版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/qq_27529917/article/details/78454929
IOC，inversion of control 控制反转，有时候也叫DI，dependency injection 依赖注入，是一种代码解耦的方式。
在一个类中定义一个属性，正常情况下需要在此类中有对此属性赋值的代码，如setter方法，或者在构造函数中赋值，因此类和属性之间的关系已经确定下来了，类和属性间就有了一定的耦合关系。而IOC的精髓就是解耦，类中没有显式的对属性的赋值代码，同时属性的实际类型和值在运行时有系统动态的赋值，Spring对IOC做了很大的扩展，使用者可以很灵活的控制注入的对象类型及值。
Spring内IOC的最直接体现就是@Autowired注解，最常用的方式就是表示在属性上，Spring容器在启动时会将容器内类型是标识了@Autowired的属性类型或者其子类，实现类的Bean通过反射的形式赋值给此属性，或者叫注入的此类中。
Spring中对@Autowired注解的解析是通过一个叫AutowiredAnnotationBeanPostProcessor的BeanPostProcessor(Bean的后处理器)来进行的。BeanPostProcessor负责Bean的一系列处理，实例化前后，初始化前后灯阶段执行相应方法，具体方法的执行顺序如下:
BeanPostPrecessors Invoke Procedure(执行流程)
1.InstantiationAwareBeanPostProcessor.postProcessBeforeInstantiation(Class<?> beanClass, String beanName);
invoke before doCreateBean(),if the method returns not null,the return Object will as the Bean, avoid invoke doCreateBean() ;
2.SmartInstantiationAwareBeanPostProcessor.determineCandidateConstructors(Class<?> beanClass, String beanName);
used to determine to use which Constructor to instance Bean;
3.MergedBeanDefinitionPostProcessor.postProcessMergedBeanDefinition(RootBeanDefinition beanDefinition, Class<?> beanType, String beanName);
the Bean gets instantiated, modify the RootBeanDefinition ,such as adding PropertyValue etc. before populate Bean;
4.SmartInstantiationAwareBeanPostProcessor.getEarlyBeanReference(Object bean, String beanName);
invoke before populate Bean,used for resolve Circular Reference;
5.InstantiationAwareBeanPostProcessor.postProcessAfterInstantiation(Object bean, String beanName);
invoke in populate Bean, but before fill in values，return false will break the opreation for filling values in Bean;
6.InstantiationAwareBeanPostProcessor.postProcessPropertyValues(PropertyValues pvs, PropertyDescriptor[] pds, Object bean, String beanName);
invoke in populate Bean, but before fill in values，modify the PropertyValues which will be apply for the Bean,
return null will break the opreation for filling values in Bean;
7.BeanPostProcessor.postProcessBeforeInitialization(Object bean, String beanName);
invoked before invokeInitMethods, modify the finally Bean instance
8.BeanPostProcessor.postProcessAfterInitialization(Object bean, String beanName);
invoked after invokeInitMethods, modify the finally Bean instance使用者可以自己定义自己的BeanPostProcessor，将其以Bean的形式注册到Spring容器中，Spring容器在启动时会执行其相应方法。
AutowiredAnnotationBeanPostProcessor在Spring容器初始化时会最先执行determineCandidateConstructors方法，先看起源码:
public class AutowiredAnnotationBeanPostProcessor extends InstantiationAwareBeanPostProcessorAdapter
implements MergedBeanDefinitionPostProcessor, PriorityOrdered, BeanFactoryAware {
//此方法的含义是解析Bean类里构造函数上的@Autowired注解，如果有合适的标识了@Autowired的构造函数,
//则在实例化此Bean时会使用此构造函数，@Import和@ComponentScan得到的Bean会如此解析，
//@Bean标识方法生成的Bean不会如此
public Constructor<?>[] determineCandidateConstructors(Class<?> beanClass, final String beanName)
throws BeanCreationException {
// Let's check for lookup methods here..
if (!this.lookupMethodsChecked.contains(beanName)) {
try {
ReflectionUtils.doWithMethods(beanClass, new ReflectionUtils.MethodCallback() {
@Override
public void doWith(Method method) throws IllegalArgumentException, IllegalAccessException {
Lookup lookup = method.getAnnotation(Lookup.class);
if (lookup != null) {
LookupOverride override = new LookupOverride(method, lookup.value());
try {
RootBeanDefinition mbd = (RootBeanDefinition) beanFactory.getMergedBeanDefinition(beanName);
mbd.getMethodOverrides().addOverride(override);
}
catch (NoSuchBeanDefinitionException ex) {
throw new BeanCreationException(beanName,
"Cannot apply @Lookup to beans without corresponding bean definition");
}
}
}
});
}
catch (IllegalStateException ex) {
throw new BeanCreationException(beanName, "Lookup method resolution failed", ex);
}
catch (NoClassDefFoundError err) {
throw new BeanCreationException(beanName, "Failed to introspect bean class [" + beanClass.getName() +
"] for lookup method metadata: could not find class that it depends on", err);
}
this.lookupMethodsChecked.add(beanName);
}
// Quick check on the concurrent map first, with minimal locking.
