spring 声明式事务原理解读

Posted 秋风兮月

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了spring 声明式事务原理解读相关的知识,希望对你有一定的参考价值。

在Spring中,声明式事务是通过事务属性(transaction attribute)来定义的。事务属性描述了事务策略如何应用到方法上。事务属性包含5个方面:

  • 传播行为
  • 隔离级别
  • 是否只读
  • 事务超时
  • 回滚规则
尽管Spring提供了多种声明式事务的机制,但是所有的方式都依赖这五个参数来控制如何管理事务策略。 (上述内容参考《Spring In Action》第三版)。 对于声明式事务是使用Spring提供的tx配置命名空间。其中一些声明式事务配置元素依赖于部分Spring的AOP配置元素。 下面是一个具体配置的例子:
<tx:advice id="txAdviceHibernate" transaction-manager="txManagerHibernate">  
        <tx:attributes>  
            <tx:method name="add*" propagation="REQUIRED" rollback-for="Exception" no-rollback-for="" />         
            <tx:method name="batch*" propagation="REQUIRED" rollback-for="Exception" />  
            <tx:method name="delete*" propagation="REQUIRED" rollback-for="Exception" />  
            <tx:method name="execute*" propagation="REQUIRED" rollback-for="Exception" />  
            <tx:method name="export*" propagation="REQUIRED" rollback-for="Exception" />  
            <tx:method name="import*" propagation="REQUIRED" rollback-for="Exception" />  
            <tx:method name="insert*" propagation="REQUIRED" rollback-for="Exception" />  
            <tx:method name="save*" propagation="REQUIRED" rollback-for="Exception" />  
            <tx:method name="update*" propagation="REQUIRED" rollback-for="Exception" />  
            <tx:method name="count*" propagation="NOT_SUPPORTED" read-only="true" />  
            <tx:method name="find*" propagation="NOT_SUPPORTED" read-only="true" />  
            <tx:method name="get*" propagation="NOT_SUPPORTED" read-only="true" />  
            <tx:method name="list*" propagation="NOT_SUPPORTED" read-only="true" />  
            <tx:method name="load*" propagation="NOT_SUPPORTED" read-only="true" />  
            <tx:method name="page*" propagation="NOT_SUPPORTED" read-only="true" />  
            <tx:method name="query*" propagation="NOT_SUPPORTED" read-only="true" />  
        </tx:attributes>  
</tx:advice>  
  
<!--此处应想办法实现动态加载-->  
<aop:config>   
	<aop:advisor pointcut="execution(* com.cetc.datamc.app.collect.bs.*.*(..)) 
		or execution(* com.cetc.datamc.app.mdms.bs.*.*(..))" advice-ref="txAdviceHibernate" />   
</aop:config>  

其中<tx:advice>是用来声明事务性策略的。对于<tx:advice>来说,事务属性定义在<tx:attributes>元素中,该元素包含了一个或多个的<tx:method>元素。<tx:method>元素为某个(或某些)name属性(使用通配符)指定的方法定义事务的参数。 当使用<tx:advice>来声明事务时,你还需要一个事务管理器,就像使用TransactionProxyFactoryBean那样。根据约定优于配置,<tx:advice>假定事务管理器被声明为一个id为transactionManager的Bean。如果碰巧为事务管理器配置了一个不同的id(如txManager),则需要在transaction-manager属性中明确指定事务管理器的id:
	<tx:advice id="txAdviceHibernate" transaction-manager="txManager">  
        ...
    </tx:advice>
<tx:advice>只是定义了AOP通知,用于把事务边界通知给方法。但是这只是事务通知,而不是完整的事务性切面。我们在<tx:advice>中没有声明哪些Bean应该被通知——我们需要一个切点来做这件事。为了完整定义事务性切面,我们必须定义一个通知器(advisor)。这就涉及aop命名空间了。通过上面的配置示例可以看到:
这里的pointcut属性使用了AspectJ切入点表达式来表明通知器适用于com.cetc.datamc.app.collect.bs和com.cetc.datamc.app.mdms.bs包下的所有类的所有方法。哪些方法应该真正运行在事务中以及方法的事务属性都是由这个事务通知来定义的。而事务通知是advice-ref属性来指定的,它引用了名为txAdviceHibernate的通知。
在这里有一点需要注意的是对于切面的织入,不是说只有在<tx:method>中配置的方法才会被织入切面。根据切面的原理,是在pointcut中配置的符合条件的目标对象都会被织入切面。Spring中是基于动态代理的方式来实现切面的,所以Spring中的切面是由包裹了目标对象的代理类实现。代理类处理方法的调用,执行额外的切面逻辑。直到应用需要被代理的Bean时,Spring才创建代理对象。如果使用的是ApplicationContext,在ApplicationContext从BeanFactory中加载所有Bean时,Spring创建被代理的对象。这时候,所有符合在pointcut中配置的类的相关方法就会被织入切面代码,并且返回相关的动态代理对象。所以说不管有没有在<tx:method>中配置相关方法,只要符合pointcut中的条件,方法就会被织入切面代码。 因此,在上面的例子中,所有的com.cetc.datamc.app.collect.bs和com.cetc.datamc.app.mdms.bs包中所有类的所有方法都会被织入事务管理的相关切面代码。 在创建好动态代理后,之前所有依赖注入的相关接口的实现类就会替换为创建的动态代理类。这样程序在使用接口调用时,就会调用到织入切面代码的动态代理类实现。 因此,在使用声明式事务的时候也需要注意,所有需要管理事务的类必须通过spring context管理,直接new的对象不行。 所有需要事务管理的类必须使用interface,这样动态代理才有用。 例如假设在com.cetc.datamc.app.collect.bs包中有下面的类
public interface IDataCollect 
	public String addDataItem(String data);
 

public class DataCollectService implements IDataCollect  
	private IMetadataService metaService;
	
	public void setMetaService(IMetadataService metaService) 
		this.metaService = metaService;
	
	
	public String addDataItem(String data) 
		metaService.deleteMeta(data);
		return metaService.addMeta(data);
	


public interface IMetadataService 
	public void deleteMeta(String data);
	public String addMeta(String data);
	public String queryMetaId(String data);
 

public class MetadataService implements IMetadataService 
	public void deleteMeta(String data) 
		System.out.println("delete metadata");
	
	
	public String addMeta(String data) 
		System.out.println("add metadata");
		String id = queryMetaId(data);
		return data;
	
	
	public String queryMetaId(String data) 
		return "id";
	
这样,在DataCollectService中有一个 IMetadataService类型的对象。那么在实际中该类型则是被Spring替换为动态代理对象,而不是原来的MetaService对象。所以调用了IMetadataService中的deleteMeta和addMeta方法,这两个方法都是调用的织入事务切面代码的函数。而不是原来的 deleteMeta和addMeta方法。 另外在MetaService还可以看到addMeta中调用了queryMetaId方法。这个方法调用的就是没有织入切面代码的原来的函数。因为这个不是使用接口调用的,而是直接调用的对象本身的方法。所以就是该对象原来的queryMetaId方法。这就是为什么需要事务管理的类必须使用interface

事务管理的切面属于<aop:around>(AOP环绕通知)类型的切面。
---------------------------------TODO------------------------------------------------ org.springframework.transaction.interceptor.TransactionAspectSupport类
/**
	 * General delegate for around-advice-based subclasses, delegating to several other template
	 * methods on this class. Able to handle @link CallbackPreferringPlatformTransactionManager
	 * as well as regular @link PlatformTransactionManager implementations.
	 * @param method the Method being invoked
	 * @param targetClass the target class that we're invoking the method on
	 * @param invocation the callback to use for proceeding with the target invocation
	 * @return the return value of the method, if any
	 * @throws Throwable propagated from the target invocation
	 */
	protected Object invokeWithinTransaction(Method method, Class targetClass, 
						final InvocationCallback invocation)
			throws Throwable 

		// If the transaction attribute is null, the method is non-transactional.
		//这里返回的即是在tx:method中配置的属性值
		//包括propagation, isolation, readonly, timeout, readonly等属性的配置值。
		//如果一个方法没有在tx:method中配置,那么txAttr为null
		final TransactionAttribute txAttr = getTransactionAttributeSource().getTransactionAttribute(method, targetClass);
		final PlatformTransactionManager tm = determineTransactionManager(txAttr);
		final String joinpointIdentification = methodIdentification(method, targetClass);

		if (txAttr == null || !(tm instanceof CallbackPreferringPlatformTransactionManager)) 
			// Standard transaction demarcation with getTransaction and commit/rollback calls.
			//该方法主要就是根据配置来采取合适的事务策略。
			//理解了该方法的调用原理,基本上spring的配置事务的各属性的涵义也就理解了
			TransactionInfo txInfo = createTransactionIfNecessary(tm, txAttr, joinpointIdentification);
			Object retVal = null;
			try 
				// This is an around advice: Invoke the next interceptor in the chain.
				// This will normally result in a target object being invoked.
				retVal = invocation.proceedWithInvocation();
			
			catch (Throwable ex) 
				// target invocation exception
				completeTransactionAfterThrowing(txInfo, ex);
				throw ex;
			
			finally 
				cleanupTransactionInfo(txInfo);
			
			commitTransactionAfterReturning(txInfo);
			return retVal;
		

		else 
			// It's a CallbackPreferringPlatformTransactionManager: pass a TransactionCallback in.
			try 
				Object result = ((CallbackPreferringPlatformTransactionManager) tm).execute(txAttr,
						new TransactionCallback<Object>() 
							public Object doInTransaction(TransactionStatus status) 
								TransactionInfo txInfo = prepareTransactionInfo(
													tm, txAttr, joinpointIdentification, status);
								try 
									return invocation.proceedWithInvocation();
								
								catch (Throwable ex) 
									if (txAttr.rollbackOn(ex)) 
										// A RuntimeException: will lead to a rollback.
										if (ex instanceof RuntimeException) 
											throw (RuntimeException) ex;
										
										else 
											throw new ThrowableHolderException(ex);
										
									
									else 
										// A normal return value: will lead to a commit.
										return new ThrowableHolder(ex);
									
								
								finally 
									cleanupTransactionInfo(txInfo);
								
							
						);

				// Check result: It might indicate a Throwable to rethrow.
				if (result instanceof ThrowableHolder) 
					throw ((ThrowableHolder) result).getThrowable();
				
				else 
					return result;
				
			
			catch (ThrowableHolderException ex) 
				throw ex.getCause();
			
		
	
/**
	 * Create a transaction if necessary based on the given TransactionAttribute.
	 * <p>Allows callers to perform custom TransactionAttribute lookups through
	 * the TransactionAttributeSource.
	 * @param txAttr the TransactionAttribute (may be @code null)
	 * @param joinpointIdentification the fully qualified method name
	 * (used for monitoring and logging purposes)
	 * @return a TransactionInfo object, whether or not a transaction was created.
	 * The @code hasTransaction() method on TransactionInfo can be used to
	 * tell if there was a transaction created.
	 * @see #getTransactionAttributeSource()
	 */
	@SuppressWarnings("serial")
	protected TransactionInfo createTransactionIfNecessary(
			PlatformTransactionManager tm, TransactionAttribute txAttr, final String joinpointIdentification) 

		// If no name specified, apply method identification as transaction name.
		if (txAttr != null && txAttr.getName() == null) 
			txAttr = new DelegatingTransactionAttribute(txAttr) 
				@Override
				public String getName() 
					return joinpointIdentification;
				
			;
		

		TransactionStatus status = null;
		if (txAttr != null) 
			if (tm != null) 
				//重点看此函数
				status = tm.getTransaction(txAttr);
			
			else 
				if (logger.isDebugEnabled()) 
					logger.debug("Skipping transactional joinpoint [" + joinpointIdentification +
							"] because no transaction manager has been configured");
				
			
		
		//如果没有在tx:method中配置,那么txAttr=null, 这样返回的TransactionStatus也为null
		return prepareTransactionInfo(tm, txAttr, joinpointIdentification, status);
	

org.springframework.transaction.support.AbstractPlatformTransactionManager类
/**
	 * This implementation handles propagation behavior. Delegates to
	 * @code doGetTransaction, @code isExistingTransaction
	 * and @code doBegin.
	 * @see #doGetTransaction
	 * @see #isExistingTransaction
	 * @see #doBegin
	 */
	public final TransactionStatus getTransaction(TransactionDefinition definition) throws TransactionException 
		//doGetTransaction在各继承该类的子类中实现。
		//这里假设使用的是Hibernate的事务,那么实现类是HibernateTransactionManager类
		//这里的object即表示我们通常所说的Transaction,即事务的类的定义
		//对于Hibernate Transaction的具体类是HibernateTransactionObject 
		Object transaction = doGetTransaction();

		// Cache debug flag to avoid repeated checks.
		boolean debugEnabled = logger.isDebugEnabled();

		if (definition == null) 
			// Use defaults if no transaction definition given.
			definition = new DefaultTransactionDefinition();
		
		//根据得到的transaction对象判断是否当前已经存在事务
		//如果存在事务,那么调用handleExistingTransaction来根据配置的策略处理事务
		if (isExistingTransaction(transaction)) 
			// Existing transaction found -> check propagation behavior to find out how to behave.
			return handleExistingTransaction(definition, transaction, debugEnabled);
		

		// Check definition settings for new transaction.
		if (definition.getTimeout() < TransactionDefinition.TIMEOUT_DEFAULT) 
			throw new InvalidTimeoutException("Invalid transaction timeout", definition.getTimeout());
		
		//如果当前不存在事务,那么根据不同的策略配置来产生具体的事务行为。
		//这里即是各种不同配置策略行为的具体代码实现过程
		// No existing transaction found -> check propagation behavior to find out how to proceed.
		if (definition.getPropagationBehavior() == TransactionDefinition.PROPAGATION_MANDATORY) 
			throw new IllegalTransactionStateException(
					"No existing transaction found for transaction marked with propagation 'mandatory'");
		
		else if (definition.getPropagationBehavior() == TransactionDefinition.PROPAGATION_REQUIRED ||
				definition.getPropagationBehavior() == TransactionDefinition.PROPAGATION_REQUIRES_NEW ||
			definition.getPropagationBehavior() == TransactionDefinition.PROPAGATION_NESTED) 
			SuspendedResourcesHolder suspendedResources = suspend(null);
			if (debugEnabled) 
				logger.debug("Creating new transaction with name [" + definition.getName() + "]: " + definition);
			
			try 
				boolean newSynchronization = (getTransactionSynchronization() != SYNCHRONIZATION_NEVER);
				DefaultTransactionStatus status = newTransactionStatus(
						definition, transaction, true, newSynchronization, debugEnabled, suspendedResources);
				//开启新事务
				doBegin(transaction, definition);
				prepareSynchronization(status, definition);
				return status;
			
			catch (RuntimeException ex) 
				resume(null, suspendedResources);
				throw ex;
			
			catch (Error err) 
				resume(null, suspendedResources);
				throw err;
			
		
		else 
			// Create "empty" transaction: no actual transaction, but potentially synchronization.
			boolean newSynchronization = (getTransactionSynchronization() == SYNCHRONIZATION_ALWAYS);
			return prepareTransactionStatus(definition, null, true, newSynchronization, debugEnabled, null);
		
	
下面我们暂时先抛开如何doGetTransaction以及如何判断isExistingTransaction不谈,先来看看如何doBegin一个新事务。 在org.springframework.orm.hibernate3.HibernateTransactionManager类中的doBegin实现如下:
@Override
	protected void doBegin(Object transaction, TransactionDefinition definition) 
		HibernateTransactionObject txObject = (HibernateTransactionObject) transaction;

		if (txObject.hasConnectionHolder() && !txObject.getConnectionHolder().isSynchronizedWithTransaction()) 
			throw new IllegalTransactionStateException(
					"Pre-bound JDBC Connection found! HibernateTransactionManager does not support " +
					"running within DataSourceTransactionManager if told to manage the DataSource itself. " +
					"It is recommended to use a single HibernateTransactionManager for all transactions " +
					"on a single DataSource, no matter whether Hibernate or JDBC access.");
		

		Session session = null;

		try 
			//如果当前txObject没有已有的Session,那么打开一个新的Session.
			//txObject也可能已经有了Session,
			//例如如果配置了OpenSessionInViewFilter,且当前的线程是一个request线程,
			//那么就会在每个request线程中就先openSession,并且绑定到当前线程中。
			//这样在doGetTransaction就会获取到该Session。详见doGetTransaction函数和OpenSessionInViewFilter实现
			if (txObject.getSessionHolder() == null || txObject.getSessionHolder().isSynchronizedWithTransaction()) 
				Interceptor entityInterceptor = getEntityInterceptor();
				//这里是openSession,而不是getCurrentSession
				Session newSession = (entityInterceptor != null ?
						getSessionFactory().openSession(entityInterceptor) : getSessionFactory().openSession());
				if (logger.isDebugEnabled()) 
					logger.debug("Opened new Session [" + SessionFactoryUtils.toString(newSession) +
							"] for Hibernate transaction");
				
				//设置newSessionHolder和newSession状态为true
				txObject.setSession(newSession);
			

			session = txObject.getSessionHolder().getSession();

			if (this.prepareConnection && isSameConnectionForEntireSession(session)) 
				// We're allowed to change the transaction settings of the JDBC Connection.
				if (logger.isDebugEnabled()) 
					logger.debug(
							"Preparing JDBC Connection of Hibernate Session [" + SessionFactoryUtils.toString(session) + "]");
				
				Connection con = session.connection();
				Integer previousIsolationLevel = DataSourceUtils.prepareConnectionForTransaction(con, definition);
				txObject.setPreviousIsolationLevel(previousIsolationLevel);
			
			else 
				// Not allowed to change the transaction settings of the JDBC Connection.
				if (definition.getIsolationLevel() != TransactionDefinition.ISOLATION_DEFAULT) 
					// We should set a specific isolation level but are not allowed to...
					throw new InvalidIsolationLevelException(
							"HibernateTransactionManager is not allowed to support custom isolation levels: " +
							"make sure that its 'prepareConnection' flag is on (the default) and that the " +
							"Hibernate connection release mode is set to 'on_close' (SpringTransactionFactory's default). " +
							"Make sure that your LocalSessionFactoryBean actually uses SpringTransactionFactory: Your " +
							"Hibernate properties should *not* include a 'hibernate.transaction.factory_class' property!");
				
				if (logger.isDebugEnabled()) 
					logger.debug(
							"Not preparing JDBC Connection of Hibernate Session [" + SessionFactoryUtils.toString(session) + "]");
				
			
			//如果是一个new Session并且事务是只读的,则设置该Session flushMode为FlushMode.MANUAL
			if (definition.isReadOnly() && txObject.isNewSession()) 
				// Just set to MANUAL in case of a new Session for this transaction.
				session.setFlushMode(FlushMode.MANUAL);
			
