Merge branch 'develop' of codeup.aliyun.com:redbearai/python/redbear-mem-open into develop

api/app/controllers/knowledge_controller.py

@@ -1,26 +1,28 @@
-from typing import Optional
 import datetime
 import json
+from typing import Optional
 import uuid
+
 from fastapi import APIRouter, Depends, HTTPException, status, Query
 from sqlalchemy import or_
 from sqlalchemy.orm import Session
 
+from app.celery_app import celery_app
+from app.core.logging_config import get_api_logger
+from app.core.rag.common import settings
+from app.core.rag.llm.chat_model import Base
+from app.core.rag.nlp import rag_tokenizer, search
+from app.core.rag.prompts.generator import graph_entity_types
+from app.core.rag.vdb.elasticsearch.elasticsearch_vector import ElasticSearchVectorFactory
+from app.core.response_utils import success
 from app.db import get_db
 from app.dependencies import get_current_user
 from app.models.user_model import User
 from app.models import knowledge_model, document_model, file_model
 from app.schemas import knowledge_schema
 from app.schemas.response_schema import ApiResponse
-from app.core.response_utils import success
 from app.services import knowledge_service, document_service
-from app.core.rag.llm.chat_model import Base
-from app.core.rag.prompts.generator import graph_entity_types
-from app.core.rag.vdb.elasticsearch.elasticsearch_vector import ElasticSearchVectorFactory
-from app.core.logging_config import get_api_logger
-from app.core.rag.nlp import rag_tokenizer, search
-from app.core.rag.common import settings
-from app.celery_app import celery_app
+from app.services.model_service import ModelConfigService
 
 # Obtain a dedicated API logger
 api_logger = get_api_logger()
@@ -47,6 +49,45 @@ def get_parser_types():
     return success(msg="Successfully obtained the knowledge parser type", data=list(knowledge_model.ParserType))
 
 
+@router.get("/knowledge_graph_entity_types", response_model=ApiResponse)
+async def get_knowledge_graph_entity_types(
+        llm_id: uuid.UUID,
+        scenario: str,
+        db: Session = Depends(get_db),
+        current_user: User = Depends(get_current_user)
+):
+    """
+    get knowledge graph entity types based on llm_id
+    """
+    api_logger.info(f"Obtain details of the knowledge graph: llm_id={llm_id}, username: {current_user.username}")
+
+    try:
+        # 1. Check whether the model exists
+        api_logger.debug(f"Check whether the model exists: {llm_id}")
+        config = ModelConfigService.get_model_by_id(db=db, model_id=llm_id)
+
+        if not config:
+            api_logger.warning(
+                f"The model does not exist or you do not have permission to access it: llm_id={llm_id}")
+            raise HTTPException(
+                status_code=status.HTTP_404_NOT_FOUND,
+                detail="The model does not exist or you do not have permission to access it"
+            )
+        # 2. Prepare to configure chat_mdl information
+        chat_model = Base(
+            key=config.api_keys[0].api_key,
+            model_name=config.api_keys[0].model_name,
+            base_url=config.api_keys[0].api_base
+        )
+        response = graph_entity_types(chat_model, scenario)
+        return success(data=response, msg="Successfully obtained knowledge graph entity types")
+    except HTTPException:
+        raise
+    except Exception as e:
+        api_logger.error(f"get knowledge graph entity types failed: llm_id={llm_id} - {str(e)}")
+        raise
+
+
 @router.get("/knowledges", response_model=ApiResponse)
 async def get_knowledges(
         parent_id: Optional[uuid.UUID] = Query(None, description="parent folder id"),
@@ -379,7 +420,7 @@ async def delete_knowledge_graph(
         current_user: User = Depends(get_current_user)
 ):
     """
-    Soft-delete knowledge graph
+    delete knowledge graph
     """
     api_logger.info(f"Request to delete knowledge graph: knowledge_id={knowledge_id}, username: {current_user.username}")
 
@@ -442,42 +483,3 @@ async def rebuild_knowledge_graph(
     except Exception as e:
         api_logger.error(f"Failed to rebuild knowledge graph: knowledge_id={knowledge_id} - {str(e)}")
         raise
-
-
-@router.get("/{knowledge_id}/knowledge_graph_entity_types", response_model=ApiResponse)
-async def get_knowledge_graph_entity_types(
-        knowledge_id: uuid.UUID,
-        scenario: str,
-        db: Session = Depends(get_db),
-        current_user: User = Depends(get_current_user)
-):
-    """
-    get knowledge graph entity types based on knowledge_id
-    """
-    api_logger.info(f"Obtain details of the knowledge graph: knowledge_id={knowledge_id}, username: {current_user.username}")
-
-    try:
-        # 1. Check whether the knowledge base exists
-        api_logger.debug(f"Check whether the knowledge base exists: {knowledge_id}")
-        db_knowledge = knowledge_service.get_knowledge_by_id(db, knowledge_id=knowledge_id, current_user=current_user)
-
-        if not db_knowledge:
-            api_logger.warning(
-                f"The knowledge base does not exist or you do not have permission to access it: knowledge_id={knowledge_id}")
-            raise HTTPException(
-                status_code=status.HTTP_404_NOT_FOUND,
-                detail="The knowledge base does not exist or you do not have permission to access it"
-            )
-        # 2. Prepare to configure chat_mdl information
-        chat_model = Base(
-            key=db_knowledge.llm.api_keys[0].api_key,
-            model_name=db_knowledge.llm.api_keys[0].model_name,
-            base_url=db_knowledge.llm.api_keys[0].api_base
-        )
-        response = graph_entity_types(chat_model, scenario)
-        return success(data=response, msg="Successfully obtained knowledge graph entity types")
-    except HTTPException:
-        raise
-    except Exception as e:
-        api_logger.error(f"get knowledge graph entity types failed: knowledge_id={knowledge_id} - {str(e)}")
-        raise

api/app/core/rag/deepdoc/parser/excel_parser.py

@@ -48,7 +48,6 @@ class RAGExcelParser:
                     logging.info(f"pandas with default engine load error: {ex}, try calamine instead")
                     file_like_object.seek(0)
                     df = pd.read_excel(file_like_object, engine="calamine")
-                    print("lxc1")
                     return RAGExcelParser._dataframe_to_workbook(df)
             except Exception as e_pandas:
                 raise Exception(f"pandas.read_excel error: {e_pandas}, original openpyxl error: {e}")
@@ -215,19 +214,35 @@ class RAGExcelParser:
                 continue
             if not rows:
                 continue
+            # 获取表头
             ti = list(rows[0])
-            for r in list(rows[1:]):
-                fields = []
-                for i, c in enumerate(r):
-                    if not c.value:
-                        continue
-                    t = str(ti[i].value) if i < len(ti) else ""
-                    t += ("：" if t else "") + str(c.value)
-                    fields.append(t)
-                line = "; ".join(fields)
-                if sheetname.lower().find("sheet") < 0:
-                    line += " ——" + sheetname
-                res.append(line)
+            header_fields = []
+            for cell in ti:
+                if cell.value:  # 只添加有值的表头
+                    header_fields.append(str(cell.value))
+
+            # 如果有数据行，处理数据行；否则只处理表头
+            data_rows = rows[1:]
+            if data_rows:
+                for r in data_rows:
+                    fields = []
+                    for i, c in enumerate(r):
+                        if not c.value:
+                            continue
+                        t = str(ti[i].value) if i < len(ti) else ""
+                        t += ("：" if t else "") + str(c.value)
+                        fields.append(t)
+                    line = "; ".join(fields)
+                    if sheetname.lower().find("sheet") < 0:
+                        line += " ——" + sheetname
+                    res.append(line)
+            else:
+                # 只有表头的情况
+                if header_fields:
+                    line = "; ".join(header_fields)
+                    if sheetname.lower().find("sheet") < 0:
+                        line += " ——" + sheetname
+                    res.append(line)
         return res
 
     @staticmethod

api/app/core/workflow/nodes/end/node.py

@@ -61,7 +61,7 @@ class EndNode(BaseNode):
             引用的节点 ID 列表
         """
         # 匹配 {{node_id.xxx}} 格式
-        pattern = r'\{\{([a-zA-Z0-9_]+)\.[a-zA-Z0-9_]+\}\}'
+        pattern = r'\{\{([a-zA-Z0-9_-]+)\.[a-zA-Z0-9_]+\}\}'
         matches = re.findall(pattern, template)
         return list(set(matches))  # 去重
 

api/app/models/data_config_model.py

@@ -51,8 +51,8 @@ class DataConfig(Base):
     # 自我反思配置
     enable_self_reflexion = Column(Boolean, default=False, comment="是否启用自我反思")
     iteration_period = Column(String, default="3", comment="反思迭代周期")
-    reflexion_range = Column(String, default="retrieval", comment="反思范围：部分/全部")
-    baseline = Column(String, default="time", comment="基线：时间/事实/时间和事实")
+    reflexion_range = Column(String, default="partial", comment="反思范围：部分/全部")
+    baseline = Column(String, default="TIME", comment="基线：时间/事实/时间和事实")
     reflection_model_id = Column(String, nullable=True, comment="反思模型ID")
     memory_verify = Column(Boolean, default=True, comment="记忆验证")
     quality_assessment = Column(Boolean, default=True, comment="质量评估")

api/app/templates/workflows/simple_qa/template.yml

@@ -41,8 +41,6 @@ nodes:
             - 使用友好、礼貌的语气
             - 适当使用格式化（如列表、段落）提高可读性
         
-        - role: user
-          content: "{{sys.message}}"
       
       model_id: null
       temperature: 0.7