deepseek app引用来源变更

aidso_geo/core/routes/interface.py

@@ -158,7 +158,7 @@ def task_commit():
         type = data.get('type',"")
         platform = data.get('platform',"")
         prompt = data.get('prompt',"")
-        data["pt"] = datetime.now().strftime("%Y%m%d")
+
         required_fields = ["prompt", "taskId", "reqId", "platform", "type"]
         missing_fields = [field for field in required_fields if field not in data]
         if missing_fields:

aidso_geo/core/routes/third_interface.py

@@ -458,7 +458,8 @@ def mt_task_commit():
 
     prompt = data.get("prompt")
     plat_form = data.get("platform")
-
+    if len(prompt) > 256:
+        return err(400, "prompt长度不能超过256个字符")
     if not prompt:
         return err(400, "missing required field: prompt")
 

aidso_geo/models/deepseek_android_data_process.py

 import json
-
+import re
 from aidso_geo.models import spider_save_tos
-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
 
@@ -17,6 +17,7 @@ def deepseek_android_process_original_data(data):
     think_bool = False
     suggestions = []
     response_bool = False
+    url_id_map = {}
 
 
     try:
@@ -34,6 +35,8 @@ def deepseek_android_process_original_data(data):
                     # 提取并解析JSON数据
                     data_str = item.split("data: ")[1]
                     json_data = json.loads(data_str)
+
+
                 except (IndexError, json.JSONDecodeError):
                     continue
                 if isinstance(json_data.get('v'), dict):
@@ -95,6 +98,50 @@ def deepseek_android_process_original_data(data):
                     if json_data.get('p') == 'response/fragments' and json_data.get('v')[0].get('type') == 'TIP':
                         response_bool = False
                         continue
+
+
+                    if isinstance(json_data.get('v'),list):
+                        v_json_data = json_data.get('v')
+                        has_tool_search = False
+                        tool_search_id = None
+
+                        for item in v_json_data:
+                            if item.get('p') == 'references':
+                                references = item.get('v') or []
+
+                                if isinstance(references, list):
+                                    for reference in references:
+                                        if (
+                                                isinstance(reference, dict)
+                                                and reference.get('type') == 'TOOL_OPEN'
+                                        ):
+                                            has_tool_search = True
+                                            tool_search_id = reference.get('id')
+                                            break
+                        for v in v_json_data:
+                            if v.get('p') == 'content' and has_tool_search:
+                                p_content = v.get('v') or ''
+                                p_content = re.sub(
+                                    r'\[reference:\d+\]',
+                                    f'[reference:{tool_search_id}]',
+                                    p_content
+                                )
+                                response_content+=p_content
+
+                            if v.get('p') == 'fragments':
+                                if v.get('p') == 'fragments':
+                                    fragments = v.get('v') or []
+
+                                    if isinstance(fragments, list):
+                                        for fragment in fragments:
+                                            if (
+                                                    isinstance(fragment, dict)
+                                                    and fragment.get('type') == 'TOOL_OPEN'
+                                                    and fragment.get('result') is not None
+                                            ):
+                                                url_id_map[fragment.get('result').get('url')] = fragment.get('id')
+
+
                     if json_data.get('p') == 'response':
                         response_bool = False
                         continue
@@ -149,6 +196,25 @@ def deepseek_android_process_original_data(data):
                     if response_bool:
                         if isinstance(json_data.get('v'),str):
                             response_content += json_data.get('v')
+
+        used_ids = set(url_id_map.values())
+        next_id = 1
+
+        for item in url_list:
+            url = item.get('url')
+
+            if url in url_id_map:
+                item['cite_index'] = url_id_map[url]
+
+            elif item.get('cite_index') is None:
+                while next_id in used_ids:
+                    next_id += 1
+
+                item['cite_index'] = next_id
+                used_ids.add(next_id)
+                next_id += 1
+
+
         spider_save_tos.process_and_save_files(file_path, search_keyword, url_list, think_content, response_content,
                                                suggestions)
         return (file_path, search_keyword, url_list, think_content, response_content, suggestions)
@@ -176,9 +242,29 @@ def deepseek_android_process_original_data(data):
     return (file_path, search_keyword, url_list, think_content, response_content, suggestions)
 
 if __name__ == '__main__':
-    file_path1 = 'geo/b1480658-f379-4ca7-8897-51ffef64b2d0/DPA/original.text'
-    file_path2 = 'geo/799bfc72-60dc-4333-b551-3e1d284c7570/DPA/original.text'
-    # file_path1 = "geo/0f59c196-aff1-46a5-aa50-5c792ebf481e/DPA/original.text"
-    # file_path = "geo/0c0abba3-9122244-wwwsss4e2dddd1ssssaaass-96aaaa2wwww1-10086/DPA/original.text"
-    # file_path2 = "geo/0a715fc6-ecc8-4e48-baf1-abc0c9477c1e/DPA/original.text"
-    deepseek_android_process_original_data(file_path2)
+    data_list = bh_utils.query_data(f"select * from geo_commit_task  where taskId = 'dd37fa86e6dd49c4b0617ceb47cea25c' and  platform  = 'DPA'")
+
+    # # #
+    # # #
+    # # # # #
+    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 deepseek_android_process_original_data(i)
+
+
+    if data_list:
+        for i in data_list:
+            try:
+                handle_item(i)
+            except Exception as e:
+                ...

