Fix temp mail registration flow

src/core/register.py

@@ -214,22 +214,42 @@ class RegistrationEngine:
 
     def _get_device_id(self) -> Optional[str]:
         """获取 Device ID"""
-        try:
-            if not self.oauth_start:
-                return None
-
-            response = self.session.get(
-                self.oauth_start.auth_url,
-                timeout=15
-            )
-            did = self.session.cookies.get("oai-did")
-            self._log(f"Device ID: {did}")
-            return did
-
-        except Exception as e:
-            self._log(f"获取 Device ID 失败: {e}", "error")
+        if not self.oauth_start:
             return None
 
+        max_attempts = 3
+        for attempt in range(1, max_attempts + 1):
+            try:
+                if not self.session:
+                    self.session = self.http_client.session
+
+                response = self.session.get(
+                    self.oauth_start.auth_url,
+                    timeout=20
+                )
+                did = self.session.cookies.get("oai-did")
+
+                if did:
+                    self._log(f"Device ID: {did}")
+                    return did
+
+                self._log(
+                    f"获取 Device ID 失败: 未返回 oai-did Cookie (HTTP {response.status_code}, 第 {attempt}/{max_attempts} 次)",
+                    "warning" if attempt < max_attempts else "error"
+                )
+            except Exception as e:
+                self._log(
+                    f"获取 Device ID 失败: {e} (第 {attempt}/{max_attempts} 次)",
+                    "warning" if attempt < max_attempts else "error"
+                )
+
+            if attempt < max_attempts:
+                time.sleep(attempt)
+                self.http_client.close()
+                self.session = self.http_client.session
+
+        return None
+
     def _check_sentinel(self, did: str) -> Optional[str]:
         """检查 Sentinel 拦截"""
         try:
@@ -860,4 +880,4 @@ class RegistrationEngine:
 
         except Exception as e:
             self._log(f"保存到数据库失败: {e}", "error")
-            return False
+            return False

src/services/temp_mail.py

@@ -8,7 +8,12 @@ import re
 import time
 import json
 import logging
-from typing import Optional, Dict, Any
+from email import message_from_string
+from email.header import decode_header, make_header
+from email.message import Message
+from email.policy import default as email_policy
+from html import unescape
+from typing import Optional, Dict, Any, List
 
 from .base import BaseEmailService, EmailServiceError, EmailServiceType
 from ..core.http_client import HTTPClient, RequestConfig
@@ -63,6 +68,97 @@ class TempMailService(BaseEmailService):
         # 邮箱缓存：email -> {jwt, address}
         self._email_cache: Dict[str, Dict[str, Any]] = {}
 
+    def _decode_mime_header(self, value: str) -> str:
+        """解码 MIME 头，兼容 RFC 2047 编码主题。"""
+        if not value:
+            return ""
+        try:
+            return str(make_header(decode_header(value)))
+        except Exception:
+            return value
+
+    def _extract_body_from_message(self, message: Message) -> str:
+        """从 MIME 邮件对象中提取可读正文。"""
+        parts: List[str] = []
+
+        if message.is_multipart():
+            for part in message.walk():
+                if part.get_content_maintype() == "multipart":
+                    continue
+
+                content_type = (part.get_content_type() or "").lower()
+                if content_type not in ("text/plain", "text/html"):
+                    continue
+
+                try:
+                    payload = part.get_payload(decode=True)
+                    charset = part.get_content_charset() or "utf-8"
+                    text = payload.decode(charset, errors="replace") if payload else ""
+                except Exception:
+                    try:
+                        text = part.get_content()
+                    except Exception:
+                        text = ""
+
+                if content_type == "text/html":
+                    text = re.sub(r"<[^>]+>", " ", text)
+                parts.append(text)
+        else:
+            try:
+                payload = message.get_payload(decode=True)
+                charset = message.get_content_charset() or "utf-8"
+                body = payload.decode(charset, errors="replace") if payload else ""
+            except Exception:
+                try:
+                    body = message.get_content()
+                except Exception:
+                    body = str(message.get_payload() or "")
+
+            if "html" in (message.get_content_type() or "").lower():
+                body = re.sub(r"<[^>]+>", " ", body)
+            parts.append(body)
+
+        return unescape("\n".join(part for part in parts if part).strip())
+
+    def _extract_mail_fields(self, mail: Dict[str, Any]) -> Dict[str, str]:
+        """统一提取邮件字段，兼容 raw MIME 和不同 Worker 返回格式。"""
+        sender = str(
+            mail.get("source")
+            or mail.get("from")
+            or mail.get("from_address")
+            or mail.get("fromAddress")
+            or ""
+        ).strip()
+        subject = str(mail.get("subject") or mail.get("title") or "").strip()
+        body_text = str(
+            mail.get("text")
+            or mail.get("body")
+            or mail.get("content")
+            or mail.get("html")
+            or ""
+        ).strip()
+        raw = str(mail.get("raw") or "").strip()
+
+        if raw:
+            try:
+                message = message_from_string(raw, policy=email_policy)
+                sender = sender or self._decode_mime_header(message.get("From", ""))
+                subject = subject or self._decode_mime_header(message.get("Subject", ""))
+                parsed_body = self._extract_body_from_message(message)
+                if parsed_body:
+                    body_text = f"{body_text}\n{parsed_body}".strip() if body_text else parsed_body
+            except Exception as e:
+                logger.debug(f"解析 TempMail raw 邮件失败: {e}")
+                body_text = f"{body_text}\n{raw}".strip() if body_text else raw
+
+        body_text = unescape(re.sub(r"<[^>]+>", " ", body_text))
+        return {
+            "sender": sender,
+            "subject": subject,
+            "body": body_text,
+            "raw": raw,
+        }
+
     def _admin_headers(self) -> Dict[str, str]:
         """构造 admin 请求头"""
         return {
@@ -224,14 +320,12 @@ class TempMailService(BaseEmailService):
 
