Spring在事务管理时，对事务的处理做了极致的抽象，即PlatformTransactionManager。对事务的操作，简单地来说，只有三步操作：获取事务，提交事务，回滚事务。

public interface PlatformTransactionManager {// 获取事务TransactionStatus getTransaction(@Nullable TransactionDefinition definition) throws TransactionException;// 提交事务void commit(TransactionStatus status) throws TransactionException;// 回滚事务void rollback(TransactionStatus status) throws TransactionException;}

当然Spring不会仅仅只提供一个接口，同时会有一个抽象模版类，实现了事务管理的具体骨架。AbstractPlatformTransactionManager类可以说是Spring事务管理的控制台，决定事务如何创建，提交和回滚。

在Spring事务管理(二)-TransactionProxyFactoryBean原理中，分析TransactionInterceptor增强时，在invoke方法中最重要的三个操作：

1. 创建事务 createTransactionIfNecessary
2. 异常后事务处理 completeTransactionAfterThrowing
3. 方法执行成功后事务提交 commitTransactionAfterReturning

在具体操作中，最后都是通过事务管理器PlatformTransactionManager的接口实现来执行的，其实也就是上面列出的三个接口方法。我们分别介绍这三个方法的实现，并以DataSourceTransactionManager为实现类观察JDBC方式事务的具体实现。

1. 获取事务

getTransaction方法根据事务定义来获取事务状态，事务状态中记录了事务定义，事务对象及事务相关的资源信息。对于事务的获取，除了调用事务管理器的实现来获取事务对象本身外，另外的很重要的一点是处理了事务的传播方式。

public final TransactionStatus getTransaction(@Nullable TransactionDefinition definition) throws TransactionException {// 1.获取事务对象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();}// 2.如果已存在事务，根据不同的事务传播方式处理获取事务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());}// 3. 如果当前没有事务，不同的事务传播方式不同处理方式// 3.1 事务传播方式为mandatory(强制必须有事务)，则抛出异常// 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'");}// 3.2 事务传播方式为required或required_new或nested(嵌套)，创建一个新的事务状态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);// 创建新的事务状态对象DefaultTransactionStatus status = newTransactionStatus(definition, transaction, true, newSynchronization, debugEnabled, suspendedResources);// 事务初始化doBegin(transaction, definition);// 准备其他同步操作prepareSynchronization(status, definition);return status;}catch (RuntimeException | Error ex) {resume(null, suspendedResources);throw ex;}}// 3.3 其他事务传播方式，返回一个事务对象为null的事务状态对象else {// Create "empty" transaction: no actual transaction, but potentially synchronization.if (definition.getIsolationLevel() != TransactionDefinition.ISOLATION_DEFAULT && logger.isWarnEnabled()) {logger.warn("Custom isolation level specified but no actual transaction initiated; " +"isolation level will effectively be ignored: " + definition);}boolean newSynchronization = (getTransactionSynchronization() == SYNCHRONIZATION_ALWAYS);return prepareTransactionStatus(definition, null, true, newSynchronization, debugEnabled, null);}
}

获取事务的方法主要做两件事情：

1. 获取事务对象
2. 根据事务传播方式返回事务状态对象

获取事务对象，在DataSourceTransactionManager的实现中，返回一个DataSourceTransactionObject对象

protected Object doGetTransaction() {DataSourceTransactionObject txObject = new DataSourceTransactionObject();txObject.setSavepointAllowed(isNestedTransactionAllowed());ConnectionHolder conHolder =(ConnectionHolder) // 从事务同步管理器中根据DataSource获取数据库连接资源     TransactionSynchronizationManager.getResource(obtainDataSource());txObject.setConnectionHolder(conHolder, false);return txObject;
}

