["status", "state"])
if state is not None:
lowered = str(state).strip().lower()
if lowered in {"ok", "success", "succeed", "done"}:
return True
if lowered in {"failed", "fail", "error", "denied", "blocked"}:
return False
return None
def poll_updates(
self, timeout_seconds: int = 25
) -> Tuple[List[ILinkIncomingMessage], Optional[str], Dict[str, Any]]:
"""
执行一次 iLink 长轮询。
返回值同时包含归一化后的消息、最新游标和原始结果摘要,
便于上层在推进 sync_buf 的同时记录调试信息。
"""
if not self.bot_token:
return [], self.sync_buf, {"success": False, "message": "bot token 未配置"}
url = f"{self.base_url}/ilink/bot/getupdates"
request_body = self._with_base_info({"get_updates_buf": self.sync_buf or ""})
resp = RequestUtils(
headers=self._headers(auth_required=True),
timeout=timeout_seconds + 10,
).post(url, json=request_body)
payload = self._json(resp)
explicit_success = self._resolve_poll_success(payload)
has_canonical_shape = self._has_canonical_poll_shape(payload)
# 某些 iLink 部署不会返回 ret/success但顶层 msgs + sync_buf 已经足够表明
# 这是一次有效的轮询响应;只有出现显式失败信号时才应拒绝消费。
success = bool(payload) and (
explicit_success is True
or (explicit_success is None and has_canonical_shape)
)
last_message = None
if payload and explicit_success is False:
last_message = self._find_first_value(
payload, ["errmsg", "message", "error", "error_msg", "detail"]
) or self._short_text(payload)
if not payload:
return [], self.sync_buf, self._build_poll_result(
success=False,
message="轮询返回空响应",
)
if not success:
return [], self.sync_buf, self._build_poll_result(
success=False,
payload=payload,
message=last_message or "轮询响应未明确成功",
)
if not has_canonical_shape:
logger.warning(
"getupdates 返回非官方结构,已拒绝消费: %s",
self._short_text(payload),
)
return [], self.sync_buf, self._build_poll_result(
success=False,
payload=payload,
message="轮询响应结构非官方,缺少顶层 msgs 字段",
)
items, sync_buf = self._extract_updates(payload)
parsed: List[ILinkIncomingMessage] = []
for item in items:
message = self._parse_incoming(item)
if message:
parsed.append(message)
if sync_buf is not None:
self.sync_buf = str(sync_buf)
return parsed, self.sync_buf, self._build_poll_result(
success=True,
payload=payload,
item_count=len(items),
parsed_count=len(parsed),
)
def test_connection(self) -> Tuple[bool, str]:
"""测试与 iLink 服务端的连接连通性。"""
if not self.bot_token:
return False, "未登录,缺少 bot token"
url = f"{self.base_url}/ilink/bot/getconfig"
resp = RequestUtils(
headers=self._headers(auth_required=True), timeout=self.timeout
).post(url, json={})
payload = self._json(resp)
if self._ok(payload):
return True, "连接正常"
return False, payload.get("errmsg") or payload.get("message") or "连接失败"
class WechatClawBot:
"""微信 ClawBot 渠道客户端。"""
_default_base_url = "https://ilinkai.weixin.qq.com"
_qrcode_ttl_seconds = 240
_active_target_ttl_seconds = 24 * 60 * 60
@classmethod
def _build_cache_key(cls, config_name: str) -> str:
"""根据配置名称构建缓存键。"""
safe_name = hashlib.md5(str(config_name or "wechatclawbot").encode("utf-8")).hexdigest()[:12]
return f"__wechatclawbot_state_{safe_name}__"
def __init__(
self,
WECHATCLAWBOT_BASE_URL: Optional[str] = None,
WECHATCLAWBOT_DEFAULT_TARGET: Optional[str] = None,
WECHATCLAWBOT_ADMINS: Optional[str] = None,
WECHATCLAWBOT_POLL_TIMEOUT: Optional[int] = None,
name: Optional[str] = None,
auto_start_polling: bool = True,
**kwargs,
):
"""初始化微信 ClawBot 实例及相关参数。"""
self._config_name = name or "wechatclawbot"
self._base_url = (WECHATCLAWBOT_BASE_URL or self._default_base_url).rstrip("/")
self._default_target = (WECHATCLAWBOT_DEFAULT_TARGET or "").strip() or None
self._auto_start_polling = bool(auto_start_polling)
self._admins = [
admin.strip()
for admin in str(WECHATCLAWBOT_ADMINS or "").split(",")
if admin.strip()
]
try:
self._poll_timeout = max(10, int(WECHATCLAWBOT_POLL_TIMEOUT or 25))
except Exception:
self._poll_timeout = 25
self._cache_key = self._build_cache_key(self._config_name)
