元宝修复 app 网页 对话内容修复

aidso_geo/models/yuanbao_android_data_process.py

@@ -31,21 +31,22 @@ def yuanbao_android_process_original_data(data):
                     # 提取并解析JSON数据
                     data_str = i.split("data: ")[1]
                     json_data = json.loads(data_str)
+
                 except (IndexError, json.JSONDecodeError):
                     continue  # 跳过格式错误的数据
                 if json_data.get('type') == 'searchGuid':
                     url_list = json_data.get('docs')
                 if json_data.get('type') == 'think':
                     think_content += json_data.get('content')
+
                 if json_data.get('type') == 'text' and json_data.get('msg') is not None:
                     response_content += json_data.get('msg')
-                    response_process += json_data.get('msg')
                 if json_data.get('type') == 'deepSearch':
                     if json_data.get('contents'):
                         if json_data.get('contents')[0].get('msg'):
-                            think_content += json_data.get('contents')[0].get('msg')
-                            if not response_process:
-                                response_content += json_data.get('contents')[0].get('msg')
+                            if json_data.get('contents')[0].get('type') == 'text':
+                                think_content += json_data.get('contents')[0].get('msg')
+
                         if json_data.get('contents')[0].get('docs'):
                             url_list.extend(json_data.get('contents')[0].get('docs'))
                 if json_data.get('type') == 'image':
@@ -72,6 +73,7 @@ def yuanbao_android_process_original_data(data):
                         mini_program_str = json.dumps(mini_program, ensure_ascii=False)
                         response_content += mini_program_str
                         response_content += '\n\nrender_ecom_card_widget_miniprogram_end:'
+
         spider_save_tos.process_and_save_files(file_path, search_keyword, url_list, think_content, response_content,
                                                suggestions,rich_media_block)
         return (file_path, search_keyword, url_list, think_content, response_content, suggestions)
@@ -105,7 +107,7 @@ if __name__ == '__main__':
     # yuanbao_android_process_original_data(file_path2)
 
 
-    data_list = bh_utils.query_data(f"select * from geo_commit_task  where taskId = 'f9018d5c-2d0a-4e6a-9476-4b92334a0715' and  platform  = 'TXYBA'")
+    data_list = bh_utils.query_data(f"select * from geo_commit_task  where taskId = '539a80f8-cd2d-4ae7-a45c-2b2d41fa738e' and  platform  = 'TXYBA'")
 
     # # #
     # # #

aidso_geo/models/yuanbao_data_process.py

 import  json
 
-from aidso_geo.utils import robot_utils
+from aidso_geo.utils import robot_utils, bh_utils
 from aidso_geo.utils.ai_interface import get_parse_sse_result
 from aidso_geo.utils.tos_utils import get_string_from_tos
 
@@ -30,6 +30,7 @@ def yuanbao_process_original_data(data):
                     # 提取并解析JSON数据
                     data_str = i.split("data: ")[1]
                     json_data = json.loads(data_str)
+
                 except (IndexError, json.JSONDecodeError):
                     continue  # 跳过格式错误的数据
 
@@ -42,9 +43,10 @@ def yuanbao_process_original_data(data):
                 if json_data.get('type') == 'deepSearch':
                     if json_data.get('contents'):
                         if json_data.get('contents')[0].get('msg'):
-                            think_content += json_data.get('contents')[0].get('msg')
-                            if not response_process:
-                                response_content += json_data.get('contents')[0].get('msg')
+                            if json_data.get('contents')[0].get('type') =='text':
+                                think_content += json_data.get('contents')[0].get('msg')
+                            # if not response_process:
+                            #     response_content += json_data.get('contents')[0].get('msg')
                 if json_data.get('type') == 'image':
                     response_content = "生成了图片"
         spider_save_tos.process_and_save_files(file_path, search_keyword, url_list, think_content, response_content,
@@ -74,7 +76,29 @@ def yuanbao_process_original_data(data):
     return (file_path, search_keyword, url_list, think_content, response_content, suggestions)
 
 if __name__ == '__main__':
-    # https://tcdn.aidso.com/geo/c7581554-435b-47ca-a2e5-d52bea763b8a/DOUBA/search_keyword.txt?secret=11049cbfb1c7018597a2085455505304520f0507080c0c4b514e09
-    file_path = 'geo/b55be837-48a9-476a-8290-1339ea2ce3f9/TXYB/original.text'
-    yuanbao_process_original_data(file_path)
+    data_list = bh_utils.query_data(f"select * from geo_commit_task  where taskId = '69189ba0-e443-4e7c-a465-cabc07973540' and  platform  = 'TXYB'")
+
+    # # #
+    # # #
+    # # # # #
+    def handle_item(i):
+        if i.get('comWordsMap'):
+            i['comWordsMap'] = json.loads(i.get('comWordsMap'))
+        if i.get('brandWords'):
+            i['brandWords'] = json.loads(i.get('brandWords'))
+        if i.get('comWords'):
+            i['comWords'] = json.loads(i.get('comWords'))
+        if i.get('keywords'):
+            i['keywords'] = json.loads(i.get('keywords'))
+        if i.get('productWordsMap'):
+            i['productWordsMap'] = json.loads(i.get('productWordsMap'))
+
+        return yuanbao_process_original_data(i)
+
 
+    if data_list:
+        for i in data_list:
+            try:
+                handle_item(i)
+            except Exception as e:
+                ...