aidso_geo/models/doubao_android_data_process.py

@@ -343,7 +343,7 @@ if __name__ == '__main__':
     # file_path = 'geo/c7eb465e-f385-4aa2-89c4-a7cf11897f45/KIMI/1(1).txt'
 
     data_list = bh_utils.query_data(
-        f"select * from geo_commit_task where taskId = '79ac3438-a362-4eaa-b954-070791ba634b' and platform =  'DOUBA'")
+        f"select * from geo_commit_task where taskId = '3f53b599-3710-4312-9b50-73f07bdba631' and platform =  'DOUBA'")
 
 
 
@@ -352,7 +352,6 @@ if __name__ == '__main__':
     # # # # #
     def handle_item(i):
         task_id = i.get('taskId')
-        logger.success(f"{task_id}处理完成")
         if i.get('comWordsMap'):
             i['comWordsMap'] = json.loads(i.get('comWordsMap'))
         if i.get('brandWords'):

aidso_geo/models/eco_data_process.py

@@ -120,6 +120,7 @@ def save_eco_data_to_bh(data,eco_type,eco_list):
     if platform == 'TXYBA':
         if eco_type == 'jdeco':
             eco_result =yuanbao_android_process_jd_eco(data,eco_list)
+
     bh_utils.insert_data('geo_eco_data',eco_result)
 
 

aidso_geo/models/process.py

@@ -1598,8 +1598,14 @@ def result_v2(response_content, data):
         if sentiment_result:
             result['sentiment'] = sentiment_result
 
-    if 'render_ecom_card_widget_' in response_content:
-        result['hasGoods'] = 1
+    ecom_markers = (
+        "render_ecom_card_widget_taobao_start",
+        "render_ecom_card_widget_damai_start",
+        "render_ecom_card_widget_product_start",
+        "render_ecom_card_widget_jd_start",
+    )
+
+    result["hasGoods"] = int(any(marker in response_content for marker in ecom_markers))
     # ---------------------------------结果整合-----------------------------
 
     # ---------------------------------上传结果-----------------------------
@@ -2003,10 +2009,10 @@ if __name__ == '__main__':
     #
     # data_list = bh_utils.query_data(f"select * from geo_commit_task where reqId in ({req_id_sql}) ")
 
-    data_list = bh_utils.query_data("select * from geo_commit_task where status = 'ING' ")
+    data_list = bh_utils.query_data("select * from geo_commit_task where platform = 'DPA' and thinking_enabled = 1 and insertime > 1782021666 order by  insertime desc")
+    # data_list = bh_utils.query_data("select * from geo_commit_task where pt = '20260720' and platform = 'TYQW' and type = 'success' ")
 
     def handle_item(i):
-
         i["comWordsMap"] = safe_json_loads(i.get("comWordsMap"), [])
         i["brandWords"] = safe_json_loads(i.get("brandWords"), [])
         i["comWords"] = safe_json_loads(i.get("comWords"), [])
@@ -2016,7 +2022,8 @@ if __name__ == '__main__':
         # type_t = 'batch'
         # commit_task(i,'ING')
         # return task_send_queue(i,type_t)
-        return platform_process(i)
+        return deepseek_android_data_process.deepseek_android_process_original_data(i)
+        # return platform_process(i)
     #
     if data_list:
         with ThreadPoolExecutor(max_workers=50) as executor:

aidso_geo/models/qianwen_android_data_process.py

 import json
+import traceback
+
 from aidso_geo.models import spider_save_tos
 from aidso_geo.models.eco_data_process import save_eco_data_to_bh
 from aidso_geo.utils import robot_utils, bh_utils
@@ -372,7 +374,6 @@ def qianwen_android_process_original_data(task_data):
                                                suggestions, rich_media_block)
         return (file_path, search_keyword, url_list, think_content, response_content, suggestions)
     except Exception as e:
-
         parts = file_path.split('/')
         platform = parts[2]
         task_id = parts[1]
@@ -407,7 +408,7 @@ if __name__ == '__main__':
 
