元注解是指注解的注解，包括@Retention @Target @Document @Inherited四种。
1.@Retention: 定义注解的保留策略
@Retention(RetentionPolicy.SOURCE) //注解仅存在于源码中，在class字节码文件中不包含
@Retention(RetentionPolicy.CLASS) // 默认的保留策略，注解会在class字节码文件中存在，但运行时无法获得，
@Retention(RetentionPolicy.RUNTIME) // 注解会在class字节码文件中存在，在运行时可以通过反射获取到
首 先要明确生命周期长度 SOURCE < CLASS < RUNTIME ，所以前者能作用的地方后者一定也能作用。一般如果需要在运行时去动态获取注解信息，那只能用 RUNTIME 注解；如果要在编译时进行一些预处理操作，比如生成一些辅助代码（如 ButterKnife），就用 CLASS注解；如果只是做一些检查性的操作，比如 @Override 和 @SuppressWarnings，则可选用 SOURCE 注解。
2.@Target：定义注解的作用目标
源码为：
@Documented
@Retention(RetentionPolicy.RUNTIME)
@Target(ElementType.ANNOTATION_TYPE)
public @interface Target {
ElementType[] value();
}
@Target(ElementType.TYPE) //接口、类、枚举、注解
@Target(ElementType.FIELD) //字段、枚举的常量
@Target(ElementType.METHOD) //方法
@Target(ElementType.PARAMETER) //方法参数
@Target(ElementType.CONSTRUCTOR) //构造函数
@Target(ElementType.LOCAL_VARIABLE)//局部变量
@Target(ElementType.ANNOTATION_TYPE)//注解
@Target(ElementType.PACKAGE) ///包
3.@Document：说明该注解将被包含在javadoc中
4.@Inherited：说明子类可以继承父类中的该注解
2.反射获取注解
举例：
// 适用类、接口（包括注解类型）或枚举
@Retention(RetentionPolicy.RUNTIME)
@Target(ElementType.TYPE)
public @interface ClassInfo {
String value();
}
// 适用field属性，也包括enum常量
@Retention(RetentionPolicy.RUNTIME)
@Target(ElementType.FIELD)
public @interface FieldInfo {
int[] value();
}
// 适用方法
@Retention(RetentionPolicy.RUNTIME)
@Target(ElementType.METHOD)
public @interface MethodInfo {
String name() default "long";
String data();
int age() default 27;
}
这3个注解分别适用于不同的元素，并都带有不同的属性，在使用注解是需要设置这些属性值。
再定义一个测试类来使用这些注解：
@ClassInfo("Test Class")
public class TestRuntimeAnnotation {
@FieldInfo(value = {1, 2})
public String fieldInfo = "FiledInfo";
@FieldInfo(value = {10086})
public int i = 100;
@MethodInfo(name = "BlueBird", data = "Big")
public static String getMethodInfo() {
return TestRuntimeAnnotation.class.getSimpleName();
}
}
在代码中获取注解信息：
private void _testRuntimeAnnotation() {
StringBuffer sb = new StringBuffer();
Class<?> cls = TestRuntimeAnnotation.class;
Constructor<?>[] constructors = cls.getConstructors();
// 获取指定类型的注解
sb.append("Class注解：").append("\n");
ClassInfo classInfo = cls.getAnnotation(ClassInfo.class);
if (classInfo != null) {
sb.append(Modifier.toString(cls.getModifiers())).append(" ")
.append(cls.getSimpleName()).append("\n");
sb.append("注解值: ").append(classInfo.value()).append("\n\n");
}
sb.append("Field注解：").append("\n");
Field[] fields = cls.getDeclaredFields();
for (Field field : fields) {
FieldInfo fieldInfo = field.getAnnotation(FieldInfo.class);
if (fieldInfo != null) {
sb.append(Modifier.toString(field.getModifiers())).append(" ")
.append(field.getType().getSimpleName()).append(" ")
.append(field.getName()).append("\n");
sb.append("注解值: ").append(Arrays.toString(fieldInfo.value())).append("\n\n");
}
}
sb.append("Method注解：").append("\n");
Method[] methods = cls.getDeclaredMethods();
for (Method method : methods) {
MethodInfo methodInfo = method.getAnnotation(MethodInfo.class);
if (methodInfo != null) {
sb.append(Modifier.toString(method.getModifiers())).append(" ")
.append(method.getReturnType().getSimpleName()).append(" ")
.append(method.getName()).append("\n");
sb.append("注解值: ").append("\n");
sb.append("name: ").append(methodInfo.name()).append("\n");
sb.append("data: ").append(methodInfo.data()).append("\n");
sb.append("age: ").append(methodInfo.age()).append("\n");
}
}
System.out.print(sb.toString());
}
3.Java源代码如何处理注解
（转载自：https://www.jianshu.com/p/e9329c8a59c2 作者：测试你个头）
jdk中是通过AnnotatedElement（package java.lang.reflect）接口实现对注解的解析，我们的Class类实现了AnnotatedElement接口
public final class Class<T> implements java.io.Serializable,
GenericDeclaration,
Type,
AnnotatedElement {
......
