gmakstutis/LightRAG
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

feat: add warning logs for deleting non-completed documents

Browse source
This commit is contained in:
yangdx 2025-08-05 12:21:08 +08:00
parent 793e82ae89
commit 01bce8c26e
1 changed files with 20 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

20
lightrag/lightrag.py
View file

@ -2087,6 +2087,26 @@ class LightRAG:
file_path="", file_path="",
) )
# Check document status and log warning for non-completed documents
doc_status = doc_status_data.get("status")
if doc_status != DocStatus.PROCESSED:
if doc_status == DocStatus.PENDING:
warning_msg = f"WARNING: Deleting PENDING document {doc_id} ('{file_path}') - document was never processed"
elif doc_status == DocStatus.PROCESSING:
warning_msg = f"WARNING: Deleting PROCESSING document {doc_id} ('{file_path}') - legacy processing state detected"
elif doc_status == DocStatus.FAILED:
error_msg = doc_status_data.get("error_msg", "Unknown error")
warning_msg = f"WARNING: Deleting FAILED document {doc_id} ('{file_path}') - processing failed: {error_msg}"
else:
warning_msg = f"WARNING: Deleting document {doc_id} ('{file_path}') with unexpected status: {doc_status}"
logger.warning(warning_msg)
# Update pipeline status for monitoring
async with pipeline_status_lock:
pipeline_status["latest_message"] = warning_msg
pipeline_status["history_messages"].append(warning_msg)
# 2. Get chunk IDs from document status # 2. Get chunk IDs from document status
chunk_ids = set(doc_status_data.get("chunks_list", [])) chunk_ids = set(doc_status_data.get("chunks_list", []))