學(xué)習(xí)筆記是學(xué)習(xí)了 小馬哥在慕課網(wǎng)的 《Spring Boot 2.0深度實(shí)踐之核心技術(shù)篇》根據(jù)自己的需要和理解做的筆記皱坛。

應(yīng)用上下文初始器(ApplicationContextInitializer)

利用Spring工廠機(jī)制,實(shí)現(xiàn)類org.springframework.core.io.support.SpringFactoriesLoader

實(shí)例化 ApplicationContextInitializer 實(shí)現(xiàn)類着裹，并排序?qū)ο蠹稀?/p>

自己追了一下源碼,發(fā)現(xiàn)Spring-boot加載應(yīng)用文上下初始器的順序是

org.springframework.boot.SpringApplication#SpringApplication()->org.springframework.core.io.support.SpringFactoriesLoader#loadFactoryNames->org.springframework.boot.SpringApplicatior#run()

具體代碼實(shí)現(xiàn)org.springframework.boot.SpringApplication#SpringApplication()

1.spring-boot使用setInitializers方法加載所有實(shí)現(xiàn)ApplicationContextInitializer.class的類，即全部上下文實(shí)現(xiàn)類助被。

/**
 * Create a new {@link SpringApplication} instance. The application context will load
 * beans from the specified primary sources (see {@link SpringApplication class-level}
 * documentation for details. The instance can be customized before calling
 * {@link #run(String...)}.
 * @param resourceLoader the resource loader to use
 * @param primarySources the primary bean sources
 * @see #run(Class, String[])
 * @see #setSources(Set)
 */
@SuppressWarnings({ "unchecked", "rawtypes" })
public SpringApplication(ResourceLoader resourceLoader, Class<?>... primarySources) {
   this.resourceLoader = resourceLoader;
   Assert.notNull(primarySources, "PrimarySources must not be null");
   this.primarySources = new LinkedHashSet<>(Arrays.asList(primarySources));
   this.webApplicationType = deduceWebApplicationType();
   //加載上下文初始器 ApplicationContextInitializer.class 的實(shí)現(xiàn)類
   setInitializers((Collection) getSpringFactoriesInstances(
         ApplicationContextInitializer.class));
   //加載監(jiān)聽初始器 ApplicationListener.class 的實(shí)現(xiàn)類
   setListeners((Collection) getSpringFactoriesInstances(ApplicationListener.class));
   this.mainApplicationClass = deduceMainApplicationClass();
}

2.我們來看一下加載上下文實(shí)體的方法

org.springframework.boot.SpringApplication#getSpringFactoriesInstances()

private <T> Collection<T> getSpringFactoriesInstances(Class<T> type,
      Class<?>[] parameterTypes, Object... args) {
   ClassLoader classLoader = Thread.currentThread().getContextClassLoader();
   // Use names and ensure unique to protect against duplicates
   //在這里調(diào)用了 SpringFactoriesLoader 中的loadFactoryNames方法蠢涝，
    //來插入想要實(shí)現(xiàn)的上下文
   Set<String> names = new LinkedHashSet<>(
         SpringFactoriesLoader.loadFactoryNames(type, classLoader));
   List<T> instances = createSpringFactoriesInstances(type, parameterTypes,
         classLoader, args, names);
   AnnotationAwareOrderComparator.sort(instances);
   return instances;
}

在這里注意一點(diǎn)就是spring-boot 使用了 Thread.currentThread().getContextClassLoader()來獲取了當(dāng)前類加載器鲸阻，這里就要介紹一下SPI機(jī)制(我也是在看這段代碼中,無法理解這段跋涣，在網(wǎng)上查了一些資料，簡(jiǎn)單的理解了一下這里為什么用這個(gè)方法鸟悴。)

SPI機(jī)制（Service Provider Interface)其實(shí)源自服務(wù)提供者框架（Service Provider Framework陈辱，參考【EffectiveJava】page6)，是一種將服務(wù)接口與服務(wù)實(shí)現(xiàn)分離以達(dá)到解耦细诸、大大提升了程序可擴(kuò)展性的機(jī)制沛贪。引入服務(wù)提供者就是引入了spi接口的實(shí)現(xiàn)者，通過本地的注冊(cè)發(fā)現(xiàn)獲取到具體的實(shí)現(xiàn)類震贵，輕松可插拔利赋。

這里的ClassLoader是spring-boot的類加載器，在這突然又想到了JVM的雙親委派模型,感興趣的同學(xué)可以了解一下猩系。那么這個(gè)類加載器的SPI機(jī)制實(shí)現(xiàn)類都有哪些媚送？ 就是配置資源META-INF/spring.factories 中的內(nèi)容。

