寫在前面

Springboot對Spring做了很好的封裝十气，僅通過添加依賴，以及一些有必要的配置外喧伞，就能夠完成項(xiàng)目的啟動(dòng)。下面我將通過對一個(gè)簡單的項(xiàng)目進(jìn)行調(diào)試追蹤，來發(fā)現(xiàn)Springboot是怎樣完成一個(gè)項(xiàng)目的啟動(dòng)的棚赔。

啟動(dòng)項(xiàng)目

SpringApplication.run(）作為項(xiàng)目啟動(dòng)的唯一入口，接下來就到SpringApplication類中查看該方法執(zhí)行哪些事眶痰。

public static ConfigurableApplicationContext run(Class<?> primarySource, String... args) {
        return run(new Class[]{primarySource}, args);
    }

public static ConfigurableApplicationContext run(Class<?>[] primarySources, String[] args) {
        return (new SpringApplication(primarySources)).run(args);
    }

可以看到本姥，通過將啟動(dòng)類的class對象作為參數(shù)傳進(jìn)來滓侍，通過SpringApplicatioin構(gòu)造方法新建一個(gè)對象裂逐，再調(diào)用其run方法。本篇先介紹SpringApplication對象的新建過程矮燎，下一篇再對run方法進(jìn)行介紹刊苍。

SpringApplication

public SpringApplication(Class<?>... primarySources) {
        this((ResourceLoader)null, primarySources);
    }

public SpringApplication(ResourceLoader resourceLoader, Class<?>... primarySources) {
        this.sources = new LinkedHashSet();
        this.bannerMode = Mode.CONSOLE;
        this.logStartupInfo = true;
        this.addCommandLineProperties = true;
        this.addConversionService = true;
        this.headless = true;
        this.registerShutdownHook = true;
        this.additionalProfiles = new HashSet();
        this.isCustomEnvironment = false;
        this.lazyInitialization = false;
        this.resourceLoader = resourceLoader;
        Assert.notNull(primarySources, "PrimarySources must not be null");
        this.primarySources = new LinkedHashSet(Arrays.asList(primarySources));
        this.webApplicationType = WebApplicationType.deduceFromClasspath();
        this.setInitializers(this.getSpringFactoriesInstances(ApplicationContextInitializer.class));
        this.setListeners(this.getSpringFactoriesInstances(ApplicationListener.class));
        this.mainApplicationClass = this.deduceMainApplicationClass();
    }

主要看關(guān)鍵的幾步穗酥，首先是this.webApplicationType = WebApplicationType.deduceFromClasspath();這一步是推斷應(yīng)用類型百揭。判斷出是servlet、reactive请毛、還是none方仿。這一步比較好理解委粉，就不從源碼上進(jìn)行分析了栗涂。

設(shè)置初始化器

第二步是設(shè)置初始化器墨吓，即setInitializers()方法聪廉，該方法又通過getSpringFactoriesInstances()方法進(jìn)行設(shè)置瞬痘。理解了getSpringFactoriesInstances()方法，就能理解設(shè)置初始化器的過程板熊。以下是getSpringFactoriesInstances()方法的源碼框全。

private <T> Collection<T> getSpringFactoriesInstances(Class<T> type, Class<?>[] parameterTypes, Object... args) {
        ClassLoader classLoader = this.getClassLoader();
        Set<String> names = new LinkedHashSet(SpringFactoriesLoader.loadFactoryNames(type, classLoader));
        List<T> instances = this.createSpringFactoriesInstances(type, parameterTypes, classLoader, args, names);
        AnnotationAwareOrderComparator.sort(instances);
        return instances;
    }

在這一步里，通過類加載器去讀取factoryNames干签，再通過factoryNames構(gòu)建factory實(shí)例津辩。

1. 讀取factoryNames

public static List<String> loadFactoryNames(Class<?> factoryType, @Nullable ClassLoader classLoader) {
        String factoryTypeName = factoryType.getName();
        return (List)loadSpringFactories(classLoader).getOrDefault(factoryTypeName, Collections.emptyList());
    }

private static Map<String, List<String>> loadSpringFactories(@Nullable ClassLoader classLoader) {
        MultiValueMap<String, String> result = (MultiValueMap)cache.get(classLoader);
        if (result != null) {
            return result;
        } else {
            try {
                Enumeration<URL> urls = classLoader != null ? classLoader.getResources("META-INF/spring.factories") : ClassLoader.getSystemResources("META-INF/spring.factories");
                LinkedMultiValueMap result = new LinkedMultiValueMap();

                while(urls.hasMoreElements()) {
                    URL url = (URL)urls.nextElement();
                    UrlResource resource = new UrlResource(url);
                    Properties properties = PropertiesLoaderUtils.loadProperties(resource);
                    Iterator var6 = properties.entrySet().iterator();

                    while(var6.hasNext()) {
                        Entry<?, ?> entry = (Entry)var6.next();
                        String factoryTypeName = ((String)entry.getKey()).trim();
                        String[] var9 = StringUtils.commaDelimitedListToStringArray((String)entry.getValue());
                        int var10 = var9.length;

