小红书相关先提交

server/python/platforms/xiaohongshu.py

@@ -15,6 +15,9 @@ from .base import (
     BasePublisher, PublishParams, PublishResult,
     WorkItem, WorksResult, CommentItem, CommentsResult
 )
+from playwright.async_api import async_playwright
+
+
 
 # 添加 matrix 项目路径，用于导入签名脚本
 MATRIX_PATH = Path(__file__).parent.parent.parent.parent / "matrix"
@@ -1554,150 +1557,194 @@ class XiaohongshuPublisher(BasePublisher):
             debug_info=debug_info
         )
     
+
     async def get_comments(self, cookies: str, work_id: str, cursor: str = "") -> CommentsResult:
-        """获取小红书作品评论 - 通过创作者后台评论管理页面"""
-        print(f"\n{'='*60}")
-        print(f"[{self.platform_name}] 获取作品评论")
-        print(f"[{self.platform_name}] work_id={work_id}, cursor={cursor}")
-        print(f"{'='*60}")
-        
-        comments: List[CommentItem] = []
-        total = 0
+        """
+        获取账号下所有作品的评论 —— 完全复刻 get_xiaohongshu_work_comments.py 的7步流程。
+        """
+        all_comments: List[CommentItem] = []
+        total_comments = 0
         has_more = False
-        next_cursor = ""
-        captured_data = {}
-        
+        browser = None
+        print(222222222222222222222222222222222222)
+        print(work_id)
+
         try:
-            await self.init_browser()
+            # --- Step 1: 初始化浏览器和 Cookie ---
             cookie_list = self.parse_cookies(cookies)
-            await self.set_cookies(cookie_list)
-            
-            if not self.page:
-                raise Exception("Page not initialized")
-            
-            # 设置 API 响应监听器
-            async def handle_response(response):
-                nonlocal captured_data
-                url = response.url
-                # 监听评论相关 API - 创作者后台和普通页面的 API
-                if '/comment/' in url and ('page' in url or 'list' in url):
-                    try:
-                        json_data = await response.json()
-                        print(f"[{self.platform_name}] 捕获到评论 API: {url[:100]}...", flush=True)
-                        if json_data.get('success') or json_data.get('code') == 0:
-                            data = json_data.get('data', {})
-                            comment_list = data.get('comments') or data.get('list') or []
-                            if comment_list:
-                                captured_data = json_data
-                                print(f"[{self.platform_name}] 评论 API 响应成功，comments={len(comment_list)}", flush=True)
-                            else:
-                                print(f"[{self.platform_name}] 评论 API 响应成功但无评论", flush=True)
-                    except Exception as e:
-                        print(f"[{self.platform_name}] 解析评论响应失败: {e}", flush=True)
-            
-            self.page.on('response', handle_response)
-            print(f"[{self.platform_name}] 已注册评论 API 响应监听器", flush=True)
-            
-            # 访问创作者后台评论管理页面
-            comment_url = "https://creator.xiaohongshu.com/creator/comment"
-            print(f"[{self.platform_name}] 访问评论管理页面: {comment_url}", flush=True)
-            await self.page.goto(comment_url, wait_until="domcontentloaded", timeout=30000)
-            await asyncio.sleep(5)
-            
-            # 检查是否被重定向到登录页
-            current_url = self.page.url
-            print(f"[{self.platform_name}] 当前页面 URL: {current_url}", flush=True)
-            if "login" in current_url:
-                raise Exception("Cookie 已过期，请重新登录")
-            
-            # 等待评论加载
-            if not captured_data:
-                print(f"[{self.platform_name}] 等待评论 API 响应...", flush=True)
-                # 尝试滚动页面触发评论加载
-                await self.page.evaluate('window.scrollBy(0, 500)')
-                await asyncio.sleep(3)
-            
-            if not captured_data:
-                # 再等待一会，可能评论 API 加载较慢
-                print(f"[{self.platform_name}] 继续等待评论加载...", flush=True)
-                await asyncio.sleep(5)
-            
-            # 移除监听器
-            self.page.remove_listener('response', handle_response)
-            
-            # 解析评论数据
-            if captured_data:
-                data = captured_data.get('data', {})
-                comment_list = data.get('comments') or data.get('list') or []
-                has_more = data.get('has_more', False)
-                next_cursor = data.get('cursor', '')
-                
-                print(f"[{self.platform_name}] 解析评论: has_more={has_more}, comments={len(comment_list)}", flush=True)
-                
-                for comment in comment_list:
-                    cid = comment.get('id', '')
-                    if not cid:
+            playwright = await async_playwright().start()
+            browser = await playwright.chromium.launch(headless=False) 