                     seen_mail_ids.add(mail_id)
 
-                    sender = str(mail.get("source", "")).lower()
-                    subject = str(mail.get("subject", ""))
-                    body_text = str(mail.get("text", "") or mail.get("html", "") or "")
-
-                    # 去除简单 HTML 标签
-                    body_clean = re.sub(r"<[^>]+>", " ", body_text)
-
-                    content = f"{sender} {subject} {body_clean}"
+                    parsed = self._extract_mail_fields(mail)
+                    sender = parsed["sender"].lower()
+                    subject = parsed["subject"]
+                    body_text = parsed["body"]
+                    raw_text = parsed["raw"]
+                    content = f"{sender}\n{subject}\n{body_text}\n{raw_text}".strip()
 
                     # 只处理 OpenAI 邮件
                     if "openai" not in sender and "openai" not in content.lower():
@@ -252,6 +346,88 @@ class TempMailService(BaseEmailService):
         logger.warning(f"等待 TempMail 验证码超时: {email}")
         return None
 
+    def list_emails(self, limit: int = 100, offset: int = 0, **kwargs) -> List[Dict[str, Any]]:
+        """
+        列出邮箱
+
+        Args:
+            limit: 返回数量上限
+            offset: 分页偏移
+            **kwargs: 额外查询参数，透传给 admin API
+
+        Returns:
+            邮箱列表
+        """
+        params = {
+            "limit": limit,
+            "offset": offset,
+        }
+        params.update({k: v for k, v in kwargs.items() if v is not None})
+
+        try:
+            response = self._make_request("GET", "/admin/mails", params=params)
+            mails = response.get("results", [])
+            if not isinstance(mails, list):
+                raise EmailServiceError(f"API 返回数据格式错误: {response}")
+
+            emails: List[Dict[str, Any]] = []
+            for mail in mails:
+                address = (mail.get("address") or "").strip()
+                mail_id = mail.get("id") or address
+                email_info = {
+                    "id": mail_id,
+                    "service_id": mail_id,
+                    "email": address,
+                    "subject": mail.get("subject"),
+                    "from": mail.get("source"),
+                    "created_at": mail.get("createdAt") or mail.get("created_at"),
+                    "raw_data": mail,
+                }
+                emails.append(email_info)
+
+                if address:
+                    cached = self._email_cache.get(address, {})
+                    self._email_cache[address] = {**cached, **email_info}
+
+            self.update_status(True)
+            return emails
+        except Exception as e:
+            logger.warning(f"列出 TempMail 邮箱失败: {e}")
+            self.update_status(False, e)
+            return list(self._email_cache.values())
+
+    def delete_email(self, email_id: str) -> bool:
+        """
+        删除邮箱
+
+        Note:
+            当前 TempMail admin API 文档未见删除地址接口，这里先从本地缓存移除，
+            以满足统一接口并避免服务实例化失败。
+        """
+        removed = False
+        emails_to_delete = []
+
+        for address, info in self._email_cache.items():
+            candidate_ids = {
+                address,
+                info.get("id"),
+                info.get("service_id"),
+            }
+            if email_id in candidate_ids:
+                emails_to_delete.append(address)
+
+        for address in emails_to_delete:
+            self._email_cache.pop(address, None)
+            removed = True
+
+        if removed:
+            logger.info(f"已从 TempMail 缓存移除邮箱: {email_id}")
+            self.update_status(True)
+        else:
+            logger.info(f"TempMail 缓存中未找到邮箱: {email_id}")
+
+        return removed
+
     def check_health(self) -> bool:
         """检查服务健康状态"""
         try:

src/web/routes/registration.py

@@ -182,6 +182,30 @@ def task_to_response(task: RegistrationTask) -> RegistrationTaskResponse:
     )
 
 
+def _normalize_email_service_config(
+    service_type: EmailServiceType,
+    config: Optional[dict],
+    proxy_url: Optional[str] = None
+) -> dict:
+    """按服务类型兼容旧字段名，避免不同服务的配置键互相污染。"""
+    normalized = config.copy() if config else {}
+
+    if 'api_url' in normalized and 'base_url' not in normalized:
+        normalized['base_url'] = normalized.pop('api_url')
+
+    if service_type == EmailServiceType.CUSTOM_DOMAIN:
+        if 'domain' in normalized and 'default_domain' not in normalized:
+            normalized['default_domain'] = normalized.pop('domain')
+    elif service_type == EmailServiceType.TEMP_MAIL:
+        if 'default_domain' in normalized and 'domain' not in normalized:
+            normalized['domain'] = normalized.pop('default_domain')
+
+    if proxy_url and 'proxy_url' not in normalized:
+        normalized['proxy_url'] = proxy_url
+
+    return normalized
+
+
 def _run_sync_registration_task(task_uuid: str, email_service_type: str, proxy: Optional[str], email_service_config: Optional[dict], email_service_id: Optional[int] = None, log_prefix: str = "", batch_id: str = "", auto_upload_cpa: bool = False, cpa_service_id: Optional[int] = None):
     """
     在线程池中执行的同步注册任务
@@ -236,15 +260,11 @@ def _run_sync_registration_task(task_uuid: str, email_service_type: str, proxy:
                 ).first()
 
                 if db_service:
-                    config = db_service.config.copy() if db_service.config else {}
-                    # 兼容旧版字段名 api_url -> base_url
-                    if 'api_url' in config and 'base_url' not in config:
-                        config['base_url'] = config.pop('api_url')
-                    if 'domain' in config and 'default_domain' not in config:
-                        config['default_domain'] = config.pop('domain')
+                    service_type = EmailServiceType(db_service.service_type)
+                    config = _normalize_email_service_config(service_type, db_service.config, actual_proxy_url)
                     # 更新任务关联的邮箱服务
                     crud.update_registration_task(db, task_uuid, email_service_id=db_service.id)
-                    logger.info(f"使用数据库邮箱服务: {db_service.name} (ID: {db_service.id})")
+                    logger.info(f"使用数据库邮箱服务: {db_service.name} (ID: {db_service.id}, 类型: {service_type.value})")
                 else:
                     raise ValueError(f"邮箱服务不存在或已禁用: {email_service_id}")
             else:
@@ -265,12 +285,7 @@ def _run_sync_registration_task(task_uuid: str, email_service_type: str, proxy:
                     ).order_by(EmailServiceModel.priority.asc()).first()
 
                     if db_service and db_service.config:
-                        config = db_service.config.copy()
-                        # 兼容旧版字段名 api_url -> base_url
-                        if 'api_url' in config and 'base_url' not in config:
-                            config['base_url'] = config.pop('api_url')
-                        if 'domain' in config and 'default_domain' not in config:
-                            config['default_domain'] = config.pop('domain')
+                        config = _normalize_email_service_config(service_type, db_service.config, actual_proxy_url)
                         crud.update_registration_task(db, task_uuid, email_service_id=db_service.id)
                         logger.info(f"使用数据库自定义域名服务: {db_service.name}")
                     elif settings.custom_domain_base_url and settings.custom_domain_api_key: