之前分析过一篇Dubbo的SPI自适应扩展 。今天我们从源码的层次来分析是如何实现的自适应扩展
先定义一个SPI接口（被@SPI标注）：
import org.apache.dubbo.common.URL;
import org.apache.dubbo.common.extension.Adaptive;
import org.apache.dubbo.common.extension.SPI;
@SPI
public interface SpiIf {
@Adaptive
void test1(URL url);
@Adaptive
void test2(ObjHasUrl ohu);
void test3(URL url);
void test4(String name);
}
这里的ObjHasUrl是一个内部有URL属性的对象，为什么要有URL属性的原因下面会说到。
下一步，定义两个实现类Spi1和Spi2：
public class Spi1 implements SpiIf {
@Override
public void test1(URL url) {
System.out.println("This is Spi1:test1");
}
@Override
public void test2(ObjHasUrl ohu) {
System.out.println("This is Spi1:test2");
}
@Override
public void test3(URL url) {
System.out.println("This is Spi1:test3");
}
@Override
public void test4(String name) {
System.out.println("This is Spi1:test4");
}
}
public class Spi2 implements SpiIf {
@Override
public void test1(URL url) {
System.out.println("This is Spi2:test1");
}
@Override
public void test2(ObjHasUrl ohu) {
System.out.println("This is Spi2:test2");
}
@Override
public void test3(URL url) {
System.out.println("This is Spi2:test3");
}
@Override
public void test4(String name) {
System.out.println("This is Spi2:test4");
}
}
最后一步，定义一个Runner测试启动器：
public class Runner {
public static void main(String[] args) {
URL url = new URL("dubbo", "123", 999);
url = url.addParameter("spi.if", "S2"); //设置url值，来获取 SpiTest的自适应扩展 S2。
SpiIf spiIf = ExtensionLoader.getExtensionLoader(SpiIf.class).getAdaptiveExtension();
spiIf.test1(url);
ObjHasUrl ohu = new ObjHasUrl(url);
spiIf.test2(ohu);
url = url.addParameter("spi.if", "S1");
SpiIf spiIf2 = ExtensionLoader.getExtensionLoader(SpiIf.class).getAdaptiveExtension();
spiIf.test2(ohu);
}
}
以上类可以直接拷贝到自己的本地工程DEBUG用。
当我们启功Runner的时候，第一步先看getExtensionLoader方法，这里开始进入Dubbo的代码：
public static <T> ExtensionLoader<T> getExtensionLoader(Class<T> type) {
if (type == null) {
throw new IllegalArgumentException("Extension type == null");
}
//验证是否是接口
if (!type.isInterface()) {
throw new IllegalArgumentException("Extension type (" + type + ") is not an interface!");
}
//验证是否有SPI注解
if (!withExtensionAnnotation(type)) {
throw new IllegalArgumentException("Extension type (" + type +
") is not an extension, because it is NOT annotated with @" + SPI.class.getSimpleName() + "!");
}
ExtensionLoader<T> loader = (ExtensionLoader<T>) EXTENSION_LOADERS.get(type);
if (loader == null) {
EXTENSION_LOADERS.putIfAbsent(type, new ExtensionLoader<T>(type));
loader = (ExtensionLoader<T>) EXTENSION_LOADERS.get(type);
}
return loader;
}
这一步主要是获取ExtensionLoader对象，主要是对接口类做一些验证，确认是扩展点（有SPI注解）。
第二步，进入ExtensionLoader的getAdaptiveExtension方法：
public T getAdaptiveExtension() {
//先找缓存
Object instance = cachedAdaptiveInstance.get();
if (instance == null) {
if (createAdaptiveInstanceError != null) {
throw new IllegalStateException("Failed to create adaptive instance: " +
createAdaptiveInstanceError.toString(),
createAdaptiveInstanceError);
}
//缓存没有，则开始创建
synchronized (cachedAdaptiveInstance) {
instance = cachedAdaptiveInstance.get();
if (instance == null) {
try {
instance = createAdaptiveExtension();
cachedAdaptiveInstance.set(instance);
} catch (Throwable t) {
createAdaptiveInstanceError = t;
throw new IllegalStateException("Failed to create adaptive instance: " + t.toString(), t);
}
}
}
}
return (T) instance;
}
这一步主要是获取接口的实现对象实列，继续分析实际创建扩展点的方法：
private T createAdaptiveExtension() {
try {
return injectExtension((T) getAdaptiveExtensionClass().newInstance());
} catch (Exception e) {
throw new IllegalStateException("Can't create adaptive extension " + type + ", cause: " + e.getMessage(), e);
}
}
这一步拆分成两步，第一步：getAdaptiveExtensionClass，第二步：injectExtension
第一步：
private Class<?> getAdaptiveExtensionClass() {
getExtensionClasses();
if (cachedAdaptiveClass != null) {
return cachedAdaptiveClass;
}
return cachedAdaptiveClass = createAdaptiveExtensionClass();
}
private Class<?> createAdaptiveExtensionClass() {
