refactor:knowledge api refactor

1.knowledge python file refactor
2.source embedding return vector_ids

assets/schema/knowledge_management.sql

@@ -8,7 +8,7 @@ CREATE TABLE `knowledge_space` (
   `gmt_modified` TIMESTAMP DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP COMMENT 'update time',
   PRIMARY KEY (`id`),
   KEY `idx_name` (`name`) COMMENT 'index:idx_name'
-) ENGINE=InnoDB AUTO_INCREMENT=100001 DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci COMMENT='knowledge space table';
+) ENGINE=InnoDB AUTO_INCREMENT=100001 DEFAULT CHARSET=utf8mb4 COMMENT='knowledge space table';
 
 CREATE TABLE `knowledge_document` (
   `id` int NOT NULL AUTO_INCREMENT COMMENT 'auto increment id',
@@ -25,7 +25,7 @@ CREATE TABLE `knowledge_document` (
   `gmt_modified` TIMESTAMP DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP COMMENT 'update time',
   PRIMARY KEY (`id`),
   KEY `idx_doc_name` (`doc_name`) COMMENT 'index:idx_doc_name'
-) ENGINE=InnoDB AUTO_INCREMENT=100001 DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci COMMENT='knowledge document table';
+) ENGINE=InnoDB AUTO_INCREMENT=100001 DEFAULT CHARSET=utf8mb4 COMMENT='knowledge document table';
 
 CREATE TABLE `document_chunk` (
   `id` int NOT NULL AUTO_INCREMENT COMMENT 'auto increment id',
@@ -38,4 +38,4 @@ CREATE TABLE `document_chunk` (
   `gmt_modified` timestamp NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP COMMENT 'update time',
   PRIMARY KEY (`id`),
   KEY `idx_document_id` (`document_id`) COMMENT 'index:document_id'
-) ENGINE=InnoDB AUTO_INCREMENT=100001 DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci COMMENT='knowledge document chunk detail'
+) ENGINE=InnoDB AUTO_INCREMENT=100001 DEFAULT CHARSET=utf8mb4 COMMENT='knowledge document chunk detail'

pilot/embedding_engine/source_embedding.py

@@ -59,7 +59,7 @@ class SourceEmbedding(ABC):
         self.vector_client = VectorStoreConnector(
             CFG.VECTOR_STORE_TYPE, self.vector_store_config
         )
-        self.vector_client.load_document(docs)
+        return self.vector_client.load_document(docs)
 
     @register
     def similar_search(self, doc, topk):

pilot/server/knowledge/service.py

@@ -5,15 +5,15 @@ from pilot.configs.config import Config
 from pilot.configs.model_config import LLM_MODEL_CONFIG
 from pilot.embedding_engine.knowledge_embedding import KnowledgeEmbedding
 from pilot.logs import logger
-from pilot.server.knowledge.chunk_dao import (
+from pilot.server.knowledge.chunk_db import (
     DocumentChunkEntity,
     DocumentChunkDao,
 )
-from pilot.server.knowledge.document_dao import (
+from pilot.server.knowledge.document_db import (
     KnowledgeDocumentDao,
     KnowledgeDocumentEntity,
 )
-from pilot.server.knowledge.space_dao import (
+from pilot.server.knowledge.space_db import (
     KnowledgeSpaceDao,
     KnowledgeSpaceEntity,
 )
@@ -195,7 +195,8 @@ class KnowledgeService:
             vector_ids = client.knowledge_embedding_batch(chunk_docs)
             doc.status = SyncStatus.FINISHED.name
             doc.result = "document embedding success"
-            doc.vector_ids = ",".join(vector_ids)
+            if vector_ids is not None:
+                doc.vector_ids = ",".join(vector_ids)
             logger.info(f"async document embedding, success:{doc.doc_name}")
         except Exception as e:
             doc.status = SyncStatus.FAILED.name