Redis counter class definition and usage

Redis is an open source log-type Key-Value database written in ANSI C language, supports network, can be memory-based and persistent, and provides APIs in multiple languages. This article mainly introduces the definition and usage of the PHP-based redis counter class. It analyzes the redis counter class defined by PHP and its related usage techniques in detail in the form of examples. Friends in need can refer to it. I hope it can help you.

Here, use its incr (increment), get (get), delete (clear) methods to implement the counter class.

1.Redis counter class code and demonstration example

RedisCounter.class.php

<?php
/**
 * PHP基于Redis计数器类
 * Date:  2017-10-28
 * Author: fdipzone
 * Version: 1.0
 *
 * Descripton:
 * php基于Redis实现自增计数，主要使用redis的incr方法，并发执行时保证计数自增唯一。
 *
 * Func:
 * public incr  执行自增计数并获取自增后的数值
 * public get   获取当前计数
 * public reset  重置计数
 * private connect 创建redis连接
 */
class RedisCounter{ // class start
  private $_config;
  private $_redis;
  /**
   * 初始化
   * @param Array $config redis连接设定
   */
  public function __construct($config){
    $this->_config = $config;
    $this->_redis = $this->connect();
  }
  /**
   * 执行自增计数并获取自增后的数值
   * @param String $key 保存计数的键值
   * @param Int  $incr 自增数量，默认为1
   * @return Int
   */
  public function incr($key, $incr=1){
    return intval($this->_redis->incr($key, $incr));
  }
  /**
   * 获取当前计数
   * @param String $key 保存计数的健值
   * @return Int
   */
  public function get($key){
    return intval($this->_redis->get($key));
  }
  /**
   * 重置计数
   * @param String $key 保存计数的健值
   * @return Int
   */
  public function reset($key){
    return $this->_redis->delete($key);
  }
  /**
   * 创建redis连接
   * @return Link
   */
  private function connect(){
    try{
      $redis = new Redis();
      $redis->connect($this->_config['host'],$this->_config['port'],$this->_config['timeout'],$this->_config['reserved'],$this->_config['retry_interval']);
      if(empty($this->_config['auth'])){
        $redis->auth($this->_config['auth']);
      }
      $redis->select($this->_config['index']);
    }catch(RedisException $e){
      throw new Exception($e->getMessage());
      return false;
    }
    return $redis;
  }
} // class end
?>

demo.php

<?php
Require &#39;RedisCounter.class.php&#39;;
// redis连接设定
$config = array(
  &#39;host&#39; => 'localhost',
  'port' => 6379,
  'index' => 0,
  'auth' => '',
  'timeout' => 1,
  'reserved' => NULL,
  'retry_interval' => 100,
);
// 创建RedisCounter对象
$oRedisCounter = new RedisCounter($config);
// 定义保存计数的健值
$key = 'mycounter';
// 执行自增计数，获取当前计数，重置计数
echo $oRedisCounter->get($key).PHP_EOL; // 0
echo $oRedisCounter->incr($key).PHP_EOL; // 1
echo $oRedisCounter->incr($key, 10).PHP_EOL; // 11
echo $oRedisCounter->reset($key).PHP_EOL; // 1
echo $oRedisCounter->get($key).PHP_EOL; // 0
?>

Output:

0
1
11
1
0

2. Concurrent call counter, check the uniqueness of the count

The test code is as follows:

<?php
Require &#39;RedisCounter.class.php&#39;;
// redis连接设定
$config = array(
  &#39;host&#39; => 'localhost',
  'port' => 6379,
  'index' => 0,
  'auth' => '',
  'timeout' => 1,
  'reserved' => NULL,
  'retry_interval' => 100,
);
// 创建RedisCounter对象
$oRedisCounter = new RedisCounter($config);
// 定义保存计数的健值
$key = 'mytestcounter';
// 执行自增计数并返回自增后的计数，记录入临时文件
file_put_contents('/tmp/mytest_result.log', $oRedisCounter->incr($key).PHP_EOL, FILE_APPEND);
?>

To test the concurrent execution, we use the ab tool to test, and set the execution to 150 times, 15 concurrency.

ab -c 15 -n 150 http://localhost/test.php

Execution results:

ab -c 15 -n 150 http://localhost/test.php
This is ApacheBench, Version 2.3 <$Revision: 1554214 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/
Benchmarking home.rabbit.km.com (be patient).....done
Server Software:    nginx/1.6.3
Server Hostname:    localhost
Server Port:      80
Document Path:     /test.php
Document Length:    0 bytes
Concurrency Level:   15
Time taken for tests:  0.173 seconds
Complete requests:   150
Failed requests:    0
Total transferred:   24150 bytes
HTML transferred:    0 bytes
Requests per second:  864.86 [#/sec] (mean)
Time per request:    17.344 [ms] (mean)
Time per request:    1.156 [ms] (mean, across all concurrent requests)
Transfer rate:     135.98 [Kbytes/sec] received
Connection Times (ms)
       min mean[+/-sd] median  max
Connect:    0  0  0.2   0    1
Processing:   3  16  3.2   16   23
Waiting:    3  16  3.2   16   23
Total:     4  16  3.1   17   23
Percentage of the requests served within a certain time (ms)
 50%   17
 66%   18
 75%   18
 80%   19
 90%   20
 95%   21
 98%   22
 99%   22
 100%   23 (longest request)

Check whether the count is unique

生成的总计数

wc -l /tmp/mytest_result.log
   150 /tmp/mytest_result.log

生成的唯一计数

sort -u /tmp/mytest_result.log | wc -l
   150

You can see that in the case of concurrent calls, the generated count is also guaranteed to be unique.

Related recommendations:

Introduction to redis counter methods to prevent brushing orders

The above is the detailed content of Redis counter class definition and usage. For more information, please follow other related articles on the PHP Chinese website!