			//如果事务不是只读的,且已有一个Session时,重新设置该Session的flushMode
			if (!definition.isReadOnly() && !txObject.isNewSession()) 
				// We need AUTO or COMMIT for a non-read-only transaction.
				FlushMode flushMode = session.getFlushMode();
				if (flushMode.lessThan(FlushMode.COMMIT)) 
					session.setFlushMode(FlushMode.AUTO);
					txObject.getSessionHolder().setPreviousFlushMode(flushMode);
				
			

			Transaction hibTx;
			//设置timeout 然后beginTransaction
			// Register transaction timeout.
			int timeout = determineTimeout(definition);
			if (timeout != TransactionDefinition.TIMEOUT_DEFAULT) 
				// Use Hibernate's own transaction timeout mechanism on Hibernate 3.1+
				// Applies to all statements, also to inserts, updates and deletes!
				hibTx = session.getTransaction();
				hibTx.setTimeout(timeout);
				hibTx.begin();
			
			else 
				// Open a plain Hibernate transaction without specified timeout.
				hibTx = session.beginTransaction();
			

			// Add the Hibernate transaction to the session holder.
			txObject.getSessionHolder().setTransaction(hibTx);

			// Register the Hibernate Session's JDBC Connection for the DataSource, if set.
			if (getDataSource() != null) 
				Connection con = session.connection();
				ConnectionHolder conHolder = new ConnectionHolder(con);
				if (timeout != TransactionDefinition.TIMEOUT_DEFAULT) 
					conHolder.setTimeoutInSeconds(timeout);
				
				if (logger.isDebugEnabled()) 
					logger.debug("Exposing Hibernate transaction as JDBC transaction [" + con + "]");
				
				//将当前的conHolder绑定到线程上,可以用来判断当前线程是否已有事务
				TransactionSynchronizationManager.bindResource(getDataSource(), conHolder);
				txObject.setConnectionHolder(conHolder);
			

			// Bind the session holder to the thread.
			if (txObject.isNewSessionHolder()) 
				//将当前的SessionHolder绑定到当前线程上,可以用来判断当前线程是否已有事务
				TransactionSynchronizationManager.bindResource(getSessionFactory(), txObject.getSessionHolder());
			
			txObject.getSessionHolder().setSynchronizedWithTransaction(true);
		

		catch (Throwable ex) 
			if (txObject.isNewSession()) 
				try 
					if (session.getTransaction().isActive()) 
						session.getTransaction().rollback();
					
				
				catch (Throwable ex2) 
					logger.debug("Could not rollback Session after failed transaction begin", ex);
				
				finally 
					SessionFactoryUtils.closeSession(session);
					txObject.setSessionHolder(null);
				
			
			throw new CannotCreateTransactionException("Could not open Hibernate Session for transaction", ex);
		
	
好了,知道了doBegin函数的实现原理,让我们回过头来看doGetTransaction和isExistingTransaction函数的实现。这样就很容易理解了。
@Override
	protected Object doGetTransaction() 
		HibernateTransactionObject txObject = new HibernateTransactionObject();
		txObject.setSavepointAllowed(isNestedTransactionAllowed());
		
		//获取当前线程中是否已经有已绑定的SessionHolder。如果当前线程已经存在事务,那么就会存在SessionHolder
		//具体可见doBegin函数的实现
		SessionHolder sessionHolder =
				(SessionHolder) TransactionSynchronizationManager.getResource(getSessionFactory());
		if (sessionHolder != null) 
			if (logger.isDebugEnabled()) 
				logger.debug("Found thread-bound Session [" +
						SessionFactoryUtils.toString(sessionHolder.getSession()) + "] for Hibernate transaction");
			
			//如果已经有了SessionHolder,那么直接就把当前已有的SessionHolder赋给新的transaction Object
			txObject.setSessionHolder(sessionHolder);
		
		else if (this.hibernateManagedSession) 
			try 
				//当前线程中已经绑定了session,例如openSessionInViewFilter的情况
				Session session = getSessionFactory().getCurrentSession();
				if (logger.isDebugEnabled()) 
					logger.debug("Found Hibernate-managed Session [" +
							SessionFactoryUtils.toString(session) + "] for Spring-managed transaction");
				
				txObject.setExistingSession(session);
			
			catch (HibernateException ex) 
				throw new DataAccessResourceFailureException(
						"Could not obtain Hibernate-managed Session for Spring-managed transaction", ex);
			
