Files
dify/api/tasks/remove_document_from_index_task.py
T
+40 5bbf685035 feat: fix i18n missing keys and merge upstream/main (#24615)
Signed-off-by: -LAN- <[email protected]>
Signed-off-by: kenwoodjw <[email protected]>
Signed-off-by: Yongtao Huang <[email protected]>
Signed-off-by: yihong0618 <[email protected]>
Signed-off-by: zhanluxianshen <[email protected]>
Co-authored-by: -LAN- <[email protected]>
Co-authored-by: GuanMu <[email protected]>
Co-authored-by: Davide Delbianco <[email protected]>
Co-authored-by: NeatGuyCoding <[email protected]>
Co-authored-by: kenwoodjw <[email protected]>
Co-authored-by: Yongtao Huang <[email protected]>
Co-authored-by: Yongtao Huang <[email protected]>
Co-authored-by: Qiang Lee <[email protected]>
Co-authored-by: 李强04 <[email protected]>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Asuka Minato <[email protected]>
Co-authored-by: Matri Qi <[email protected]>
Co-authored-by: huayaoyue6 <[email protected]>
Co-authored-by: Bowen Liang <[email protected]>
Co-authored-by: znn <[email protected]>
Co-authored-by: crazywoola <[email protected]>
Co-authored-by: crazywoola <[email protected]>
Co-authored-by: Copilot <[email protected]>
Co-authored-by: yihong <[email protected]>
Co-authored-by: Muke Wang <[email protected]>
Co-authored-by: wangmuke <[email protected]>
Co-authored-by: Wu Tianwei <[email protected]>
Co-authored-by: quicksand <[email protected]>
Co-authored-by: 非法操作 <[email protected]>
Co-authored-by: zxhlyh <[email protected]>
Co-authored-by: Eric Guo <[email protected]>
Co-authored-by: Zhedong Cen <[email protected]>
Co-authored-by: jiangbo721 <[email protected]>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: hjlarry <[email protected]>
Co-authored-by: lxsummer <[email protected]>
Co-authored-by: 湛露先生 <[email protected]>
Co-authored-by: Guangdong Liu <[email protected]>
Co-authored-by: QuantumGhost <[email protected]>
Co-authored-by: Claude <[email protected]>
Co-authored-by: Yessenia-d <[email protected]>
Co-authored-by: huangzhuo1949 <[email protected]>
Co-authored-by: huangzhuo <[email protected]>
Co-authored-by: 17hz <[email protected]>
Co-authored-by: Amy <[email protected]>
Co-authored-by: Joel <[email protected]>
Co-authored-by: Nite Knite <[email protected]>
Co-authored-by: Yeuoly <[email protected]>
Co-authored-by: Petrus Han <[email protected]>
Co-authored-by: iamjoel <[email protected]>
Co-authored-by: Kalo Chin <[email protected]>
Co-authored-by: Ujjwal Maurya <[email protected]>
Co-authored-by: Maries <[email protected]>
2025-08-27 15:07:28 +08:00

76 lines
2.7 KiB
Python

import logging
import time
import click
from celery import shared_task
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
logger = logging.getLogger(__name__)
@shared_task(queue="dataset")
def remove_document_from_index_task(document_id: str):
"""
Async Remove document from index
:param document_id: document id
Usage: remove_document_from_index.delay(document_id)
"""
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
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
indexing_cache_key = f"document_{document.id}_indexing"
try:
dataset = document.dataset
if not dataset:
raise Exception("Document has no dataset")
index_processor = IndexProcessorFactory(document.doc_form).init_index_processor()
segments = db.session.query(DocumentSegment).where(DocumentSegment.document_id == document.id).all()
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()