chatgpt-mirai-qq-bot-web-search 0.2.6__py3-none-any.whl → 0.2.8__py3-none-any.whl

{chatgpt_mirai_qq_bot_web_search-0.2.6.dist-info → chatgpt_mirai_qq_bot_web_search-0.2.8.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: chatgpt-mirai-qq-bot-web-search
-Version: 0.2.6
+Version: 0.2.8
 Summary: WebSearch adapter for lss233/chatgpt-mirai-qq-bot
 Home-page: https://github.com/chuanSir123/web_search
 Author: chuanSir

{chatgpt_mirai_qq_bot_web_search-0.2.6.dist-info → chatgpt_mirai_qq_bot_web_search-0.2.8.dist-info}/RECORD

@@ -1,11 +1,11 @@
 web_search/__init__.py,sha256=bpu1gpf9tq3mOAoaGMM-8S6GBw2GzRAG6ClHGlp-CVw,4607
 web_search/blocks.py,sha256=QuXt3KMuY4hUW-ucleNYXFvW6YbUAB4Xu4m_SxdCd-U,9152
 web_search/config.py,sha256=DhLiERBJR2V5Boglf7Aq9Rbc4vsvLIh67CrLDIPeqA0,398
-web_search/web_searcher.py,sha256=760WTBGusxYntigGi5Wsbm2rRHEWq3sTwfNlf3ew3b0,20919
+web_search/web_searcher.py,sha256=-wNuPzG0oZAr9-ihI5EdAexB9QXLUvc6bGyoVkWTlvw,21814
 web_search/example/roleplayWithWebSearch.yaml,sha256=C-dGy3z8gcRcmxzurssP-kPRLqMf1TYR-nnNUaJjISE,7468
-chatgpt_mirai_qq_bot_web_search-0.2.6.dist-info/LICENSE,sha256=ILBn-G3jdarm2w8oOrLmXeJNU3czuJvVhDLBASWdhM8,34522
-chatgpt_mirai_qq_bot_web_search-0.2.6.dist-info/METADATA,sha256=88y-xtYK-se7dwVVU2_fcXx_rGCUEgIxWMJvjI4Dfwg,1734
-chatgpt_mirai_qq_bot_web_search-0.2.6.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
-chatgpt_mirai_qq_bot_web_search-0.2.6.dist-info/entry_points.txt,sha256=o3kRDSdSmSdnCKlK6qS57aN0WpI4ab-Nxub2NwUrjf0,64
-chatgpt_mirai_qq_bot_web_search-0.2.6.dist-info/top_level.txt,sha256=PoNm8MJYw_y8RTMaNlY0ePLoNHxVUAE2IHDuL5fFubI,11
-chatgpt_mirai_qq_bot_web_search-0.2.6.dist-info/RECORD,,
+chatgpt_mirai_qq_bot_web_search-0.2.8.dist-info/LICENSE,sha256=ILBn-G3jdarm2w8oOrLmXeJNU3czuJvVhDLBASWdhM8,34522
+chatgpt_mirai_qq_bot_web_search-0.2.8.dist-info/METADATA,sha256=8L4yju3nxv1fAs4phYBFXJjNpsC1jIh1rhB5P9aiovI,1734
+chatgpt_mirai_qq_bot_web_search-0.2.8.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+chatgpt_mirai_qq_bot_web_search-0.2.8.dist-info/entry_points.txt,sha256=o3kRDSdSmSdnCKlK6qS57aN0WpI4ab-Nxub2NwUrjf0,64
+chatgpt_mirai_qq_bot_web_search-0.2.8.dist-info/top_level.txt,sha256=PoNm8MJYw_y8RTMaNlY0ePLoNHxVUAE2IHDuL5fFubI,11
+chatgpt_mirai_qq_bot_web_search-0.2.8.dist-info/RECORD,,

web_search/web_searcher.py

@@ -31,7 +31,7 @@ class WebSearcher:
         self.video_ids = self._load_video_ids()
         self.search_engines = {
             'bing': {
-                'url': 'https://www.bing.com/search?q={}',
+                'url': 'https://cn.bing.com/search?q={}',
                 'selectors': ['.b_algo', '#b_results .b_algo', 'main .b_algo'],
                 'title_selector': 'h2',
                 'link_selector': 'h2 a',
@@ -72,7 +72,7 @@ class WebSearcher:
 