self._filecache = FileCache()
self._lock = threading.Lock()
self._stop_event = threading.Event()
self._poll_thread: Optional[threading.Thread] = None
self._state = self._load_state()
self._message_endpoint = (
f"http://127.0.0.1:{settings.PORT}/api/v1/message?token={settings.API_TOKEN}&source={quote(self._config_name, safe='')}"
)
if self._state.get("bot_token") and self._auto_start_polling:
self._start_polling()
def _load_state(self) -> Dict[str, Any]:
"""从文件缓存加载登录状态。"""
content = self._filecache.get(self._cache_key)
if not content:
return {
"bot_token": None,
"account_id": None,
"sync_buf": None,
"qrcode": {},
"known_targets": {},
"user_context_tokens": {},
"base_url": self._base_url,
}
try:
data = json.loads(content.decode("utf-8"))
if not isinstance(data, dict):
raise ValueError("invalid state")
data.setdefault("qrcode", {})
data.setdefault("known_targets", {})
data.setdefault("user_context_tokens", {})
data.setdefault("base_url", self._base_url)
return data
except Exception as err:
logger.warning(f"加载登录状态失败,已重置缓存:{err}")
return {
"bot_token": None,
"account_id": None,
"sync_buf": None,
"qrcode": {},
"known_targets": {},
"user_context_tokens": {},
"base_url": self._base_url,
}
def _save_state(self) -> None:
"""将当前状态持久化到文件缓存。"""
self._state["base_url"] = self._base_url
self._filecache.set(
self._cache_key,
json.dumps(self._state, ensure_ascii=False).encode("utf-8"),
)
def _build_client(self) -> ILinkClient:
"""根据当前持久化状态构建一次性的 iLink 客户端。"""
return ILinkClient(
base_url=self._state.get("base_url") or self._base_url,
bot_token=self._state.get("bot_token"),
account_id=self._state.get("account_id"),
sync_buf=self._state.get("sync_buf"),
timeout=max(self._poll_timeout, 20),
)
def _update_state(self, **kwargs) -> None:
"""更新并保存持久化状态。"""
with self._lock:
self._state.update(kwargs)
self._save_state()
def _clear_login_state(self) -> None:
"""清理当前的登录状态。"""
with self._lock:
self._state["bot_token"] = None
self._state["account_id"] = None
self._state["sync_buf"] = None
self._state["user_context_tokens"] = {}
self._save_state()
def _qrcode_expired(self, updated_at: Optional[int]) -> bool:
"""检查二维码是否过期。"""
if not updated_at:
return True
return int(time.time()) - int(updated_at) > self._qrcode_ttl_seconds
def _remember_target(
self, user_id: str, username: Optional[str], context_token: Optional[str]
) -> None:
"""记录已知消息目标,用于消息发送时定位。"""
if not user_id:
return
now_ts = int(time.time())
with self._lock:
known_targets = self._state.setdefault("known_targets", {})
known_targets[str(user_id)] = {
"username": username or str(user_id),
"last_active": now_ts,
}
if context_token:
tokens = self._state.setdefault("user_context_tokens", {})
tokens[str(user_id)] = str(context_token)
self._save_state()
def _get_context_token(self, user_id: str) -> Optional[str]:
"""获取特定用户的上下文 Token。"""
tokens = self._state.get("user_context_tokens") or {}
token = tokens.get(str(user_id))
return str(token) if token else None
def _get_targets(self, userid: Optional[str] = None) -> List[str]:
"""获取消息发送的目标列表。"""
if userid:
return [str(userid)]
if self._default_target:
return [self._default_target]
now_ts = int(time.time())
known_targets = self._state.get("known_targets") or {}
active_targets = [
target
for target, data in known_targets.items()
if isinstance(data, dict)
and now_ts - int(data.get("last_active") or 0) <= self._active_target_ttl_seconds
]
if active_targets:
return sorted(active_targets)
return sorted(known_targets.keys())
@staticmethod
def _short_text(value: Any, max_len: int = 240) -> str:
"""将内容缩短用于日志记录。"""
if value is None:
return ""
if isinstance(value, (dict, list)):
try:
text = json.dumps(value, ensure_ascii=False)
except Exception:
text = str(value)
else:
text = str(value)
text = text.replace("\n", " ").replace("\r", " ").strip()
if len(text) > max_len:
return f"{text[:max_len]}..."