		
		//如果当前线程已经绑定了ConnectionHolder,如果当前线程中已经开启了事务,就会存在ConnectionHolder
		//直接将已有的ConnectionHolder赋值给该transaction
		if (getDataSource() != null) 
			ConnectionHolder conHolder = (ConnectionHolder)
					TransactionSynchronizationManager.getResource(getDataSource());
			txObject.setConnectionHolder(conHolder);
		

		return txObject;
	
@Override
protected boolean isExistingTransaction(Object transaction) 
	HibernateTransactionObject txObject = (HibernateTransactionObject) transaction;
	//spring管理的事务或者是Hibernate自己管理的事务
	return (txObject.hasSpringManagedTransaction() ||
			(this.hibernateManagedSession && txObject.hasHibernateManagedTransaction()));
isExistingTransaction函数里面的的hasSpringManagedTransaction的实现如下:
public boolean hasSpringManagedTransaction() 
			//如果当前的transaction Object的sessionHolder不为null,说明当前存在事务
			return (this.sessionHolder != null && this.sessionHolder.getTransaction() != null);
		
hasHibernateManagedTransaction函数实现如下:
public boolean hasHibernateManagedTransaction() 
            return this.sessionHolder != null && this.sessionHolder.getSession().getTransaction().isActive();
        
这两个函数都很简单,结合doGetTransaction函数的实现原理,就很容易理解了。 好了,最后再来看一下AbstractPlatformTransactionManager类中的AbstractPlatformTransactionManager函数的实现原理。这个函数也是主要根据不同的配置策略产生不同的事务行为。理解了该函数,也能对事务不同策略的配置理解的更加透彻了。
/**
	 * Create a TransactionStatus for an existing transaction.
	 */
	private TransactionStatus handleExistingTransaction(
			TransactionDefinition definition, Object transaction, boolean debugEnabled)
			throws TransactionException 
		//配置的策略为当前方法不应该运行在事务上下文中,如果当前正有一个事务运行,则会抛出异常
		if (definition.getPropagationBehavior() == TransactionDefinition.PROPAGATION_NEVER) 
			throw new IllegalTransactionStateException(
					"Existing transaction found for transaction marked with propagation 'never'");
		
		//配置的策略为当前方法不应该运行在事务中,如果存在当前事务,在该方法运行期间,当前事务将被挂起
		if (definition.getPropagationBehavior() == TransactionDefinition.PROPAGATION_NOT_SUPPORTED) 
			if (debugEnabled) 
				logger.debug("Suspending current transaction");
			
			Object suspendedResources = suspend(transaction);
			boolean newSynchronization = (getTransactionSynchronization() == SYNCHRONIZATION_ALWAYS);
			return prepareTransactionStatus(
					definition, null, false, newSynchronization, debugEnabled, suspendedResources);
		
		//配置的策略为当前方法必须运行在它自己的事务中,一个新的事务将被启动。
		//如果存在当前事务,在该方法运行期间,当前事务将被挂起
		if (definition.getPropagationBehavior() == TransactionDefinition.PROPAGATION_REQUIRES_NEW) 
			if (debugEnabled) 
				logger.debug("Suspending current transaction, creating new transaction with name [" +
						definition.getName() + "]");
			
