Dubbo源码分析(十三)------------Serialization之序列化实现

Dubbo的版本是基于2.7.3.Release. Dubbo内部的序列化默认实现是用hessian2的序列化框架实现,同时叶预留SPI的扩展点给开发者去实现.下面就来看下-Serialization的api的设计.

/**
 * Serialization strategy interface that specifies a serializer. (SPI, Singleton, ThreadSafe)
 * The default extension is hessian2 and the default serialization implementation of the dubbo protocol.
@SPI("hessian2")
public interface Serialization {
    /**
     * Get content type unique id, recommended that custom implementations use values greater than 20.
     */
    byte getContentTypeId();
    /**
     * 获取内容的类型
     */
    String getContentType();
    /**
     * Get a serialization implementation instance
     * 序列化
     */
    @Adaptive
    ObjectOutput serialize(URL url, OutputStream output) throws IOException;

    /**
     * Get a deserialization implementation instance
     * 反序列化的
     */
    @Adaptive
    ObjectInput deserialize(URL url, InputStream input) throws IOException;
}

Hession实现层就是实现了`Serialization的接口. 实现也比较简单,序列化和反序列化都是 直接new了一个Hession2ObjectOutput和Hession2ObjectInpout对象.

 */
public class Hessian2Serialization implements Serialization {
    @Override
    public byte getContentTypeId() {
        return HESSIAN2_SERIALIZATION_ID;
    }

    @Override
    public String getContentType() {
        return "x-application/hessian2";
    }
    //序列化方法
    @Override
    public ObjectOutput serialize(URL url, OutputStream out) throws IOException {
        return new Hessian2ObjectOutput(out);
    }
   //反序列化
    @Override
    public ObjectInput deserialize(URL url, InputStream is) throws IOException {
        return new Hessian2ObjectInput(is);
    }
}

Duobb框架内部实现的序列化组件的主要是以下9种:

测试用例

    protected URL url = new URL("protocol", "1.1.1.1", 1234);
    protected ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream();
    @Test
    public void test_boolArray_withType() throws Exception {
        boolean[] data = new boolean[]{true, false, true};

        ObjectOutput objectOutput = serialization.serialize(url, byteArrayOutputStream);
        objectOutput.writeObject(data);
        objectOutput.flushBuffer();

        ByteArrayInputStream byteArrayInputStream = new ByteArrayInputStream(
                byteArrayOutputStream.toByteArray());
        ObjectInput deserialize = serialization.deserialize(url, byteArrayInputStream);

        assertTrue(Arrays.equals(data, (boolean[]) deserialize.readObject(boolean[].class)));

        try {
            deserialize.readObject(boolean[].class);
            fail();
        } catch (ArrayIndexOutOfBoundsException e) {
        }
        // NOTE: Hessian2 throws ArrayIndexOutOfBoundsException instead of IOException, let's live with this.
    }

总结: 今天主要了解下dubbo内部实现了序列化集成的8中组件,后面对于dubboCodec中怎么应用序列化的着重分析.