return text
@staticmethod
def _split_content(content: str, max_bytes: int = 3000) -> List[str]:
"""将长消息内容拆分为符合大小限制的块。"""
if not content:
return []
chunks: List[str] = []
current = bytearray()
for line in content.splitlines():
encoded = (line + "\n").encode("utf-8")
if len(encoded) > max_bytes:
if current:
chunks.append(current.decode("utf-8", errors="replace").strip())
current = bytearray()
start = 0
while start < len(encoded):
end = min(start + max_bytes, len(encoded))
while start < end < len(encoded) and (encoded[end] & 0xC0) == 0x80:
end -= 1
chunks.append(encoded[start:end].decode("utf-8", errors="replace").strip())
start = end
continue
if len(current) + len(encoded) > max_bytes:
chunks.append(current.decode("utf-8", errors="replace").strip())
current = bytearray()
current += encoded
if current:
chunks.append(current.decode("utf-8", errors="replace").strip())
return [chunk for chunk in chunks if chunk]
@staticmethod
def _compose_text(
title: Optional[str] = None,
text: Optional[str] = None,
link: Optional[str] = None,
) -> str:
"""组合标题、文本和链接,生成消息内容。"""
parts = []
if title:
parts.append(str(title).strip())
if text:
parts.append(str(text).replace("\n\n", "\n"))
if link:
parts.append(f"查看详情:{link}")
return "\n\n".join(part for part in parts if part).strip()
@staticmethod
def _guess_mime_type(file_path: Path, file_bytes: bytes) -> str:
"""猜测文件 MIME 类型。"""
guessed = mimetypes.guess_type(file_path.name)[0]
if guessed:
return guessed
if file_bytes.startswith(b"\x89PNG\r\n\x1a\n"):
return "image/png"
if file_bytes.startswith(b"\xff\xd8\xff"):
return "image/jpeg"
if file_bytes.startswith((b"GIF87a", b"GIF89a")):
return "image/gif"
return "application/octet-stream"
@staticmethod
def _load_remote_image(image: str) -> Optional[bytes]:
"""加载远程图片并返回二进制数据。"""
image_url = str(image or "").strip()
if not image_url:
return None
if image_url.startswith("data:image/"):
try:
_, raw = image_url.split(",", 1)
return base64.b64decode(raw)
except Exception:
return None
if image_url.startswith("/"):
image_url = settings.MP_DOMAIN(image_url)
if not image_url.lower().startswith("http"):
return None
try:
resp = RequestUtils(
timeout=20,
proxies=settings.PROXY,
ua=settings.USER_AGENT,
).get_res(image_url)
if resp and resp.status_code == 200 and resp.content:
content_type = (resp.headers.get("Content-Type") or "").lower()
if not content_type or "image" in content_type:
return resp.content
except Exception as err:
logger.warning(f"加载图片失败:{err}")
return None
def get_state(self) -> bool:
"""获取当前登录状态。"""
return bool(self._state.get("bot_token"))
def stop(self) -> None:
"""停止消息轮询。"""
self._stop_event.set()
if self._poll_thread and self._poll_thread.is_alive():
self._poll_thread.join(timeout=5)
self._poll_thread = None
def _start_polling(self) -> None:
"""启动消息轮询线程。"""
if not self._state.get("bot_token"):
return
if self._poll_thread and self._poll_thread.is_alive():
return
self._stop_event.clear()
self._poll_thread = threading.Thread(target=self._poll_loop, daemon=True)
self._poll_thread.start()
logger.info("消息轮询线程已启动")
def _poll_loop(self) -> None:
"""
持续拉取微信消息并转发到本地消息入口。
轮询异常时按退避策略重试;本地消息入口失败只记日志,不回滚 sync_buf
由上层的消息ID去重兜底避免单次本地异常导致后续消息全部阻塞。
"""
consecutive_failures = 0
backoff = [1, 2, 5, 10, 30]
while not self._stop_event.is_set() and self._state.get("bot_token"):
try:
client = self._build_client()
