Python Redis数据结构服务器

首先下载redis：

pip install redis

在开始使用redis时会出现以下的问题：

原因是redis没有启动，解决方法是安装了redis-tool，运行redis-cli

如果此时发生如下错误：

你要安装redis-server：

sudo apt-get install redis-server

再次运行redis-cli时出现如下的效果即为正常启动：

接下来进行字符串的一些操作：set、get、getset、getrange、setrange、mset、delete、incr、incrbyfloat、decr

>>> import redis>>> conn = redis.Redis()>>> conn.keys('*')[]>>> conn.set('secret','ni!')True>>> conn.set('carats',24)True>>> conn.set('fever','101.5')True>>> conn.get('secret')'ni!'>>> conn.incr('carats')25>>> conn.getset('secret','icky-icky-icky-ptang-zoop-boing!')'ni!'>>> conn.getrange('secret',-6,-1)'boing!'>>> conn.setrange('secret',0,'ICKY')32>>> conn.get('secret')'ICKY-icky-icky-ptang-zoop-boing!'>>> conn.delete('fever')1>>> conn.set('fever','101.5')True>>> conn.incrbyfloat('fever')102.5

列表操作：

>>> conn.lpush('zoo','bear')1L>>> conn.lpush('zoo','alligator','duck')3L>>> conn.linsert('zoo','before','bear','beaver')4>>> conn.linsert('zoo','after','bear','cassowary')5>>> conn.lset('zoo',2,'marmoset')True>>> conn.rpush('zoo','yak')6L>>> conn.lindex('zoo',3)'bear'>>> conn.lrange('zoo',0,2)['duck', 'alligator', 'marmoset']>>> conn.ltrim('zoo',1,4)True>>> conn.lrange('zoo',0,-1)['alligator', 'marmoset', 'bear', 'cassowary']

哈希表操作：创建一个Redis的哈希表song，并对它进行操作

>>> conn.hmset('song',{'do':'a deer','re':'about a deer'})True>>> conn.hset('song','mi','a note to follow re')1L>>> conn.hget('song','mi')'a note to follow re'>>> conn.hmget('song','re','do')['about a deer', 'a deer']>>> conn.hkeys('song')['do', 're', 'mi']>>> conn.hvals('song')['a deer', 'about a deer', 'a note to follow re']>>> conn.hlen('song')3>>> conn.hgetall('song'){'do': 'a deer', 'mi': 'a note to follow re', 're': 'about a deer'}>>> conn.hsetnx('song','fa','a note that rhymes wirh la')1L

集合操作：

如果重新用conn.sadd('zoo','duck','goat','turkey')，会出现冲突问题

>>> conn.sadd('boo','duck','goat','turkey')3>>> conn.scard('boo')3>>> conn.smembers('boo')set(['turkey', 'goat', 'duck'])>>> conn.srem('boo','turkey')1>>> conn.sadd('better_boo','tiger','wolf','duck')3>>> conn.sinter('boo','better_boo')set(['duck'])>>> conn.sinterstore('fowl_boo','boo','better_boo')1>>> conn.sunion('boo','better_boo')set(['tiger', 'wolf', 'goat', 'duck'])>>> conn.sunionstore('fabulous_boo','boo','better_boo')4>>> conn.sdiff('boo','better_boo')set(['goat'])

有序集合：   以时间序列为例子，利用时间戳跟踪用户的登录。这里时间用unix的epoch值

>>> import time>>> now=time.time()>>> now1454420094.849293>>> conn.zadd('logins','smeagol',now)1>>> conn.zadd('logins','sauron',now+(5*60))1>>> conn.zadd('logins','bilbo',now+(2*60*60))1>>> conn.zadd('logins','treebeard',now+(24*60*60))1>>> conn.zrank('logins','bilbo')2>>> conn.zscore('logins','bilbo')1454427294.849293>>> conn.zrange('logins',0,-1)['smeagol', 'sauron', 'bilbo', 'treebeard']>>> conn.zrange('logins',0,-1,withscores=True)[('smeagol', 1454420094.849293), ('sauron', 1454420394.849293), ('bilbo', 1454427294.849293), ('treebeard', 1454506494.849293)]

位图：省空间且快速的处理超大集合数字的方式

>>> days=['2016-01-31','2016-02-01','2016-02-02']>>> big_spender=1089>>> tire_kicker=40459>>> late_joiner=550212>>> conn.setbit(days[0],big_spender,1)0>>> conn.setbit(days[0],tire_kicker,1)0>>> conn.setbit(days[1],big_spender,1)0>>> conn.setbit(days[2],big_spender,1)0>>> conn.setbit(days[2],late_joiner,1)0>>> for day in days  File "<stdin>", line 1    for day in days                  ^SyntaxError: invalid syntax>>> for day in days:... conn.bitcount(day)  File "<stdin>", line 2    conn.bitcount(day)       ^IndentationError: expected an indented block>>> for day in days:...     conn.bitcount(day)... 212>>> conn.getbit(days[1],tire_kicker)0>>> conn.bitop('and','everyday',*days)68777L>>> conn.bitcount('everyday')1>>> conn.getbit('everyday',big_spender)1>>> conn.bitop('or','everyday',*days)68777L>>> conn.bitcount('alldays')0>>> conn.bitop('or','alldays',*days)68777L>>> conn.bitcount('alldays')3

缓存和过期：expire（）函数