手写Spring Config，最终一战，来瞅瞅撒

配置分析

为什么要提供配置的方式呢，之前的内容中我们测试的时候都是通过代码来进行的：

GeneralBeanDefinition bd = new GeneralBeanDefinition();bd.setBeanClass(Lad.class);List args = new ArrayList<>();args.add("sunwukong");args.add(new BeanReference("magicGril"));bd.setConstructorArgumentValues(args);bf.registerBeanDefinition("swk", bd);bd = new GeneralBeanDefinition();bd.setBeanClass(MagicGril.class);args = new ArrayList<>();args.add("bAIgujing");bd.setConstructorArgumentValues(args);bf.registerBeanDefinition("magicGril", bd);

下面看下平时使用的时候，通过配置是什么样的：

Springframework.org/schema/beans"xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"xsi:schemaLocation="http://www.springframework.org/schema/beanshttp://www.springframework.org/schema/beans/spring-beans.XSD">

可以看出，提供配置的方式的优点：

• 实用简单，改动起来比较灵活
• 而且不需要改动代码

常用的配置方式，就是XML和注解的形式，它们的工作过程如下：

配置的工作过程定义XML标记和注解

需要定义什么样的XML标记和注解呢？通过之前的内容知道，配置的内容就是Bean定义信息，那么Bean定义的内容就是需要配置的内容

首先来看下Bean定义接口中有哪些信息：

XML配置的方式，首先需要定义一个DTD或者XSD文档，来定义一套标记信息，去指定Bean定义

可以看出，bean的配置指定的内容就是Bean定义接口中的信息

注解的方式，需要定义一套注解，那么需要哪些注解呢，也是Bean定义接口中的内容：

• 指定类、指定BeanName、指定scope、指定工厂方法、指定工厂Bean、指定init method、指定destroy method，这些在我们使用Spring的时候是通过@Component来实现的
• 指定构造参数的依赖：@Autowired、@Qualifier
• 指定属性依赖：@Value

Bean配置的解析过程，需要单独的接口来实现，而不是在BeanFactory中来做，要做到单一职责原则，所以需要定义单独的接口来解析Bean配置，然后再向BeanFactory注册Bean定义

ApplicationContext接口

ApplicationContext这个接口就是用来完成Bean配置解析的，上面说到实现配置的方式有XML和注解，所以会有两个实现类来实现ApplicationContext接口

1. XML方式的实现：

• XML文件可能存在多个，所以这里使用了list
• 需要完成：加载xml、解析xml、创建Bean定义、注册Bean定义的任务

1. 注解方式的实现

• 扫描的包也会存在多个，这里也使用list
• 需要完成：扫描包、获取注解、创建Bean定义、注册Bean定义的任务

因为需要创建和注册Bean定义，所以会使用到BeanFactory和BeanDefinitionRegistry接口，那么这部分代码在子类中分别实现的话就会重复，所以抽象出来放在父类中：

用户在使用的使用需要知道哪些接口和类呢？

1. 指定配置相关：xml、注解
2. 获取bean相关：BeanFactory

那么可以使用外观模式，让用户只需要知道ApplicationContext和其子类就行了，ApplicationContext可以继承BeanFactory，继而把两个接口合在一起:

ApplicationContext接口：

* @className: ApplicationContext* 用来构建整个应用环境的接口，用来完成Bean的配置和解析* 1：为了减少用户对框架类接口的依赖，扩展了BeanFactory接口，* Bean的配置和Bean的获取都可以通过ApplicationContext接口来完成* 2：配置资源的方式有xml和注解，所以存在xml和注解两种子类的实现* 3. Bean配置解析首先需要加载，所以实现了配置资源Resource的加载接口ResourceLoader* @author: TRpublic interface ApplicationContext extends ResourceLoader,BeanFactory {ApplicationContext的抽象类实现* @className: abstractApplicationContext* @description: ApplicationContext的抽象类实现* @author: TRpublic abstract class AbstractApplicationContext implements ApplicationContext {/** 用组合的方式来持有BeanFactory，完成BeanFactory接口的方法 */protected BeanFactory beanFactory;public AbstractApplicationContext() {super();this.beanFactory = new PreBuildBeanFactory();public AbstractApplicationContext(BeanFactory beanFactory) {super();this.beanFactory = beanFactory;@Overridepublic Object getBean(String beanName) throws Exception {return this.beanFactory.getBean(beanName);@Overridepublic void registerBeanPostProcessor(BeanPostProcessor beanPostProcessor) {this.beanFactory.registerBeanPostProcessor(beanPostProcessor);xml配置方式的ApplicationContext实现类* @className: XmlApplicationContext* @description: xml配置方式的ApplicationContext实现类* @author: TRpublic class XmlApplicationContext extends AbstractApplicationContext {注解配置方式的ApplicationContext实现类* @className: AnnotationApplicationContext* @description: 注解配置方式的ApplicationContext实现类* @author: TRpublic class AnnotationApplicationContext extends AbstractApplicationContext {配置的实现XML方式XML文件来源的处理

