fix(browser): make URL validation robust to redirects

browser/instance.py

@@ -6,6 +6,7 @@ from browser.navigation import handle_successful_navigation
 from camoufox.sync_api import Camoufox
 from utils.paths import logs_dir
 from utils.common import parse_headless_mode, ensure_dir
+from utils.url_helper import extract_url_path
 
 
 def run_browser_instance(config):
@@ -144,9 +145,13 @@ def run_browser_instance(config):
                 logger.error("检测到Google登录页面（需要输入邮箱）。Cookie已完全失效。")
                 page.screenshot(path=os.path.join(screenshot_dir, f"FAIL_identifier_page_{diagnostic_tag}.png"))
                 return
-            elif expected_url.split('?')[0] in final_url:
-                
-                logger.info("URL正确。现在等待页面完成初始加载...")
+
+            # 提取路径部分进行匹配（允许域名重定向）
+            expected_path = extract_url_path(expected_url).split('?')[0]
+            final_path = extract_url_path(final_url)
+
+            if expected_path and expected_path in final_path:
+                logger.info(f"URL验证通过。预期路径: {expected_path}, 最终URL: {final_url}")
 
                 # --- NEW ROBUST STRATEGY: Wait for the loading spinner to disappear ---
                 # This is the key to solving the race condition. The error message or
@@ -197,7 +202,10 @@ def run_browser_instance(config):
                 page.screenshot(path=os.path.join(screenshot_dir, f"FAIL_chooser_click_failed_{diagnostic_tag}.png"))
                 return
             else:
-                logger.error(f"导航到了一个意外的URL: {final_url}")
+                logger.error(f"导航到了意外的URL。")
+                logger.error(f"  预期路径: {expected_path}")
+                logger.error(f"  最终URL: {final_url}")
+                logger.error(f"  最终路径: {final_path}")
                 page.screenshot(path=os.path.join(screenshot_dir, f"FAIL_unexpected_url_{diagnostic_tag}.png"))
                 return
 

utils/url_helper.py

@@ -0,0 +1,46 @@
+"""
+URL处理辅助函数
+
+提供URL解析和路径提取功能，用于导航验证中的域名无关匹配。
+"""
+
+from urllib.parse import urlparse
+
+
+def extract_url_path(url: str) -> str:
+    """
+    提取URL的路径和查询参数部分，忽略协议和域名差异
+
+    用于验证导航是否到达正确页面，允许域名重定向。
+
+    Args:
+        url: 完整URL字符串
+
+    Returns:
+        路径+查询参数+片段（例如："/apps/drive/123?param=value#section"）
+        如果URL为空或无效，返回空字符串
+
+    Examples:
+        >>> extract_url_path("https://ai.studio/apps/drive/123?param=value")
+        '/apps/drive/123?param=value'
+
+        >>> extract_url_path("https://aistudio.google.com/apps/drive/123")
+        '/apps/drive/123'
+
+        >>> extract_url_path("https://example.com/path")
+        '/path'
+    """
+    if not url:
+        return ""
+
+    try:
+        parsed = urlparse(url)
+        result = parsed.path
+        if parsed.query:
+            result += '?' + parsed.query
+        if parsed.fragment:
+            result += '#' + parsed.fragment
+        return result
+    except Exception:
+        # 如果URL格式无效，返回空字符串
+        return ""