Fix/upload limit (#2521)

Co-authored-by: jyong <jyong@dify.ai>
Co-authored-by: StyleZhang <jasonapring2015@outlook.com>
This commit is contained in:
Jyong
2024-02-22 17:16:22 +08:00
committed by GitHub
parent 52b12ed7eb
commit 97fe817186
12 changed files with 97 additions and 14 deletions

View File

@@ -4,10 +4,12 @@ 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 Document
from models.dataset import Dataset, Document
from services.feature_service import FeatureService
@shared_task(queue='dataset')
@@ -21,6 +23,35 @@ def document_indexing_task(dataset_id: str, document_ids: list):
"""
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.utcnow()
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'))