每次执行doGetTransaction方法，即会创建一个DataSourceTransactionObject对象txObject，并从事务同步管理器中根据DataSource获取数据库连接持有对象ConnectionHolder，然后存入txObject中。**事务同步管理类持有一个ThreadLocal级别的resources对象，存储DataSource和ConnectionHolder的映射关系。**因此返回的txObject中持有的ConnectionHolder可能有值，也可能为空。而不同的事务传播方式下，事务管理的处理根据txObejct中是否存在事务有不同的处理方式。

关于关注事务传播方式的实现，很多人对事务传播方式都是一知半解，只是因为没有了解源码的实现。现在就来看看具体的实现。事务传播方式的实现分为两种情况，事务不存在和事务已经存在。isExistingTransaction方法判断事务是否存在，默认在AbstractPlatformTransactionManager抽象类中返回false，而在DataSourceTransactionManager实现中，则根据是否有数据库连接来决定。

protected boolean isExistingTransaction(Object transaction) {DataSourceTransactionObject txObject = (DataSourceTransactionObject) transaction;return (txObject.hasConnectionHolder() && txObject.getConnectionHolder().isTransactionActive());
}

当前无事务

如果当前没有事务，则不同事务传播方式的处理如下：

1. 事务传播方式为mandatory(强制必须有事务)，当前没有事务，即抛出异常。
2. 事务传播方式为required或required_new或nested(嵌套)，当前没有事务，即会创建一个新的事务状态。
3. 其他事务传播方式时，直接返回事务对象为null的事务状态对象，即不在事务中执行。

如何创建一个新的事务状态

// 1. 新建事务状态,返回DefaultTransactionStatus对象
DefaultTransactionStatus status = newTransactionStatus(definition, transaction, true, newSynchronization, debugEnabled, suspendedResources);
// 2. 事务初始化
doBegin(transaction, definition);
// 3. 准备同步操作
prepareSynchronization(status, definition);
return status;

第一步，新建事务状态，就是构建一个DefaultTransactionStatus对象

protected DefaultTransactionStatus newTransactionStatus(TransactionDefinition definition, @Nullable Object transaction, boolean newTransaction,boolean newSynchronization, boolean debug, @Nullable Object suspendedResources) {boolean actualNewSynchronization = newSynchronization &&!TransactionSynchronizationManager.isSynchronizationActive();return new DefaultTransactionStatus(transaction, newTransaction, actualNewSynchronization,definition.isReadOnly(), debug, suspendedResources);
}

第二步，事务初始化，AbstractPlatformTransactionManager没有实现，来看DataSourceTransactionManager的实现：获取一个新的数据库连接并开启事务，完成事务的基本属性设置。

protected void doBegin(Object transaction, TransactionDefinition definition) {DataSourceTransactionObject txObject = (DataSourceTransactionObject) transaction;Connection con = null;try {// 如果当前没有事务，由DataSource获取一个新的数据库连接，并赋予txObjectif (!txObject.hasConnectionHolder() ||txObject.getConnectionHolder().isSynchronizedWithTransaction()) {Connection newCon = obtainDataSource().getConnection();if (logger.isDebugEnabled()) {logger.debug("Acquired Connection [" + newCon + "] for JDBC transaction");}txObject.setConnectionHolder(new ConnectionHolder(newCon), true);}// 设置数据库连接与事务同步txObject.getConnectionHolder().setSynchronizedWithTransaction(true);con = txObject.getConnectionHolder().getConnection();// 设置事务隔离级别Integer previousIsolationLevel = DataSourceUtils.prepareConnectionForTransaction(con, definition);txObject.setPreviousIsolationLevel(previousIsolationLevel);// Switch to manual commit if necessary. This is very expensive in some JDBC drivers,// so we don't want to do it unnecessarily (for example if we've explicitly// configured the connection pool to set it already).// 非常重要的一点，JDBC通过设置自动提交为false，开启一个新的事务if (con.getAutoCommit()) {txObject.setMustRestoreAutoCommit(true);if (logger.isDebugEnabled()) {logger.debug("Switching JDBC Connection [" + con + "] to manual commit");}con.setAutoCommit(false);}// 如果设置事务只读属性，执行Statement设置只读prepareTransactionalConnection(con, definition);// 激活事务状态txObject.getConnectionHolder().setTransactionActive(true);// 设置超时时间int timeout = determineTimeout(definition);if (timeout != TransactionDefinition.TIMEOUT_DEFAULT) {txObject.getConnectionHolder().setTimeoutInSeconds(timeout);}// Bind the connection holder to the thread.// 如果为新连接，绑定DataSource和数据库连接持有者的映射关系if (txObject.isNewConnectionHolder()) {TransactionSynchronizationManager.bindResource(obtainDataSource(), txObject.getConnectionHolder());}}catch (Throwable ex) {if (txObject.isNewConnectionHolder()) {DataSourceUtils.releaseConnection(con, obtainDataSource());txObject.setConnectionHolder(null, false);}throw new CannotCreateTransactionException("Could not open JDBC Connection for transaction", ex);}
}

