youngkingdom/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-04-27 05:58:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge remote main and resolve conflicts for summaryindex feature

Browse Source 
- Resolved conflicts in 9 task files by adopting session_factory pattern from main
- Preserved all summaryindex functionality including enable/disable logic
- Updated all task files to use session_factory.create_session() instead of db.session
- Merged new features from main (FileService, DocumentBatchDownloadZipPayload, etc.)
This commit is contained in:
FFXN
2026-01-21 16:03:54 +08:00
parent 63d33fe93f ed0e068a47
commit 5d732edbb0
822 changed files with 60654 additions and 10354 deletions
Download Patch File Download Diff File Expand all files Collapse all files

115
api/tasks/remove_document_from_index_task.py
View File

@ -5,8 +5,8 @@ import click
from celery import shared_task
from sqlalchemy import select
from core.db.session_factory import session_factory
from core.rag.index_processor.index_processor_factory import IndexProcessorFactory
from extensions.ext_database import db
from extensions.ext_redis import redis_client
from libs.datetime_utils import naive_utc_now
from models.dataset import Document, DocumentSegment
@ -25,67 +25,70 @@ def remove_document_from_index_task(document_id: str):
logger.info(click.style(f"Start remove document segments from index: {document_id}", fg="green"))
start_at = time.perf_counter()
document = db.session.query(Document).where(Document.id == document_id).first()
if not document:
logger.info(click.style(f"Document not found: {document_id}", fg="red"))
db.session.close()
return
with session_factory.create_session() as session:
document = session.query(Document).where(Document.id == document_id).first()
if not document:
logger.info(click.style(f"Document not found: {document_id}", fg="red"))
return
if document.indexing_status != "completed":
logger.info(click.style(f"Document is not completed, remove is not allowed: {document_id}", fg="red"))
db.session.close()
return
if document.indexing_status != "completed":
logger.info(click.style(f"Document is not completed, remove is not allowed: {document_id}", fg="red"))
return
indexing_cache_key = f"document_{document.id}_indexing"
indexing_cache_key = f"document_{document.id}_indexing"
try:
dataset = document.dataset
try:
dataset = document.dataset
if not dataset:
raise Exception("Document has no dataset")
if not dataset:
raise Exception("Document has no dataset")
index_processor = IndexProcessorFactory(document.doc_form).init_index_processor()
index_processor = IndexProcessorFactory(document.doc_form).init_index_processor()
segments = db.session.scalars(select(DocumentSegment).where(DocumentSegment.document_id == document.id)).all()
segments = session.scalars(select(DocumentSegment).where(DocumentSegment.document_id == document.id)).all()
# Disable summary indexes for all segments in this document
from services.summary_index_service import SummaryIndexService
# Disable summary indexes for all segments in this document
from services.summary_index_service import SummaryIndexService
segment_ids_list = [segment.id for segment in segments]
if segment_ids_list:
try:
SummaryIndexService.disable_summaries_for_segments(
dataset=dataset,
segment_ids=segment_ids_list,
disabled_by=document.disabled_by,
segment_ids_list = [segment.id for segment in segments]
if segment_ids_list:
try:
SummaryIndexService.disable_summaries_for_segments(
dataset=dataset,
segment_ids=segment_ids_list,
disabled_by=document.disabled_by,
)
except Exception as e:
logger.warning("Failed to disable summaries for document %s: %s", document.id, str(e))
index_node_ids = [segment.index_node_id for segment in segments]
if index_node_ids:
try:
index_processor.clean(dataset, index_node_ids, with_keywords=True, delete_child_chunks=False)
except Exception:
logger.exception("clean dataset %s from index failed", dataset.id)
# update segment to disable
session.query(DocumentSegment).where(DocumentSegment.document_id == document.id).update(
{
DocumentSegment.enabled: False,
DocumentSegment.disabled_at: naive_utc_now(),
DocumentSegment.disabled_by: document.disabled_by,
DocumentSegment.updated_at: naive_utc_now(),
}
)
session.commit()
end_at = time.perf_counter()
logger.info(
click.style(
f"Document removed from index: {document.id} latency: {end_at - start_at}",
fg="green",
)
except Exception as e:
logger.warning("Failed to disable summaries for document %s: %s", document.id, str(e))
index_node_ids = [segment.index_node_id for segment in segments]
if index_node_ids:
try:
index_processor.clean(dataset, index_node_ids, with_keywords=True, delete_child_chunks=False)
except Exception:
logger.exception("clean dataset %s from index failed", dataset.id)
# update segment to disable
db.session.query(DocumentSegment).where(DocumentSegment.document_id == document.id).update(
{
DocumentSegment.enabled: False,
DocumentSegment.disabled_at: naive_utc_now(),
DocumentSegment.disabled_by: document.disabled_by,
DocumentSegment.updated_at: naive_utc_now(),
}
)
db.session.commit()
end_at = time.perf_counter()
logger.info(click.style(f"Document removed from index: {document.id} latency: {end_at - start_at}", fg="green"))
except Exception:
logger.exception("remove document from index failed")
if not document.archived:
document.enabled = True
db.session.commit()
finally:
redis_client.delete(indexing_cache_key)
db.session.close()
)
except Exception:
logger.exception("remove document from index failed")
if not document.archived:
document.enabled = True
session.commit()
finally:
redis_client.delete(indexing_cache_key)