			SuspendedResourcesHolder suspendedResources = suspend(transaction);
			try 
				boolean newSynchronization = (getTransactionSynchronization() != SYNCHRONIZATION_NEVER);
				DefaultTransactionStatus status = newTransactionStatus(
						definition, transaction, true, newSynchronization, debugEnabled, suspendedResources);
				doBegin(transaction, definition);
				prepareSynchronization(status, definition);
				return status;
			
			catch (RuntimeException beginEx) 
				resumeAfterBeginException(transaction, suspendedResources, beginEx);
				throw beginEx;
			
			catch (Error beginErr) 
				resumeAfterBeginException(transaction, suspendedResources, beginErr);
				throw beginErr;
			
		
		//配置的策略为如果当前已经存在一个事务,那么该方法将会在嵌套事务中运行。
		//嵌套的事务可以独立于当前事务进行单独地提交或回滚。如果当前事务不存在,
		//那么其行为与PROPAGATION_REQUIRED一样
		if (definition.getPropagationBehavior() == TransactionDefinition.PROPAGATION_NESTED) 
			if (!isNestedTransactionAllowed()) 
				throw new NestedTransactionNotSupportedException(
						"Transaction manager does not allow nested transactions by default - " +
						"specify 'nestedTransactionAllowed' property with value 'true'");
			
			if (debugEnabled) 
				logger.debug("Creating nested transaction with name [" + definition.getName() + "]");
			
			if (useSavepointForNestedTransaction()) 
				// Create savepoint within existing Spring-managed transaction,
				// through the SavepointManager API implemented by TransactionStatus.
				// Usually uses JDBC 3.0 savepoints. Never activates Spring synchronization.
				DefaultTransactionStatus status =
						prepareTransactionStatus(definition, transaction, false, false, debugEnabled, null);
				status.createAndHoldSavepoint();
				return status;
			
			else 
				// Nested transaction through nested begin and commit/rollback calls.
				// Usually only for JTA: Spring synchronization might get activated here
				// in case of a pre-existing JTA transaction.
				boolean newSynchronization = (getTransactionSynchronization() != SYNCHRONIZATION_NEVER);
				DefaultTransactionStatus status = newTransactionStatus(
						definition, transaction, true, newSynchronization, debugEnabled, null);
				doBegin(transaction, definition);
				prepareSynchronization(status, definition);
				return status;
			
		

		// Assumably PROPAGATION_SUPPORTS or PROPAGATION_REQUIRED.
		if (debugEnabled) 
			logger.debug("Participating in existing transaction");
		
		if (isValidateExistingTransaction()) 
			if (definition.getIsolationLevel() != TransactionDefinition.ISOLATION_DEFAULT) 
				Integer currentIsolationLevel = TransactionSynchronizationManager.getCurrentTransactionIsolationLevel();
				if (currentIsolationLevel == null || currentIsolationLevel != definition.getIsolationLevel()) 
					Constants isoConstants = DefaultTransactionDefinition.constants;
					throw new IllegalTransactionStateException("Participating transaction with definition [" +
							definition + "] specifies isolation level which is incompatible with existing transaction: " +
							(currentIsolationLevel != null ?
									isoConstants.toCode(currentIsolationLevel, DefaultTransactionDefinition.PREFIX_ISOLATION) :
									"(unknown)"));
				
			
			if (!definition.isReadOnly()) 
				if (TransactionSynchronizationManager.isCurrentTransactionReadOnly()) 
					throw new IllegalTransactionStateException("Participating transaction with definition [" +
							definition + "] is not marked as read-only but existing transaction is");
				
			
		
		boolean newSynchronization = (getTransactionSynchronization() != SYNCHRONIZATION_NEVER);
		return prepareTransactionStatus(definition, transaction, false, newSynchronization, debugEnabled, null);
	

以上是关于spring 声明式事务原理解读的主要内容,如果未能解决你的问题,请参考以下文章

spring中声明式事务 配置好后,在java代码中怎么使用

Spring-声明式事物

什么叫做spring的声明式事务

SpringBoot事务的使用(编程式声明式)及 @Transactional 工作原理失效处理

Spring的声明式事务处理怎么配置,可以在提交事务后关闭数据库连接?

声明式事务的介绍