coinex // 撮合引擎逻辑流程 (两种数据源初始化源和前端源)

tenc1239

已于 2023-04-19 22:22:38 修改

阅读量491

点赞数

文章标签： java 开发语言

于 2023-04-19 03:13:13 首次发布

本文链接：https://blog.csdn.net/tenc1239/article/details/130234618

版权

1 生产者数据源

1.1. match-server 一启动初始化数据自动查询数据库查询level2要展示的数据

1.2 match-server接收前端发给Exchange-server的数据

2. 将查询/接受的数据EntrustOrder 转成 Order 解耦过滤掉不要的属性

3.Order转成 OrderEvent

4. 分配序号发布数据到ringbuffer,管理序号和栅栏(前提是ringbuffer构建好了)

5 分配序号发布数据到ringbuffer

5.1. 初始化数据源实现的代码循环每一条数据转换分配序号发布数据到ringbuffer

5.2 前端数据源

1 生产者数据源

1.1 match-server 一启动初始化数据自动查询数据库查询level2要展示的数据

List<EntrustOrder> entrustOrders = entrustOrderMapper.selectList(
                new LambdaQueryWrapper<EntrustOrder>()
                        .eq(EntrustOrder::getStatus, 0)
                        .orderByAsc(EntrustOrder::getCreated)

1.2 match-server接收前端发给Exchange-server的数据

 @StreamListener("order_in") // "order_in" 在 Sink中
    public void handleMessage(EntrustOrder entrustOrder) { // 消息监听}

2. 将查询/接受的数据EntrustOrder 转成 Order 解耦过滤掉不要的属性

public static  Order entrustOrder2Order(EntrustOrder entrustOrder) {
        Order order = new Order();
        order.setOrderId(entrustOrder.getId().toString());

        order.setPrice(entrustOrder.getPrice());
        order.setAmount(entrustOrder.getVolume().subtract(entrustOrder.getDeal())); // 交易的数量= 总数量- 已经成交的数量

        order.setSymbol(entrustOrder.getSymbol());
        order.setOrderDirection(OrderDirection.getOrderDirection(entrustOrder.getType().intValue())); // 交易side
        order.setTime(entrustOrder.getCreated().getTime());

        return order ;

    }

3.Order转成 OrderEvent

// 使用事件转换器的好处，1环形队列获取序号， 2拿到事件填充数据， 3再发布序号 省了从2给 事件填充数据 private static final EventTranslatorOneArg<OrderEvent, Order> TRANSLATOR = new EventTranslatorOneArg<OrderEvent, Order>() { // Order 转化成 OrderEvent
        // 现在上面实例化 然后 在下面 实现

        /**
         * Translate a data representation into fields set in given event 将数据表示转换为给定事件中设置的字段
         * Params:参数
         * event
         *  into which the data should be translated. 要转成什么data
         * sequence 序列
         * that is assigned to event. 分配给事件的序列。
         * arg0
         * The first user specified argument to the translator 转换器的第一个用户指定参数
         */
        public void translateTo(OrderEvent event, long sequence, Order input) {
            event.setSource(input);
        } // setSource 是 source的 set方法
    };

4. 分配序号发布数据到ringbuffer,管理序号和栅栏(前提是ringbuffer构建好了)

/**
     * 我们使用DisruptorTemplate 时,就使用它的onData方法
     * @param input
     *
     * public <A> void publishEvent(EventTranslatorOneArg<E, A> translator, A arg0)
     *     {
     *         final long sequence = sequencer.next();
     *         translateAndPublish(translator, sequence, arg0);
     *     }
     * private <A> void translateAndPublish(EventTranslatorOneArg<E, A> translator, long sequence, A arg0)
     *     {
     *         try
     *         {
     *             translator.translateTo(get(sequence), sequence, arg0);
     *         }
     *         finally
     *         {
     *             sequencer.publish(sequence);
     *         }
     *     }
     *
     *     void publish(long sequence);
     *
     */
    public void onData(Order input) {
        ringBuffer.publishEvent(TRANSLATOR, input); // 这个就是分配序号 然后发布数据到ringbuffer
    }

5 分配序号发布数据到ringbuffer

5.1. 初始化数据源从数据查询数据循环每一条数据转换分配序号发布数据到ringbuffer

for (EntrustOrder entrustOrder : entrustOrders) {
            disruptorTemplate.onData(BeanUtils.entrustOrder2Order(entrustOrder)); // 往ringbuffer 中放 // BeanUtils.entrustOrder2Order 数据转换
        } // BeanUtils.entrustOrder2Order(entrustOrder) 是数据转换 就是从entrustOrder  筛选出 Order 需要的属性数据

5.2 前端数据源高频多生产者每次一条数据

exchange-service 发送消息到match

disruptorTemplate.onData(BeanUtils.entrustOrder2Order(entrustOrder));

6 disruptorTemplate.onData 源码

onData

public void onData(Order input) {
         ringBuffer.publishEvent(TRANSLATOR, input); // 这个就是分配序号 然后发布数据到ringbuffer
          }

TRANSLATOR

// 先用 EventTranslatorOneArg方法做了个 TRANSLATOR 再调用publishEvent方法
private static final EventTranslatorOneArg<OrderEvent, Order> TRANSLATOR = new EventTranslatorOneArg<OrderEvent, Order>() { // Order 转化成 OrderEvent

        public void translateTo(OrderEvent event, long sequence, Order input) {
            event.setSource(input);
        } // setSource 是 source的 set方法
    };

publishEvent

@Override
    public <A> void publishEvent(EventTranslatorOneArg<E, A> translator, A arg0)
    {
        final long sequence = sequencer.next(); // 单生成者就是 但生产者中得next算法 1:43:28 // https://www.bilibili.com/video/BV1zM4y1L7Q9/?spm_id_from=333.337.search-card.all.click&vd_source=ff8b7f852278821525f11666b36f180a
        translateAndPublish(translator, sequence, arg0);
    }

translateAndPublish

private <A> void translateAndPublish(EventTranslatorOneArg<E, A> translator, long sequence, A arg0)
    {
        try
        {
            translator.translateTo(get(sequence), sequence, arg0);
        }
        finally
        {
            sequencer.publish(sequence);
        }
    }

EventTranslatorOneArg

public interface EventTranslatorOneArg<T, A>
{
    /**
     * Translate a data representation into fields set in given event
     *
     * @param event    into which the data should be translated.
     * @param sequence that is assigned to event.
     * @param arg0     The first user specified argument to the translator
     */
    void translateTo(T event, long sequence, A arg0);
}

translateTo 自定义还是调用?

private static final EventTranslatorOneArg<OrderEvent, Order> TRANSLATOR = new EventTranslatorOneArg<OrderEvent, Order>() { // Order 转化成 OrderEvent
        public void translateTo(OrderEvent event, long sequence, Order input) {
            event.setSource(input);
        } // setSource 是 source的 set方法
    };

Sequencer

public interface Sequencer extends Cursored, Sequenced
{
}

Sequenced ->publish

public interface Sequenced
{
    /**
     * Publishes a sequence. Call when the event has been filled.
     *
     * @param sequence the sequence to be published.
     */
    void publish(long sequence); // 操作: publish 序号
}