zlei6/ragflow
1
0
Fork 0
mirror of https://github.com/infiniflow/ragflow.git synced 2026-08-12 20:03:40 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: optimize dataflow indexing and logs (#17737)

Browse Source
This commit is contained in:
buua436
2026-08-03 19:15:44 +08:00
committed by GitHub
parent e997fd655a
commit 1d141aff18
9 changed files with 103 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
rag/utils/es_conn.py
View File

@@ -334,7 +334,7 @@ class ESConnection(ESConnectionBase):
self.logger.error(f"ESConnection.search timeout for {ATTEMPT_TIME} times!")
raise Exception("ESConnection.search timeout.")
def insert(self, documents: list[dict], index_name: str, knowledgebase_id: str = None) -> list[str]:
def insert(self, documents: list[dict], index_name: str, knowledgebase_id: str = None, refresh: str | bool = "wait_for") -> list[str]:
# Refers to https://www.elastic.co/guide/en/elasticsearch/reference/current/docs-bulk.html
operations = []
for d in documents:
@@ -351,7 +351,7 @@ class ESConnection(ESConnectionBase):
for _ in range(ATTEMPT_TIME):
try:
res = []
r = self.es.bulk(index=index_name, operations=operations, refresh="wait_for", timeout="60s")
r = self.es.bulk(index=index_name, operations=operations, refresh=refresh, timeout="60s")
if re.search(r"False", str(r["errors"]), re.IGNORECASE):
return res