refactor(kb-zhihu-metadata): Simplify to only support zvideo and zhuanlan

- Remove question, answer, user types (not needed for knowledge base)
- Keep only zvideo (知乎视频) and zhuanlan (知乎专栏文章)
- Update URL patterns and extraction logic
- Simplified CLI interface

skills/kb-zhihu-metadata/zhihu_extractor.py

@@ -78,9 +78,8 @@ class ZhihuPlaywrightExtractor:
     """知乎内容提取器 - 使用 Playwright"""
 
     URL_PATTERNS = {
-        'answer': r'zhihu\.com/question/\d+/answer/(\d+)',
+        'zvideo': r'zhihu\.com/zvideo/(\d+)',
-        'article': r'zhuanlan\.zhihu\.com/p/(\d+)',
+        'zhuanlan': r'zhuanlan\.zhihu\.com/p/(\d+)',
-        'question': r'zhihu\.com/question/(\d+)(?:$|/)',
     }
 
     def __init__(self, headless: bool = True, timeout: int = 30):
@@ -133,10 +132,10 @@ class ZhihuPlaywrightExtractor:
             return ''
         return element.inner_html() if hasattr(element, 'inner_html') else ''
 
-    def _extract_answer_content(self) -> Dict[str, Any]:
+    def _extract_zvideo_content(self) -> Dict[str, Any]:
-        """提取回答内容"""
+        """提取知乎视频内容"""
         result = {
-            'type': 'answer',
+            'type': 'zvideo',
             'title': '',
             'author': '',
             'content': '',
@@ -149,72 +148,49 @@ class ZhihuPlaywrightExtractor:
         page_title = self.browser.page.title()
         print(f"页面标题: {page_title}")
 
-        # 尝试保存页面快照用于调试
+        # 提取视频标题
-        try:
+        title_elem = self.browser.query_selector('.VideoPlaylistInfo-title')
-            import os
+        if title_elem:
-            debug_dir = '/home/admin/.openclaw/workspace/debug'
+            result['title'] = self._extract_text_content(title_elem)
-            os.makedirs(debug_dir, exist_ok=True)
+        
-            self.browser.page.screenshot(path=os.path.join(debug_dir, 'zhihu_screenshot.png'))
+        # 备用：尝试其他选择器
-            with open(os.path.join(debug_dir, 'zhihu_page.html'), 'w', encoding='utf-8') as f:
+        if not result['title']:
-                f.write(self.browser.page.content())
+            title_elem = self.browser.query_selector('h1')
-            print(f"页面快照已保存到 {debug_dir}")
-        except Exception as e:
-            print(f"保存页面快照时出错: {e}")
-
-        # 尝试多个可能的选择器
-        selectors = {
-            'title': ['h1.QuestionHeader-title', 'h1', '.QuestionHeader-main'],
-            'author': ['.AuthorInfo-name', '.AuthorInfo .Text', '[itemprop="name"]'],
-            'content': ['.RichContent-inner', '.RichContent', '.List-itemText', '[data-zop-feedlist] article'],
-            'time': ['.ContentItem-time', '[itemprop="dateCreated"]'],
-            'vote': ['.VoteButton--up .VoteButton-label', '.VoteButton--up']
-        }
-
-        for selector in selectors['title']:
-            title_elem = self.browser.query_selector(selector)
             if title_elem:
                 result['title'] = self._extract_text_content(title_elem)
-                print(f"找到标题（使用选择器 {selector}）: {result['title'][:50]}...")
-                break
 
-        for selector in selectors['author']:
+        # 提取作者信息
-            author_elem = self.browser.query_selector(selector)
+        author_elem = self.browser.query_selector('.AuthorInfo-name')
-            if author_elem:
+        if author_elem:
-                result['author'] = self._extract_text_content(author_elem)
+            result['author'] = self._extract_text_content(author_elem)
-                print(f"找到作者（使用选择器 {selector}）: {result['author']}")
-                break
 
-        for selector in selectors['content']:
+        # 提取视频描述
-            content_elem = self.browser.query_selector(selector)
+        desc_elem = self.browser.query_selector('.Video-desc')
-            if content_elem:
+        if desc_elem:
-                html_content = self._extract_html_content(content_elem)
+            html_content = self._extract_html_content(desc_elem)
-                result['content'] = html_content
+            result['content'] = html_content
-                result['content_markdown'] = self.markdown_converter.convert(html_content)
+            result['content_markdown'] = self.markdown_converter.convert(html_content)
-                print(f"找到内容（使用选择器 {selector}），长度: {len(result['content_markdown'])}")
-                break
 
