File size: 4,040 Bytes
20f348c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 |
import datetime
import logging
import time
import click
from celery import shared_task # type: ignore
from core.indexing_runner import IndexingRunner
from core.rag.index_processor.index_processor_factory import IndexProcessorFactory
from extensions.ext_database import db
from extensions.ext_redis import redis_client
from models.dataset import Dataset, Document, DocumentSegment
from services.feature_service import FeatureService
@shared_task(queue="dataset")
def retry_document_indexing_task(dataset_id: str, document_ids: list[str]):
"""
Async process document
:param dataset_id:
:param document_ids:
Usage: retry_document_indexing_task.delay(dataset_id, document_id)
"""
documents: list[Document] = []
start_at = time.perf_counter()
dataset = db.session.query(Dataset).filter(Dataset.id == dataset_id).first()
if not dataset:
raise ValueError("Dataset not found")
for document_id in document_ids:
retry_indexing_cache_key = "document_{}_is_retried".format(document_id)
# check document limit
features = FeatureService.get_features(dataset.tenant_id)
try:
if features.billing.enabled:
vector_space = features.vector_space
if 0 < vector_space.limit <= vector_space.size:
raise ValueError(
"Your total number of documents plus the number of uploads have over the limit of "
"your subscription."
)
except Exception as e:
document = (
db.session.query(Document).filter(Document.id == document_id, Document.dataset_id == dataset_id).first()
)
if document:
document.indexing_status = "error"
document.error = str(e)
document.stopped_at = datetime.datetime.now(datetime.timezone.utc).replace(tzinfo=None)
db.session.add(document)
db.session.commit()
redis_client.delete(retry_indexing_cache_key)
return
logging.info(click.style("Start retry document: {}".format(document_id), fg="green"))
document = (
db.session.query(Document).filter(Document.id == document_id, Document.dataset_id == dataset_id).first()
)
if not document:
logging.info(click.style("Document not found: {}".format(document_id), fg="yellow"))
return
try:
# clean old data
index_processor = IndexProcessorFactory(document.doc_form).init_index_processor()
segments = db.session.query(DocumentSegment).filter(DocumentSegment.document_id == document_id).all()
if segments:
index_node_ids = [segment.index_node_id for segment in segments]
# delete from vector index
index_processor.clean(dataset, index_node_ids, with_keywords=True, delete_child_chunks=True)
for segment in segments:
db.session.delete(segment)
db.session.commit()
document.indexing_status = "parsing"
document.processing_started_at = datetime.datetime.now(datetime.timezone.utc).replace(tzinfo=None)
db.session.add(document)
db.session.commit()
indexing_runner = IndexingRunner()
indexing_runner.run([document])
redis_client.delete(retry_indexing_cache_key)
except Exception as ex:
document.indexing_status = "error"
document.error = str(ex)
document.stopped_at = datetime.datetime.now(datetime.timezone.utc).replace(tzinfo=None)
db.session.add(document)
db.session.commit()
logging.info(click.style(str(ex), fg="yellow"))
redis_client.delete(retry_indexing_cache_key)
pass
end_at = time.perf_counter()
logging.info(click.style("Retry dataset: {} latency: {}".format(dataset_id, end_at - start_at), fg="green"))
|