更新今天的数据

product/playwright-get-data.py

@@ -22,6 +22,8 @@ class ProductHuntScraper:
         self.debug_port = debug_port
         self.browser = None
         self.page = None
+        self.click_records = []  # 记录点击行为
+        self.dom_selection_records = []  # 记录DOM选取行为
         
     async def connect_to_existing_chrome(self):
         """连接到已运行的Chrome实例"""
@@ -58,7 +60,47 @@ class ProductHuntScraper:
         except Exception as e:
             logger.error(f"连接Chrome失败: {e}")
             return False
-    
+
+    async def record_click(self, x, y, selector="", description=""):
+        """记录点击行为"""
+        click_record = {
+            "timestamp": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
+            "type": "click",
+            "x": x,
+            "y": y,
+            "selector": selector,
+            "description": description
+        }
+        self.click_records.append(click_record)
+        logger.info(f"记录点击: {description} - 坐标({x}, {y}) - 选择器: {selector}")
+
+    async def record_dom_selection(self, selector, description=""):
+        """记录DOM选取行为"""
+        dom_record = {
+            "timestamp": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
+            "type": "dom_selection",
+            "selector": selector,
+            "description": description
+        }
+        self.dom_selection_records.append(dom_record)
+        logger.info(f"记录DOM选取: {description} - 选择器: {selector}")
+
+    async def save_behavior_records(self):
+        """保存行为记录到文件"""
+        import json
+        
+        records = {
+            "click_records": self.click_records,
+            "dom_selection_records": self.dom_selection_records
+        }
+        
+        filename = f"playwright_behavior_records_{datetime.now().strftime('%Y%m%d_%H%M%S')}.json"
+        
+        with open(filename, "w", encoding="utf-8") as f:
+            json.dump(records, f, ensure_ascii=False, indent=2)
+        
+        logger.success(f"行为记录已保存到: {filename}")
+
     async def navigate_to_producthunt(self, url):
         """导航到ProductHunt页面"""
         if not self.page:
@@ -115,6 +157,7 @@ class ProductHuntScraper:
             logger.info("模拟点击制作人链接...")
             
             # 查找包含制作人信息的div容器（class="flex flex-col gap-1"）
+            await self.record_dom_selection('div.flex.flex-col.gap-1', "制作人信息容器")
             div_container = await self.page.query_selector('div.flex.flex-col.gap-1')
             if not div_container:
                 logger.warning("未找到class='flex flex-col gap-1'的div容器，使用备用方法")
@@ -156,9 +199,17 @@ class ProductHuntScraper:
             
             logger.info(f"点击位置: ({center_x:.1f}, {center_y:.1f})")
             
+            # 记录点击行为
+            await self.record_click(center_x, center_y, 'div.flex.flex-col.gap-1', "制作人链接点击")
+            
+            # 先模拟点击，然后监听新窗口打开事件
+            # 添加动态点击效果：先移动到位置，短暂停留，然后点击
+            await self.page.mouse.move(center_x, center_y)
+            await self.page.wait_for_timeout(2000)  # 短暂停留2000毫秒，模拟用户移动鼠标
+            
             # 监听新窗口打开事件
             async with self.page.context.expect_page() as new_page_info:
-                # 模拟点击计算出的位置
+                # 执行点击操作
                 await self.page.mouse.click(center_x, center_y)
             
             # 获取新页面
@@ -171,11 +222,13 @@ class ProductHuntScraper:
             logger.success("新窗口已加载完成")
             
             # 抓取第一个section的tag
+            await self.record_dom_selection('section', "新窗口第一个section标签")
             first_section = await new_page.query_selector('section')
             if first_section:
                 logger.success("找到第一个section标签")
                 
                 # 在section下面找一个没有任何class的div标签
+                await self.record_dom_selection('div:not([class])', "section下无class的div标签")
                 div_without_class = await first_section.query_selector('div:not([class])')
                 if div_without_class:
                     logger.success("找到无class的div标签")
@@ -216,17 +269,20 @@ class ProductHuntScraper:
             new_page = await self.browser.new_page()
             
             # 导航到制作人页面
-            await new_page.goto(maker_link, wait_until="domcontentloaded", timeout=3000000)
+            await new_page.goto(maker_link, wait_until="domcontentloaded", timeout=30000)
             
             # 等待页面加载
-            await new_page.wait_for_timeout(3000)
+            await new_page.wait_for_timeout(15000)
+            logger.info("页面加载等待完成，开始提取内容...")
             
             # 抓取第一个section的tag
+            await self.record_dom_selection('section', "备用方法-新窗口第一个section标签")
             first_section = await new_page.query_selector('section')
             if first_section:
                 logger.success("找到第一个section标签")
                 
                 # 在section下面找一个没有任何class的div标签
+                await self.record_dom_selection('div:not([class])', "备用方法-section下无class的div标签")
                 div_without_class = await first_section.query_selector('div:not([class])')
                 if div_without_class:
                     logger.success("找到无class的div标签")
@@ -248,6 +304,10 @@ class ProductHuntScraper:
                 result = maker_text
                 logger.info(f"制作人发言(回退a标签): {maker_text[:200]}...")
             
+            # 添加充分延迟，确保内容完全加载
+            logger.info("等待内容完全稳定...")
+            await new_page.wait_for_timeout(3000)
+            
             # 关闭新页面
             await new_page.close()
             logger.info("新窗口已关闭")
@@ -269,6 +329,7 @@ class ProductHuntScraper:
             product_info = {}
             
             # 提取产品名称（h1标签）
+            await self.record_dom_selection("h1", "产品名称")
             name_element = await self.page.query_selector("h1")
             if name_element:
                 product_info["name"] = (await name_element.text_content()).strip()
@@ -277,6 +338,7 @@ class ProductHuntScraper:
             # 提取产品简介（class为"relative text-16 font-normal text-gray-700"的div）
             logger.info("正在提取产品简介...")
             try:
+                await self.record_dom_selection('div.relative.text-16.font-normal.text-gray-700', "产品简介")
                 intro_div = await self.page.query_selector('div.relative.text-16.font-normal.text-gray-700')
                 if intro_div:
                     product_info["introduction"] = (await intro_div.text_content()).strip()
@@ -290,6 +352,7 @@ class ProductHuntScraper:
             logger.info("等待制作人发言动态加载...")
             try:
                 # 等待section标签出现，最长等待60秒
+                await self.record_dom_selection('section.flex.flex-col.gap-2', "制作人发言区域")
                 section_element = await self.page.wait_for_selector(
                     'section.flex.flex-col.gap-2', 
                     timeout=60000
@@ -298,6 +361,7 @@ class ProductHuntScraper:
                     logger.success("制作人发言区域已加载")
                     
                     # 查找section标签下面的第一个a标签
+                    await self.record_dom_selection('a', "制作人链接")
                     a_element = await section_element.query_selector('a')
                     if a_element:
                         # 提取a标签的文本内容
@@ -412,6 +476,10 @@ async def main():
             with open("product_info.json", "w", encoding="utf-8") as f:
                 json.dump(product_info, f, ensure_ascii=False, indent=2)
             logger.info("产品信息已保存到 product_info.json")
+            
+            # 保存点击和DOM选取行为记录
+            await scraper.save_behavior_records()
+            logger.info("行为记录已保存到 behavior_records.json")
         else:
             logger.warning("未能提取到产品信息")