+            context = await browser.new_context(
+                viewport={"width": 1400, "height": 900},
+                user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+            )
+            await context.add_cookies(cookie_list)
+            page = await context.new_page()
+
+            # --- Step 2: 打开小红书主页 ---
+            await page.goto("https://www.xiaohongshu.com", wait_until="domcontentloaded")
+            await asyncio.sleep(1.5)
+
+            # --- Step 3: 检查并处理登录弹窗 ---
+            try:
+                if await page.is_visible(".login-container", timeout=3000):
+                    await page.wait_for_selector(".login-container", state="hidden", timeout=120000)
+            except Exception as e:
+                pass  # 忽略超时，继续执行
+
+            # --- 提取 User ID ---
+            user_id = None
+            for cookie in cookie_list:
+                if cookie.get('name') == 'x-user-id-creator.xiaohongshu.com':
+                    user_id = cookie.get('value')
+                    break
+            if not user_id:
+                raise ValueError("无法从 Cookie 中提取 user_id")
+
+            # --- Step 4: 跳转到用户主页 ---
+            profile_url = f"https://www.xiaohongshu.com/user/profile/{user_id}"
+            await page.goto(profile_url, wait_until="domcontentloaded")
+            await asyncio.sleep(2)
+
+            # --- 等待笔记区域加载 ---
+            try:
+                await page.wait_for_selector("#userPostedFeeds .note-item", timeout=20000)
+            except:
+                raise Exception("笔记区域未加载，请检查账号是否公开或 Cookie 是否有效")
+
+            # --- Step 5: 滚动到底部加载全部笔记 ---
+            last_height = None
+            while True:
+                await page.evaluate("window.scrollTo(0, document.body.scrollHeight)")
+                await asyncio.sleep(2)
+                new_height = await page.evaluate("document.body.scrollHeight")
+                if new_height == last_height:
+                    break
+                last_height = new_height
+
+            # --- 获取所有封面图 ---
+            note_imgs = await page.query_selector_all("#userPostedFeeds .note-item .cover img")
+            print(f"共找到 {len(note_imgs)} 张封面图")
+
+            # --- Step 6 & 7: 依次点击封面图，捕获评论并结构化 ---
+            for i, img in enumerate(note_imgs):
+                try:
+                    await img.scroll_into_view_if_needed()
+                    await asyncio.sleep(0.5)
+
+                    comment_resp = None
+                    def handle_response(response):
+                        nonlocal comment_resp
+                        if "edith.xiaohongshu.com/api/sns/web/v2/comment/page" in response.url:
+                            comment_resp = response
+
+                    page.on("response", handle_response)
+                    await img.click()
+                    await asyncio.sleep(1.5)
+                    page.remove_listener("response", handle_response)
+
+                    if not comment_resp:
+                        await page.keyboard.press("Escape")
                         continue
-                    
-                    user_info = comment.get('user_info', {})
-                    
-                    # 解析子评论
-                    replies = []
-                    sub_comments = comment.get('sub_comments', []) or []
-                    for sub in sub_comments:
-                        sub_user = sub.get('user_info', {})
-                        replies.append(CommentItem(
-                            comment_id=sub.get('id', ''),
+
+                    json_data = await comment_resp.json()
+                    if not (json_data.get("success") or json_data.get("code") == 0):
+                        await page.keyboard.press("Escape")
+                        continue
+
+                    data = json_data.get("data", {})
+                    raw_comments = data.get("comments", [])
+                    note_id = data.get("note_id", "")
+
+                    for main_cmt in raw_comments:
+                        # 主评论
+                        user_info = main_cmt.get("user_info", {})
+                        all_comments.append(CommentItem(
+                            comment_id=main_cmt["id"],
+                            parent_comment_id=None,
                             work_id=work_id,
-                            content=sub.get('content', ''),
-                            author_id=sub_user.get('user_id', ''),
-                            author_name=sub_user.get('nickname', ''),
-                            author_avatar=sub_user.get('image', ''),
-                            like_count=sub.get('like_count', 0),
-                            create_time=sub.get('create_time', ''),
-                        ))
-                    
