opensource/dify
3
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-09-24 23:22:26 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix(api): bind scoped lookups to nested owner refs (#38470)

Browse Source
This commit is contained in:
WH-2099
2026-07-15 09:31:09 +00:00
committed by GitHub
parent b721e7a32d
commit 6a511da325
28 changed files with 708 additions and 215 deletions
Download Patch File Download Diff File Expand all files Collapse all files
api/core/app/apps/pipeline/pipeline_runner.py
+44 -19
View File
@@ -27,9 +27,10 @@ from graphon.graph_events import GraphEngineEvent, GraphRunFailedEvent
from graphon.runtime import GraphRuntimeState, VariablePool
from graphon.variable_loader import VariableLoader
from graphon.variables.variables import RAGPipelineVariable, RAGPipelineVariableInput
from models.dataset import Document, Pipeline
from models.dataset import Pipeline
from models.model import EndUser
from models.workflow import Workflow
from services.dataset_ref_service import DatasetRefService, DocumentRef
logger = logging.getLogger(__name__)
@@ -93,9 +94,38 @@ class PipelineRunner(WorkflowBasedAppRunner):
user_id = self.application_generate_entity.user_id
pipeline = session.get(Pipeline, app_config.app_id)
if not pipeline:
if not pipeline or pipeline.tenant_id != app_config.tenant_id:
raise ValueError("Pipeline not found")
dataset = pipeline.retrieve_dataset(session)
if (
not dataset
or dataset.tenant_id != pipeline.tenant_id
or dataset.id != self.application_generate_entity.dataset_id
):
raise ValueError("Pipeline dataset not found")
document_id = self.application_generate_entity.document_id
original_document_id = self.application_generate_entity.original_document_id
dataset_ref = DatasetRefService.create_dataset_ref(dataset)
document_ref = (
DatasetRefService.create_document_ref_from_id(
dataset_ref,
document_id,
)
if document_id
else None
)
if document_ref and DatasetRefService.get_document_by_ref(document_ref, session=session) is None:
raise ValueError("Pipeline document not found")
if original_document_id and original_document_id != document_id:
original_document_ref = DatasetRefService.create_document_ref_from_id(
dataset_ref,
original_document_id,
)
if DatasetRefService.get_document_by_ref(original_document_ref, session=session) is None:
raise ValueError("Pipeline original document not found")
workflow = self.get_workflow(session=session, pipeline=pipeline, workflow_id=app_config.workflow_id)
if not workflow:
raise ValueError("Workflow not initialized")
@@ -206,9 +236,7 @@ class PipelineRunner(WorkflowBasedAppRunner):
generator = workflow_entry.run()
for event in generator:
self._update_document_status(
event, self.application_generate_entity.document_id, self.application_generate_entity.dataset_id
)
self._update_document_status(event, document_ref)
self._handle_event(workflow_entry, event)
def get_workflow(self, session: Session, pipeline: Pipeline, workflow_id: str) -> Workflow | None:
@@ -295,17 +323,14 @@ class PipelineRunner(WorkflowBasedAppRunner):
return graph
def _update_document_status(self, event: GraphEngineEvent, document_id: str | None, dataset_id: str | None) -> None:
"""
Update document status
"""
if isinstance(event, GraphRunFailedEvent):
if document_id and dataset_id:
with create_session() as session, session.begin():
document = session.scalar(
select(Document).where(Document.id == document_id, Document.dataset_id == dataset_id).limit(1)
)
if document:
document.indexing_status = "error"
document.error = event.error or "Unknown error"
session.add(document)
def _update_document_status(self, event: GraphEngineEvent, document_ref: DocumentRef | None) -> None:
"""Set an owner-bound document to error after a failed graph run, if it exists."""
if not isinstance(event, GraphRunFailedEvent) or document_ref is None:
return
with create_session() as session, session.begin():
document = DatasetRefService.get_document_by_ref(document_ref, session=session)
if document:
document.indexing_status = "error"
document.error = event.error or "Unknown error"
session.add(document)
api/core/rag/index_processor/index_processor.py
+37 -11
View File
@@ -65,14 +65,22 @@ class IndexProcessor:
*,
session: Session,
) -> IndexingResultDict:
document = session.scalar(select(Document).where(Document.id == document_id).limit(1))
if not document:
raise KnowledgeIndexNodeError(f"Document {document_id} not found.")
dataset = session.scalar(select(Dataset).where(Dataset.id == dataset_id).limit(1))
if not dataset:
raise KnowledgeIndexNodeError(f"Dataset {dataset_id} not found.")
document = session.scalar(
select(Document)
.where(
Document.id == document_id,
Document.dataset_id == dataset.id,
Document.tenant_id == dataset.tenant_id,
)
.limit(1)
)
if not document:
raise KnowledgeIndexNodeError(f"Document {document_id} not found.")
dataset_name_value = dataset.name
document_name_value = document.name
created_at_value = document.created_at
@@ -83,7 +91,11 @@ class IndexProcessor:
index_processor = IndexProcessorFactory(dataset.chunk_structure).init_index_processor()
if original_document_id:
segments = session.scalars(
select(DocumentSegment).where(DocumentSegment.document_id == original_document_id)
select(DocumentSegment).where(
DocumentSegment.document_id == original_document_id,
DocumentSegment.dataset_id == dataset.id,
DocumentSegment.tenant_id == dataset.tenant_id,
)
).all()
if segments:
index_node_ids = [segment.index_node_id for segment in segments if segment.index_node_id]
@@ -97,7 +109,11 @@ class IndexProcessor:
dataset, index_node_ids, with_keywords=True, delete_child_chunks=True, session=session
)
session.commit()
segment_delete_stmt = delete(DocumentSegment).where(DocumentSegment.document_id == original_document_id)
segment_delete_stmt = delete(DocumentSegment).where(
DocumentSegment.document_id == original_document_id,
DocumentSegment.dataset_id == dataset.id,
DocumentSegment.tenant_id == dataset.tenant_id,
)
session.execute(segment_delete_stmt)
session.commit()
@@ -113,6 +129,7 @@ class IndexProcessor:
select(func.sum(DocumentSegment.word_count)).where(
DocumentSegment.document_id == document_id,
DocumentSegment.dataset_id == dataset_id,
DocumentSegment.tenant_id == dataset.tenant_id,
)
)
) or 0
@@ -125,6 +142,7 @@ class IndexProcessor:
.where(
DocumentSegment.document_id == document_id,
DocumentSegment.dataset_id == dataset_id,
DocumentSegment.tenant_id == dataset.tenant_id,
)
.values(
status="completed",
@@ -156,15 +174,23 @@ class IndexProcessor:
session: Session,
) -> Preview:
doc_language = None
if document_id:
document = session.scalar(select(Document).where(Document.id == document_id).limit(1))
else:
document = None
dataset = session.scalar(select(Dataset).where(Dataset.id == dataset_id).limit(1))
if not dataset:
raise KnowledgeIndexNodeError(f"Dataset {dataset_id} not found.")
if document_id:
document = session.scalar(
select(Document)
.where(
Document.id == document_id,
Document.dataset_id == dataset.id,
Document.tenant_id == dataset.tenant_id,
)
.limit(1)
)
else:
document = None
if summary_index_setting is None:
summary_index_setting = dataset.summary_index_setting