python_study.txt

python 学习笔记
1. 爬虫 翻页 参考项目
	https://github.com/whuhacker/Sitedossier-Crawler

2. 实时用户统计
	# -*- coding: utf-8 -*-
	import time
	from redis import Redis
	from datetime import datetime
	ONLINE_LAST_MINUTES = 5
	redis = Redis(host='', port=, db=, password='')
	def mark_online(user_id):   # 讲一个用户标记为 online
	    now = int(time.time())  # 当前的uinx 时间戳
	    expires = now + (ONLINE_LAST_MINUTES * 60) + 10     # 过期 unix 时间戳
	    all_users_key = 'online-users/%d' % (now // 60)     # 集合名 包含分钟信息
	    user_key = 'user_activity/%s' % user_id
	    p = redis.pipeline();
	    p.sadd(all_users_key, user_id)      # 将用户id 插入到包含分钟信息的集合中
	    p.set(user_key, now)                # 记录用户标记时间
	    p.expireat(all_users_key, expires)  # 设定集合过期时间为Unix 时间戳
	    p.expireat(user_key, expires)
	    p.execute()

	def get_user_last_activity(user_id):        # 获取用户最后活跃时间
	    last_active = redis.get('user_activity/%s' % user_id)
	    if last_active is None:
		return None
	    return datetime.utcfromtimestamp(int(last_active))

	def get_online_users():
	    current = int(time.time()) // 60
	    minutes = xrange(ONLINE_LAST_MINUTES)
	    return redis.sunion(['online-users/%d' % (current - x) for x in minutes])