优化数据处理逻辑，更新工单数据文件路径，增加多个工单数据集的创建与上传功能，同时调整查询参数以提高检索效果。

rag2_0/dify/AnalysisDifyAnswer.py

@@ -1,8 +1,5 @@
 import pandas as pd
 import json
-
-from regex import search
-
 import sys
 import os
 sys.path.append(os.getcwd())

rag2_0/dify/DifyQueryRetrieval_api.py

@@ -94,7 +94,7 @@ async def retrieve(request: RetrieveRequest):
             request.original_query, 
             query_list, 
             data_set_list,
-            top_k=3
+            top_k=5
         )
         end_time = time.time()
         

rag2_0/dify/WorkorderToDify.py

@@ -5,25 +5,41 @@ sys.path.append(os.getcwd())
 import rag2_0.dify.dify_client.dify_api as DifyApi
 
 import pandas as pd
-pd_data = pd.read_excel("data/excel/2025年5月30日到6月10号对话记录_转工单.xlsx")
+pd_data = pd.read_excel("data/excel/工单汇总（给AI）_2.xlsx")
 
 
 dify_api = DifyApi.DifyApi()
-dataset_id = dify_api.get_or_create_dataset_by_name("工单问答数据")
+peiwang_dataset_id = dify_api.get_or_create_dataset_by_name("配网工单数据")
-document_id = dify_api.upload_text_to_document(text_name="5月30日到6月10号对话工单", text="", dataset_id=dataset_id)
+zhuwang_dataset_id = dify_api.get_or_create_dataset_by_name("主网工单数据")
+jianga_dataset_id = dify_api.get_or_create_dataset_by_name("技改工单数据")
+chuneng_dataset_id = dify_api.get_or_create_dataset_by_name("储能工单数据")
 
-segments_list=[]
+
+soft_segments_list={}
 for index, row in pd_data.iterrows():
     query = row["客户问题"]
     answer = row["解决方案"]
-    if "存在抱怨" in answer:
+    skill_group = row["技能组"]
-        answer = answer.split("存在抱怨")[0]
     
     content = f"问题：{query}\n回答：{answer}"
-    segments_list.append({
+    if skill_group not in soft_segments_list:
+        soft_segments_list[skill_group]=[]
+    soft_segments_list[skill_group].append({
         "content": str(content),
         "answer": "",
         "keywords": []
     })
 
-dify_api.add_document_segments(dataset_id=dataset_id, document_id=document_id, segments_list=segments_list)
+for skill_group, segments_list in soft_segments_list.items():
+    if skill_group == "配网":
+        dataset_id = peiwang_dataset_id
+    elif skill_group == "主网":
+        dataset_id = zhuwang_dataset_id
+    elif skill_group == "技改":
+        dataset_id = jianga_dataset_id
+    elif skill_group == "储能":
+        dataset_id = chuneng_dataset_id
+    document_id = dify_api.get_document_id(dataset_id=dataset_id, document_name=f"{skill_group}工单数据")
+    if not document_id:
+        document_id = dify_api.upload_text_to_document(text_name=f"{skill_group}工单数据", text="", dataset_id=dataset_id)
+    dify_api.add_document_segments(dataset_id=dataset_id, document_id=document_id, segments_list=segments_list)