messages, sync_buf, result = client.poll_updates(
timeout_seconds=self._poll_timeout
)
if sync_buf is not None and sync_buf != self._state.get("sync_buf"):
self._update_state(sync_buf=sync_buf)
if not result.get("success"):
raise RuntimeError(result.get("message") or "poll failed")
for message in messages:
self._remember_target(
user_id=message.user_id,
username=message.username,
context_token=message.context_token,
)
response = None
try:
response = RequestUtils(timeout=15).post_res(
self._message_endpoint,
json=message.to_message_payload(),
)
if response is None:
logger.error(
f"转发微信 ClawBot 消息失败message_id={message.message_id}, "
"本地消息入口无响应"
)
elif response.status_code != 200:
logger.error(
"转发微信 ClawBot 消息失败:"
f"message_id={message.message_id}, status={response.status_code}, "
f"body={self._short_text(response.text)}"
)
except Exception as err:
logger.error(
f"转发微信 ClawBot 消息失败message_id={message.message_id}, error={err}"
)
finally:
if response is not None:
try:
response.close()
except Exception:
pass
consecutive_failures = 0
except Exception as err:
consecutive_failures += 1
delay = backoff[min(consecutive_failures - 1, len(backoff) - 1)]
logger.warning(f"轮询异常,{delay}s 后重试:{err}")
if consecutive_failures >= 10:
logger.error("轮询连续失败,已清理登录状态")
self._clear_login_state()
break
self._stop_event.wait(delay)
def _build_known_targets(self) -> List[Dict[str, Any]]:
known_targets = self._state.get("known_targets") or {}
items = []
for userid, data in known_targets.items():
if not isinstance(data, dict):
continue
items.append(
{
"userid": userid,
"username": data.get("username") or userid,
"last_active": data.get("last_active"),
}
)
return sorted(items, key=lambda item: item.get("last_active") or 0, reverse=True)
def refresh_qrcode(self) -> Dict[str, Any]:
if self._state.get("bot_token"):
return self.get_status(refresh_remote=False)
client = ILinkClient(
base_url=self._base_url,
timeout=max(self._poll_timeout, 20),
)
result = client.get_qrcode()
if not result.get("success"):
return result
qrcode = {
"qrcode": result.get("qrcode"),
"qrcode_url": result.get("qrcode_url"),
"status": "waiting",
"updated_at": int(time.time()),
}
self._update_state(qrcode=qrcode)
return self.get_status(refresh_remote=False)
def get_status(
self,
refresh_remote: bool = True,
auto_generate_qrcode: bool = False,
) -> Dict[str, Any]:
qrcode = self._state.get("qrcode") or {}
if (
auto_generate_qrcode
and not self._state.get("bot_token")
and (not qrcode.get("qrcode") or self._qrcode_expired(qrcode.get("updated_at")))
):
self.refresh_qrcode()
qrcode = self._state.get("qrcode") or {}
if refresh_remote and not self._state.get("bot_token") and qrcode.get("qrcode"):
client = ILinkClient(
base_url=self._base_url,
timeout=max(self._poll_timeout, 20),
)
result = client.get_qrcode_status(str(qrcode.get("qrcode")))
updated_qrcode = dict(qrcode)
updated_qrcode["status"] = result.get("status") or updated_qrcode.get("status") or "waiting"
updated_qrcode["updated_at"] = int(time.time())
if result.get("qrcode_url"):
updated_qrcode["qrcode_url"] = result.get("qrcode_url")
update_payload: Dict[str, Any] = {"qrcode": updated_qrcode}
if result.get("token"):
update_payload.update(
{
"bot_token": result.get("token"),
"account_id": str(result.get("account_id") or "") or None,
"sync_buf": None,
"base_url": (result.get("base_url") or self._base_url).rstrip("/"),