xml配置文件的来源会有多种，比如：

不同来源的XML文件，它的加载方式是不一样的，但是在解析的过程中，最后都希望获取到InputStream

这里也需要设计一套接口，对于不同来源的XML文件分别进行处理

InputStreamSource接口* @className: InputStreamSource* @description: 配置方式的最终统一接口* @author: TRpublic interface InputStreamSource {* 最终要获取的就是输入流* @return: JAVA.io.InputStreamInputStream getInputStream() throws IOException;Resource接口* @className: Resource* @description: 输入流的资源扩展接口* @author: TRpublic interface Resource extends InputStreamSource {//classpath形式的xml配置文件String CLASS_PATH_PREFIX = "classpath:";//系统文件形式的xml配置文件String File_SYSTEM_PREFIX = "file:";* 判断资源是否存在* @return: booleanboolean exists();* 是否可读* @return: booleanboolean isReadable();* 是否打开* @return: booleanboolean isOpen();* 获取资源文件* @return: java.io.FileFile getFile();InputStreamSource接口的实现类

FileSystemResource实现类：

* @className: FileSystemResource* @description: 系统文件类型的资源实现类* @author: TRpublic class FileSystemResource implements Resource {/** 文件资源对象 */private File file;public FileSystemResource(String fileName) {super();this.file = new File(fileName);public FileSystemResource(File file) {super();this.file = file;@Overridepublic boolean exists() {return this.file == null ? false : this.file.exists();@Overridepublic boolean isReadable() {return this.file == null ? false : this.file.canRead();@Overridepublic boolean isOpen() {return false;@Overridepublic File getFile() {return file;@Overridepublic InputStream getInputStream() throws IOException {return new FileInputStream(this.file);

ClassPathResource实现类：

* @className: ClassPathResource* @description: classpath形式的资源实现类* @author: TRpublic class ClassPathResource implements Resource {//classpath所需要的信息private String path;private Class clazz;private ClassLoader classLoader;public ClassPathResource(String path) {this(path, null );public ClassPathResource(String path, Class clazz) {this(path, clazz, null);public ClassPathResource(String path, Class clazz, ClassLoader classLoader) {super();this.path = path;this.clazz = clazz;this.classLoader = classLoader;public String getPath() {return path;public void setPath(String path) {this.path = path;public Class getClazz() {return clazz;public void setClazz(Class clazz) {this.clazz = clazz;public ClassLoader getClassLoader() {return classLoader;public void setClassLoader(ClassLoader classLoader) {this.classLoader = classLoader;@Overridepublic boolean exists() {if (StringUtils.isNotBlank(path)) {if (this.clazz != null) {return this.clazz.getResource(path) != null;if (this.classLoader != null) {return this.classLoader.getResource(path.startsWith("/") ? path.substring(1) : path) != null;return this.getClass().getResource(path) != null;return false;@Overridepublic boolean isReadable() {return exists();@Overridepublic boolean isOpen() {return false;@Overridepublic File getFile() {return null;@Overridepublic InputStream getInputStream() throws IOException {if (StringUtils.isNotBlank(path)) {if (this.clazz != null) {return this.clazz.getResourceAsStream(path);if (this.classLoader != null) {return this.classLoader.getResourceAsStream(path.startsWith("/") ? path.substring(1) : path);return this.getClass().getResourceAsStream(path);return null;

