zlei6/ragflow
1
0
Fork 0
mirror of https://github.com/infiniflow/ragflow.git synced 2026-08-15 05:04:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: exclude disabled documents from dataset structures (#18041)

Browse Source
This commit is contained in:
buua436
2026-08-10 17:50:34 +08:00
committed by GitHub
parent 23369586c0
commit 6f50e478e0
3 changed files with 14 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
api/apps/services/dataset_api_service.py
View File

@@ -607,6 +607,10 @@ def run_index(dataset_id: str, tenant_id: str, index_type: str):
types=[],
suffix=[],
)
# Disabled documents must not participate in a dataset-level structure
# rebuild. Keep them out of the fan-out task itself; the merger also
# applies a database-backed filter as a defense in depth.
documents = [document for document in documents if str(document.get("status", "1")) != "0"]
if not documents:
return False, f"No documents in Dataset {dataset_id}"
@@ -2053,17 +2057,6 @@ async def get_dataset_structure(dataset_id: str, tenant_id: str, kind: str, keyw
except Exception:
logging.exception("get_dataset_structure: bucket build failed for kb=%s template=%s", dataset_id, tid)
continue
if scope_kwd == "dataset" and not entities and not relations:
try:
entities, relations = await sgc.build_bucket(
index_nm,
dataset_id,
{"compilation_template_ids": [tid], "scope_kwd": ["doc"], "doc_id": sorted(active_doc_ids)},
excluded_doc_ids=disabled_doc_ids,
)
except Exception:
logging.exception("get_dataset_structure: doc fallback bucket build failed for kb=%s template=%s", dataset_id, tid)
continue
if resolved_kind in {"knowledge_graph", "mind_map", "timeline"}:
entities = sgc.filter_entities_with_relations(entities, relations)
if not entities:

2
api/db/services/document_service.py
View File

@@ -45,7 +45,7 @@ class DocumentService(CommonService):
@classmethod
@DB.connection_context()
def get_disabled_doc_ids_by_kb_id(cls, kb_id) -> set[str]:
return {str(doc_id) for doc_id in cls.model.select(cls.model.id).where((cls.model.kb_id == kb_id) & (cls.model.status == "0")).tuples()}
return {str(doc_id) for (doc_id,) in cls.model.select(cls.model.id).where((cls.model.kb_id == kb_id) & (cls.model.status == "0")).tuples()}
@classmethod
def get_cls_model_fields(cls):