}
)
self._base_url = (result.get("base_url") or self._base_url).rstrip("/")
self._update_state(**update_payload)
if self._auto_start_polling:
self._start_polling()
else:
self._update_state(**update_payload)
qrcode = self._state.get("qrcode") or {}
return {
"success": True,
"connected": bool(self._state.get("bot_token")),
"account_id": self._state.get("account_id"),
"qrcode": qrcode.get("qrcode"),
"qrcode_url": qrcode.get("qrcode_url"),
"qrcode_status": qrcode.get("status"),
"qrcode_updated_at": qrcode.get("updated_at"),
"known_targets": self._build_known_targets(),
"default_target": self._default_target,
"base_url": self._base_url,
}
def logout(self) -> Dict[str, Any]:
self.stop()
with self._lock:
self._state["bot_token"] = None
self._state["account_id"] = None
self._state["sync_buf"] = None
self._state["qrcode"] = {}
self._state["user_context_tokens"] = {}
self._save_state()
return {"success": True, "message": "已退出微信 ClawBot 登录"}
def test_connection(self) -> Tuple[bool, str]:
if not self._state.get("bot_token"):
return False, "未登录,请先扫码完成绑定"
return self._build_client().test_connection()
@classmethod
def migrate_cached_state(
cls,
old_name: Optional[str],
new_name: Optional[str],
cleanup_old: bool = False,
overwrite: bool = False,
) -> Tuple[bool, str]:
"""迁移配置重命名后的登录缓存,默认复制而不是删除旧缓存。"""
source_name = str(old_name or "").strip()
target_name = str(new_name or "").strip()
if not source_name or not target_name:
return False, "旧名称或新名称不能为空"
if source_name == target_name:
return True, "通知名称未变化,无需迁移登录缓存"
cache = FileCache()
source_key = cls._build_cache_key(source_name)
target_key = cls._build_cache_key(target_name)
source_state = cache.get(source_key)
if not source_state:
return True, "未找到可迁移的微信 ClawBot 登录缓存"
if cache.exists(target_key) and not overwrite:
return True, "新名称下已存在登录缓存,跳过迁移"
cache.set(target_key, source_state)
if cleanup_old:
cache.delete(source_key)
return True, f"已将微信 ClawBot 登录缓存从 {source_name} 迁移到 {target_name}"
@staticmethod
def _decode_ref_payload(ref: str, kind: str) -> Optional[Dict[str, Any]]:
prefix = f"wxclaw://{kind}/"
if not ref or not ref.startswith(prefix):
return None
encoded = ref.replace(prefix, "", 1)
padding = "=" * (-len(encoded) % 4)
try:
return json.loads(
base64.urlsafe_b64decode((encoded + padding).encode("ascii")).decode(
"utf-8"
)
)
except Exception:
return None
@staticmethod
def _decode_aes_key(aeskey: Optional[str]) -> Optional[bytes]:
if not aeskey:
return None
value = str(aeskey).strip()
if not value:
return None
if re.fullmatch(r"[0-9a-fA-F]{32}", value):
try:
return bytes.fromhex(value)
except Exception:
return None
for candidate in (value, value + "=" * (-len(value) % 4)):
try:
decoded = base64.b64decode(candidate)
if len(decoded) == 16:
return decoded
if len(decoded) == 32 and re.fullmatch(rb"[0-9a-fA-F]{32}", decoded):
return bytes.fromhex(decoded.decode("ascii"))
except Exception:
continue
return None
@staticmethod
def _unpad_bytes(content: bytes) -> Optional[bytes]:
if not content:
return None
padding_len = content[-1]
if padding_len <= 0 or padding_len > 16:
return None
if content[-padding_len:] != bytes([padding_len]) * padding_len:
return None
return content[:-padding_len]
@classmethod
def _decrypt_if_needed(
cls,
content: bytes,
aeskey: Optional[str],
mime_type: Optional[str] = None,
) -> bytes:
aes_bytes = cls._decode_aes_key(aeskey)
if not aes_bytes or not content:
return content
candidates = [content]
try:
candidates.append(AES.new(aes_bytes, AES.MODE_ECB).decrypt(content))
except Exception:
pass
try:
candidates.append(AES.new(aes_bytes, AES.MODE_CBC, aes_bytes[:16]).decrypt(content))
except Exception:
pass
normalized_candidates: List[bytes] = []
for candidate in candidates:
normalized_candidates.append(candidate)
unpadded = cls._unpad_bytes(candidate)
if unpadded:
normalized_candidates.append(unpadded)
preferred_prefixes = []
mime_value = (mime_type or "").lower()
if mime_value.startswith("image/png"):
preferred_prefixes.append(b"\x89PNG\r\n\x1a\n")
elif mime_value.startswith("image/jpeg"):
preferred_prefixes.append(b"\xff\xd8\xff")
elif mime_value.startswith("image/gif"):
preferred_prefixes.extend([b"GIF87a", b"GIF89a"])
elif mime_value.startswith("audio/"):
preferred_prefixes.extend([b"#!AMR", b"ID3", b"OggS", b"RIFF"])
for candidate in normalized_candidates:
if any(candidate.startswith(prefix) for prefix in preferred_prefixes):
return candidate
for candidate in normalized_candidates:
if candidate.startswith((b"\x89PNG\r\n\x1a\n", b"\xff\xd8\xff", b"GIF87a", b"GIF89a", b"PK\x03\x04", b"#!AMR", b"ID3", b"OggS", b"RIFF")):
return candidate
return content
@staticmethod
def _guess_binary_mime(content: bytes, default: str = "application/octet-stream") -> str:
if not content:
return default
if content.startswith(b"\x89PNG\r\n\x1a\n"):
return "image/png"
if content.startswith(b"\xff\xd8\xff"):
return "image/jpeg"
if content.startswith((b"GIF87a", b"GIF89a")):
return "image/gif"
if content.startswith(b"#!AMR"):
return "audio/amr"
if content.startswith(b"OggS"):
return "audio/ogg"
if content.startswith(b"ID3"):
return "audio/mpeg"
if content.startswith(b"RIFF") and b"WAVE" in content[:16]:
return "audio/wav"
if content.startswith(b"PK\x03\x04"):
return "application/zip"
return default
def _download_ref_bytes(self, ref: str, kind: str) -> Optional[Tuple[bytes, str, Optional[str]]]:
payload = self._decode_ref_payload(ref=ref, kind=kind)
if not payload:
return None
download_url = payload.get("url")
if not download_url:
return None
try:
resp = RequestUtils(timeout=30).get_res(download_url)
except Exception as err:
logger.error(f"下载 {kind} 失败:{err}")
return None
if not resp or not resp.content:
return None
content = self._decrypt_if_needed(
content=resp.content,
aeskey=payload.get("aeskey"),
mime_type=payload.get("mime_type"),
)
mime_type = payload.get("mime_type") or self._guess_binary_mime(
content,
(resp.headers.get("Content-Type") or "application/octet-stream").split(
";", 1
)[0],
)
return content, mime_type, payload.get("name")
def download_image_to_data_url(self, image_ref: str) -> Optional[str]:
result = self._download_ref_bytes(ref=image_ref, kind="image")
if not result:
return None
content, mime_type, _ = result
return f"data:{mime_type};base64,{base64.b64encode(content).decode()}"
def download_media_bytes(self, media_ref: str) -> Optional[bytes]:
kind = None
if media_ref.startswith("wxclaw://file/"):
kind = "file"
elif media_ref.startswith("wxclaw://voice/"):
kind = "voice"
if not kind:
return None
result = self._download_ref_bytes(ref=media_ref, kind=kind)
return result[0] if result else None
def send_msg(
self,
title: str,
text: Optional[str] = None,
