flask扩展 flask-celery

共计 1581 个字符，预计需要花费 4 分钟才能阅读完成。

一、flask-celery 安装

pip install celery

pip install redis==2.10.6

pip install flask-celery-helper

二、创建和加载

• 工程目录下创建 run_celery.py 文件

  # 定义创建 celery 对象的方法 
  from celery import Celery
  
  def make_celery(app):
      celery = Celery(
          app.import_name,
          backend=app.config['CELERY_RESULT_BACKEND'],
          broker=app.config['CELERY_BROKER_URL']
      )
      celery.conf.update(app.config)
  
      TaskBase = celery.Task
      class ContextTask(celery.Task):
          abstract = True
          def __call__(self, *args, **kwargs):
              with app.app_context():
                  return TaskBase.__call__(self, *args, **kwargs)
  
      celery.Task = ContextTask
      return celery
  

• manage.py

  from flask_script import Manager
  from flask_migrate import Migrate, MigrateCommand
  
  from app import create_app
  from exts import db
  
  from run_celery import make_celery
  
  # 将要生成表的模型类导入 
  from myApp.models import *
  
  app = create_app(__name__)
  celery = make_celery(app)
  
  # 创建迁移管理对象 
  migrate = Migrate(app, db)
  manager = Manager(app)
  manager.add_command("db", MigrateCommand)
  
  if __name__ == '__main__':
      manager.run()
  

• exts/ext_celery

  from flask_celery import Celery
  celery = Celery()
  from .ext_celery import celery
  

• 关联 manage.py 中的 celery 与 ext_celery 中的 cel

  app.py

  from exts import celery
  celery.init_app(app)
  

三、配置

# celery
CELERY_RESULT_BACKEND = "redis://:@127.0.0.1:6379/5"
CELERY_BROKER_URL = "redis://:@127.0.0.1:6379/6"
# celery 的工人一直启动可能会造成内存泄露，该参数规定每个工人执行了多少个任务后就会被杀死 
# CELERY_MAX_TASKS_PER_CHILD = 50

四、封装任务

在应用目录下创建名为 tasks 的包目录

# from manage import celery
from exts import celery

@celery.task()
def mail():
    print("---- 开始耗时操作 ----")
    import time
    time.sleep(5)
    print("---- 结束耗时操作 ----")

五、添加到队列

from myApp.tasks import mail
@myApp.route("/sendMail/")
def sendMail():
    # 将耗时任务添加到队列 
    mail.delay()
    return "邮件发送成功"

六、启动 flask 服务

python manage.py runserver -d -r

七、启动工人

celery worker -A manage.celery --loglevel=info