}
AnnotatedElement代码：
AnnotatedElement的注释：
Represents an annotated element of the program currently running in this VM. This interface allows annotations to be read reflectively
翻译过来就是：AnnotatedElement代表了jvm中一个正在运行的被注解元素，这个接口允许通过反射的方式读取注解
可以看下Class类中对于AnnotatedElement接口都是如何实现的：
/**
* @throws NullPointerException {@inheritDoc}
* @since 1.5
*/
@SuppressWarnings("unchecked")
public <A extends Annotation> A getAnnotation(Class<A> annotationClass) {
Objects.requireNonNull(annotationClass);
return (A) annotationData().annotations.get(annotationClass);
}
/**
* {@inheritDoc}
* @throws NullPointerException {@inheritDoc}
* @since 1.5
*/
@Override
public boolean isAnnotationPresent(Class<? extends Annotation> annotationClass) {
return GenericDeclaration.super.isAnnotationPresent(annotationClass);
}
/**
* @throws NullPointerException {@inheritDoc}
* @since 1.8
*/
@Override
public <A extends Annotation> A[] getAnnotationsByType(Class<A> annotationClass) {
Objects.requireNonNull(annotationClass);
AnnotationData annotationData = annotationData();
return AnnotationSupport.getAssociatedAnnotations(annotationData.declaredAnnotations,
this,
annotationClass);
}
/**
* @since 1.5
*/
public Annotation[] getAnnotations() {
return AnnotationParser.toArray(annotationData().annotations);
}
/**
* @throws NullPointerException {@inheritDoc}
* @since 1.8
*/
@Override
@SuppressWarnings("unchecked")
public <A extends Annotation> A getDeclaredAnnotation(Class<A> annotationClass) {
Objects.requireNonNull(annotationClass);
return (A) annotationData().declaredAnnotations.get(annotationClass);
}
/**
* @throws NullPointerException {@inheritDoc}
* @since 1.8
*/
@Override
public <A extends Annotation> A[] getDeclaredAnnotationsByType(Class<A> annotationClass) {
Objects.requireNonNull(annotationClass);
return AnnotationSupport.getDirectlyAndIndirectlyPresent(annotationData().declaredAnnotations,
annotationClass);
}
/**
* @since 1.5
*/
public Annotation[] getDeclaredAnnotations() {
return AnnotationParser.toArray(annotationData().declaredAnnotations);
}
上面的接口实现中，大致的原理都是一致的，我们挑选其中的getAnnotation方法来讲解：
- getAnnotation
根据注解的class实例从类的注解缓存数据中获取匹配的注解类型
Controller是注解类型，Controller.getClass()获取到的就是Class实例 1、代码中annotationData().annotations是一个Map（key为注解的Class实例，value为注解类型），源码为：
// annotation data that might get invalidated when JVM TI RedefineClasses() is called
private static class AnnotationData {
final Map<Class<? extends Annotation>, Annotation> annotations;
final Map<Class<? extends Annotation>, Annotation> declaredAnnotations;
// Value of classRedefinedCount when we created this AnnotationData instance
final int redefinedCount;
AnnotationData(Map<Class<? extends Annotation>, Annotation> annotations,
Map<Class<? extends Annotation>, Annotation> declaredAnnotations,
int redefinedCount) {
this.annotations = annotations;
this.declaredAnnotations = declaredAnnotations;
this.redefinedCount = redefinedCount;
}
}
2、annotationData()的源码是：
private AnnotationData annotationData() {
while (true) { // retry loop
AnnotationData annotationData = this.annotationData;
int classRedefinedCount = this.classRedefinedCount;
if (annotationData != null &&
annotationData.redefinedCount == classRedefinedCount) {
return annotationData;
}
// null or stale annotationData -> optimistically create new instance
