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

Fix regex to match multiline content in extraction parsing

Browse source 
• Remove non-greedy quantifier
• Add DOTALL flag for multiline matching
• Apply to both parsing functions
• Enable cross-line content extraction
This commit is contained in:
yangdx 2025-09-01 10:35:06 +08:00
parent 7baeb186c6
commit 8bbf307aeb
1 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
lightrag/operate.py
View file

@ -835,7 +835,7 @@ async def _parse_extraction_result(
[context_base["record_delimiter"], context_base["completion_delimiter"]], [context_base["record_delimiter"], context_base["completion_delimiter"]],
) )
for record in records: for record in records:
record = re.search(r"\((.*?)\)", record) record = re.search(r"\((.*)\)", record, re.DOTALL)
if record is None: if record is None:
continue continue
record = record.group(1) record = record.group(1)
@ -1749,7 +1749,7 @@ async def extract_entities(
) )
for record in records: for record in records:
record = re.search(r"\((.*?)\)", record) record = re.search(r"\((.*)\)", record, re.DOTALL)
if record is None: if record is None:
continue continue
record = record.group(1) record = record.group(1)