-        for selector in selectors['time']:
+        # 提取发布时间
-            time_elem = self.browser.query_selector(selector)
+        time_elem = self.browser.query_selector('.Video-meta time')
-            if time_elem:
+        if time_elem:
-                result['publish_time'] = time_elem.get_attribute('data-tooltip') or self._extract_text_content(time_elem)
+            result['publish_time'] = time_elem.get_attribute('datetime') or self._extract_text_content(time_elem)
-                break
 
-        for selector in selectors['vote']:
+        # 提取点赞数
-            vote_elem = self.browser.query_selector(selector)
+        vote_elem = self.browser.query_selector('.VideoLikeButton-count')
-            if vote_elem:
+        if vote_elem:
-                vote_text = self._extract_text_content(vote_elem)
+            vote_text = self._extract_text_content(vote_elem)
-                try:
+            try:
-                    result['vote_up'] = int(vote_text.replace(',', ''))
+                result['vote_up'] = int(vote_text.replace(',', ''))
-                except (ValueError, AttributeError):
+            except (ValueError, AttributeError):
-                    pass
+                pass
-                break
 
         return result
 
-    def _extract_article_content(self) -> Dict[str, Any]:
+    def _extract_zhuanlan_content(self) -> Dict[str, Any]:
-        """提取文章内容"""
+        """提取知乎专栏文章内容"""
         result = {
-            'type': 'article',
+            'type': 'zhuanlan',
             'title': '',
             'author': '',
             'content': '',
@@ -223,11 +199,21 @@ class ZhihuPlaywrightExtractor:
             'vote_up': 0,
         }
 
+        # 打印页面标题用于调试
+        page_title = self.browser.page.title()
+        print(f"页面标题: {page_title}")
+
         # 提取文章标题
         title_elem = self.browser.query_selector('.Post-Title')
         if title_elem:
             result['title'] = self._extract_text_content(title_elem)
 
+        # 备用
+        if not result['title']:
+            title_elem = self.browser.query_selector('h1')
+            if title_elem:
+                result['title'] = self._extract_text_content(title_elem)
+
         # 提取作者信息
         author_elem = self.browser.query_selector('.AuthorInfo-name')
         if author_elem:
@@ -256,37 +242,6 @@ class ZhihuPlaywrightExtractor:
 
         return result
 
-    def _extract_question_content(self) -> Dict[str, Any]:
-        """提取问题内容（问题页本身）"""
-        result = {
-            'type': 'question',
-            'title': '',
-            'author': '',
-            'content': '',
-            'content_markdown': '',
-            'publish_time': '',
-            'vote_up': 0,
-        }
-
-        # 提取问题标题
-        title_elem = self.browser.query_selector('h1.QuestionHeader-title')
-        if title_elem:
-            result['title'] = self._extract_text_content(title_elem)
-
-        # 提取问题详情
-        detail_elem = self.browser.query_selector('.QuestionHeader-detail')
-        if detail_elem:
-            html_content = self._extract_html_content(detail_elem)
-            result['content'] = html_content
-            result['content_markdown'] = self.markdown_converter.convert(html_content)
-
-        # 问题的作者通常是提问者
-        author_elem = self.browser.query_selector('.AuthorInfo-name')
-        if author_elem:
-            result['author'] = self._extract_text_content(author_elem)
-
-        return result
-
     def extract(self, url: str) -> Optional[Dict[str, Any]]:
         """
         从 URL 提取知乎内容
@@ -297,7 +252,7 @@ class ZhihuPlaywrightExtractor:
         Returns:
             包含提取内容的字典，结构如下:
             {
-                'type': 'answer' | 'article' | 'question',
+                'type': 'zvideo' | 'zhuanlan',
                 'title': str,
                 'author': str,
                 'content': str (HTML),
@@ -328,25 +283,20 @@ class ZhihuPlaywrightExtractor:
             import time
             time.sleep(3)  # 等待 JavaScript 执行
 
