File size: 2,983 Bytes
4304c6d |
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 |
import datetime
import logging
import time
import click
from celery import shared_task
from flask import current_app
from core.indexing_runner import DocumentIsPausedException, IndexingRunner
from extensions.ext_database import db
from models.dataset import Dataset, Document
from services.feature_service import FeatureService
@shared_task(queue='dataset')
def document_indexing_task(dataset_id: str, document_ids: list):
"""
Async process document
:param dataset_id:
:param document_ids:
Usage: document_indexing_task.delay(dataset_id, document_id)
"""
documents = []
start_at = time.perf_counter()
dataset = db.session.query(Dataset).filter(Dataset.id == dataset_id).first()
# check document limit
features = FeatureService.get_features(dataset.tenant_id)
try:
if features.billing.enabled:
vector_space = features.vector_space
count = len(document_ids)
batch_upload_limit = int(current_app.config['BATCH_UPLOAD_LIMIT'])
if count > batch_upload_limit:
raise ValueError(f"You have reached the batch upload limit of {batch_upload_limit}.")
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:
for document_id in document_ids:
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()
return
for document_id in document_ids:
logging.info(click.style('Start process document: {}'.format(document_id), fg='green'))
document = db.session.query(Document).filter(
Document.id == document_id,
Document.dataset_id == dataset_id
).first()
if document:
document.indexing_status = 'parsing'
document.processing_started_at = datetime.datetime.now(datetime.timezone.utc).replace(tzinfo=None)
documents.append(document)
db.session.add(document)
db.session.commit()
try:
indexing_runner = IndexingRunner()
indexing_runner.run(documents)
end_at = time.perf_counter()
logging.info(click.style('Processed dataset: {} latency: {}'.format(dataset_id, end_at - start_at), fg='green'))
except DocumentIsPausedException as ex:
logging.info(click.style(str(ex), fg='yellow'))
except Exception:
pass
|