ConnectionPool实现redis在python中的连接

这篇文章主要介绍了Python与Redis的连接教程,Redis是一个高性能的基于内存的数据库,需要的朋友可以参考下

 

今天在写zabbix storm job监控脚本的时候用到了python的redis模块，之前也有用过，但是没有过多的了解，今天看了下相关的api和源码,看到有ConnectionPool的实现，这里简单说下。 在ConnectionPool之前，如果需要连接redis，我都是用StrictRedis这个类，在源码中可以看到这个类的具体解释：   redis.StrictRedis Implementation of the Redis protocol.This abstract class provides a Python interface to all Redis commands and an implementation of the Redis protocol.Connection and Pipeline derive from this, implementing how the commands are sent and received to the Redis server

使用的方法：

?

1 2	r=redis.StrictRedis(host=xxxx, port=xxxx, db=xxxx) r.xxxx()

有了ConnectionPool这个类之后，可以使用如下方法

?

1 2	pool = redis.ConnectionPool(host=xxx, port=xxx, db=xxxx) r = redis.Redis(connection_pool=pool)

这里Redis是StrictRedis的子类 简单分析如下： 在StrictRedis类的__init__方法中，可以初始化connection_pool这个参数，其对应的是一个ConnectionPool的对象：

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

class StrictRedis(object): ........   def __init__(self, host='localhost', port=6379,          db=0, password=None, socket_timeout=None,          socket_connect_timeout=None,          socket_keepalive=None, socket_keepalive_options=None,          connection_pool=None, unix_socket_path=None,          encoding='utf-8', encoding_errors='strict',          charset=None, errors=None,          decode_responses=False, retry_on_timeout=False,          ssl=False, ssl_keyfile=None, ssl_certfile=None,          ssl_cert_reqs=None, ssl_ca_certs=None):      if not connection_pool:        ..........        connection_pool = ConnectionPool(**kwargs)      self.connection_pool = connection_pool

在StrictRedis的实例执行具体的命令时会调用execute_command方法，这里可以看到具体实现是从连接池中获取一个具体的连接，然后执行命令，完成后释放连接：

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

# COMMAND EXECUTION AND PROTOCOL PARSING def execute_command(self, *args, **options):   "Execute a command and return a parsed response"   pool = self.connection_pool   command_name = args[0]   connection = pool.get_connection(command_name, **options) #调用ConnectionPool.get_connection方法获取一个连接   try:     connection.send_command(*args) #命令执行，这里为Connection.send_command     return self.parse_response(connection, command_name, **options)   except (ConnectionError, TimeoutError) as e:     connection.disconnect()     if not connection.retry_on_timeout and isinstance(e, TimeoutError):       raise     connection.send_command(*args)     return self.parse_response(connection, command_name, **options)   finally:     pool.release(connection) #调用ConnectionPool.release释放连接

在来看看ConnectionPool类：

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

class ConnectionPool(object):     ...........   def __init__(self, connection_class=Connection, max_connections=None,          **connection_kwargs):  #类初始化时调用构造函数     max_connections = max_connections or 2 ** 31     if not isinstance(max_connections, (int, long)) or max_connections < 0: #判断输入的max_connections是否合法       raise ValueError('"max_connections" must be a positive integer')     self.connection_class = connection_class #设置对应的参数     self.connection_kwargs = connection_kwargs     self.max_connections = max_connections     self.reset() #初始化ConnectionPool 时的reset操作   def reset(self):     self.pid = os.getpid()     self._created_connections = 0 #已经创建的连接的计数器     self._available_connections = []  #声明一个空的数组，用来存放可用的连接     self._in_use_connections = set() #声明一个空的集合，用来存放已经在用的连接     self._check_lock = threading.Lock() .......   def get_connection(self, command_name, *keys, **options): #在连接池中获取连接的方法     "Get a connection from the pool"     self._checkpid()     try:       connection = self._available_connections.pop() #获取并删除代表连接的元素，在第一次获取connectiong时，因为_available_connections是一个空的数组，       会直接调用make_connection方法     except IndexError:       connection = self.make_connection()     self._in_use_connections.add(connection)  #向代表正在使用的连接的集合中添加元素     return connection    def make_connection(self): #在_available_connections数组为空时获取连接调用的方法     "Create a new connection"     if self._created_connections >= self.max_connections:  #判断创建的连接是否已经达到最大限制，max_connections可以通过参数初始化       raise ConnectionError("Too many connections")     self._created_connections += 1  #把代表已经创建的连接的数值+1     return self.connection_class(**self.connection_kwargs)   #返回有效的连接，默认为Connection(**self.connection_kwargs)   def release(self, connection): #释放连接，链接并没有断开，只是存在链接池中     "Releases the connection back to the pool"     self._checkpid()     if connection.pid != self.pid:       return     self._in_use_connections.remove(connection)  #从集合中删除元素     self._available_connections.append(connection) #并添加到_available_connections 的数组中   def disconnect(self): #断开所有连接池中的链接     "Disconnects all connections in the pool"     all_conns = chain(self._available_connections,              self._in_use_connections)     for connection in all_conns:       connection.disconnect()

execute_command最终调用的是Connection.send_command方法，关闭链接为 Connection.disconnect方法，而Connection类的实现：

?

1 2 3 4 5 6 7

class Connection(object):   "Manages TCP communication to and from a Redis server"   def __del__(self):  #对象删除时的操作，调用disconnect释放连接     try:       self.disconnect()     except Exception:       pass

核心的链接建立方法是通过socket模块实现：

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

def _connect(self):     err = None     for res in socket.getaddrinfo(self.host, self.port, 0,                    socket.SOCK_STREAM):       family, socktype, proto, canonname, socket_address = res       sock = None       try:         sock = socket.socket(family, socktype, proto)         # TCP_NODELAY         sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)         # TCP_KEEPALIVE         if self.socket_keepalive:  #构造函数中默认 socket_keepalive=False，因此这里默认为短连接           sock.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1)           for k, v in iteritems(self.socket_keepalive_options):             sock.setsockopt(socket.SOL_TCP, k, v)         # set the socket_connect_timeout before we connect         sock.settimeout(self.socket_connect_timeout) #构造函数中默认socket_connect_timeout=None，即连接为blocking的模式         # connect         sock.connect(socket_address)         # set the socket_timeout now that we're connected         sock.settimeout(self.socket_timeout) #构造函数中默认socket_timeout=None         return sock       except socket.error as _:         err = _         if sock is not None:           sock.close() .....

关闭链接的方法：

?

1 2 3 4 5 6 7 8 9 10 11

def disconnect(self):   "Disconnects from the Redis server"   self._parser.on_disconnect()   if self._sock is None:     return   try:     self._sock.shutdown(socket.SHUT_RDWR) #先shutdown再close     self._sock.close()   except socket.error:     pass   self._sock = None

可以小结如下 1）默认情况下每创建一个Redis实例都会构造出一个ConnectionPool实例，每一次访问redis都会从这个连接池得到一个连接，操作完成后会把该连接放回连接池（连接并没有释放）,可以构造一个统一的ConnectionPool，在创建Redis实例时，可以将该ConnectionPool传入，那么后续的操作会从给定的ConnectionPool获得连接，不会再重复创建ConnectionPool。 2）默认情况下没有设置keepalive和timeout，建立的连接是blocking模式的短连接。 3）不考虑底层tcp的情况下，连接池中的连接会在ConnectionPool.disconnect中统一销毁。