celery4+django2定时任务的实现代码

.
├── linux_news
│  ├── celery.py
│  ├── __init__.py
│  ├── settings.py
│  ├── urls.py
│  └── wsgi.py
├── manage.py
├── news
│  ├── admin.py
│  ├── apps.py
│  ├── __init__.py
│  ├── migrations
│  ├── models
│  ├── tasks.py
│  ├── tests.py
│  └── views
└── start-celery.sh

import os
from celery import Celery

# set the default Django settings module for the 'celery' program.
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'linux_news.settings')

app = Celery('linux_news')

# 'django.conf:settings'表示django,conf.settings也就是django项目的配置，celery会根据前面设置的环境变量自动查找并导入
# - namespace表示在settings.py中celery配置项的名字的统一前缀，这里是'CELERY_'，配置项的名字也需要大写
app.config_from_object('django.conf:settings', namespace='CELERY')

# Load task modules from all registered Django app configs.
app.autodiscover_tasks()

from .celery import app as celery_app

from linux_news.celery import celery_app as app
from .models import *
import time
import feedparser
import pytz
import html


@app.task(ignore_result=True)
def fetch_news(origin_name):
  """
  fetch all news from origin_name
  """
  origin = get_feeds_origin(origin_name)
  feeds = feedparser.parse(origin.feed_link)
  for item in feeds['entries']:
    entry = NewsEntry()
    entry.title = item.title
    entry.origin = origin
    entry.author = item.author
    entry.link = item.link
    # add timezone
    entry.publish_time = item.time.replace(tzinfo=pytz.utc)
    entry.summary = html.escape(item.summary)

    entry.save()


@app.task(ignore_result=True)
def fetch_all_news():
  """
  这是我们的定时任务
  fetch all origins' news to db
  """
  origins = NewsOrigin.objects.all()
  for origin in origins:
    fetch_news.delay(origin.origin_name)

# redis
REDIS_PORT = 6379
REDIS_DB = 0
# 从环境变量中取得redis服务器地址
REDIS_HOST = os.environ.get('REDIS_ADDR', 'redis')

# celery settings
# 这两项必须设置，否则不能正常启动celery beat
CELERY_ENABLE_UTC = True
CELERY_TIMEZONE = TIME_ZONE
# 任务队列配置
CELERY_BROKER_URL = f'redis://{REDIS_HOST}:{REDIS_PORT}/{REDIS_DB}'
CELERY_ACCEPT_CONTENT = ['application/json', ]
CELERY_RESULT_BACKEND = f'redis://{REDIS_HOST}:{REDIS_PORT}/{REDIS_DB}'
CELERY_TASK_SERIALIZER = 'json'

from celery.schedules import crontab
CELERY_BEAT_SCHEDULE={
    'fetch_news_every-1-hour': {
      'task': 'news.tasks.fetch_all_news',
      'schedule': crontab(minute=0, hour='*/1'),
    }
}

export REDIS_ADDR=127.0.0.1

celery -A linux_news worker -l info -B -f /path/to/log