Project

A Django background task app. By Bgtask, we can conveniently execute asynchronous tasks, and obtain the return data. The actuator runs in a separate process communicating with Django via TCP, ensuring it does not impact the main logic.

Usage

Install bgtask

pip install -U bgtask4django

Configurations

Add bgtask to INSTALLED_APPS in settings.py.

# settings.py
INSTALLED_APPS = [
    'bgtask',
    '...'
]
# Configuration should be read first from the Django settings, 
# and if not found, it should be looked up from the system environment variables.
BGTASK_SERVER_IP = "127.0.0.1"
BGTASK_SERVER_PORT = 23323
BGTASK_WORK_THREADS = 10

Migrate

python manage.py migrate bgtask

Run server

bgtask run in a separate process, which is the same runtime environment as the Django instance.

python manage.py bgtaskserver

How to use in Django?

one demo

# views.py
import time
import pickle

from bgtask.client import submit
from bgtask.models import BgTask, StateEnum, ResultEnum


def send_email(address, title, message=None):
    # some codes...
    time.sleep(5)
    # other codes...


def register(request, *args, **kwargs):
    # `submit` will put function's import-path into task-queue by tcp,
    # if no exception, it will return `task_id`.
    task_id = submit(send_email, "*@*.com", "test", message="this is a test")
    if not task_id:
        print("send email fail, check log for reason")
    ...

    # In other function or logic
    # if we know task_id, we can get 'Bgtask' object, and then get task state and return data
    task = BgTask.objects.get(pk=task_id)
    print(task.state)
    if task.state == StateEnum.FINISH and task.result == ResultEnum.SUCCESS:
        print(pickle.loads(task.return_value))