UrlResource实现类：

* @className: UrlResource* @description: URL形式的资源实现类* @author: TRpublic class UrlResource implements Resource {/** url的资源对象 */private URL url;public UrlResource(String url) throws IOException {this.url = new URL(url);public UrlResource(URL url) {super();this.url = url;public URL getUrl() {return url;public void setUrl(URL url) {this.url = url;@Overridepublic boolean exists() {return this.url != null;@Overridepublic boolean isReadable() {return exists();@Overridepublic boolean isOpen() {return false;@Overridepublic File getFile() {return null;@Overridepublic InputStream getInputStream() throws IOException {return null;XML资源加载器

用户给定资源时是一个字符串，上面有三种资源，那么谁去负责创建这些资源呢

这里需要定义一个资源加载器，去分辨不同的资源，然后进行加载，这部分工作是由ApplicationContext来完成的，所以ApplicationContext需要继承ResourceLoader接口

ResourceLoader接口：

* @className: ResourceLoader* 配置资源加载接口* 不同的配置方式，加载过程不一样，所以需要抽象出来一个接口应对变化的部分* 虽然加载的方式不一样，但是返回的资源结果是一样的，都是Resource* @author: TRpublic interface ResourceLoader {* 加载资源* @param location:* @return: demo.context.ResourceResource getResource(String location) throws IOException;

在这里，还需要区分用户给的字符串代表的是哪种资源，所以需要定义字符串的规则：

注解方式如何扫描的

扫描的包有哪些呢？

需要到指定的包目录下找出所有的类文件，而且要包含子孙包下的

需要定义一个资源路径的匹配行为

扫描的结果

扫描到了包下的class文件后，需要的是类名，而且扫描的是class文件，直接使用上面的FileResource即可

扫描的类ClassPathBeanDefinitionScanner

* @className: ClassPathBeanDefinitionScanner* @description: 扫描class文件* @author: TRpublic class ClassPathBeanDefinitionScanner {private static Log logger = LogFactory.getLog(ClassPathBeanDefinitionScanner.class);private BeanDefinitionRegistry registry;private BeanDefinitionReader reader;private PathMatcher pathMatcher = new AntPathMatcher();private String resourcePatter = "**/*.class";public ClassPathBeanDefinitionScanner(BeanDefinitionRegistry registry) {super();this.registry = registry;this.reader = new AnnotationBeanDefinitionReader(registry);* 扫描包的方法* @param basePackages:* @return: voidpublic void scan(String... basePackages) throws Throwable {if (basePackages != null && basePackages.length > 0) {for (String b : basePackages) {this.reader.loadBeanDefintions(doScan(b));* 将扫描的class转为Resource* @param basePackage:* @return: demo.context.Resource[]private Resource[] doScan(String basePackage) throws IOException {// 扫描包下的类// 构造初步匹配模式串，= 给入的包串 + / + **/*.class，替换里面的.为/String pathPattern = StringUtils.replace(basePackage, ".", "/") + "/" + this.resourcePatter;if (pathPattern.charAt(0) != '/') {pathPattern = "/" + pathPattern;// 找出模式的根包路径String rootPath = this.determineRootDir(pathPattern);// 得到文件名匹配的绝对路径模式String fullPattern = this.getClass().getResource("/").toString() + pathPattern;// 根据根包理解得到根包对应的目录File rootDir = new File(this.getClass().getResource(rootPath).toString());// 存放找到的类文件的resource集合Set scanedClassFileResources = new HashSet<>();// 调用doRetrieveMatchingFiles来扫描class文件this.doRetrieveMatchingFiles(fullPattern, rootDir, scanedClassFileResources);return (Resource[]) scanedClassFileResources.toArray();private String determineRootDir(String location) {int rootDirEnd = location.length();rootDirEnd = location.indexOf('*');int zi = location.indexOf('?');if (zi != -1 && zi < rootDirEnd) {rootDirEnd = location.lastIndexOf('/', zi);if (rootDirEnd != -1) {return location.substring(0, rootDirEnd);} else {return location;* 递归找指定目录下的所有类，匹配模式的加入到结果中。* @param fullPattern* @param dir* @param result* @throws IOExceptionprotected void doRetrieveMatchingFiles(String fullPattern, File dir, Set result) throws IOException {if (logger.isTraceEnabled()) {logger.trace("Searching directory [" + dir.getAbsolutePath() + "] for files matching pattern ["+ fullPattern + "]");for (File content : listDirectory(dir)) {String currPath = StringUtils.replace(content.getAbsolutePath(), File.separator, "/");if (content.isDirectory() && getPathMatcher().matchStart(fullPattern, currPath + "/")) {if (!content.canRead()) {if (logger.isDebugEnabled()) {logger.debug("Skipping subdirectory [" + dir.getAbsolutePath()+ "] because the application is not allowed to read the directory");} else {doRetrieveMatchingFiles(fullPattern, content, result);if (getPathMatcher().match(fullPattern, currPath)) {result.add(new FileSystemResource(content));protected File[] listDirectory(File dir) {File[] files = dir.listFiles();if (files == null) {if (logger.isInfoEnabled()) {logger.info("Could not retrieve contents of directory [" + dir.getAbsolutePath() + "]");return new File[0];Arrays.sort(files, Comparator.comparing(File::getName));return files;public BeanDefinitionRegistry getRegistry() {return registry;public void setRegistry(BeanDefinitionRegistry registry) {this.registry = registry;public BeanDefinitionReader getReader() {return reader;public void setReader(BeanDefinitionReader reader) {this.reader = reader;public PathMatcher getPathMatcher() {return pathMatcher;public void setPathMatcher(PathMatcher pathMatcher) {this.pathMatcher = pathMatcher;public String getResourcePatter() {return resourcePatter;public void setResourcePatter(String resourcePatter) {this.resourcePatter = resourcePatter;解析成Bean定义

