Apache DBUtils使用总结
 
DBUtils是个小巧的JDBC轻量级封装的工具包，其最核心的特性是结果集的封装，可以直接将查询出来的结果集封装成JavaBean，这就为我们做了最枯燥乏味、最容易出错的一大部分工作。
 
在使用DBUtils之前，应该注意一些问题：
 
1、DBUtils是JDBC的简单封装，可以和JDBC混合使用。
 
2、DBUtils对结果集自动封装为JavaBean是有着苛刻要求的：必须满足JavaBean的规范，其次Bean的getter与setter方法的名字与结果集的列名一一对应，而不要求JavaBean的私有成员与表结果集列名一一对应。
比如：
person表中有个字段叫：address，那么对应的JavaBean的Person类中必须有getAddress和setAddress两个方法，而Person类中可以将address属性命名为add，这是没问题的。
 
3、DBUtils可以将结果集封装为各种类型，主要有：Bean/List<Bean>，Map/List<Map>/Map<Map>，数组/List<数组>，列/List<列>，这些类型。
对于Map<Map>的类型使用KeyedHandler作为结果集处理器，内层的Map是“列名-值"对，外层的Map是“主键-内层Map的引用”，但此处的主键不一定就是数据库的主键，可以随意指定，比如：
ResultSetHandler h = new KeyedHandler("id");
Map found = (Map) queryRunner.query("select id, name, age from person", h);
Map jane = (Map) found.get(new Long(1)); // jane's id is 1
String janesName = (String) jane.get("name");
Integer janesAge = (Integer) jane.get("age");
 
4、DBUtils执行插入操作的时候，无法返回自增主键，这是一个很严重的问题，当然不能怪DBUtils，可以通过变通的方法来实现，比如在MySQL中，执行完了一个插入SQL后，接着执行SELECT LAST_INSERT_ID()语句，就可以获取到自增主键。
 
5、DBUtils的性能和JDBC性能是一样，测试过程中没发现性能损失，拥有了很高性能的同时，而不失JDBC的灵活性。
 
6、对于JavaBean的成员类型定义，有一条原则那就是：尽可能使用包装类型，而不要使用基本类型。很多人不理解为什么，包括我见到一些傻B（，我提出让改为包装类型，他还信誓旦旦的说他的代码多牛多叼），都干好多年了，这个简单的道理还不知道:
        //错误
        int a1 = (Integer) null;
        boolean x1 = (Boolean)null;
        //正确
        Integer a2 = (Integer) null;
        Boolean x2 = (Boolean)null;
 
实际上就是为了保证在查询结果为null的时候，也不会因为给基本类型赋null值而发生错误。
 
下面给出一个简单例子，作为以后写代码时候有点参考：
 
环境：
MySQL5.4
JDK1.5
 
建表SQL：
CREATE TABLE person (
     id bigint(20) NOT NULL AUTO_INCREMENT,
     name varchar(24) DEFAULT NULL,
     age int(11) DEFAULT NULL,
     address varchar(120) DEFAULT NULL,
     PRIMARY KEY (id)
) ENGINE=MyISAM AUTO_INCREMENT=1 DEFAULT CHARSET=gbk
 
Java代码：
public class Person {
        private Long id;
        private String sdf;
        private String address2;
        private Integer age;

        public Person() {
        }

        public Person(String sdf) {
                this.sdf = sdf;
        }

        public Person(String sdf, Integer age, String address) {
                this.sdf = sdf;
                this.age = age;
                this.address2 = address;
        }

        public Long getId() {
                return id;
        }

        public void setId(Long id) {
                this.id = id;
        }

        public String getSdf() {
                return sdf;
        }

        public void setSdf(String sdf) {
                this.sdf = sdf;
        }

        public Integer getAge() {
                return age;
        }

        public void setAge(Integer age) {
                this.age = age;
        }

        public String getAddress() {
                return address2;
        }

        public void setAddress(String address2) {
                this.address2 = address2;
        }
}
 
测试
package com.lavasoft.dbstu;

import com.lavasoft.common.DBToolkit;
import org.apache.commons.dbutils.QueryRunner;
import org.apache.commons.dbutils.handlers.BeanHandler;
import org.apache.commons.dbutils.handlers.BeanListHandler;
import org.apache.commons.dbutils.handlers.MapHandler;
import org.apache.commons.dbutils.handlers.ScalarHandler;

