Team Ai
Apppublic

Underground-Digital/Workflow-Engine

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes
disable_segment_from_index_task.py65 linesDownload Raw Back to tasks
1import logging2import time3 4import click5from celery import shared_task6from werkzeug.exceptions import NotFound7 8from core.rag.index_processor.index_processor_factory import IndexProcessorFactory9from extensions.ext_database import db10from extensions.ext_redis import redis_client11from models.dataset import DocumentSegment12 13 14@shared_task(queue="dataset")15def disable_segment_from_index_task(segment_id: str):16    """17    Async disable segment from index18    :param segment_id:19 20    Usage: disable_segment_from_index_task.delay(segment_id)21    """22    logging.info(click.style("Start disable segment from index: {}".format(segment_id), fg="green"))23    start_at = time.perf_counter()24 25    segment = db.session.query(DocumentSegment).filter(DocumentSegment.id == segment_id).first()26    if not segment:27        raise NotFound("Segment not found")28 29    if segment.status != "completed":30        raise NotFound("Segment is not completed , disable action is not allowed.")31 32    indexing_cache_key = "segment_{}_indexing".format(segment.id)33 34    try:35        dataset = segment.dataset36 37        if not dataset:38            logging.info(click.style("Segment {} has no dataset, pass.".format(segment.id), fg="cyan"))39            return40 41        dataset_document = segment.document42 43        if not dataset_document:44            logging.info(click.style("Segment {} has no document, pass.".format(segment.id), fg="cyan"))45            return46 47        if not dataset_document.enabled or dataset_document.archived or dataset_document.indexing_status != "completed":48            logging.info(click.style("Segment {} document status is invalid, pass.".format(segment.id), fg="cyan"))49            return50 51        index_type = dataset_document.doc_form52        index_processor = IndexProcessorFactory(index_type).init_index_processor()53        index_processor.clean(dataset, [segment.index_node_id])54 55        end_at = time.perf_counter()56        logging.info(57            click.style("Segment removed from index: {} latency: {}".format(segment.id, end_at - start_at), fg="green")58        )59    except Exception:60        logging.exception("remove segment from index failed")61        segment.enabled = True62        db.session.commit()63    finally:64        redis_client.delete(indexing_cache_key)65