zlei6/ragflow
1
0
Fork 0
mirror of https://github.com/infiniflow/ragflow.git synced 2026-06-29 23:41:12 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update chunk/metadata cli (#15055)

Browse Source 
### What problem does this PR solve?

Update chunk/metadata cli

### Type of change

- [ ] Refactoring
This commit is contained in:
qinling0210
2026-05-20 20:32:06 +08:00
committed by GitHub
parent 90c76e73d0
commit dbef3e361f
17 changed files with 602 additions and 147 deletions
Download Patch File Download Diff File Expand all files Collapse all files

38
rag/utils/infinity_conn.py
View File

@@ -294,10 +294,13 @@ class InfinityConnection(InfinityConnectionBase):
df_list = list()
assert isinstance(knowledgebase_ids, list)
table_list = list()
if index_name.startswith("ragflow_doc_meta_"):
table_names_to_search = [index_name]
else:
table_names_to_search = [f"{index_name}_{kb_id}" for kb_id in knowledgebase_ids]
if not knowledgebase_ids:
self.logger.warning("INFINITY get called with empty knowledgebase_ids for index %s", index_name)
return None
table_names_to_search = [f"{index_name}_{kb_id}" for kb_id in knowledgebase_ids if kb_id]
if not table_names_to_search:
self.logger.warning("INFINITY get has only blank knowledgebase_ids for index %s", index_name)
return None
for table_name in table_names_to_search:
table_list.append(table_name)
try:
@@ -455,7 +458,8 @@ class InfinityConnection(InfinityConnectionBase):
d[k] = v if v else "{}"
else:
d[k] = v
for k in ["docnm_kwd", "title_tks", "title_sm_tks", "important_kwd", "important_tks", "content_with_weight",
for k in ["docnm_kwd", "title_tks", "title_sm_tks", "important_kwd", "important_tks",
"content_with_weight",
"content_ltks", "content_sm_ltks", "authors_tks", "authors_sm_tks", "question_kwd",
"question_tks"]:
if k in d:
@@ -575,7 +579,8 @@ class InfinityConnection(InfinityConnectionBase):
else:
new_value[k] = v
for k in ["docnm_kwd", "title_tks", "title_sm_tks", "important_kwd", "important_tks", "content_with_weight",
"content_ltks", "content_sm_ltks", "authors_tks", "authors_sm_tks", "question_kwd", "question_tks"]:
"content_ltks", "content_sm_ltks", "authors_tks", "authors_sm_tks", "question_kwd",
"question_tks"]:
if k in new_value:
del new_value[k]
@@ -583,7 +588,8 @@ class InfinityConnection(InfinityConnectionBase):
if removeValue:
col_to_remove = list(removeValue.keys())
row_to_opt = table_instance.output(col_to_remove + ["id"]).filter(filter).to_df()
self.logger.debug(f"INFINITY search table {str(table_name)}, filter {filter}, result: {str(row_to_opt[0])}")
self.logger.debug(
f"INFINITY search table {str(table_name)}, filter {filter}, result: {str(row_to_opt[0])}")
row_to_opt = self.get_fields(row_to_opt, col_to_remove)
for id, old_v in row_to_opt.items():
for k, remove_v in removeValue.items():
@@ -608,15 +614,15 @@ class InfinityConnection(InfinityConnectionBase):
return True
def adjust_chunk_pagerank_fea(
self,
chunk_id: str,
index_name: str,
knowledgebase_id: str,
delta: int,
min_weight: int,
max_weight: int,
row_id: int | None = None,
max_retries: int = 2,
self,
chunk_id: str,
index_name: str,
knowledgebase_id: str,
delta: int,
min_weight: int,
max_weight: int,
row_id: int | None = None,
max_retries: int = 2,
) -> bool:
"""Adjust pagerank_fea on one chunk row in Infinity.