xinyin025/LangBot
1
0
Fork 0
mirror of https://github.com/langbot-app/LangBot.git synced 2026-07-22 12:26:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: file be deleted twice

Browse Source
This commit is contained in:
Junyan Qin
2025-07-12 17:47:53 +08:00
parent 1e85d02ae4
commit 1ef0193028
1 changed files with 22 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files
pkg/rag/knowledge/mgr.py
+22 -14
View File
@@ -21,8 +21,12 @@ class RAGManager:
self.parser = FileParser() self.parser = FileParser()
self.chunker = Chunker() self.chunker = Chunker()
# Initialize Embedder with targeted model type and name # Initialize Embedder with targeted model type and name
self.embedder = Embedder(model_type='third_party_api', model_name_key='bge-m3', chroma_manager=self.chroma_manager) self.embedder = Embedder(
self.retriever = Retriever(model_type='third_party_api', model_name_key='bge-m3', chroma_manager=self.chroma_manager) model_type='third_party_api', model_name_key='bge-m3', chroma_manager=self.chroma_manager
)
self.retriever = Retriever(
model_type='third_party_api', model_name_key='bge-m3', chroma_manager=self.chroma_manager
)
async def initialize_rag_system(self): async def initialize_rag_system(self):
"""Initializes the RAG system by creating database tables.""" """Initializes the RAG system by creating database tables."""
@@ -143,9 +147,7 @@ class RAGManager:
self.ap.logger.error(f'Error retrieving all files: {str(e)}', exc_info=True) self.ap.logger.error(f'Error retrieving all files: {str(e)}', exc_info=True)
return [] return []
async def store_data( async def store_data(self, file_path: str, kb_id: str, file_type: str, file_id: str = None):
self, file_path: str, kb_id: str, file_type: str, file_id: str = None
):
""" """
Parses, chunks, embeds, and stores data from a given file into the RAG system. Parses, chunks, embeds, and stores data from a given file into the RAG system.
Associates the file with a knowledge base using kb_id in the File table. Associates the file with a knowledge base using kb_id in the File table.
@@ -165,9 +167,7 @@ class RAGManager:
file_name = os.path.basename(file_path) file_name = os.path.basename(file_path)
text = await self.parser.parse(file_path) text = await self.parser.parse(file_path)
if not text: if not text:
self.ap.logger.warning( self.ap.logger.warning(f'No text extracted from file {file_path}. ')
f'No text extracted from file {file_path}. '
)
return return
chunks_texts = await self.chunker.chunk(text) chunks_texts = await self.chunker.chunk(text)
@@ -222,8 +222,9 @@ class RAGManager:
try: try:
await self.ap.storage_mgr.storage_provider.delete(file_id) await self.ap.storage_mgr.storage_provider.delete(file_id)
except Exception as e: except Exception as e:
self.ap.logger.error(f'Error deleting file from storage for file_id {file_id}: {str(e)}', exc_info=True) self.ap.logger.error(
await self.ap.storage_mgr.storage_provider.delete(file_id) f'Error deleting file from storage for file_id {file_id}: {str(e)}', exc_info=True
)
self.ap.logger.info(f'Deleted file record for file_id: {file_id}') self.ap.logger.info(f'Deleted file record for file_id: {file_id}')
else: else:
self.ap.logger.warning( self.ap.logger.warning(
@@ -326,7 +327,14 @@ class RAGManager:
return return
self.ap.logger.info(f'File with ID {file_id} exists, proceeding with association.') self.ap.logger.info(f'File with ID {file_id} exists, proceeding with association.')
# add new file record # add new file record
file_to_update = File(id=file_id, kb_id=kb.id, file_name=file_id, path=os.path.join('data', 'storage', file_id), file_type=os.path.splitext(file_id)[1].lstrip('.'), status=0) file_to_update = File(
id=file_id,
kb_id=kb.id,
file_name=file_id,
path=os.path.join('data', 'storage', file_id),
file_type=os.path.splitext(file_id)[1].lstrip('.'),
status=0,
)
session.add(file_to_update) session.add(file_to_update)
session.commit() session.commit()
self.ap.logger.info( self.ap.logger.info(
@@ -342,12 +350,12 @@ class RAGManager:
# 进行文件解析 # 进行文件解析
try: try:
await self.store_data( await self.store_data(
file_path = os.path.join('data', 'storage', file_id), file_path=os.path.join('data', 'storage', file_id),
kb_id=knowledge_base_uuid, kb_id=knowledge_base_uuid,
file_type=os.path.splitext(file_id)[1].lstrip('.'), file_type=os.path.splitext(file_id)[1].lstrip('.'),
file_id=file_id file_id=file_id,
) )
except Exception as store_e: except Exception:
# 如果存储数据时出错，更新文件状态为失败 # 如果存储数据时出错，更新文件状态为失败
file_obj = session.query(File).filter_by(id=file_id).first() file_obj = session.query(File).filter_by(id=file_id).first()
if file_obj: if file_obj: