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

Merge branch 'context-builder'

Browse source
This commit is contained in:
yangdx 2025-07-23 16:14:44 +08:00
commit 00d7bc80bf
1 changed files with 73 additions and 64 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

17
lightrag/operate.py
View file

@ -1329,6 +1329,16 @@ async def merge_nodes_and_edges(
for edge_key, edges in all_edges.items(): for edge_key, edges in all_edges.items():
tasks.append(asyncio.create_task(_locked_process_edges(edge_key, edges))) tasks.append(asyncio.create_task(_locked_process_edges(edge_key, edges)))
# Check if there are any tasks to process
if not tasks:
log_message = f"No entities or relationships to process for {file_path}"
logger.info(log_message)
if pipeline_status_lock is not None:
async with pipeline_status_lock:
pipeline_status["latest_message"] = log_message
pipeline_status["history_messages"].append(log_message)
return
# Execute all tasks in parallel with semaphore control and early failure detection # Execute all tasks in parallel with semaphore control and early failure detection
done, pending = await asyncio.wait(tasks, return_when=asyncio.FIRST_EXCEPTION) done, pending = await asyncio.wait(tasks, return_when=asyncio.FIRST_EXCEPTION)
@ -2020,7 +2030,6 @@ async def _build_query_context(
# Unified token control system - Apply precise token limits to entities and relations # Unified token control system - Apply precise token limits to entities and relations
tokenizer = text_chunks_db.global_config.get("tokenizer") tokenizer = text_chunks_db.global_config.get("tokenizer")
if tokenizer:
# Get new token limits from query_param (with fallback to global_config) # Get new token limits from query_param (with fallback to global_config)
max_entity_tokens = getattr( max_entity_tokens = getattr(
query_param, query_param,
@ -2145,9 +2154,9 @@ async def _build_query_context(
if chunks: if chunks:
all_chunks.extend(chunks) all_chunks.extend(chunks)
# Apply token processing to chunks if tokenizer is available # Apply token processing to chunks
text_units_context = [] text_units_context = []
if tokenizer and all_chunks: if all_chunks:
# Calculate dynamic token limit for text chunks # Calculate dynamic token limit for text chunks
entities_str = json.dumps(entities_context, ensure_ascii=False) entities_str = json.dumps(entities_context, ensure_ascii=False)
relations_str = json.dumps(relations_context, ensure_ascii=False) relations_str = json.dumps(relations_context, ensure_ascii=False)
@ -2600,7 +2609,7 @@ async def _get_edge_data(
combined = { combined = {
"src_id": k["src_id"], "src_id": k["src_id"],
"tgt_id": k["tgt_id"], "tgt_id": k["tgt_id"],
"rank": edge_degrees_dict.get(pair, k.get("rank", 0)), "rank": edge_degrees_dict.get(pair, 0),
"created_at": k.get("created_at", None), "created_at": k.get("created_at", None),
**edge_props, **edge_props,
} }