技术分享

Spring Boot 中 @EnableXXX 注解的驱动逻辑

@Enable驱动逻辑

前言

工作中经常用到,如下注解:@EnableEurekaClient@EnableFeignClients@EnableCircuitBreaker@EnableHystrix他们都是@Enable开头,各自实现不同的功能,解析这种@Enable的逻辑是什么呢?

@Enable驱动逻辑

找入口

@Enable的模块驱动,依赖于@Import实现。@Import作用是装载导入类,主要包括@Configuration class,ImportSelector实现类,ImportBeanDefinitionRegistrar实现类。XML时代,经常是@Import,<context:component-scan>,<context:annotation-config>.一起使用。<context:annotation-config>(注解配置)中大概率有我们需要找的逻辑。根据 Spring Framework2.0引入的可扩展的XML编程机制,XML Schema命名空间需要与Handler建立映射关系。该关系配置在相对于classpath下的/META-INF/spring.handlers中。
查看ContextNamespaceHandler 源码

public class ContextNamespaceHandler extends NamespaceHandlerSupport {    @Override    public void init() {        //省略其他代码     registerBeanDefinitionParser("annotation-config",                      new AnnotationConfigBeanDefinitionParser());   } }复制代码

<context:annotation-config> 对应AnnotationConfigBeanDefinitionParser这个就是要找的入口


找核心类

从AnnotationConfigBeanDefinitionParser的parse方法开始一路向下,找到AnnotationConfigUtils.registerAnnotationConfigProcessors中注册了ConfigurationClassPostProcessor。


ConfigurationClassPostProcessor类注释说明

1. 用于的引导处理@Configuration类2. <context:annotation-config/>或 <context:component-scan/>时会注册否则需要手工编程3. ConfigurationClassPostProcessor第一优先级,保证@Configuration}类中声明@Bean,在其他 BeanFactoryPostProcessor执行之前被注册

扩展

AnnotationConfigApplicationContext中new AnnotationBeanDefinitionReader也调用了 AnnotationConfigUtils .registerAnnotationConfigProcessors

从类注释中,可以看出ConfigurationClassPostProcessor就是要找的核心类找核心方法查看 ConfigurationClassPostProcessor 的层级关系为

Aware系列注入相应资源,Ordered设置优先级,值得关注的就是postProcessBeanDefinitionRegistry了。postProcessBeanDefinitionRegistry其内部有2个方法

  1. postProcessBeanDefinitionRegistry在BeanDefinition注册之后,BeanFactoryPostProcessor执行之前,修改或重写BeanDefinition

  2. 继承自BeanFactoryPostProcessor的postProcessBeanFactory,BeanDefinition加载之后,Bean实例化之前,重写或添加BeanDefinition,修改BeanFactory

浏览2个方法,都有processConfigBeanDefinitions,从名称可以看出是处理配置类Bean定义

ConfigurationClassPostProcessor#processConfigBeanDefinitions就是要找的核心方法


梳理流程

public void processConfigBeanDefinitions(BeanDefinitionRegistry registry) {    List<BeanDefinitionHolder> configCandidates = new ArrayList<>();    String[] candidateNames = registry.getBeanDefinitionNames();    for (String beanName : candidateNames) {       BeanDefinition beanDef = registry.getBeanDefinition(beanName);       if (ConfigurationClassUtils.isFullConfigurationClass(beanDef) ||             ConfigurationClassUtils.isLiteConfigurationClass(beanDef)) {          if (logger.isDebugEnabled()) {             logger.debug("Bean definition has already been processed as a configuration class: " + beanDef);          }       }       else if (ConfigurationClassUtils.checkConfigurationClassCandidate(beanDef, this.metadataReaderFactory)) {          configCandidates.add(new BeanDefinitionHolder(beanDef, beanName));       }    }    // 没有找到 @Configuration classes 立即返回    if (configCandidates.isEmpty()) {       return;    }    //根据@Order 值进行排序    configCandidates.sort((bd1, bd2) -> {       int i1 = ConfigurationClassUtils.getOrder(bd1.getBeanDefinition());       int i2 = ConfigurationClassUtils.getOrder(bd2.getBeanDefinition());       return Integer.compare(i1, i2);    });    //通过封闭的应用程序上下文, 检测任何自定义bean名称生成策略    supplied through the enclosing application context    SingletonBeanRegistry sbr = null;    if (registry instanceof SingletonBeanRegistry) {       sbr = (SingletonBeanRegistry) registry;       if (!this.localBeanNameGeneratorSet) {          BeanNameGenerator generator = (BeanNameGenerator) sbr.getSingleton(CONFIGURATION_BEAN_NAME_GENERATOR);          if (generator != null) {             this.componentScanBeanNameGenerator = generator;             this.importBeanNameGenerator = generator;          }       }    }    if (this.environment == null) {       this.environment = new StandardEnvironment();    }    // 解析@Configuration class    ConfigurationClassParser parser = new ConfigurationClassParser(          this.metadataReaderFactory, this.problemReporter, this.environment,          this.resourceLoader, this.componentScanBeanNameGenerator, registry);    Set<BeanDefinitionHolder> candidates = new LinkedHashSet<>(configCandidates);    Set<ConfigurationClass> alreadyParsed = new HashSet<>(configCandidates.size());    do {       parser.parse(candidates);       parser.validate();       Set<ConfigurationClass> configClasses = new LinkedHashSet<>(parser.getConfigurationClasses());       configClasses.removeAll(alreadyParsed);       // 读ConfigurationClass的信息,创建BeanDefinition       if (this.reader == null) {          this.reader = new ConfigurationClassBeanDefinitionReader(                registry, this.sourceExtractor, this.resourceLoader, this.environment,                this.importBeanNameGenerator, parser.getImportRegistry());       }       this.reader.loadBeanDefinitions(configClasses);       alreadyParsed.addAll(configClasses);       candidates.clear();       if (registry.getBeanDefinitionCount() > candidateNames.length) {          String[] newCandidateNames = registry.getBeanDefinitionNames();          Set<String> oldCandidateNames = new HashSet<>(Arrays.asList(candidateNames));          Set<String> alreadyParsedClasses = new HashSet<>();          for (ConfigurationClass configurationClass : alreadyParsed) {             alreadyParsedClasses.add(configurationClass.getMetadata().getClassName());          }          for (String candidateName : newCandidateNames) {             if (!oldCandidateNames.contains(candidateName)) {                BeanDefinition bd = registry.getBeanDefinition(candidateName);                if (ConfigurationClassUtils.checkConfigurationClassCandidate(bd, this.metadataReaderFactory) &&                      !alreadyParsedClasses.contains(bd.getBeanClassName())) {                   candidates.add(new BeanDefinitionHolder(bd, candidateName));                }             }          }          candidateNames = newCandidateNames;       }    }    while (!candidates.isEmpty());    // 将ImportRegistry注册为bean以支持importware@Configuration类    if (sbr != null && !sbr.containsSingleton(IMPORT_REGISTRY_BEAN_NAME)) {       sbr.registerSingleton(IMPORT_REGISTRY_BEAN_NAME, parser.getImportRegistry());    }    if (this.metadataReaderFactory instanceof CachingMetadataReaderFactory) {       // Clear cache in externally provided MetadataReaderFactory; this is a no-op       // for a shared cache since it'll be cleared by the ApplicationContext.       ((CachingMetadataReaderFactory) this.metadataReaderFactory).clearCache();    } }复制代码

ConfigurationClassPostProcessor#processConfigBeanDefinitions核心如下:

  1. 根据@Order 值进行排序

  2. 解析@Configuration class 为ConfigurationClass对象

  3. 读ConfigurationClass的信息,创建BeanDefinition

  4. 将ImportRegistry注册为bean以支持importware@Configuration类

重点关注解析方法

ConfigurationClassParser#parse方法负责解析@Configuration class 为ConfigurationClass对象查阅其源码如下:ConfigurationClassParser#doProcessConfigurationClass代码如下:

protected final SourceClass doProcessConfigurationClass(ConfigurationClass configClass, SourceClass sourceClass)       throws IOException {    if (configClass.getMetadata().isAnnotated(Component.class.getName())) {       // Recursively process any member (nested) classes first       processMemberClasses(configClass, sourceClass);    }    // Process any @PropertySource annotations    for (AnnotationAttributes propertySource : AnnotationConfigUtils.attributesForRepeatable(          sourceClass.getMetadata(), PropertySources.class,          org.springframework.context.annotation.PropertySource.class)) {       if (this.environment instanceof ConfigurableEnvironment) {          processPropertySource(propertySource);       }       else {          logger.info("Ignoring @PropertySource annotation on [" + sourceClass.getMetadata().getClassName() +                "]. Reason: Environment must implement ConfigurableEnvironment");       }    }    // Process any @ComponentScan annotations    Set<AnnotationAttributes> componentScans = AnnotationConfigUtils.attributesForRepeatable(          sourceClass.getMetadata(), ComponentScans.class, ComponentScan.class);    if (!componentScans.isEmpty() &&          !this.conditionEvaluator.shouldSkip(sourceClass.getMetadata(), ConfigurationPhase.REGISTER_BEAN)) {       for (AnnotationAttributes componentScan : componentScans) {          // The config class is annotated with @ComponentScan -> perform the scan immediately          Set<BeanDefinitionHolder> scannedBeanDefinitions =                this.componentScanParser.parse(componentScan, sourceClass.getMetadata().getClassName());          // Check the set of scanned definitions for any further config classes and parse recursively if needed          for (BeanDefinitionHolder holder : scannedBeanDefinitions) {             BeanDefinition bdCand = holder.getBeanDefinition().getOriginatingBeanDefinition();             if (bdCand == null) {                bdCand = holder.getBeanDefinition();             }             if (ConfigurationClassUtils.checkConfigurationClassCandidate(bdCand, this.metadataReaderFactory)) {                parse(bdCand.getBeanClassName(), holder.getBeanName());             }          }       }    }    // Process any @Import annotations    processImports(configClass, sourceClass, getImports(sourceClass), true);    // Process any @ImportResource annotations    AnnotationAttributes importResource =          AnnotationConfigUtils.attributesFor(sourceClass.getMetadata(), ImportResource.class);    if (importResource != null) {       String[] resources = importResource.getStringArray("locations");       Class<? extends BeanDefinitionReader> readerClass = importResource.getClass("reader");       for (String resource : resources) {          String resolvedResource = this.environment.resolveRequiredPlaceholders(resource);          configClass.addImportedResource(resolvedResource, readerClass);       }    }    // Process individual @Bean methods    Set<MethodMetadata> beanMethods = retrieveBeanMethodMetadata(sourceClass);    for (MethodMetadata methodMetadata : beanMethods) {       configClass.addBeanMethod(new BeanMethod(methodMetadata, configClass));    }    // Process default methods on interfaces    processInterfaces(configClass, sourceClass);    // Process superclass, if any    if (sourceClass.getMetadata().hasSuperClass()) {       String superclass = sourceClass.getMetadata().getSuperClassName();       if (superclass != null && !superclass.startsWith("java") &&             !this.knownSuperclasses.containsKey(superclass)) {          this.knownSuperclasses.put(superclass, configClass);          // Superclass found, return its annotation metadata and recurse          return sourceClass.getSuperClass();       }    }    // No superclass -> processing is complete    return null; }复制代码

ConfigurationClassParser#doProcessConfigurationClass(ConfigurationClass,AnnatationMetatdata)将@PropertySource, @ComponentScan, @Import,@ImportResource,@Bean等一起处理了。看到这里基本逻辑已经理清 了,但是有一个疑问@Configuration中的@Bean没有其他特殊处理吗?


浏览代码解决疑问

从上边浏览的代码可以看到完全模式,会被AOP增强那什么是完全模式呢? 在ConfigurationClassUtils找到如下方法:

public class ConfigurationClassUtils{ //省略其他方法 public static boolean isFullConfigurationCandidate(AnnotationMetadata metadata) {    return metadata.isAnnotated(Configuration.class.getName()); } }复制代码

即 @Configuration class是完全模式,@Component,@Bean是轻量级模式那AOP增强了作用是什么呢?查看 ConfigurationClassEnhancer 的类注释如下:


/*** Enhances {@linkConfiguration} classes by generating a CGLIB subclass which* interacts with the Spring container to respect bean scoping semantics for* {@code @Bean} methods. Each such {@code @Bean} method will be overridden in* the generated subclass, only delegating to the actual {@code @Bean} method* implementation if the container actually requests the construction of a new* instance. Otherwise, a call to such an {@code @Bean} method serves as a* reference back to the container, obtaining the corresponding bean by name.** @author Chris Beams* @author Juergen Hoeller* @since 3.0* @see #enhance* @seeConfigurationClassPostProcessor*/class ConfigurationClassEnhancer {


大概意思如下:通过CGLIB增强@Configuration class。每个@Bean方法会生成子类。首次被调用时,@Bean方法会被执行用于创建bean实例;再次被调用时,不会再执行创建bean实例,而是根据bean名称返回首次该方法被执行时创建的bean实例。


总结

1.ConfigurationClassPostProcessor负责筛选@Component Class、@Configuration Class以及@Bean定义的Bean,**2.ConfigurationClassParser从候选的Bean定义中解析出ConfigurationClass集合,随后被3.ConfigurationClassBeanDefinitionReader转换为BeanDefinition4.ConfigurationClassParser的解析顺序,@PropertySource->@ComponentScan->@Import->@ImportResource->@Bean->接口的默认方法->处理父类5.@Configuration class是完全模式,@Component,@Bean是轻量级模式6.CGLIB增强@Configuration class。每个@Bean方法会生成子类。首次被调用时,@Bean方法会被执行用于创建bean实例;再次被调用时,不会再执行创建bean实例,而是根据bean名称返回首次该方法被执行时创建的bean实例。