String code = new AdaptiveClassCodeGenerator(type, cachedDefaultName).generate();
ClassLoader classLoader = findClassLoader();
org.apache.dubbo.common.compiler.Compiler compiler = ExtensionLoader.getExtensionLoader(org.apache.dubbo.common.compiler.Compiler.class).getAdaptiveExtension();
return compiler.compile(code, classLoader);
}
重点来了，创建class的时候会生成一个code（实际是SpiIf$Adaptive->SpiIf一个实现类）：
import org.apache.dubbo.common.extension.ExtensionLoader;
public class SpiIf$Adaptive implements SpiIf {
public void test2(com.zf.server.authserver.spi.dubbospitest2.ObjHasUrl arg0) {
if (arg0 == null) throw new IllegalArgumentException("com.zf.server.authserver.spi.dubbospitest2.ObjHasUrl argument == null");
if (arg0.getUrl() == null) throw new IllegalArgumentException("com.zf.server.authserver.spi.dubbospitest2.ObjHasUrl argument getUrl() == null");
org.apache.dubbo.common.URL url = arg0.getUrl();
String extName = url.getParameter("spi.if");
if(extName == null) throw new IllegalStateException("Failed to get extension (com.zf.server.authserver.spi.dubbospitest2.SpiIf) name from url (" + url.toString() + ") use keys([spi.if])");
com.zf.server.authserver.spi.dubbospitest2.SpiIf extension = (com.zf.server.authserver.spi.dubbospitest2.SpiIf)ExtensionLoader.getExtensionLoader(com.zf.server.authserver.spi.dubbospitest2.SpiIf.class).getExtension(extName);
extension.test2(arg0);
}
public void test1(org.apache.dubbo.common.URL arg0) {
if (arg0 == null) throw new IllegalArgumentException("url == null");
org.apache.dubbo.common.URL url = arg0;
String extName = url.getParameter("spi.if");
if(extName == null) throw new IllegalStateException("Failed to get extension (com.zf.server.authserver.spi.dubbospitest2.SpiIf) name from url (" + url.toString() + ") use keys([spi.if])");
com.zf.server.authserver.spi.dubbospitest2.SpiIf extension = (com.zf.server.authserver.spi.dubbospitest2.SpiIf)ExtensionLoader.getExtensionLoader(com.zf.server.authserver.spi.dubbospitest2.SpiIf.class).getExtension(extName);
extension.test1(arg0);
}
public void test3(org.apache.dubbo.common.URL arg0) {
throw new UnsupportedOperationException("The method public abstract void com.zf.server.authserver.spi.dubbospitest2.SpiIf.test3(org.apache.dubbo.common.URL) of interface com.zf.server.authserver.spi.dubbospitest2.SpiIf is not adaptive method!");
}
public void test4(java.lang.String arg0) {
throw new UnsupportedOperationException("The method public abstract void com.zf.server.authserver.spi.dubbospitest2.SpiIf.test4(java.lang.String) of interface com.zf.server.authserver.spi.dubbospitest2.SpiIf is not adaptive method!");
}
}
仔细一看，其实就是为我们的接口生成了一个实现类。然后为Adaptive注解标注的方法生成了实际的内容（就是根据URL参数来获取实际的扩展类），这也解释的Adaptive注解的实际作用。
还有一点需要注意：test1提供的是URL的参数，test2提供的是包含URL属性的对象。它们的共同点就是都包含了一个URL，如果不提供会提示没有URL异常。具体原因可以自行分析以下方法：
new AdaptiveClassCodeGenerator(type, cachedDefaultName).generate();
最终在加载生成的实现类。
第二步：injectExtension
private T injectExtension(T instance) {
if (objectFactory == null) {
return instance;
}
try {
for (Method method : instance.getClass().getMethods()) {
if (!isSetter(method)) {
continue;
}
if (method.getAnnotation(DisableInject.class) != null) {
continue;
}
Class<?> pt = method.getParameterTypes()[0];
if (ReflectUtils.isPrimitives(pt)) {
continue;
}
try {
String property = getSetterProperty(method);
Object object = objectFactory.getExtension(pt, property);
if (object != null) {
method.invoke(instance, object);
}
} catch (Exception e) {
logger.error("Failed to inject via method " + method.getName()
+ " of interface " + type.getName() + ": " + e.getMessage(), e);
}
}
} catch (Exception e) {
logger.error(e.getMessage(), e);
}
return instance;
}
而这一步干啥呢？简单说就是循环set方法，如果参数类型也是一个自适应扩展点的话，继续上面的步骤拿到扩展点对象并反射注入，实现了Dubbo版的依赖注入。
至此，返回最终生成的对象-> SpiIf$Adaptive的实例并缓存在cachedAdaptiveInstance中，在Runner中就会根据url对应的参数值来获取对应的扩展类。
总结：
1、自适应扩展接口需要 SPI注解，方法需要Adaptive注解，Adaptive方法需要URL参数或者是有URL属性的对象参数；
2、最终会返回接口实现类对象 SpiIf$Adaptive，里面封装了根据url参数来获取扩展对象的方法；