Constructor<?>[] candidateConstructors = this.candidateConstructorsCache.get(beanClass);
if (candidateConstructors == null) {
// Fully synchronized resolution now...
synchronized (this.candidateConstructorsCache) {
//对每个类的构造函数只解析一次，解析完会存储结果，以备下次服用
candidateConstructors = this.candidateConstructorsCache.get(beanClass);
if (candidateConstructors == null) {
Constructor<?>[] rawCandidates;
try {
rawCandidates = beanClass.getDeclaredConstructors();
}
catch (Throwable ex) {
throw new BeanCreationException(beanName,
"Resolution of declared constructors on bean Class [" + beanClass.getName() +
"] from ClassLoader [" + beanClass.getClassLoader() + "] failed", ex);
}
List<Constructor<?>> candidates = new ArrayList<Constructor<?>>(rawCandidates.length);
Constructor<?> requiredConstructor = null;
Constructor<?> defaultConstructor = null;
for (Constructor<?> candidate : rawCandidates) {
//查看构造函数上是否标识@Autowired注解，获取注解的属性
AnnotationAttributes ann = findAutowiredAnnotation(candidate);
if (ann == null) {
Class<?> userClass = ClassUtils.getUserClass(beanClass);
if (userClass != beanClass) {
try {
Constructor<?> superCtor =
userClass.getDeclaredConstructor(candidate.getParameterTypes());
//如果此构造函数是重载了父类的构造函数，则寻找父类的构造函数，查看是否标识@Autowired注解，获取注解的属性
ann = findAutowiredAnnotation(superCtor);
}
catch (NoSuchMethodException ex) {
// Simply proceed, no equivalent superclass constructor found...
}
}
}
//当构造函数上存在@Autowired注解
if (ann != null) {
//@Autowired默认的required = true,当有一个required = true的@Autowired标识的构造函数时，
//不能有其他的构造函数再标识@Autowired，否则会报错
if (requiredConstructor != null) {
throw new BeanCreationException(beanName,
"Invalid autowire-marked constructor: " + candidate +
". Found constructor with 'required' Autowired annotation already: " +
requiredConstructor);
}
//@Autowired的required的属性值，true / false
boolean required = determineRequiredStatus(ann);
if (required) {
if (!candidates.isEmpty()) {
throw new BeanCreationException(beanName,
"Invalid autowire-marked constructors: " + candidates +
". Found constructor with 'required' Autowired annotation: " +
candidate);
}
requiredConstructor = candidate;
}
candidates.add(candidate);
}
else if (candidate.getParameterTypes().length == 0) {
defaultConstructor = candidate;
}
}
if (!candidates.isEmpty()) {
// Add default constructor to list of optional constructors, as fallback.
////当没有构造函数标识@Autowired时，设置默认的构造函数作为额外选择
if (requiredConstructor == null) {
if (defaultConstructor != null) {
candidates.add(defaultConstructor);
} //如果没有默认的无参构造函数，且有@Autowired（required = false）的构造函数，则发出警告信
else if (candidates.size() == 1 && logger.isWarnEnabled()) {
logger.warn("Inconsistent constructor declaration on bean with name '" + beanName +
"': single autowire-marked constructor flagged as optional - " +
"this constructor is effectively required since there is no " +
"default constructor to fall back to: " + candidates.get(0));
}
}
candidateConstructors = candidates.toArray(new Constructor<?>[candidates.size()]);
} //若只有一个构造函数，且没标识@Autowired，其参数长度>0，将其作为候选者
else if (rawCandidates.length == 1 && rawCandidates[0].getParameterTypes().length > 0) {
candidateConstructors = new Constructor<?>[] {rawCandidates[0]};
} //没有合适的构造函数
else {
candidateConstructors = new Constructor<?>[0];
}
this.candidateConstructorsCache.put(beanClass, candidateConstructors);
}
}
}
return (candidateConstructors.length > 0 ? candidateConstructors : null);
}
}AutowiredAnnotationBeanPostProcessor筛选@Autowired标识的构造函数的代码就是这些，筛选出这些构造函数之后，Spring使用ConstructorResolver这个类来择取合适的构造函数，流程如下:
- 首先对这些构造函数按修饰符优先public排序，修饰符相同再按参数的长短排序，最先解析参数最长的public修饰的构造函数，其优先级最高.
- 对构造函数的每一个参数解析，如果每一个参数均能从Spring容器中找到合适的Bean，则此将此构造函数作为最优解，如果容器内Bean不能满足所有参数，则解析下一个构造函数。
- 如果存在两个参数长度相同的构造函数，且容器内Bean均能满足参数解析，则按参数类型和Bean类型的差异性求取参数的差异权重，比如参数是接口，Bean是实现类，则差异加2，参数是集合，Bean是单个类，则转换成集合，差异加2等等，比较两个构造函数的差异权重大小，差异小的那个作为最优解。如果两个差异权重相等，则抛出含有模棱两可的构造函数的BeanCreationException。
- 当有了最优解的构造函数后，如果下一个构造函数的参数长度等于最优解，则解析此构造函数，如果参数长度小于最优解，则不再解析，直接忽略之后的所有构造函数。
- 当得到了构造函数最优解之后，将此构造函数存入此Bean的BeanDefinition中，以备下次复用，就是说只对构造函数候选者集合解析一次，下次实例化Bean的时候可以直接得到这个最优解。
- 以上的情况是基于Spring容器实例化Bean的情况，就是实例化时不会附带Arguments,就是不带参数，如果是使用者自己实例化Bean时，通过BeanFactory的 Object getBean(String name, Object… args) throws BeansException; T getBean(Class requiredType, Object… args) throws BeansException;方法实例化时，按照构造函数的顺序传入实例化参数，则Spring在找合适的构造函数时会忽略之前缓存的最优解，以同样的顺序解析构造函数候选者集合，看看哪些构造函数的参数适合传入的参数，找不到合适的则跑出BeanCreationException。
以上就是@Autowired注解在构造函数上的用法的原理及流程。
重点: BeanFactory的getBean（）方法获取scope = singleton的Bean时，不会生成新的Bean对象，
在scope为request及session的生命周期内，Bean的实例化只会触发一次，也就是说@Autowired的标识的构造函数不是每次调用getBean()均会触发执行。只有scope = prototype的Bean，才会每次均执行。