3.我們來看一下引用SpringFactoriesLoader#SpringFactoriesLoader的方法是做什么的寇甸。

/**
 * Load the fully qualified class names of factory implementations of the
 * given type from {@value #FACTORIES_RESOURCE_LOCATION}, using the given
 * class loader.
 * @param factoryClass the interface or abstract class representing the factory
 * @param classLoader the ClassLoader to use for loading resources; can be
 * {@code null} to use the default
 * @see #loadFactories
 * @throws IllegalArgumentException if an error occurs while loading factory names
 */
public static List<String> loadFactoryNames(Class<?> factoryClass, @Nullable ClassLoader classLoader) {
   String factoryClassName = factoryClass.getName();
   return loadSpringFactories(classLoader).getOrDefault(factoryClassName, Collections.emptyList());
}

在源碼注釋中可以清晰的看到該方法的作用塘偎，首先要了解到FACTORIES_RESOURCE_LOCATION全局變量的定義

/**
 * The location to look for factories.
 * <p>Can be present in multiple JAR files.
 */
public static final String FACTORIES_RESOURCE_LOCATION = "META-INF/spring.factories";

META-INF/spring.factories存放的就是所有spring-boot實(shí)現(xiàn)的各種繼承規(guī)約接口的實(shí)現(xiàn)類的類路徑。

了解了這個(gè)全局變量拿霉，那么這段注釋我們就可以明白了吟秩。

/**Load the fully qualified class names of factory implementations of the given type from {@value #FACTORIES_RESOURCE_LOCATION}*/從FACTORIES_RESOURCE_LOCATION中獲取所有繼承已知工廠類別的類全限定名。

4.既然都到這了绽淘，我們就不能不看一下SpringFactoriesLoader#loadSpringFactories(classLoader)這個(gè)方法涵防。

private static Map<String, List<String>> loadSpringFactories(@Nullable ClassLoader classLoader) {
   MultiValueMap<String, String> result = cache.get(classLoader);
   if (result != null) {
      return result;
   }

   try {
       /**
       * 獲取所有實(shí)現(xiàn)工廠接口的類的全限定名
       */
      Enumeration<URL> urls = (classLoader != null ?
            classLoader.getResources(FACTORIES_RESOURCE_LOCATION) :
            ClassLoader.getSystemResources(FACTORIES_RESOURCE_LOCATION));
      result = new LinkedMultiValueMap<>();
      while (urls.hasMoreElements()) {
         URL url = urls.nextElement();
         UrlResource resource = new UrlResource(url);
         Properties properties = PropertiesLoaderUtils.loadProperties(resource);
         for (Map.Entry<?, ?> entry : properties.entrySet()) {
            List<String> factoryClassNames = Arrays.asList(
                  StringUtils.commaDelimitedListToStringArray((String) entry.getValue()));
            result.addAll((String) entry.getKey(), factoryClassNames);
         }
      }
      cache.put(classLoader, result);
      return result;
   }
   catch (IOException ex) {
      throw new IllegalArgumentException("Unable to load factories from location [" +
            FACTORIES_RESOURCE_LOCATION + "]", ex);
   }
}

通過 classLoader.getResources和ClassLoader.getSystemResources來獲取META-INF/spring.factories下的內(nèi)容。比如:

# Initializers
org.springframework.context.ApplicationContextInitializer=\
org.springframework.boot.autoconfigure.SharedMetadataReaderFactoryContextInitializer,\
org.springframework.boot.autoconfigure.logging.ConditionEvaluationReportLoggingListener

然后通過解析Properties內(nèi)容來獲取全限定名收恢。

最后將想獲取的全限定名返回給

org.springframework.boot.SpringApplication#getSpringFactoriesInstances()方法:

private <T> Collection<T> getSpringFactoriesInstances(Class<T> type,
      Class<?>[] parameterTypes, Object... args) {
   ClassLoader classLoader = Thread.currentThread().getContextClassLoader();
   // Use names and ensure unique to protect against duplicates
   Set<String> names = new LinkedHashSet<>(
         SpringFactoriesLoader.loadFactoryNames(type, classLoader));
    //使用反射機(jī)制生成實(shí)例對(duì)象
   List<T> instances = createSpringFactoriesInstances(type, parameterTypes,
         classLoader, args, names);
    //進(jìn)行實(shí)例集合排序
   AnnotationAwareOrderComparator.sort(instances);
   return instances;
}

通過#createSpringFactoriesInstances 方法生成實(shí)例以及調(diào)用AnnotationAwareOrderComparator.sort（）進(jìn)行排序武学，而排序需要實(shí)現(xiàn)org.springframework.core.Ordered接口或者標(biāo)注org.springframework.core.Ordered注解，排序不是強(qiáng)制性的伦意。就像META-INF/spring.factories中應(yīng)用上下文初始器的實(shí)現(xiàn)類 org.springframework.boot.autoconfigure.SharedMetadataReaderFactoryContextInitializer和

org.springframework.boot.autoconfigure.logging.ConditionEvaluationReportLoggingListener都沒有實(shí)現(xiàn)排序的接口或者注解火窒。

說了這么多，接下來讓我們自己實(shí)現(xiàn)一個(gè)上下文初始器吧驮肉。