AnnotationData newAnnotationData = createAnnotationData(classRedefinedCount);
// try to install it
if (Atomic.casAnnotationData(this, annotationData, newAnnotationData)) {
// successfully installed new AnnotationData
return newAnnotationData;
}
}
}
核心的逻辑是：当this.annotationData为空，解析类中的annotationData并写入this.annotationData，最后都会返回this.annotationData
3、其中Atomic.casAnnotationData(this, annotationData, newAnnotationData)的作用便是将解析到的annotationData写入this.annotationData：
static <T> boolean casAnnotationData(Class<?> clazz,
AnnotationData oldData,
AnnotationData newData) {
return unsafe.compareAndSwapObject(clazz, annotationDataOffset, oldData, newData);
}
其中unsafe.compareAndSwapObject是一个native方法
4、而createAnnotationData(classRedefinedCount)的作用是解析类中用到的annotationData
private AnnotationData createAnnotationData(int classRedefinedCount) {
Map<Class<? extends Annotation>, Annotation> declaredAnnotations =
AnnotationParser.parseAnnotations(getRawAnnotations(), getConstantPool(), this);
Class<?> superClass = getSuperclass();
Map<Class<? extends Annotation>, Annotation> annotations = null;
if (superClass != null) {
Map<Class<? extends Annotation>, Annotation> superAnnotations =
superClass.annotationData().annotations;
for (Map.Entry<Class<? extends Annotation>, Annotation> e : superAnnotations.entrySet()) {
Class<? extends Annotation> annotationClass = e.getKey();
if (AnnotationType.getInstance(annotationClass).isInherited()) {
if (annotations == null) { // lazy construction
annotations = new LinkedHashMap<>((Math.max(
declaredAnnotations.size(),
Math.min(12, declaredAnnotations.size() + superAnnotations.size())
) * 4 + 2) / 3
);
}
annotations.put(annotationClass, e.getValue());
}
}
}
if (annotations == null) {
// no inherited annotations -> share the Map with declaredAnnotations
annotations = declaredAnnotations;
} else {
// at least one inherited annotation -> declared may override inherited
annotations.putAll(declaredAnnotations);
}
return new AnnotationData(annotations, declaredAnnotations, classRedefinedCount);
}
整个的处理逻辑是：
1、获取类本身的declaredAnnotations
2、获取父类的annotations
3、将declaredAnnotations+annotations整合，返回
Annotation解析的范例代码：
@Component
public class SSHClient {
......
}
public class AnnotationHelper {
public static void main(String[] args) {
Annotation[] annotations = new Annotation[0];
annotations = SSHClient.class.getAnnotations();
for (Annotation annotation : annotations) {
System.out.println(annotation.toString());
System.out.println(annotation.annotationType());
System.out.println(annotation.getClass().getName());
System.out.println(annotation.getClass().getTypeName());
System.out.println(annotation.getClass().toString());
}
if (SSHClient.class.isAnnotationPresent(Component.class)) {
System.out.println("find Component annotation");
}
Annotation annotation = SSHClient.class.getAnnotation(Component.class);
System.out.println(annotation.toString());
System.out.println(annotation.annotationType());
System.out.println(annotation.getClass().getName());
System.out.println(annotation.getClass().getTypeName());
System.out.println(annotation.getClass().toString());
}
}
执行结果:
可以看到，通过Annotation接口中定义的annotationType()可以获取Annotation的类型
实际应用中，比如spring框架中对注解的解析有专门的工具类，但是都是基于AnnotatedElement中定义的方法来实现的
以上，就是整个元注解和注解解析相关的讲解。
