MyBatis之SqlSession介绍

2321321hhh

SqlSession的创建

Sqlsession对应着一次数据库会话。由于数据库回话不是永久的，因此Sqlsession的生命周期也不应该是永久的，相反，在你每次访问数据库时都需要创建它（当然并不是说在Sqlsession里只能执行一次sql，你可以执行多次，当一旦关闭了Sqlsession就需要重新创建它）。创建Sqlsession的地方只有一个，那就是SqlsessionFactory的openSession方法：

package org.apache.ibatis.session;

import java.sql.Connection;

public interface SqlSessionFactory {

  SqlSession openSession();

  SqlSession openSession(boolean autoCommit);
  SqlSession openSession(Connection connection);
  SqlSession openSession(TransactionIsolationLevel level);

  SqlSession openSession(ExecutorType execType);
  SqlSession openSession(ExecutorType execType, boolean autoCommit);
  SqlSession openSession(ExecutorType execType, TransactionIsolationLevel level);
  SqlSession openSession(ExecutorType execType, Connection connection);

  Configuration getConfiguration();

}

实际创建SqlSession的地方是openSessionFromDataSource，如下：

package org.apache.ibatis.session.defaults;

import java.sql.Connection;
import java.sql.SQLException;

import org.apache.ibatis.exceptions.ExceptionFactory;
import org.apache.ibatis.executor.ErrorContext;
import org.apache.ibatis.executor.Executor;
import org.apache.ibatis.mapping.Environment;
import org.apache.ibatis.session.Configuration;
import org.apache.ibatis.session.ExecutorType;
import org.apache.ibatis.session.SqlSession;
import org.apache.ibatis.session.SqlSessionFactory;
import org.apache.ibatis.session.TransactionIsolationLevel;
import org.apache.ibatis.transaction.Transaction;
import org.apache.ibatis.transaction.TransactionFactory;
import org.apache.ibatis.transaction.managed.ManagedTransactionFactory;

public class DefaultSqlSessionFactory implements SqlSessionFactory {

  private final Configuration configuration;

  public DefaultSqlSessionFactory(Configuration configuration) {
    this.configuration = configuration;
  }

  public SqlSession openSession() {
    return openSessionFromDataSource(configuration.getDefaultExecutorType(), null, false);
  }
  
  private SqlSession openSessionFromDataSource(ExecutorType execType, TransactionIsolationLevel level, boolean autoCommit) {
    Transaction tx = null;
    try {
      final Environment environment = configuration.getEnvironment();
      final TransactionFactory transactionFactory = getTransactionFactoryFromEnvironment(environment);
      tx = transactionFactory.newTransaction(environment.getDataSource(), level, autoCommit);
      final Executor executor = configuration.newExecutor(tx, execType, autoCommit);
      return new DefaultSqlSession(configuration, executor);
    } catch (Exception e) {
      closeTransaction(tx); // may have fetched a connection so lets call close()
      throw ExceptionFactory.wrapException("Error opening session.  Cause: " + e, e);
    } finally {
      ErrorContext.instance().reset();
    }
  }
}

可以看出，创建sqlsession经过了以下几个主要步骤：

1)       从配置中获取Environment；

2)       从Environment中取得DataSource；

3)       从Environment中取得TransactionFactory；

4)       从DataSource里获取数据库连接对象Connection；

5)       在取得的数据库连接上创建事务对象Transaction；

6)       创建Executor对象（该对象非常重要，事实上sqlsession的所有操作都是通过它完成的）；

7)       创建sqlsession对象。


Executor的创建

Sqlsession只是个门面，真正干事的是Executor，Sqlsession对数据库的操作都是通过Executor来完成的。与Sqlsession一样，Executor也是动态创建的：

final Executor executor = configuration.newExecutor(tx, execType, autoCommit);

Configuration类的实现：

public class Configuration {

  protected Environment environment;
  
  public Executor newExecutor(Transaction transaction, ExecutorType executorType, boolean autoCommit) {
    executorType = executorType == null ? defaultExecutorType : executorType;
    executorType = executorType == null ? ExecutorType.SIMPLE : executorType;
    Executor executor;
    if (ExecutorType.BATCH == executorType) {
      executor = new BatchExecutor(this, transaction);
    } else if (ExecutorType.REUSE == executorType) {
      executor = new ReuseExecutor(this, transaction);
    } else {
      executor = new SimpleExecutor(this, transaction);
    }
    if (cacheEnabled) {
      executor = new CachingExecutor(executor, autoCommit);
    }
    executor = (Executor) interceptorChain.pluginAll(executor);
    return executor;
  }
}

Executor的定义：

package org.apache.ibatis.executor;

import java.sql.SQLException;
import java.util.List;

import org.apache.ibatis.cache.CacheKey;
import org.apache.ibatis.mapping.BoundSql;
import org.apache.ibatis.mapping.MappedStatement;
import org.apache.ibatis.reflection.MetaObject;
import org.apache.ibatis.session.ResultHandler;
import org.apache.ibatis.session.RowBounds;
import org.apache.ibatis.transaction.Transaction;

public interface Executor {

  ResultHandler NO_RESULT_HANDLER = null;

  int update(MappedStatement ms, Object parameter) throws SQLException;

  <E> List<E> query(MappedStatement ms, Object parameter, RowBounds rowBounds, ResultHandler resultHandler, CacheKey cacheKey, BoundSql boundSql) throws SQLException;