 
     # data_list = bh_utils.query_data(f"select * from geo_commit_task  where taskId = '0001e01d-a8ea-405e-acef-93e4f55abbff' and  platform  = 'TYQWA'")
-    data_list = bh_utils.query_data(f"select * from geo_commit_task where taskId = '9dfd84a4-f551-4d74-b1ee-9367aed53a8c' and platform = 'TYQWA'")
+    data_list = bh_utils.query_data(f"select * from geo_commit_task where taskId = 'b347cfe7-0f74-47c4-8328-3c92b9cad1ea' and platform = 'TYQWA'")
 
     def handle_item(i):
 

aidso_geo/models/qianwen_data_process.py

@@ -52,6 +52,7 @@ def qianwen_process_original_data(data):
                                                 try:
                                                     if isinstance(json.loads(pluginResult), list):
                                                         if len(json.loads(pluginResult))>=2:
+
                                                             url_list = json.loads(pluginResult)[1].get(
                                                                 'search_results')
                                                     if isinstance(json.loads(pluginResult), dict):
@@ -85,7 +86,9 @@ def qianwen_process_original_data(data):
                                                         think_content = mu.get('content').get('think_content')
 
                         if json_data.get("data").get('messages'):
+
                             messages  = json_data.get("data").get('messages')
+
                             for ms in messages:
 
                                 if ms.get('mime_type') =='multi_load/iframe' and ms.get('status') == 'complete':
@@ -103,11 +106,29 @@ def qianwen_process_original_data(data):
                                             if mu.get('type') == 'ref_source_inline':
 
                                                 mu_content = mu.get('content')
+
                                                 if mu_content:
                                                     if mu_content.get('query_list'):
                                                         search_keyword.extend(mu_content.get('query_list'))
                                                     if mu_content.get('list'):
                                                         url_list_batch.extend(mu_content.get('list'))
+                                            if mu.get('type') == 'source_group_web':
+                                                mu_content = mu.get('content')
+                                                mu_source_seq = mu.get('source_seq')
+                                                if mu_content.get('list'):
+                                                    mu_content_list = mu_content.get('list')
+                                                    for mu_list in mu_content_list:
+                                                        if mu_list.get('type') == 'source':
+                                                            mu_list_list = (mu_list.get('content').get('list'))
+                                                            for mu_list_l in mu_list_list:
+                                                                ...
+                                                                # 处理引用文章
+                                                                # print(mu_list_l)
+                                                                # print(mu_source_seq)
+                                                                # print(mu_source_seq)
+                                                                # print('----')
+                                                                # print('----')
+
                                             if mu.get('type') == 'video_note_list':
                                                 rich_media_block.append(
                                                     {
@@ -119,6 +140,7 @@ def qianwen_process_original_data(data):
                                 if ms.get('mime_type') =='bar/iframe' and ms.get('status') == 'complete':
                                     if  ms.get('meta_data').get('sources'):
                                         url_list_batch = ms.get('meta_data').get('sources')[0].get('content').get('list')
+                                        # print(ms)
                                 if ms.get('mime_type') =='paa/iframe' and ms.get('status') == 'complete':
                                     paas = ms.get('meta_data').get('paas')
                                     if paas:
@@ -137,11 +159,11 @@ def qianwen_process_original_data(data):
                         "time": url.get('publish_time', ''),
                     }
                 )
-
         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)
     except Exception as e:
+        traceback.print_exc()
         parts = file_path.split('/')
         platform = parts[2]
         task_id = parts[1]
@@ -173,7 +195,7 @@ if __name__ == '__main__':
     # # file_path3 = 'geo/51a7ee04-711c-4cf0-9d4c-4b523fba7037/TYQW/original.text'
     # qianwen_process_original_data(file_path2)
     # for i in task_id:
-    data_list = bh_utils.query_data(f"select * from geo_commit_task  where taskId = '3e317dd8-2512-476c-a31a-104c0bcaca75' and  platform  = 'TYQW'")
+    data_list = bh_utils.query_data(f"select * from geo_commit_task  where taskId = '13e740b9-955c-4f9b-a03e-92e656ffaf43' and  platform  = 'TYQW'")
 
     # # #
     # # #

aidso_geo/models/yuanbao_data_process.py

@@ -79,7 +79,7 @@ def yuanbao_process_original_data(data):
     return (file_path, search_keyword, url_list, think_content, response_content, suggestions)
 
 if __name__ == '__main__':
-    data_list = bh_utils.query_data(f"select * from geo_commit_task  where taskId = '30fde623-70a3-4ca7-86fb-393b60841402' and  platform  = 'TXYB'")
+    data_list = bh_utils.query_data(f"select * from geo_commit_task  where taskId = '9dfd84a4-f551-4d74-b1ee-9367aed53a8c' and  platform  = 'TXYB'")
 
     # # #
     # # #

aidso_geo/utils/ai_interface.py

@@ -328,8 +328,8 @@ def get_yuanbao_message(promp, think, max_retry=3):
                 data=payload.encode("utf-8"),
                 timeout=300
             )
-
             response.raise_for_status()
+
             json_response = response.json()
 
 
@@ -502,5 +502,5 @@ def get_parse_sse_result(platform, task_id):
 
 if __name__ == '__main__':
 
-    print(get_deepseek_message("防晒霜推荐",1))
+    print(get_yuanbao_message("防晒霜推荐",'0'))
     # print(get_doubao_message("防晒霜推荐","disabled"))