-            # 尝试多个可能的选择器
+            # 根据类型提取内容
-            if url_type == 'answer':
+            if url_type == 'zvideo':
-                selectors = ['.RichContent-inner', '.RichContent', '.List-itemText', '[data-zop-feedlist]']
+                result = self._extract_zvideo_content()
-                print(f"尝试查找选择器: {selectors}")
+            elif url_type == 'zhuanlan':
-                result = self._extract_answer_content()
+                result = self._extract_zhuanlan_content()
-            elif url_type == 'article':
+            else:
-                selectors = ['.Post-RichText', '.Post-Title']
+                print(f"不支持的类型: {url_type}")
-                print(f"尝试查找选择器: {selectors}")
+                return None
-                result = self._extract_article_content()
-            else:  # question
-                selectors = ['.QuestionHeader-title', 'h1']
-                print(f"尝试查找选择器: {selectors}")
-                result = self._extract_question_content()
 
             # 添加额外信息
             result['url'] = url
             result['content_id'] = self.extract_id(url)
 
-            print(f"提取成功! 标题: {result['title'][:50]}...")
+            print(f"提取成功! 标题: {result.get('title', '')[:50]}...")
             return result
 
         except Exception as e:

skills/kb-zhihu-metadata/zhihu_metadata.py

@@ -2,7 +2,9 @@
 """
 知乎元数据获取工具
 
-支持获取知乎问题、回答、用户的元数据信息。
+仅支持两种内容类型：
+- zvideo: 知乎视频 (https://www.zhihu.com/zvideo/xxx)
+- zhuanlan: 知乎专栏文章 (https://zhuanlan.zhihu.com/p/xxx)
 """
 
 import argparse
@@ -19,52 +21,34 @@ if current_dir not in sys.path:
 from zhihu_extractor import ZhihuPlaywrightExtractor
 
 
-def extract_question(question_id: str, headless: bool = True) -> Dict[str, Any]:
+def extract_zvideo(video_id: str, headless: bool = True) -> Dict[str, Any]:
     """
-    获取知乎问题元数据
+    获取知乎视频元数据
 
     Args:
-        question_id: 问题 ID (数字字符串)
+        video_id: 视频 ID (数字字符串)
         headless: 是否使用无头浏览器
 
     Returns:
-        问题元数据字典
+        视频元数据字典
     """
-    url = f"https://www.zhihu.com/question/{question_id}"
+    url = f"https://www.zhihu.com/zvideo/{video_id}"
-    return _extract_metadata(url, "question", headless)
+    return _extract_metadata(url, "zvideo", headless)
 
 
-def extract_answer(answer_id: str, headless: bool = True) -> Dict[str, Any]:
+def extract_zhuanlan(article_id: str, headless: bool = True) -> Dict[str, Any]:
     """
-    获取知乎回答元数据
+    获取知乎专栏文章元数据
 
     Args:
-        answer_id: 回答 ID (数字字符串)
+        article_id: 文章 ID (数字字符串)
         headless: 是否使用无头浏览器
 
     Returns:
-        回答元数据字典
+        文章元数据字典
     """
-    # 回答 URL 格式: https://www.zhihu.com/question/{question_id}/answer/{answer_id}
+    url = f"https://zhuanlan.zhihu.com/p/{article_id}"
-    # 需要先获取问题ID，这里简化处理，尝试从回答ID构建URL
+    return _extract_metadata(url, "zhuanlan", headless)
-    # 实际上知乎回答URL需要问题ID，这里用模拟URL
-    url = f"https://www.zhihu.com/question/0/answer/{answer_id}"
-    return _extract_metadata(url, "answer", headless)
-
-
-def extract_user(url_token: str, headless: bool = True) -> Dict[str, Any]:
-    """
-    获取知乎用户信息
-
-    Args:
-        url_token: 用户 URL token
-        headless: 是否使用无头浏览器
-
-    Returns:
-        用户信息字典
-    """
-    url = f"https://www.zhihu.com/people/{url_token}"
-    return _extract_user_metadata(url, headless)
 
 
 def _extract_metadata(url: str, content_type: str, headless: bool) -> Dict[str, Any]:
@@ -73,7 +57,7 @@ def _extract_metadata(url: str, content_type: str, headless: bool) -> Dict[str,
 
     Args:
         url: 知乎 URL
-        content_type: 内容类型 (question/answer)
+        content_type: 内容类型 (zvideo/zhuanlan)
         headless: 是否使用无头浏览器
 
     Returns:
@@ -104,120 +88,33 @@ def _extract_metadata(url: str, content_type: str, headless: bool) -> Dict[str,
         }
 
