今天主要讲解下常用的Handler，Handler讲完之后，就能对dbutils整体有了了解了。

来看下ResultSetHandler接口

/** * Implementations of this interface convert ResultSets into other objects. * 此接口的实现类将结果集转化为其他的对象 * @param 
    
      the target type the input ResultSet will be converted to. */public interface ResultSetHandler
     
       {    /**     * Turn the 
      ResultSet into an Object.     * 将结果集转换成对象     * @param rs The 
      ResultSet to handle.  It has not been touched     * before being passed to this method.     *     * @return An Object initialized with 
      ResultSet data. It is     * legal for implementations to return 
      null if the     * 
      ResultSet contained 0 rows.     *     * @throws SQLException if a database access error occurs     */    T handle(ResultSet rs) throws SQLException;}
     
    

它就只有一个方法声明，很简单，就不讲什么。

在这里我会讲解：BeanHandler、ArrayHandler、BeanListHandler、MapHandler、ScalarHandler这5个Handler。剩下的AbstractKeyedHandler以及实现类和AbstractListHandler及实现类，会在下一篇里讲到。

BeanHandler源码：

/** * ResultSetHandler implementation that converts the first * ResultSet row into a JavaBean. This class is thread safe. * ResultSetHandler实现类将结果集的第一行转换成一个javabean * @param 
    
      the target bean type * @see org.apache.commons.dbutils.ResultSetHandler */public class BeanHandler
     
       implements ResultSetHandler
      
        {    /**     * The Class of beans produced by this handler.     * 这个handler要返回的bean的Class     */    private final Class
       
         type;    /**     * The RowProcessor implementation to use when converting rows     * into beans.     * 将行转换成bean时，使用RowProcessor的哪个实现类。默认使用的是BasicRowProcessor     */    private final RowProcessor convert;    /**     * Creates a new instance of BeanHandler.     * 创建一个BeanHandler对象，默认使用的是BasicRowProcessor     * @param type The Class that objects returned from 
        handle()     * are created from.     */    public BeanHandler(Class
        
          type) {    	//static final RowProcessor ROW_PROCESSOR = new BasicRowProcessor();        this(type, ArrayHandler.ROW_PROCESSOR);    }    /**     * Creates a new instance of BeanHandler.     * 创建一个BeanHandler对象。可以自定义RowProcessor     * @param type The Class that objects returned from 
         handle()     * are created from.     * @param convert The 
         RowProcessor implementation     * to use when converting rows into beans.     */    public BeanHandler(Class
         
           type, RowProcessor convert) { this.type = type; this.convert = convert; } /** * Convert the first row of the 
          ResultSet into a bean with the * 
          Class given in the constructor. * 将RS中的第一行记录转换成一个bean对象，这个bean对象通过在构造器中给定的Class创建。使用了反射。 * @param rs 
          ResultSet to process. * @return An initialized JavaBean or 
          null if there were no * rows in the 
          ResultSet. * * @throws SQLException if a database access error occurs * @see org.apache.commons.dbutils.ResultSetHandler#handle(java.sql.ResultSet) */ @Override public T handle(ResultSet rs) throws SQLException { /* * 如果有RS中有记录的时候才进行处理，否则返回null。属性convert在构造函数中赋值了 */ return rs.next() ? this.convert.toBean(rs, this.type) : null; }}
         
        
       
      
     
    

BeanHandler的实现简单，就是调用了BasicRowProcessor中的toBean方法。

BeanListHandler源码：

/** * ResultSetHandler implementation that converts a * ResultSet into a List of beans. This class is * thread safe. * 此实现类将结果集转换成bean的集合 * @param 
    
      the target bean type * @see org.apache.commons.dbutils.ResultSetHandler */public class BeanListHandler
     
       implements ResultSetHandler
      
       
        > {    /**     * The Class of beans produced by this handler.     */    private final Class
        
          type;    /**     * The RowProcessor implementation to use when converting rows     * into beans.     * 使用的还是BasicRowProcessor     */    private final RowProcessor convert;    /**     * Creates a new instance of BeanListHandler.     *     * @param type The Class that objects returned from 
         handle()     * are created from.     */    public BeanListHandler(Class
         
           type) { this(type, ArrayHandler.ROW_PROCESSOR); } /** * Creates a new instance of BeanListHandler. * * @param type The Class that objects returned from 
          handle() * are created from. * @param convert The 
          RowProcessor implementation * to use when converting rows into beans. */ public BeanListHandler(Class
          
            type, RowProcessor convert) { this.type = type; this.convert = convert; } /** * Convert the whole 
           ResultSet into a List of beans with * the 
           Class given in the constructor. * 调用了BasicRowProcessor的toBeanList方法。 * @param rs The 
           ResultSet to handle. * * @return A List of beans, never 
           null. * * @throws SQLException if a database access error occurs * @see org.apache.commons.dbutils.RowProcessor#toBeanList(ResultSet, Class) */ @Override public List
           
             handle(ResultSet rs) throws SQLException { return this.convert.toBeanList(rs, type); }}
           
          
         