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

/**
* Created by IntelliJ IDEA.
*
* @author leizhimin 2010-1-25 21:00:29
*/
public class PersonDAOImpl implements PersonDAO {
        private static PersonDAOImpl instance = new PersonDAOImpl();

        public static PersonDAOImpl getInstance() {
                return instance;
        }

        public static void main(String[] args) {
                //错误
                int a1 = (Integer) null;
                boolean x1 = (Boolean) null;
                //正确
                Integer a2 = (Integer) null;
                Boolean x2 = (Boolean) null;

                getInstance().save(null);
//                getInstance().save(null);
//                getInstance().save(null);
//                getInstance().save(null);
//                getInstance().save(null);
                getInstance().update(null);
                getInstance().load(null);
                getInstance().load4Map(null);
        }

        @Override
        public Long save(String sql) {
                Long id = null;
                String ins_sql = "INSERT INTO person (NAME, age, address) VALUES ('aaa', 21, 'address001')";
                Connection conn = DBToolkit.getConnection();
                QueryRunner qr = new QueryRunner();
                try {
                        qr.update(conn, ins_sql);
                        //获取新增记录的自增主键
                        id = (Long) qr.query(conn, "SELECT LAST_INSERT_ID()", new ScalarHandler(1));
                } catch (SQLException e) {
                        e.printStackTrace();
                } finally {
                        DBToolkit.closeConnection(conn);
                }
                return id;
        }

        @Override
        public int delete(Long id) {
                int x = 0;
                Connection conn = DBToolkit.getConnection();
                QueryRunner qr = new QueryRunner();
                try {
                        x = qr.update(conn, "DELETE FROM person WHERE id = ?", id);
                } catch (SQLException e) {
                        e.printStackTrace();
                } finally {
                        DBToolkit.closeConnection(conn);
                }
                return x;
        }

        @Override
        public int update(Person person) {
                int x = 0;
                Connection conn = DBToolkit.getConnection();
                QueryRunner qr = new QueryRunner();
                try {
                        x = qr.update(conn, "UPDATE person SET NAME = ?, age = ?, address = ? WHERE id = ?", "xxx", 23, "ttt", 5);
                } catch (SQLException e) {
                        e.printStackTrace();
                } finally {
                        DBToolkit.closeConnection(conn);
                }
                return x;

        }

        @Override
        public Person load(Long id) {
                Connection conn = DBToolkit.getConnection();
                QueryRunner qr = new QueryRunner();
                try {
                        Person person = (Person) qr.query(conn, "SELECT * FROM person where id = ?", new BeanHandler(Person.class), 3L);
                        System.out.println(person.getId() + "\t" + person.getSdf() + "\t" + person.getAge() + "\t" + person.getAddress());
                } catch (SQLException e) {
                        e.printStackTrace();
                }
                return null;
        }

        @Override
        public List<Person> findPerson(String sql) {
                Connection conn = DBToolkit.getConnection();
                QueryRunner qr = new QueryRunner();
                try {
                        List<Person> pset = (List) qr.query(conn, "SELECT * FROM person", new BeanListHandler(Person.class));
                        for (Person person : pset) {
                                System.out.println(person.getId() + "\t" + person.getSdf() + "\t" + person.getAge() + "\t" + person.getAddress());
                        }
                } catch (SQLException e) {
                        e.printStackTrace();
                }
                return null;
        }

        public Person load4Map(Long id) {
                Connection conn = DBToolkit.getConnection();
                QueryRunner qr = new QueryRunner();
                try {
                        //先将两个字段置为null
                        qr.update(conn, "update person set age = null,address =null where id =1");
                        Map<String, Object> map = qr.query(conn, "SELECT * FROM person where id = ?", new MapHandler(), 1L);
                        Person person = new Person();
                        person.setId((Long) map.get("id"));
                        person.setSdf((String) map.get("name"));
                        person.setAge((Integer) map.get("age"));
                        person.setAddress((String) map.get("address"));
                        System.out.println(person.getId() + "\t" + person.getSdf() + "\t" + person.getAge() + "\t" + person.getAddress());
                } catch (SQLException e) {
                        e.printStackTrace();
                }
                return null;
        }
}
 
注意：
从上面的过程看，每个SQL的执行都需要用到QueryRunner，这其实是一个普通的类，非线程安全。在创建这个QueryRunner对象的时候，每次都要new。
 
QueryRunner对象的创建方式很多，可以通过DateSource，也可以通过Connection来创建。从QueryRunner对象上，可以直接获取到DataSource或者Connection对象。
 
如果你想看更多例子：