 
-def _extract_user_metadata(url: str, headless: bool) -> Dict[str, Any]:
-    """
-    内部函数：提取用户信息
-
-    Args:
-        url: 用户主页 URL
-        headless: 是否使用无头浏览器
-
-    Returns:
-        用户信息字典
-    """
-    extractor = ZhihuPlaywrightExtractor(headless=headless)
-
-    try:
-        extractor.browser.start()
-        extractor.browser.goto(url, wait_until="networkidle")
-
-        import time
-        time.sleep(3)
-
-        # 提取用户名
-        username = ""
-        username_elem = extractor.browser.query_selector('.ProfileHeader-name')
-        if username_elem:
-            username = username_elem.inner_text()
-
-        # 提取个人简介
-        headline = ""
-        headline_elem = extractor.browser.query_selector('.ProfileHeader-headline')
-        if headline_elem:
-            headline = headline_elem.inner_text()
-
-        # 提取简介详情
-        description = ""
-        desc_elem = extractor.browser.query_selector('.ProfileHeader-description')
-        if desc_elem:
-            description = desc_elem.inner_text()
-
-        # 提取关注者数
-        followers = 0
-        follower_elem = extractor.browser.query_selector('.FollowshipCard-followersNum')
-        if follower_elem:
-            follower_text = follower_elem.inner_text()
-            try:
-                followers = int(follower_text.replace(',', ''))
-            except ValueError:
-                pass
-
-        # 提取关注者数
-        following = 0
-        following_elem = extractor.browser.query_selector('.FollowshipCard-followingsNum')
-        if following_elem:
-            following_text = following_elem.inner_text()
-            try:
-                following = int(following_text.replace(',', ''))
-            except ValueError:
-                pass
-
-        # 提取回答数
-        answers = 0
-        answers_elem = extractor.browser.query_selector('.Profile-mainColumn .Tabs-item span')
-        if answers_elem:
-            answers_text = answers_elem.inner_text()
-            try:
-                answers = int(answers_text.replace(',', ''))
-            except ValueError:
-                pass
-
-        metadata = {
-            "type": "user",
-            "url": url,
-            "username": username,
-            "headline": headline,
-            "description": description,
-            "followers": followers,
-            "following": following,
-            "answers_count": answers,
-        }
-        return metadata
-
-    except Exception as e:
-        return {
-            "type": "user",
-            "error": str(e),
-            "url": url
-        }
-    finally:
-        extractor.browser.close()
-
-
 def main():
     """命令行入口"""
     parser = argparse.ArgumentParser(
         description="知乎元数据获取工具",
         formatter_class=argparse.RawDescriptionHelpFormatter,
         epilog="""
+支持的类型:
+  zvideo   - 知乎视频 (https://www.zhihu.com/zvideo/123456)
+  zhuanlan - 知乎专栏文章 (https://zhuanlan.zhihu.com/p/123456)
+
 示例:
-  python zhihu_metadata.py --id "266309957" --type question
+  python zhihu_metadata.py --id "123456" --type zvideo
-  python zhihu_metadata.py --id "2003427891925509507" --type answer
+  python zhihu_metadata.py --id "123456" --type zhuanlan
-  python zhihu_metadata.py --id "example-user" --type user
         """
     )
 
     parser.add_argument(
         "--id", "-i",
         required=True,
-        help="知乎内容 ID (问题ID、回答ID或用户URL token)"
+        help="知乎内容 ID (视频ID或文章ID)"
     )
 
     parser.add_argument(
         "--type", "-t",
         required=True,
-        choices=["question", "answer", "user"],
+        choices=["zvideo", "zhuanlan"],
-        help="内容类型"
+        help="内容类型: zvideo=视频, zhuanlan=专栏文章"
     )
 
     parser.add_argument(
@@ -236,12 +133,10 @@ def main():
     args = parser.parse_args()
 
     # 提取元数据
-    if args.type == "question":
+    if args.type == "zvideo":
-        result = extract_question(args.id, headless=args.headless)
+        result = extract_zvideo(args.id, headless=args.headless)
-    elif args.type == "answer":
+    else:  # zhuanlan
-        result = extract_answer(args.id, headless=args.headless)
+        result = extract_zhuanlan(args.id, headless=args.headless)
-    else:  # user
-        result = extract_user(args.id, headless=args.headless)
 
     # 输出结果
     json_output = json.dumps(result, ensure_ascii=False, indent=2)