'commit'

dsLightRag/WxGzh/T2_GetArticleList.py

@@ -47,6 +47,20 @@ from selenium import webdriver
 from selenium.webdriver.chrome.options import Options
 from selenium.webdriver.chrome.service import Service as ChromeService
 
+async def save_article_to_db(pool, article_title, account_name, article_url, publish_time, content, account_id):
+    try:
+        async with pool.acquire() as conn:
+            # 确保account_id是整数
+            account_id_int = int(account_id) if account_id else 0
+            await conn.execute('''
+                INSERT INTO t_wechat_articles 
+                (title, source, url, publish_time, content, source_id)
+                VALUES ($1, $2, $3, $4, $5, $6)
+            ''', article_title, account_name, article_url,
+               publish_time, content, account_id_int)  # 修改为整数类型
+    except Exception as e:
+        logging.error(f"保存文章失败: {e}")
+
 if __name__ == '__main__':
     # 从文件cookies.txt中获取
     with open('cookies.txt', 'r', encoding='utf-8') as f:
@@ -157,17 +171,26 @@ if __name__ == '__main__':
         fakeid_list = query_fakeid_response.json().get('app_msg_list')
 
         for item in fakeid_list:
-            # 采集item示例
             article_url = item.get('link')
             article_title = item.get('title')
-            publish_time = datetime.datetime.fromtimestamp(int(item.get("update_time"))).strftime('%Y-%m-%d %H:%M:%S')
+            publish_time = datetime.datetime.fromtimestamp(int(item.get("update_time")))
 
-            # 直接获取并显示文章内容
             if '试卷' in article_title:  # 过滤掉试卷
                 continue
+
             print(f"正在处理文章: {article_title} ({publish_time})")
             content = get_article_content(article_url)
-            print(f"文章内容预览: {content[:200]}...")
+
+            loop = asyncio.new_event_loop()
+            asyncio.set_event_loop(loop)
+            try:
+                pool = loop.run_until_complete(init_postgres_pool())
+                loop.run_until_complete(
+                    save_article_to_db(pool, article_title, account_name, article_url, publish_time, content,
+                                       account_id))
+            finally:
+                loop.run_until_complete(pool.close())
+                loop.close()
 
             time.sleep(1)
         # 关闭浏览器