百度分布式id-UidGenerator之DefaultUidGenerator源码解读

1：百度源码与图

https://github.com/baidu/uid-generator

uid-generator支持自定义时间戳、工作机器id和序列号等各部分的位数，以应用于不同场景。默认分配方式如下。sign 默认一位符号标识，data seconds 时间，worder node id，机器id也是重启的次数，sequence 并发系列。uid-generator 支持两种获取id的方法，分别是DefaultUidGenerator/CachedUidGenerator.这篇先解读DefaultUidGenerator

2：DefaultUidGenerator

下载源码，在本地启动，我们可以看到default-uid-spring.xml，里面有个bean，初始化数据

1：afterPropertiesSet 初始化相关数据

    @Override
   public void afterPropertiesSet() throws Exception {
    // initialize bits allocator 初始化位移和最大值
    bitsAllocator = new BitsAllocator(timeBits, workerBits, seqBits);

    // initialize worker id 获取重启次数id
    workerId = workerIdAssigner.assignWorkerId();
    //重启id大于最大重启id次数，直接抛出异常
    if (workerId > bitsAllocator.getMaxWorkerId()) {
        throw new RuntimeException("Worker id " + workerId + " exceeds the max " + bitsAllocator.getMaxWorkerId());
    }

    LOGGER.info("Initialized bits(1, {}, {}, {}) for workerID:{}", timeBits, workerBits, seqBits, workerId);
}

1.1： bitsAllocator = new BitsAllocator(timeBits, workerBits, seqBits);

  进入这个方法，可以看到最大位移，位移数要等于64（sign+data seconds+worder node id+sequence），要不能就会抛出异常

1.2：workerId = workerIdAssigner.assignWorkerId(); 获取机器id，我们走进这方法，能够看到它在数据库增加了一条自增记录，拿到id

@Transactional
public long assignWorkerId() {
    // build worker node entity
    WorkerNodeEntity workerNodeEntity = buildWorkerNode();

    // add worker node for new (ignore the same IP + PORT)
    workerNodeDAO.addWorkerNode(workerNodeEntity);
    LOGGER.info("Add worker node:" + workerNodeEntity);

    return workerNodeEntity.getId();
}

2： 获取id的代码

protected synchronized long nextId() {
    //获取当前时间单位s
    long currentSecond = getCurrentSecond();
    // Clock moved backwards, refuse to generate uid 当前时间大于上次时间，就意味着时间回拨，抛出去
    if (currentSecond < lastSecond) {
        long refusedSeconds = lastSecond - currentSecond;
        throw new UidGenerateException("Clock moved backwards. Refusing for %d seconds", refusedSeconds);
    }

    // At the same second, increase sequence 当前时间相等的情况下，验证是否大于最大序列值，如果大于等待下一s
    if (currentSecond == lastSecond) {
        sequence = (sequence + 1) & bitsAllocator.getMaxSequence();
        // Exceed the max sequence, we wait the next second to generate uid
        //大于就等待下一秒
        if (sequence == 0) {
            currentSecond = getNextSecond(lastSecond);
        }

    // At the different second, sequence restart from zero
    } else {
    //如果当前时间大于上一次时间，则是从0开始
        sequence = 0L;
    }

    lastSecond = currentSecond;

    // Allocate bits for UID---根据三个时间端进行位移
    return bitsAllocator.allocate(currentSecond - epochSeconds, workerId, sequence);
}
 //获取当前最新时间s
private long getCurrentSecond() {
    long currentSecond = TimeUnit.MILLISECONDS.toSeconds(System.currentTimeMillis());
    //当前时间减去初始时间，还大于最大的时间位移时间，就直接抛出异常
    if (currentSecond - epochSeconds > bitsAllocator.getMaxDeltaSeconds()) {
        throw new UidGenerateException("Timestamp bits is exhausted. Refusing UID generate. Now: " + currentSecond);
    }

    return currentSecond;
}

3 总结

3.1 DefaultUidGenerator 单位走的是s，所以最大时间是2的28次方，只够8.7年，也可以自行变更data seconds等相关参数

3.2 注意DefaultUidGenerator 时间回拨，也是强制性抛出，要解决这个还是要看CachedUidGenerator（解决了时间回拨功能）

3.3 DefaultUidGenerator 基于雪花算法稍微变更了一点，熟悉雪花算法的（看我上一篇，简单自己写雪花算法，更容易理解），非常容易入手