fix: 模拟登录时页面跳转导致 page.content() 竞态失败（未知错误） (#6091)

app/helper/cookie.py

@@ -1,4 +1,5 @@
 import base64
+import time
 from typing import Tuple, Optional
 
 from lxml import etree
@@ -57,6 +58,36 @@ class CookieHelper:
         ]
     }
 
+    @staticmethod
+    def get_page_content(page: BrowserPage, retries: int = 3, interval: float = 1.0) -> Optional[str]:
+        """
+        获取页面源码，页面跳转中（如登录前后的重定向）会导致 page.content() 抛出
+        "Unable to retrieve content because the page is navigating" 异常，等待加载完成后重试
+        :param page: 浏览器页面
+        :param retries: 最大重试次数
+        :param interval: 重试间隔（秒）
+        :return: 页面源码
+        """
+        for i in range(retries):
+            # 等待加载失败不代表源码不可读取，最后一次等待失败时仍尝试直接获取源码
+            try:
+                page.wait_for_load_state("domcontentloaded", timeout=10 * 1000)
+            except Exception as e:
+                if i < retries - 1:
+                    logger.warning(f"等待页面加载完成失败：{str(e)}，{interval}秒后重试 ({i + 1}/{retries - 1})")
+                    time.sleep(interval)
+                    continue
+                logger.warning(f"等待页面加载完成失败：{str(e)}，尝试直接获取源码")
+            try:
+                return page.content()
+            except Exception as e:
+                if i >= retries - 1:
+                    logger.error(f"获取页面源码失败：{str(e)}")
+                    return None
+                logger.warning(f"获取页面源码失败：{str(e)}，{interval}秒后重试 ({i + 1}/{retries - 1})")
+                time.sleep(interval)
+        return None
+
     @staticmethod
     def parse_cookies(cookies: list) -> str:
         """
@@ -93,11 +124,13 @@ class CookieHelper:
             :return: Cookie和UA
             """
             # 登录页面代码
-            html_text = page.content()
+            html_text = self.get_page_content(page)
             if not html_text:
                 return None, None, "获取源码失败"
             # 查找用户名输入框
             html = etree.HTML(html_text)
+            if html is None:
+                return None, None, "解析网页源码失败"
             try:
                 username_xpath = None
                 for xpath in self._SITE_LOGIN_XPATH.get("username"):
@@ -189,7 +222,12 @@ class CookieHelper:
                 if "verify" in page.url:
                     if not otp_code:
                         return None, None, "需要二次验证码"
-                    html = etree.HTML(page.content())
+                    html_text = self.get_page_content(page)
+                    if not html_text:
+                        return None, None, "获取网页源码失败"
+                    html = etree.HTML(html_text)
+                    if html is None:
+                        return None, None, "解析网页源码失败"
                     for xpath in self._SITE_LOGIN_XPATH.get("twostep"):
                         if html.xpath(xpath):
                             try:
@@ -205,14 +243,17 @@ class CookieHelper:
                             break
 
                 # 登录后的源码
-                html_text = page.content()
+                html_text = self.get_page_content(page)
                 if not html_text:
                     return None, None, "获取网页源码失败"
                 if SiteUtils.is_logged_in(html_text):
                     return self.parse_cookies(page.context.cookies()), \
                         page.evaluate("() => window.navigator.userAgent"), ""
                 else:
-                    # 读取错误信息
+                    # 从登录后的页面读取错误信息
+                    html = etree.HTML(html_text)
+                    if html is None:
+                        return None, None, "登录失败"
                     error_xpath = None
                     for xpath in self._SITE_LOGIN_XPATH.get("error"):
                         if html.xpath(xpath):