  <E> List<E> query(MappedStatement ms, Object parameter, RowBounds rowBounds, ResultHandler resultHandler) throws SQLException;

  List<BatchResult> flushStatements() throws SQLException;

  void commit(boolean required) throws SQLException;

  void rollback(boolean required) throws SQLException;

  CacheKey createCacheKey(MappedStatement ms, Object parameterObject, RowBounds rowBounds, BoundSql boundSql);

  boolean isCached(MappedStatement ms, CacheKey key);

  void clearLocalCache();

  void deferLoad(MappedStatement ms, MetaObject resultObject, String property, CacheKey key, Class<?> targetType);

  Transaction getTransaction();

  void close(boolean forceRollback);

  boolean isClosed();

}

wKioL1h3P17iInW1AAAuMO9C85k200.jpg

可以看出，创建的Executor只是3中基础类型之一：

      BatchExecutor专门用于执行批量sql操作

      ReuseExecutor会重用statement执行sql操作

      SimpleExecutor只是简单执行sql没有什么特别的。

如果开启cache的话（默认是开启的并且没有任何理由去关闭它），就会创建CachingExecutor，它以前面创建的Executor作为唯一参数。CachingExecutor在查询数据库前先查找缓存，若没找到的话调用delegate（就是构造时传入的Executor对象）从数据库查询，并将查询结果存入缓存中。

package org.apache.ibatis.executor;

import java.sql.SQLException;
import java.util.List;

import org.apache.ibatis.cache.Cache;
import org.apache.ibatis.cache.CacheKey;
import org.apache.ibatis.cache.TransactionalCacheManager;
import org.apache.ibatis.mapping.BoundSql;
import org.apache.ibatis.mapping.MappedStatement;
import org.apache.ibatis.mapping.ParameterMapping;
import org.apache.ibatis.mapping.ParameterMode;
import org.apache.ibatis.mapping.StatementType;
import org.apache.ibatis.reflection.MetaObject;
import org.apache.ibatis.session.ResultHandler;
import org.apache.ibatis.session.RowBounds;
import org.apache.ibatis.transaction.Transaction;

public class CachingExecutor implements Executor {

  private Executor delegate;
  private boolean autoCommit; // issue #573. No need to call commit() on autoCommit sessions
  private TransactionalCacheManager tcm = new TransactionalCacheManager();

  private boolean dirty;

  public CachingExecutor(Executor delegate) {
    this(delegate, false);
  }

  public CachingExecutor(Executor delegate, boolean autoCommit) {
    this.delegate = delegate;
    this.autoCommit = autoCommit;
  }

  public Transaction getTransaction() {
    return delegate.getTransaction();
  }

  public void close(boolean forceRollback) {
    try {
      //issue #499. Unresolved session handling
      //issue #573. Autocommit sessions should commit
      if (dirty && !autoCommit) {
        tcm.rollback();
      } else {
        tcm.commit();
      }
    } finally {
      delegate.close(forceRollback);
    }
  }
  
  // 其他代码略  
  
}

Executor对象是可以被插件拦截的（拦截器Interceptor，如：

@Intercepts({@Signature(type = Executor.class, method = "query",
        args = {MappedStatement.class, Object.class, RowBounds.class, ResultHandler.class})})

），如果定义了针对Executor类型的插件，最终生成的Executor对象是被各个插件插入后的代理对象。


Mapper介绍

Mybatis官方手册建议通过mapper对象访问mybatis，因为使用mapper看起来更优雅，就像下面这样：

session = sqlSessionFactory.openSession();  
UserMapper userMapper= session.getMapper(UserMapper.class);  
UserBo user = new UserBo();  
user.setUsername("iMbatis");  
user.setPassword("iMbatis");  
userMapper.insertUser(user);

那么这个mapper到底是什么呢，它是如何创建的呢，它又是怎么与sqlsession等关联起来的呢？下面为你一一解答。

表面上看mapper是在sqlsession里创建的，但实际创建它的地方是MapperRegistry：

package org.apache.ibatis.session;

import java.io.Closeable;
import java.sql.Connection;
import java.util.List;
import java.util.Map;

import org.apache.ibatis.executor.BatchResult;

/**
* The primary Java interface for working with MyBatis.
* Through this interface you can execute commands, get mappers and manage transactions.
*
*/
public interface SqlSession extends Closeable {

  <T> T selectOne(String statement);

  <T> T selectOne(String statement, Object parameter);

  <E> List<E> selectList(String statement);

  <E> List<E> selectList(String statement, Object parameter);

  <E> List<E> selectList(String statement, Object parameter, RowBounds rowBounds);

  <K, V> Map<K, V> selectMap(String statement, String mapKey);

  <K, V> Map<K, V> selectMap(String statement, Object parameter, String mapKey);

  <K, V> Map<K, V> selectMap(String statement, Object parameter, String mapKey, RowBounds rowBounds);

  void select(String statement, Object parameter, ResultHandler handler);

  void select(String statement, ResultHandler handler);

  void select(String statement, Object parameter, RowBounds rowBounds, ResultHandler handler);

  int insert(String statement);

  int insert(String statement, Object parameter);

  int update(String statement);

  int update(String statement, Object parameter);

  int delete(String statement);


  int delete(String statement, Object parameter);

  void commit();

  void commit(boolean force);

  void rollback();

  void rollback(boolean force);

  List<BatchResult> flushStatements();

  void close();

  void clearCache();

  Configuration getConfiguration();

  <T> T getMapper(Class<T> type);

  Connection getConnection();
}

我问你

谢谢分享