第三步：准备同步操作，如果事务状态开启同步，则在事务同步管理器中设置事务基础属性

protected void prepareSynchronization(DefaultTransactionStatus status, TransactionDefinition definition) {if (status.isNewSynchronization()) {TransactionSynchronizationManager.setActualTransactionActive(status.hasTransaction());TransactionSynchronizationManager.setCurrentTransactionIsolationLevel(definition.getIsolationLevel() != TransactionDefinition.ISOLATION_DEFAULT ?definition.getIsolationLevel() : null);TransactionSynchronizationManager.setCurrentTransactionReadOnly(definition.isReadOnly());TransactionSynchronizationManager.setCurrentTransactionName(definition.getName());TransactionSynchronizationManager.initSynchronization();}
}

当前有事务

如果当前已经有事务存在，由handleExistingTransaction方法完成事务操作。

1. 传播方式为never（不允许事务），抛出异常
2. 传播方式为not_supported（不支持），则挂起当前事务，以无事务方式运行
3. 传播方式为required_new，挂起原有事务，并开启新的事务
4. 传播方式为nested（嵌套）,创建嵌套事务。这里一般方式都是通过savepoint保存点的完成嵌套，但Spring对JTA事务单独做了另一种处理。
5. 传播方式为supports或required，返回当前事务

private TransactionStatus handleExistingTransaction(TransactionDefinition definition, Object transaction, boolean debugEnabled)throws TransactionException {// 传播方式为never（不允许事务），抛出异常if (definition.getPropagationBehavior() == TransactionDefinition.PROPAGATION_NEVER) {throw new IllegalTransactionStateException("Existing transaction found for transaction marked with propagation 'never'");}// 传播方式为not_supported（不支持），则挂起当前事务，以无事务方式运行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);}// 传播方式为required_new，挂起原有事务，并开启新的事务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 | Error beginEx) {resumeAfterBeginException(transaction, suspendedResources, beginEx);throw beginEx;}}// 传播方式为nested（嵌套）,创建嵌套事务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() + "]");}// 使用保存点支持嵌套事务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;}// 只适用于JTA事务：通过嵌套的begin和commit/rollback创建嵌套事务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.// 传播方式为supports或required，返回当前事务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);
}

这里关注两个点

第一是事务的挂起，Spring并不是真的对数据库连接做了什么挂起操作，而是在逻辑上由事务同步管理器做了事务信息和状态的重置，并将原事务信息和状态返回，并记录在新的事务状态对象中，从而形成一种链式结构。

