Redis分布式缓存：微信抢红包解决方案

    /**
     * 包红包的接口
     */
    @GetMapping(value = "/set")
    public long setRedpacket(int total, int count) {
        //拆解红包
        Integer[] packet= this.splitRedPacket(total,count);
        //为红包生成全局唯一id
        long n=this.incrementId();
        //采用list存储红包
        String key=RED_PACKET_KEY+n;
        this.redisTemplate.opsForList().leftPushAll(key,packet);
        //设置3天过期
        this.redisTemplate.expire(key,3, TimeUnit.DAYS);
        log.info("拆解红包{}={}",key,packet);
        return n;
    }

/**
 * 拆解红包
 * 1.红包金额要被全部拆解完
 * 2.红包金额不能差太离谱
 * total 红包金额
 * count 红包数量
 */
public  Integer[] splitRedPacket(int total, int count) {
    int use = 0;
    Integer[] array = new Integer[count];
    Random random = new Random();
    for (int i = 0; i < count; i++) {
        if (i == count - 1)
            array[i] = total - use;
        else {
            // 红包随机金额浮动系数
            int avg = (total - use) * 2 / (count - i);
            array[i] = 1 + random.nextInt(avg - 1);
        }
        use = use + array[i];
    }
    return array;
}

    
    /**
     * 抢红包接口
     */
    @GetMapping(value = "/rob")
    public int rob(long redid,long userid) {
        //第一步：验证该用户是否抢过
        Object packet=this.redisTemplate.opsForHash().get(RED_PACKET_CONSUME_KEY+redid,String.valueOf(userid));
        if(packet==null){
            //第二步：从list队列，弹出一个红包
            Object obj=this.redisTemplate.opsForList().leftPop(RED_PACKET_KEY+redid);
            if(obj!=null){
                //第三步：抢到红包存起来
                this.redisTemplate.opsForHash().put(RED_PACKET_CONSUME_KEY+redid,String.valueOf(userid),obj);
                log.info("用户={}抢到{}",userid,obj);
                //TODO 异步把数据落地到数据库上
                
                return (Integer) obj;
            }
            //-1 代表抢完
            return -1;
        }
        //-2 代表已抢
        return -2;
    }