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

Move relationship ID sorting to before vector DB operations

Browse source 
• Remove verbose entity rebuild logging
• Sort IDs before vector DB updates
• Keep graph storage with original order
This commit is contained in:
yangdx 2025-10-28 19:13:48 +08:00
parent c81a56a113
commit 29c4a91dc3
1 changed files with 7 additions and 16 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

23
lightrag/operate.py
View file

@ -677,14 +677,6 @@ async def rebuild_knowledge_from_chunks(
entity_chunks_storage=entity_chunks_storage, entity_chunks_storage=entity_chunks_storage,
) )
rebuilt_entities_count += 1 rebuilt_entities_count += 1
status_message = (
f"Rebuild `{entity_name}` from {len(chunk_ids)} chunks"
)
logger.info(status_message)
if pipeline_status is not None and pipeline_status_lock is not None:
async with pipeline_status_lock:
pipeline_status["latest_message"] = status_message
pipeline_status["history_messages"].append(status_message)
except Exception as e: except Exception as e:
failed_entities_count += 1 failed_entities_count += 1
status_message = f"Failed to rebuild `{entity_name}`: {e}" status_message = f"Failed to rebuild `{entity_name}`: {e}"
@ -1432,10 +1424,6 @@ async def _rebuild_single_relationship(
else: else:
truncation_info = "" truncation_info = ""
# Sort src and tgt to ensure consistent ordering (smaller string first)
if src > tgt:
src, tgt = tgt, src
# Update relationship in graph storage # Update relationship in graph storage
updated_relationship_data = { updated_relationship_data = {
**current_relationship, **current_relationship,
@ -1510,6 +1498,9 @@ async def _rebuild_single_relationship(
await knowledge_graph_inst.upsert_edge(src, tgt, updated_relationship_data) await knowledge_graph_inst.upsert_edge(src, tgt, updated_relationship_data)
# Update relationship in vector database # Update relationship in vector database
# Sort src and tgt to ensure consistent ordering (smaller string first)
if src > tgt:
src, tgt = tgt, src
try: try:
rel_vdb_id = compute_mdhash_id(src + tgt, prefix="rel-") rel_vdb_id = compute_mdhash_id(src + tgt, prefix="rel-")
rel_vdb_id_reverse = compute_mdhash_id(tgt + src, prefix="rel-") rel_vdb_id_reverse = compute_mdhash_id(tgt + src, prefix="rel-")
@ -2145,10 +2136,6 @@ async def _merge_edges_then_upsert(
else: else:
logger.debug(status_message) logger.debug(status_message)
# Sort src_id and tgt_id to ensure consistent ordering (smaller string first)
if src_id > tgt_id:
src_id, tgt_id = tgt_id, src_id
# 11. Update both graph and vector db # 11. Update both graph and vector db
for need_insert_id in [src_id, tgt_id]: for need_insert_id in [src_id, tgt_id]:
if not (await knowledge_graph_inst.has_node(need_insert_id)): if not (await knowledge_graph_inst.has_node(need_insert_id)):
@ -2236,6 +2223,10 @@ async def _merge_edges_then_upsert(
weight=weight, weight=weight,
) )
# Sort src_id and tgt_id to ensure consistent ordering (smaller string first)
if src_id > tgt_id:
src_id, tgt_id = tgt_id, src_id
if relationships_vdb is not None: if relationships_vdb is not None:
rel_vdb_id = compute_mdhash_id(src_id + tgt_id, prefix="rel-") rel_vdb_id = compute_mdhash_id(src_id + tgt_id, prefix="rel-")
rel_vdb_id_reverse = compute_mdhash_id(tgt_id + src_id, prefix="rel-") rel_vdb_id_reverse = compute_mdhash_id(tgt_id + src_id, prefix="rel-")