protected final SuspendedResourcesHolder suspend(@Nullable Object transaction) throws TransactionException {if (TransactionSynchronizationManager.isSynchronizationActive()) {List<TransactionSynchronization> suspendedSynchronizations = doSuspendSynchronization();try {Object suspendedResources = null;if (transaction != null) {suspendedResources = doSuspend(transaction);}String name = TransactionSynchronizationManager.getCurrentTransactionName();TransactionSynchronizationManager.setCurrentTransactionName(null);boolean readOnly = TransactionSynchronizationManager.isCurrentTransactionReadOnly();TransactionSynchronizationManager.setCurrentTransactionReadOnly(false);Integer isolationLevel = TransactionSynchronizationManager.getCurrentTransactionIsolationLevel();TransactionSynchronizationManager.setCurrentTransactionIsolationLevel(null);boolean wasActive = TransactionSynchronizationManager.isActualTransactionActive();TransactionSynchronizationManager.setActualTransactionActive(false);return new SuspendedResourcesHolder(suspendedResources, suspendedSynchronizations, name, readOnly, isolationLevel, wasActive);}catch (RuntimeException | Error ex) {// doSuspend failed - original transaction is still active...doResumeSynchronization(suspendedSynchronizations);throw ex;}}else if (transaction != null) {// Transaction active but no synchronization active.Object suspendedResources = doSuspend(transaction);return new SuspendedResourcesHolder(suspendedResources);}else {// Neither transaction nor synchronization active.return null;}
}

第二是嵌套事务设置保存点，通常由JDBC3.0支持的savepoint API完成，然后将保存点记录在事务状态中。

DefaultTransactionStatus status =prepareTransactionStatus(definition, transaction, false, false, debugEnabled, null);
status.createAndHoldSavepoint();
return status;

至此，完成了获取事务

2.提交事务

首先要说的，commit方法并不是一定提交事务，也可能回滚。

public final void commit(TransactionStatus status) throws TransactionException {// 事务已经提交，再次提交抛出异常if (status.isCompleted()) {throw new IllegalTransactionStateException("Transaction is already completed - do not call commit or rollback more than once per transaction");}DefaultTransactionStatus defStatus = (DefaultTransactionStatus) status;// 如果事务状态设置了回滚标识，则执行回滚if (defStatus.isLocalRollbackOnly()) {if (defStatus.isDebug()) {logger.debug("Transactional code has requested rollback");}processRollback(defStatus, false);return;}// 设置全局回滚标识为true，则执行回滚if (!shouldCommitOnGlobalRollbackOnly() && defStatus.isGlobalRollbackOnly()) {if (defStatus.isDebug()) {logger.debug("Global transaction is marked as rollback-only but transactional code requested commit");}processRollback(defStatus, true);return;}// 提交事务processCommit(defStatus);
}

processCommit执行事务的提交，但事务的提交也分为几种情况：

1. 存在保存点，即嵌套事务，则释放保存点
2. 如果事务是由当前事务状态开启的，即事务传播的第一层，执行事务提交
3. 其他情况（比如事务是继承自上一层），则不做任何操作

且在processCommit方法中，不同时候设置了不同状态的触发监控，用来提示事务同步相关资源，触发需要的操作。

private void processCommit(DefaultTransactionStatus status) throws TransactionException {try {boolean beforeCompletionInvoked = false;try {boolean unexpectedRollback = false;prepareForCommit(status);// 提交前提示triggerBeforeCommit(status);// 完成前提示triggerBeforeCompletion(status);beforeCompletionInvoked = true;if (status.hasSavepoint()) {if (status.isDebug()) {logger.debug("Releasing transaction savepoint");}unexpectedRollback = status.isGlobalRollbackOnly();status.releaseHeldSavepoint();}else if (status.isNewTransaction()) {if (status.isDebug()) {logger.debug("Initiating transaction commit");}unexpectedRollback = status.isGlobalRollbackOnly();// 执行事务提交doCommit(status);}else if (isFailEarlyOnGlobalRollbackOnly()) {unexpectedRollback = status.isGlobalRollbackOnly();}// Throw UnexpectedRollbackException if we have a global rollback-only// marker but still didn't get a corresponding exception from commit.if (unexpectedRollback) {throw new UnexpectedRollbackException("Transaction silently rolled back because it has been marked as rollback-only");}}catch (UnexpectedRollbackException ex) {// can only be caused by doCommit// 回滚完成提示triggerAfterCompletion(status, TransactionSynchronization.STATUS_ROLLED_BACK);throw ex;}catch (TransactionException ex) {// can only be caused by doCommitif (isRollbackOnCommitFailure()) {doRollbackOnCommitException(status, ex);}else {// 未知状态完成提示triggerAfterCompletion(status, TransactionSynchronization.STATUS_UNKNOWN);}throw ex;}catch (RuntimeException | Error ex) {if (!beforeCompletionInvoked) {triggerBeforeCompletion(status);}doRollbackOnCommitException(status, ex);throw ex;}// Trigger afterCommit callbacks, with an exception thrown there// propagated to callers but the transaction still considered as committed.try {// 事务提交完成提示triggerAfterCommit(status);}finally {// 操作完成完成提示triggerAfterCompletion(status, TransactionSynchronization.STATUS_COMMITTED);}}finally {// 完成后清理cleanupAfterCompletion(status);}
}

