gardel/lightrag
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Deduplicate merged relation keywords

Browse Source
This commit is contained in:
yangdx
2025-05-08 15:52:18 +08:00
parent b92f9b9453
commit de40f1b5b3
1 changed files with 15 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
lightrag/operate.py
View File

@@ -388,14 +388,22 @@ async def _merge_edges_then_upsert(
) )
) )
) )
keywords = GRAPH_FIELD_SEP.join(
sorted( # Split all existing and new keywords into individual terms, then combine and deduplicate
set( all_keywords = set()
[dp["keywords"] for dp in edges_data if dp.get("keywords")] # Process already_keywords (which are comma-separated)
+ already_keywords for keyword_str in already_keywords:
if keyword_str: # Skip empty strings
all_keywords.update(k.strip() for k in keyword_str.split(",") if k.strip())
# Process new keywords from edges_data
for edge in edges_data:
if edge.get("keywords"):
all_keywords.update(
k.strip() for k in edge["keywords"].split(",") if k.strip()
) )
) # Join all unique keywords with commas
) keywords = ",".join(sorted(all_keywords))
source_id = GRAPH_FIELD_SEP.join( source_id = GRAPH_FIELD_SEP.join(
set( set(
[dp["source_id"] for dp in edges_data if dp.get("source_id")] [dp["source_id"] for dp in edges_data if dp.get("source_id")]