image: Optional[str] = None,
userid: Optional[str] = None,
link: Optional[str] = None,
**kwargs,
) -> Optional[bool]:
targets = self._get_targets(userid=userid)
if not targets:
logger.warning("未找到可发送的微信 ClawBot 目标")
return False
content = self._compose_text(title=title, text=text, link=link)
# 当前 iLink 发送实现会把图文拆成两条消息发送。
# 为避免用户看到“同一条通知被拆成文本 + 图片”两次触达,这里约定:
# 只要通知里已经有文本内容,就优先只发文本;只有纯图片通知才发送图片。
image_bytes = self._load_remote_image(image) if image and not content else None
ok = False
for target in targets:
context_token = self._get_context_token(target)
if content:
client = self._build_client()
sent = True
for chunk in self._split_content(content):
if not client.send_text(
to_user=target,
text=chunk,
context_token=context_token,
):
sent = False
break
elif image_bytes:
sent = self._build_client().send_image_png(
to_user=target,
image_bytes=image_bytes,
context_token=context_token,
)
else:
client = self._build_client()
sent = True
for chunk in self._split_content(content):
if not client.send_text(
to_user=target,
text=chunk,
context_token=context_token,
):
sent = False
break
ok = ok or bool(sent)
return ok
def send_file(
self,
file_path: str,
file_name: Optional[str] = None,
title: Optional[str] = None,
text: Optional[str] = None,
userid: Optional[str] = None,
**kwargs,
) -> Optional[bool]:
path = Path(file_path)
if not path.exists() or not path.is_file():
logger.warning(f"待发送文件不存在:{file_path}")
return False
file_bytes = path.read_bytes()
effective_name = file_name or path.name
mime_type = self._guess_mime_type(path, file_bytes)
targets = self._get_targets(userid=userid)
if not targets:
return False
ok = False
for target in targets:
context_token = self._get_context_token(target)
# send_file 的主语义是发送附件本体,附加文案可以丢弃,避免出现
# “先发一段说明,再发一个文件”的重复触达体验。
if mime_type.startswith("image/"):
sent = self._build_client().send_image_png(
to_user=target,
image_bytes=file_bytes,
context_token=context_token,
)
else:
sent = self._build_client().send_file_bytes(
to_user=target,
file_bytes=file_bytes,
file_name=effective_name,
mime_type=mime_type,
context_token=context_token,
)
ok = ok or bool(sent)
return ok
def send_medias_msg(
self, medias: List[MediaInfo], userid: Optional[str] = None
) -> Optional[bool]:
if not medias:
return False
lines = []
for index, media in enumerate(medias, start=1):
line = f"{index}. {media.title_year}"
if media.vote_average:
line += f" 评分:{media.vote_average}"
if media.detail_link:
line += f"\n{media.detail_link}"
lines.append(line)
return self.send_msg(title="媒体列表", text="\n\n".join(lines), userid=userid)
def send_torrents_msg(
self,
torrents: List[Context],
userid: Optional[str] = None,
title: Optional[str] = None,
link: Optional[str] = None,
) -> Optional[bool]:
if not torrents:
return False
lines = []
for index, context in enumerate(torrents, start=1):
torrent = context.torrent_info
meta = MetaInfo(title=torrent.title, subtitle=torrent.description)
text = (
f"{index}.【{torrent.site_name}{meta.season_episode} {meta.resource_term} "
f"{meta.video_term} {meta.release_group} {StringUtils.str_filesize(torrent.size)} "
f"{torrent.volume_factor} {torrent.seeders}"
)
text = re.sub(r"\s+", " ", text).strip()
if torrent.page_url:
text += f"\n{torrent.page_url}"
lines.append(text)
return self.send_msg(
title=title or "种子列表",
text="\n\n".join(lines),
userid=userid,
link=link,
)