                        for(int var11 = 0; var11 < var10; ++var11) {
                            String factoryImplementationName = var9[var11];
                            result.add(factoryTypeName, factoryImplementationName.trim());
                        }
                    }
                }

                cache.put(classLoader, result);
                return result;
            } catch (IOException var13) {
                throw new IllegalArgumentException("Unable to load factories from location [META-INF/spring.factories]", var13);
            }
        }
    }

loadSpringFactories()這個(gè)方法會到org.springframework.boot的jar包中讀取所有的META-INF/spring.factories里面的內(nèi)容。比如下面是spring-boot.jar包其中的Application Context Initializers內(nèi)容：

# Application Context Initializers
org.springframework.context.ApplicationContextInitializer=\
org.springframework.boot.context.ConfigurationWarningsApplicationContextInitializer,\
org.springframework.boot.context.ContextIdApplicationContextInitializer,\
org.springframework.boot.context.config.DelegatingApplicationContextInitializer,\
org.springframework.boot.rsocket.context.RSocketPortInfoApplicationContextInitializer,\
org.springframework.boot.web.context.ServerPortInfoApplicationContextInitializer

讀取這些文件后將內(nèi)容以鍵值對的形式存儲到Map<String, List<String>>中容劳。通過該方法讀取了# Application Context Initializers喘沿、# Application Listeners、# Environment Post Processors等等這樣一些寫在spring.factories文件中的factoryNames鸭蛙。再通過傳給loadFactoryNames()方法的Class<?> factoryType參數(shù)篩選出相對應(yīng)的factoryNames摹恨，即以Map的鍵名讀取值∪⑹樱回到setInitializers()方法晒哄，傳入的參數(shù)為ApplicationContextInitializer.class，因此獲取到的便是ApplicationContextInitializers的factoryNames肪获。

2. 構(gòu)建factory實(shí)例

通過createSpringFactoriesInstances()方法構(gòu)建factory實(shí)例寝凌，該方法的代碼如下：

private <T> List<T> createSpringFactoriesInstances(Class<T> type, Class<?>[] parameterTypes, ClassLoader classLoader, Object[] args, Set<String> names) {
        List<T> instances = new ArrayList(names.size());
        Iterator var7 = names.iterator();

        while(var7.hasNext()) {
            String name = (String)var7.next();

            try {
                Class<?> instanceClass = ClassUtils.forName(name, classLoader);
                Assert.isAssignable(type, instanceClass);
                Constructor<?> constructor = instanceClass.getDeclaredConstructor(parameterTypes);
                T instance = BeanUtils.instantiateClass(constructor, args);
                instances.add(instance);
            } catch (Throwable var12) {
                throw new IllegalArgumentException("Cannot instantiate " + type + " : " + name, var12);
            }
        }

        return instances;
    }

可以看到，這一步就是根據(jù)上一步得到的factory類名孝赫，通過反射機(jī)制創(chuàng)建factory實(shí)例较木。

設(shè)置監(jiān)聽器

setListeners()方法與setInitializers()方法類似，通過傳入ApplicationListener.class對象青柄，從META-INF/spring.factories中讀取相應(yīng)的ApplicationListener的factoryNames伐债。再利用反射機(jī)制對這些factory進(jìn)行實(shí)例化。需要指出的是致开，在loadSpringFactories()方法中峰锁，由于setInitializers()方法調(diào)用了一次并將結(jié)果存入cache中，因此setListeners()方法再次調(diào)用時(shí)會將cache中保存的結(jié)果直接返回双戳。

推斷主應(yīng)用類

deduceMainApplicationClass()方法通過跟蹤棧軌跡虹蒋，找出main方法所在的類，進(jìn)而推斷出我們項(xiàng)目的主應(yīng)用類。代碼如下：

private Class<?> deduceMainApplicationClass() {
        try {
            StackTraceElement[] stackTrace = (new RuntimeException()).getStackTrace();
            StackTraceElement[] var2 = stackTrace;
            int var3 = stackTrace.length;

            for(int var4 = 0; var4 < var3; ++var4) {
                StackTraceElement stackTraceElement = var2[var4];
                if ("main".equals(stackTraceElement.getMethodName())) {
                    return Class.forName(stackTraceElement.getClassName());
                }
            }
        } catch (ClassNotFoundException var6) {
        }

        return null;
    }

總結(jié)

回顧一下內(nèi)容魄衅，首先峭竣，啟動(dòng)項(xiàng)目分為兩步：創(chuàng)建SpringApplication對象、調(diào)用SpringApplication對象的run()方法晃虫。創(chuàng)建SpringApplication對象主要有四步：判斷應(yīng)用類型皆撩、設(shè)置初始化器、設(shè)置監(jiān)聽器哲银、判斷主應(yīng)用類毅访。其中設(shè)置初始化器和監(jiān)聽器都通過類加載器去讀取jar包中的META-INF/spring.factories文件，存儲到Map中盘榨，然后再通過反射機(jī)制完成factory的實(shí)例化喻粹。這就是SpringApplication對象創(chuàng)建過程完成的工作。下一篇我將對run()方法進(jìn)行介紹草巡。