spring解决循环依赖为什么要用三级缓存?

 2022-09-15
原文地址:https://cloud.tencent.com/developer/article/1718125

也许有些朋友对spring的循环依赖问题并不了解,让我们先一起看看这个例子。

    @Service
    public class AService {
    
        private BService bService;
    
        public AService(BService bService) {
            this.bService = bService;
        }
    
        public void doA() {
            System.out.println("call doA");
        }
    }
    @Service
    public class BService {
    
        private AService aService;
    
        public BService(AService aService) {
            this.aService = aService;
        }
    
        public void doB() {
            System.out.println("call doB");
        }
    }
    @RequestMapping("/test")
    @RestController
    public class TestController {
    
        @Autowired
        private AService aService;
    
        @RequestMapping("/doSameThing")
        public String doSameThing() {
            aService.doA();
            return "success";
        }
    }
    @SpringBootApplication
    public class Application {
    
        /**
         * 程序入口
         * @param args 程序输入参数
         */
        public static void main(String[] args) {
            new SpringApplicationBuilder(Application.class).web(WebApplicationType.SERVLET).run(args);
        }
    }

我们在运行Application类的main方法启动服务时,报了如下异常:

    Requested bean is currently in creation: Is there an unresolvable circular reference?

202209152258300131.png

这里提示得很明显,出现了循环依赖。

什么是循环依赖?

循环依赖是实例a依赖于实例b,实例b又依赖于实例a。

202209152258327152.png

或者实例a依赖于实例b,实例b依赖于实例c,实例c又依赖于实例a。

202209152258338123.png

像这种多个实例之间的相互依赖关系构成一个环形,就是循环依赖。

为什么会形成循环依赖?

上面的例子中AService实例化时会调用构造方法 public AService(BService bService),该构造方法依赖于BService的实例。此时BService还没有实例化,需要调用构造方法public BService(AService aService)才能完成实例化,该构造方法巧合又需要AService的实例作为参数。由于AService和BService都没有提前实例化,在实例化过程中又相互依赖对方的实例作为参数,这样构成了一个死循环,所以最终都无法再实例化了。

spring要如何解决循环依赖?

只需要将上面的例子稍微调整一下,不用构造函数注入,直接使用Autowired注入。

    @Service
    public class AService {
    
        @Autowired
        private BService bService;
    
        public AService() {
        }
    
        public void doA() {
            System.out.println("call doA");
        }
    }
    @Service
    public class BService {
    
        @Autowired
        private AService aService;
    
        public BService() {
        }
    
        public void doB() {
            System.out.println("call doB");
        }
    }

我们看到可以正常启动了,说明循环依赖被自己解决了

202209152258352364.png

spring为什么能循环依赖?

调用applicationContext.getBean(xx)方法,最终会调到AbstractBeanFactory类的doGetBean方法。由于该方法很长,我把部分不相干的代码省略掉了。

    protected <T> T doGetBean(final String name, @Nullable final Class<T> requiredType,
    @Nullable final Object[] args, boolean typeCheckOnly) throws BeansException { final String beanName = transformedBeanName(name);    Object bean;
    
        Object sharedInstance = getSingleton(beanName);
        if (sharedInstance != null && args == null) {
           省略........
          bean = getObjectForBeanInstance(sharedInstance, name, beanName, null);
        } else {
           省略........
    
            if (mbd.isSingleton()) {
              sharedInstance = getSingleton(beanName, () -> {
                try {
                  return createBean(beanName, mbd, args);
                }
                catch (BeansException ex) {
                  destroySingleton(beanName);
                  throw ex;
                }
              });
              bean = getObjectForBeanInstance(sharedInstance, name, beanName, mbd);
            }
    
            else if (mbd.isPrototype()) {
              // It's a prototype -> create a new instance.
              Object prototypeInstance = null;
              try {
                beforePrototypeCreation(beanName);
                prototypeInstance = createBean(beanName, mbd, args);
              }
              finally {
                afterPrototypeCreation(beanName);
              }
              bean = getObjectForBeanInstance(prototypeInstance, name, beanName, mbd);
            }
            else {
              String scopeName = mbd.getScope();
              final Scope scope = this.scopes.get(scopeName);
              if (scope == null) {
                throw new IllegalStateException("No Scope registered for scope name '" + scopeName + "'");
              }
              try {
                Object scopedInstance = scope.get(beanName, () -> {
                  beforePrototypeCreation(beanName);
                  try {
                    return createBean(beanName, mbd, args);
                  }
                  finally {
                    afterPrototypeCreation(beanName);
                  }
                });
                bean = getObjectForBeanInstance(scopedInstance, name, beanName, mbd);
              }
              catch (IllegalStateException ex) {
                throw new BeanCreationException(beanName,
                    "Scope '" + scopeName + "' is not active for the current thread; consider " +
                    "defining a scoped proxy for this bean if you intend to refer to it from a singleton",
                    ex);
              }
            }
          }
          catch (BeansException ex) {
            cleanupAfterBeanCreationFailure(beanName);
            throw ex;
          }
        }
        省略........
        return (T) bean;
      }

我们可以看到,该方法一进来会调用getSingleton方法从缓存获取实例,如果获取不到。会判断作用域是否为:单例,多列 或者 都不是,不同的作用域创建实例的规则不一样。接下来,我们重点看一下getSingleton方法。

      public Object getSingleton(String beanName) {
        return getSingleton(beanName, true);
      }
      protected Object getSingleton(String beanName, boolean allowEarlyReference) {
        Object singletonObject = this.singletonObjects.get(beanName);
        if (singletonObject == null && isSingletonCurrentlyInCreation(beanName)) {
          synchronized (this.singletonObjects) {
            singletonObject = this.earlySingletonObjects.get(beanName);
            if (singletonObject == null && allowEarlyReference) {
              ObjectFactory<?> singletonFactory = this.singletonFactories.get(beanName);
              if (singletonFactory != null) {
                singletonObject = singletonFactory.getObject();
                this.earlySingletonObjects.put(beanName, singletonObject);
                this.singletonFactories.remove(beanName);
              }
            }
          }
        }
        return singletonObject;
      }

我们发现有三个Map集合:

      /** Cache of singleton objects: bean name --> bean instance */
      private final Map<String, Object> singletonObjects = new ConcurrentHashMap<>(256);
    
      /** Cache of singleton factories: bean name --> ObjectFactory */
      private final Map<String, ObjectFactory<?>> singletonFactories = new HashMap<>(16);
    
      /** Cache of early singleton objects: bean name --> bean instance */
      private final Map<String, Object> earlySingletonObjects = new HashMap<>(16);

singletonObjects对应一级缓存,earlySingletonObjects对应二级缓存,singletonFactories对应三级缓存。

上面getSingleton方法的逻辑是:

  1. 先从singletonObjects(一级缓存)中获取实例,如果可以获取到则直接返回singletonObject实例。
  2. 如果从singletonObjects(一级缓存)中获取不对实例,再从earlySingletonObjects(二级缓存)中获取实例,如果可以获取到则直接返回singletonObject实例。
  3. 如果从earlySingletonObjects(二级缓存)中获取不对实例,则从singletonFactories(三级缓存)中获取singletonFactory,如果获取到则调用getObject方法创建实例,把创建好的实例放到earlySingletonObjects(二级缓存)中,并且从singletonFactories(三级缓存)删除singletonFactory实例,然后返回singletonObject实例。
  4. 如果从singletonObjects、earlySingletonObjects和singletonFactories中都获取不到实例,则singletonObject对象为空。

获取实例需要调用applicationContext.getBean("xxx")方法,第一次调用getBean方法,代码走到getSingleton方法时返回的singletonObject对象是空的。然后接着往下执行,默认情况下bean的作用域是单例的,接下来我们重点看看这段代码:

202209152258371435.png

createBean方法会调用doCreateBean方法,该方法同样比较长,我们把不相干的代码省略掉。

    protected Object doCreateBean(final String beanName, final RootBeanDefinition mbd, final @Nullable Object[] args)
          throws BeanCreationException {
    
        BeanWrapper instanceWrapper = null;
        省略......
        
        if (instanceWrapper == null) {
          instanceWrapper = createBeanInstance(beanName, mbd, args);
        }
        final Object bean = instanceWrapper.getWrappedInstance();
        省略........
    
        boolean earlySingletonExposure = (mbd.isSingleton() && this.allowCircularReferences &&
            isSingletonCurrentlyInCreation(beanName));
        if (earlySingletonExposure) {
          addSingletonFactory(beanName, () -> getEarlyBeanReference(beanName, mbd, bean));
        }
    
        Object exposedObject = bean;
        try {
          populateBean(beanName, mbd, instanceWrapper);
          exposedObject = initializeBean(beanName, exposedObject, mbd);
        }
        catch (Throwable ex) {
          省略 .....
        }
        省略 .......
        return exposedObject;
      }

该方法的主要流程是:

  1. 创建bean实例
  2. 判断作用域是否为单例,允许循环依赖,并且当前bean正在创建,还没有创建完成。如果都满足条件,则调用addSingletonFactory将bean实例放入缓存中。
  3. 调用populateBean方法进行依赖注入
  4. 调用initializeBean方法完成对象初始化和AOP增强

我们关注的重点可以先放到addSingletonFactory方法上。

      protected void addSingletonFactory(String beanName, ObjectFactory<?> singletonFactory) {
        Assert.notNull(singletonFactory, "Singleton factory must not be null");
        synchronized (this.singletonObjects) {
          if (!this.singletonObjects.containsKey(beanName)) {
            this.singletonFactories.put(beanName, singletonFactory);
            this.earlySingletonObjects.remove(beanName);
            this.registeredSingletons.add(beanName);
          }
        }
      }

该方法的逻辑是判断如果singletonObjects(一级缓存)中找不到实例,则将singletonFactory实例放到singletonFactories(三级缓存)中,并且移除earlySingletonObjects(二级缓存)中的实例。

createBean方法执行完之后,会调用外层的getSingleton方法

202209152258386936.png

我们重点看看这个getSingleton方法

    public Object getSingleton(String beanName, ObjectFactory<?> singletonFactory) {
        Assert.notNull(beanName, "Bean name must not be null");
        synchronized (this.singletonObjects) {
          Object singletonObject = this.singletonObjects.get(beanName);
          if (singletonObject == null) {
            if (this.singletonsCurrentlyInDestruction) {
              throw new BeanCreationNotAllowedException(beanName,
                  "Singleton bean creation not allowed while singletons of this factory are in destruction " +
                  "(Do not request a bean from a BeanFactory in a destroy method implementation!)");
            }
            beforeSingletonCreation(beanName);
            boolean newSingleton = false;
            boolean recordSuppressedExceptions = (this.suppressedExceptions == null);
            if (recordSuppressedExceptions) {
              this.suppressedExceptions = new LinkedHashSet<>();
            }
            try {
              singletonObject = singletonFactory.getObject();
              newSingleton = true;
            }
            catch (IllegalStateException ex) {
    
              singletonObject = this.singletonObjects.get(beanName);
              if (singletonObject == null) {
                throw ex;
              }
            }
            catch (BeanCreationException ex) {
              if (recordSuppressedExceptions) {
                for (Exception suppressedException : this.suppressedExceptions) {
                  ex.addRelatedCause(suppressedException);
                }
              }
              throw ex;
            }
            finally {
              if (recordSuppressedExceptions) {
                this.suppressedExceptions = null;
              }
              afterSingletonCreation(beanName);
            }
            if (newSingleton) {
              addSingleton(beanName, singletonObject);
            }
          }
          return singletonObject;
        }
      }

该方法逻辑很简单,就是先从singletonObjects(一级缓存)中获取实例,如果获取不到,则调用singletonFactory.getObject()方法创建一个实例,然后调用addSingleton方法放入singletonObjects缓存中。

      protected void addSingleton(String beanName, Object singletonObject) {
        synchronized (this.singletonObjects) {
          this.singletonObjects.put(beanName, singletonObject);
          this.singletonFactories.remove(beanName);
          this.earlySingletonObjects.remove(beanName);
          this.registeredSingletons.add(beanName);
        }
      }

该方法会将实例放入singletonObjects(一级缓存),并且删除singletonFactories(二级缓存),这样以后再调用getBean时,都能从singletonObjects(一级缓存)中获取到实例了。

说了这么多,再回到示例中的场景。

202209152258395347.png

spring为什么要用三级缓存,而不是二级缓存?

像示例的这种情况只用二级缓存是没有问题的。

但是假如有这种情况:a实例同时依赖于b实例和c实例,b实例又依赖于a实例,c实例也依赖于a实例。

a实例化时,先提前暴露objectFactorya到三级缓存,调用getBean(b)依赖注入b实例。b实例化之后,提前暴露objectFactoryb到三级缓存,调用getBean(a)依赖注入a实例,由于提前暴露了objectFactorya,此时可以从三级缓存中获取到a实例, b实例完成了依赖注入,升级为一级缓存。a实例化再getBean(c)依赖注入c实例,c实例化之后,提前暴露objectFactoryc到三级缓存,调用getBean(a)依赖注入a实例,由于提前暴露了objectFactorya,此时可以从三级缓存中获取到a实例。注意这里又要从三级缓存中获取a实例,我们知道三级缓存中的实例是通过调用singletonFactory.getObject()方法获取的,返回结果每次都可能不一样。如果不用二级缓存,这里会有问题,两次获取的a实例不一样。

总结:

只有单例的情况下才能解决循环依赖问题,并且allowCircularReferences要设置成true。

以下情况还是会出现循环依赖:

  1. 构造器注入
  2. 作用域非单例的情况,当然在自定义作用域,自己可以实现避免循环依赖的逻辑
  3. allowCircularReferences参数设置为false

大家喜欢这篇文章的话,烦请关注一下 :苏三说技术