XML和注解最终的输出结果都是Resource，在这里还需要把Resource解析成Bean定义信息才行

需要定义接口来进行解析：

BeanDefinitionReader接口：

* @className: BeanDefinitionReader* @description: 将Resource资源解析成Bean定义的接口* @author: TRpublic interface BeanDefinitionReader {* 解析单个资源* @param resource:* @return: voidvoid loadBeanDefintions(Resource resource) throws Throwable;* 解析多个资源* @param resource:* @return: voidvoid loadBeanDefintions(Resource... resource) throws Throwable;

AbstractBeanDefinitionReader抽象类：

* @className: AbstractBeanDefinitionReader* @description: TODO* @date: 2021/6/10 15:58* @author: jinpeng.sunpublic abstract class AbstractBeanDefinitionReader implements BeanDefinitionReader {/** 持有BeanDefinitionRegistry接口，以便完成注册到BeanFactory中 */protected BeanDefinitionRegistry beanDefinitionRegistry;public AbstractBeanDefinitionReader(BeanDefinitionRegistry beanDefinitionRegistry) {super();this.beanDefinitionRegistry = beanDefinitionRegistry;

xml配置方式的bean定义解析器:

* @className: XmlBeanDefinitionReader* @description: xml配置方式的bean定义解析器* @author: TRpublic class XmlBeanDefinitionReader extends AbstractBeanDefinitionReader {public XmlBeanDefinitionReader(BeanDefinitionRegistry beanDefinitionRegistry) {super(beanDefinitionRegistry);@Overridepublic void loadBeanDefintions(Resource resource) throws Throwable {this.loadBeanDefintions(new Resource[] {resource});@Overridepublic void loadBeanDefintions(Resource... resource) throws Throwable {if (resource != null && resource.length > 0) {for (Resource r : resource) {this.parseXml(r);private void parseXml(Resource r) {//TODO 解析xml文档，获取bean定义，创建bean定义对象，注册到BeanDefinitionRegistry中