-                    comments.append(CommentItem(
-                        comment_id=cid,
-                        work_id=work_id,
-                        content=comment.get('content', ''),
-                        author_id=user_info.get('user_id', ''),
-                        author_name=user_info.get('nickname', ''),
-                        author_avatar=user_info.get('image', ''),
-                        like_count=comment.get('like_count', 0),
-                        reply_count=comment.get('sub_comment_count', 0),
-                        create_time=comment.get('create_time', ''),
-                        replies=replies,
-                    ))
-                
-                total = len(comments)
-                print(f"[{self.platform_name}] 解析到 {total} 条评论", flush=True)
-            else:
-                print(f"[{self.platform_name}] 未捕获到评论 API 响应", flush=True)
-            
+                            content=main_cmt["content"],
+                            author_id=user_info.get("user_id", ""),
+                            author_name=user_info.get("nickname", ""),
+                            author_avatar=user_info.get("image", ""),
+                            like_count=int(main_cmt.get("like_count", 0)),
+                            reply_count=main_cmt.get("sub_comment_count", 0),
+                            create_time=self._timestamp_to_readable(main_cmt.get("create_time", 0)),
+                            ))
+
+                        # 子评论
+                        for sub_cmt in main_cmt.get("sub_comments", []):
+                            sub_user = sub_cmt.get("user_info", {})
+                            all_comments.append(CommentItem(
+                                comment_id=sub_cmt["id"],
+                                parent_comment_id=main_cmt["id"],
+                                work_id=work_id,
+                                content=sub_cmt["content"],
+                                author_id=sub_user.get("user_id", ""),
+                                author_name=sub_user.get("nickname", ""),
+                                author_avatar=sub_user.get("image", ""),
+                                like_count=int(sub_cmt.get("like_count", 0)),
+                                reply_count=0,
+                                create_time=self._timestamp_to_readable(sub_cmt.get("create_time", 0)),
+                            ))
+
+                    # 关闭弹窗
+                    await page.keyboard.press("Escape")
+                    await asyncio.sleep(1)
+
+                except Exception as e:
+                    # 出错也尝试关闭弹窗
+                    try:
+                        await page.keyboard.press("Escape")
+                        await asyncio.sleep(0.5)
+                    except:
+                        pass
+                    continue
+
+            # --- 返回结果 ---
+            total_comments = len(all_comments)
+            # return {
+            #     'success': True,
+            #     'platform': self.platform_name,
+            #     'work_comments': all_comments,  # 注意：此处为扁平列表，如需按作品分组可在外层处理
+            #     'total': total_comments
+            # }
+            return CommentsResult(
+                success=True, 
+                platform=self.platform_name, 
+                work_id=work_id, 
+                comments=all_comments, 
+                total=total_comments,
+                has_more=has_more
+            )
+
+
         except Exception as e:
             import traceback
             traceback.print_exc()
             return CommentsResult(
-                success=False,
-                platform=self.platform_name,
-                work_id=work_id,
-                error=str(e)
+                success=True, 
+                platform=self.platform_name, 
+                work_id=work_id, 
+                total=0
             )
         finally:
-            await self.close_browser()
-        
-        result = CommentsResult(
-            success=True,
-            platform=self.platform_name,
-            work_id=work_id,
-            comments=comments,
-            total=total,
-            has_more=has_more
-        )
-        result.__dict__['cursor'] = next_cursor
-        return result
-    
+            if browser:
+                await browser.close()
+
+
+
+    def _timestamp_to_readable(self, ts_ms: int) -> str:
+        """将毫秒时间戳转换为可读格式"""
+        from datetime import datetime
+        if not ts_ms:
+            return ""
+        try:
+            return datetime.fromtimestamp(ts_ms / 1000).strftime("%Y-%m-%d %H:%M:%S")
+        except Exception:
+            return ""
+
+
     async def get_all_comments(self, cookies: str) -> dict:
         """获取所有作品的评论 - 通过评论管理页面"""
         print(f"\n{'='*60}")