DataSourceTransactionManager对doCommit的实现，就是执行数据库连接的提交

protected void doCommit(DefaultTransactionStatus status) {DataSourceTransactionObject txObject = (DataSourceTransactionObject) status.getTransaction();Connection con = txObject.getConnectionHolder().getConnection();if (status.isDebug()) {logger.debug("Committing JDBC transaction on Connection [" + con + "]");}try {// 提交事务con.commit();}catch (SQLException ex) {throw new TransactionSystemException("Could not commit JDBC transaction", ex);}
}

3.回滚事务

回滚事务时也分为几种情况：

1. 存在保存点（嵌套事务），则回滚到保存点
2. 如果事务是由当前事务状态开启的，则执行回滚操作
3. 其他情况下，如果事务状态设置了回滚标识，则设置事务对象的状态也为回滚，否则不做任何操作

private void processRollback(DefaultTransactionStatus status, boolean unexpected) {try {boolean unexpectedRollback = unexpected;try {triggerBeforeCompletion(status);if (status.hasSavepoint()) {if (status.isDebug()) {logger.debug("Rolling back transaction to savepoint");}// 回滚保存点status.rollbackToHeldSavepoint();}else if (status.isNewTransaction()) {if (status.isDebug()) {logger.debug("Initiating transaction rollback");}// 回滚事务doRollback(status);}else {// Participating in larger transactionif (status.hasTransaction()) {if (status.isLocalRollbackOnly() || isGlobalRollbackOnParticipationFailure()) {if (status.isDebug()) {logger.debug("Participating transaction failed - marking existing transaction as rollback-only");}doSetRollbackOnly(status);}else {if (status.isDebug()) {logger.debug("Participating transaction failed - letting transaction originator decide on rollback");}}}else {logger.debug("Should roll back transaction but cannot - no transaction available");}// Unexpected rollback only matters here if we're asked to fail earlyif (!isFailEarlyOnGlobalRollbackOnly()) {unexpectedRollback = false;}}}catch (RuntimeException | Error ex) {triggerAfterCompletion(status, TransactionSynchronization.STATUS_UNKNOWN);throw ex;}triggerAfterCompletion(status, TransactionSynchronization.STATUS_ROLLED_BACK);// Raise UnexpectedRollbackException if we had a global rollback-only markerif (unexpectedRollback) {throw new UnexpectedRollbackException("Transaction rolled back because it has been marked as rollback-only");}}finally {cleanupAfterCompletion(status);}
}

对于DataSourceTransactionManager实现，回滚保存点和回滚事务都由JDBC的API来完成。

至此，事务管理器对事务的三种操作就简单地介绍完了，但其中事务同步资源的控制十分精妙，这里就不做详细的介绍。有兴趣的自己去研究源码。于我而言，任何强大的机制都是由一行行地源码精妙地组建出来的，深入进去，一切都将真相大白。