注解配置方式的bean定义解析器：

* @className: AnnotationBeanDefinitionReader* @description: 注解配置方式的bean定义解析器：* @author: TRpublic class AnnotationBeanDefinitionReader extends AbstractBeanDefinitionReader {public AnnotationBeanDefinitionReader(BeanDefinitionRegistry beanDefinitionRegistry) {super(beanDefinitionRegistry);@Overridepublic void loadBeanDefintions(Resource resource) throws Throwable {this.loadBeanDefintions(new Resource[] {resource});@Overridepublic void loadBeanDefintions(Resource... resource) throws Throwable {if (resource != null && resource.length > 0) {for (Resource r : resource) {this.retriveAndRegistBeanDefinition(r);private void retriveAndRegistBeanDefinition(Resource resource) {if(resource != null && resource.getFile() != null) {String className = getClassNameFormFile(resource.getFile());try {Class clazz = Class.forName(className);Component component = clazz.getAnnotation(Component.class);if (component != null) {GeneralBeanDefinition beanDefinition = new GeneralBeanDefinition();beanDefinition.setBeanClass(clazz);beanDefinition.setScope(component.scope());beanDefinition.setFactoryMethodName(component.factoryMethodName());beanDefinition.setFactoryBeanName(component.factoryBeanName());beanDefinition.setInitMethodName(component.initMethodName());beanDefinition.setDestroyMethodName(component.destroyMethodName());//获取所有的构造方法，在构造方法上找Autowired注解，如果有的话，将这个构造方法set到bdthis.handleConstructor(clazz, beanDefinition);//处理工厂方法参数依赖if(StringUtils.isNotBlank(beanDefinition.getFactoryMethodName())) {this.handleFactoryMethodArgs(clazz, beanDefinition);//处理属性依赖this.handlePropertyDi(clazz, beanDefinition);String beanName = "".equals(component.value()) ? component.name() : null;if (StringUtils.isBlank(beanName)) {// TODO 应用名称生成规则生成beanName;// 默认驼峰命名法beanName = CaseFormat.LOWER_HYPHEN.to(CaseFormat.LOWER_CAMEL, clazz.getSimpleName());// 注册bean定义this.beanDefinitionRegistry.registerBeanDefinition(beanName, beanDefinition);} catch (ClassNotFoundException | BeanDefinitionException e) {e.printStackTrace();private void handlePropertyDi(Class clazz, GeneralBeanDefinition bd) {// TODO Auto-generated method stubprivate void handleFactoryMethodArgs(Class clazz, GeneralBeanDefinition bd) {// TODO Auto-generated method stubprivate void handleConstructor(Class clazz, GeneralBeanDefinition bd) {//获取所有的构造方法，在构造方法上找Autowired注解，如果有的话，将这个构造方法set到bdConstructor[] constructors = clazz.getConstructors();if (constructors != null && constructors.length > 0) {for (Constructor c : constructors) {if (c.getAnnotation(Autowired.class) != null) {bd.setConstructor(c);Parameter[] ps = c.getParameters();//遍历获取参数上的注解，及创建参数依赖break;private int classPathAbsLength = AnnotationBeanDefinitionReader.class.getResource("/").toString().length();private String getClassNameFormFile(File file) {//返回绝对路径名字符串String absPath = file.getAbsolutePath();String name = absPath.substring(classPathAbsLength+1, absPath.indexOf("."));return StringUtils.replace(name, File.separator, ".");

完善XmlApplicationContext和AnnotationApplicationContext：

public class XmlApplicationContext extends AbstractApplicationContext {private List resources;private BeanDefinitionReader definitionReader;public XmlApplicationContext(String... locations) throws Throwable {super();load(locations);//资源解析成BeanDefinition，外派给BeanDefinitionReader接口来实现this.definitionReader = new XmlBeanDefinitionReader((BeanDefinitionRegistry) this.beanFactory);Resource[] resourceArray = new Resource[resources.size()];resources.toArray(resourceArray);//将解析后的BeanDefinition装载到BeanFactory中definitionReader.loadBeanDefintions(resourceArray);* 根据用户指定的配置文件位置，加载资源信息* @param locations:* @return: voidprivate void load(String[] locations) throws IOException {if (resources == null) {resources = new ArrayList();//完成加载，创建好Resourceif (locations != null && locations.length > 0) {for (String lo : locations) {Resource resource = getResource(lo);if (resource != null) {this.resources.add(resource);@Overridepublic Resource getResource(String location) throws IOException {if (StringUtils.isNotBlank(location)) {//根据字符串的前缀判断区分，class、系统文件、url三种资源的加载if (location.startsWith(Resource.CLASS_PATH_PREFIX)) {return new ClassPathResource(location.substring(Resource.CLASS_PATH_PREFIX.length()));} else if (location.startsWith(Resource.File_SYSTEM_PREFIX)) {return new FileSystemResource(location.substring(Resource.File_SYSTEM_PREFIX.length()));} else {return new UrlResource(location);return null;public class AnnotationApplicationContext extends AbstractApplicationContext {private ClassPathBeanDefinitionScanner scanner;public AnnotationApplicationContext(String... locations) throws Throwable {scanner = new ClassPathBeanDefinitionScanner((BeanDefinitionRegistry) this.beanFactory);scanner.scan(locations);@Overridepublic Resource getResource(String location) throws IOException {return null;