             # 合并所有选项到一个字典
             context_options = {
-                'headless': True,
+                'headless': False,
                 'chromium_sandbox': False,
                 'slow_mo': 50,  # 减慢操作速度，更像人类
                 'args': [
@@ -163,9 +163,9 @@ class WebSearcher:
             # 创建新标签页获取内容
             page = await context.new_page()
             try:
-                # 设置更严格的资源加载策略
+                # 修改资源加载策略，允许加载必要的CSS
                 await page.route("**/*", lambda route: route.abort()
-                    if route.request.resource_type in ['image', 'stylesheet', 'font', 'media']
+                    if route.request.resource_type in ['image', 'font', 'media']  # 移除 'stylesheet'
                     else route.continue_())
 
                 # 使用 domcontentloaded 而不是 networkidle
@@ -263,46 +263,64 @@ class WebSearcher:
 
             # 使用搜索引擎特定的选择器
             results = None
+            # 等待页面稳定
+            await asyncio.sleep(2)  # 添加短暂延迟
 
             # 对于Google，让页面有更多时间加载
             if engine == 'google':
                 await self.simulate_human_scroll(page)
 
+            selector_timeout = 5000
             for selector in engine_config['selectors']:
                 try:
                     logger.info(f"Trying selector: {selector}")
-                    await page.wait_for_selector(selector, timeout=8000)  # 增加等待时间
+                    await page.wait_for_selector(selector, timeout=selector_timeout)  # 增加等待时间
+                    selector_timeout = 500
                     results = await page.query_selector_all(selector)
                     if results and len(results) > 0:
                         logger.info(f"Found {len(results)} results with selector {selector}")
                         break
                 except Exception as e:
+                    selector_timeout = 500
                     logger.warning(f"Selector {selector} failed: {e}")
                     continue
 
             if not results:
-                # 尝试直接使用 JavaScript 获取元素
-                if engine == 'google':
-                    try:
-                        # 使用更通用的JavaScript选择器尝试获取结果
-                        results = await page.evaluate("""
-                            () => {
-                                const elements = document.querySelectorAll('div[data-sokoban-container], div.g, .MjjYud');
-                                return Array.from(elements).length;
-                            }
-                        """)
-                        logger.info(f"JavaScript found {results} elements")
-
-                        # 如果找到了元素，使用evaluate来处理它们
-                        if results > 0:
-                            # 自定义处理逻辑...
-                            pass
-                    except Exception as e:
-                        logger.error(f"JavaScript evaluation failed: {e}")
-
-                logger.error("No search results found with any selector")
-                await page.screenshot(path=f'search_failed_{engine}.png')
-                return "搜索结果加载失败"
+                # 添加重试机制
+                retry_count = 0
+                while not results and retry_count < max_results:
+                    logger.info(f"Retrying search, attempt {retry_count + 1}/{max_results}")
+                    # 刷新页面重试
+                    await page.goto(
+                                engine_config['url'].format(encoded_query),
+                                wait_until='load',
+                                timeout=timeout * 1000
+                            )
+                    await self.simulate_human_scroll(page)
+
+                    # 重新尝试所有选择器
+                    selector_timeout = 5000
+                    for selector in engine_config['selectors']:
+                        try:
+                            logger.info(f"Retrying selector: {selector}")
+                            await page.wait_for_selector(selector, timeout=selector_timeout)
+                            selector_timeout = 500
+                            results = await page.query_selector_all(selector)
+                            if results and len(results) > 0:
+                                logger.info(f"Found {len(results)} results with selector {selector} on retry {retry_count + 1}")
+                                break
+                        except Exception as e:
+                            selector_timeout = 500
+                            logger.warning(f"Selector {selector} failed on retry {retry_count + 1}: {e}")
+                            continue
+
+                    retry_count += 1
+
+
+                # 如果所有重试都失败了，才返回错误
+                if not results:
+                    logger.error("No search results found after all retries")
+                    return "搜索结果加载失败"
 
             logger.info(f"Found {len(results)} search results")