        
       
      
     
    

BeanListHandler跟BeanHandler实现类似。

ArrayHandler源码：

/** * ResultSetHandler implementation that converts a * ResultSet into an Object[]. This class is * thread safe. * 
 * 将ResultSet中的一行记录变成一个Object[]数组，此类是线程安全的 * 
 * @see org.apache.commons.dbutils.ResultSetHandler */public class ArrayHandler implements ResultSetHandler
    
      {    /**     * Singleton processor instance that handlers share to save memory.  Notice     * the default scoping to allow only classes in this package to use this     * instance.     */    static final RowProcessor ROW_PROCESSOR = new BasicRowProcessor();    /**     * The RowProcessor implementation to use when converting rows     * into arrays.     */    private final RowProcessor convert;    /**     * Creates a new instance of ArrayHandler using a     * 
     BasicRowProcessor for conversion.     */    public ArrayHandler() {        this(ROW_PROCESSOR);    }    /**     * Creates a new instance of ArrayHandler.     *     * @param convert The 
     RowProcessor implementation     * to use when converting rows into arrays.     */    public ArrayHandler(RowProcessor convert) {        super();        this.convert = convert;    }    /**     * Places the column values from the first row in an 
     Object[].     * 将第一行的记录编程Object[]     * @param rs 
     ResultSet to process.     * @return An Object[] or 
     null if there are no rows in the     * 
     ResultSet.     *     * @throws SQLException if a database access error occurs     * @see org.apache.commons.dbutils.ResultSetHandler#handle(java.sql.ResultSet)     */    @Override    public Object[] handle(ResultSet rs) throws SQLException {    	//有记录的话，就将第一行记录变成一个Object[]        return rs.next() ? this.convert.toArray(rs) : null;    }}
    

MapHandler源码：

/** * ResultSetHandler implementation that converts the first * ResultSet row into a Map. This class is thread * safe. * * @see org.apache.commons.dbutils.ResultSetHandler */public class MapHandler implements ResultSetHandler

ScalarHandler源码：

/** * ResultSetHandler implementation that converts one * ResultSet column into an Object. This class is thread safe. * 将结果集中第一行的第一列的值变成一个对象。即是单值的。 * @param 
    
      The type of the scalar * @see org.apache.commons.dbutils.ResultSetHandler */public class ScalarHandler
     
       implements ResultSetHandler
      
        {    /**     * The column number to retrieve.     */    private final int columnIndex;    /**     * The column name to retrieve.  Either columnName or columnIndex     * will be used but never both.     */    private final String columnName;    /**     * Creates a new instance of ScalarHandler.  The first column will     * be returned from 
       handle().     * 默认会返回结果集中第一行的第一列的值     */    public ScalarHandler() {        this(1, null);    }    /**     * Creates a new instance of ScalarHandler.     *     * @param columnIndex The index of the column to retrieve from the     * 
       ResultSet.     */    public ScalarHandler(int columnIndex) {        this(columnIndex, null);    }    /**     * Creates a new instance of ScalarHandler.     *     * @param columnName The name of the column to retrieve from the     * 
       ResultSet.     */    public ScalarHandler(String columnName) {        this(1, columnName);    }    /** Helper constructor     * @param columnIndex The index of the column to retrieve from the     * 
       ResultSet.     * @param columnName The name of the column to retrieve from the     * 
       ResultSet.     */    private ScalarHandler(int columnIndex, String columnName) {        this.columnIndex = columnIndex;        this.columnName = columnName;    }    /**     * Returns one 
       ResultSet column as an object via the     * 
       ResultSet.getObject() method that performs type     * conversions.     * @param rs 
       ResultSet to process.     * @return The column or 
       null if there are no rows in     * the 
       ResultSet.     *     * @throws SQLException if a database access error occurs     * @throws ClassCastException if the class datatype does not match the column type     *     * @see org.apache.commons.dbutils.ResultSetHandler#handle(java.sql.ResultSet)     */    // We assume that the user has picked the correct type to match the column    // so getObject will return the appropriate type and the cast will succeed.    @SuppressWarnings("unchecked")    @Override    public T handle(ResultSet rs) throws SQLException {        if (rs.next()) {        	//如果列名为null，那么就使用列码            if (this.columnName == null) {                return (T) rs.getObject(this.columnIndex);            } else {                return (T) rs.getObject(this.columnName);            }        } else {            return null;        }    }}
      
     
    

 ScalarHandler用于获取单值，常用的是计算表的记录数count(*)。下面的代码是测试代码。用的数据库还是之前的。

@Test	public void testScalarHandler() throws Exception{		String sql = "select * from t_user where id=?";		Object[] params = new Object[]{3};		ResultSetHandler
    
      rsh = new ScalarHandler
     
      (2);				System.out.println(queryRunner.query(sql, rsh, params));	}
     
    

这几个就是常用的Handler了。平时开发的话，一般是够用了。在查询方面，使用了策略模式。用户可以自定义ResultSetHandler，用户自己来实现对结果集的处理，也可以结合自定义的RowProcessor实现类来一起使用。