轻量级封装DbUtils&Mybatis之二Dbutils

DbUtils入门#

Apache出品的极为轻量级的Jdbc访问框架，核心类只有两个：QueryRunner和ResultSetHandler。
各类ResultSetHandler：
ArrayHandler：把结果集中的第一行数据转成对象数组。
ArrayListHandler：把结果集中的每一行数据都转成一个数组，再存放到List中。
BeanListHandler：将结果集中的每一行数据都封装到一个对应的JavaBean实例中，存放到List里。

请参考孤傲苍狼的博客javaweb学习总结(四十一)——Apache的DBUtils框架学习，墙裂推荐。

Detail#

测试样例代码##

package org.wit.ff.jdbc;


import org.junit.Test;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.test.context.ContextConfiguration;
import org.springframework.test.context.junit4.AbstractJUnit4SpringContextTests;
import org.wit.ff.jdbc.access.IDataAccessor;
import org.wit.ff.jdbc.converter.ParamsConverter;
import org.wit.ff.jdbc.sql.db.MySQLBuilder;

import java.util.ArrayList;
import java.util.Date;
import java.util.List;

/**
 * Created by F.Fang on 2015/3/31.
 * Version :2015/3/31
 */
@ContextConfiguration(locations = {"classpath:applicationContext.xml"})
public class CRUDTest extends AbstractJUnit4SpringContextTests{

    @Autowired
    private IDataAccessor dataAccessor;

    @Test
    public void query(){
        MySQLBuilder builder = new MySQLBuilder();
        // sql: select * from Audience limit 2
        builder.SELECT("*").FROM("Audience").PAGE(0, 2);
        String sql = builder.toString();
        final Audience audience = new Audience();
        audience.setName("ff");

        // 也可以直接填写SQL.
        List<Audience> list = dataAccessor.query(sql, null, Audience.class);
        System.out.println(list);
    }

    @Test
    public void insert(){
        // 批量新增.
        String sql = "insert into audience(id,name,pay) values(?,?,?)";
        Object[][] params = new Object[1][];
        params[0] = new Object[]{100,"ff",100.1};
        dataAccessor.insert(sql, params);
    }

    @Test
    public void insertDate(){
        // 试试日期.
        String sql = "insert into test_date(id,current) values(?,?)";
        Object[][] params = new Object[1][];
        params[0] = new Object[]{100, new Date()};
        dataAccessor.insert(sql, params);
    }

    @Test
    public void insertWithParamsConverter(){
        // 试试PramsConverter好用不.
        ParamsConverter<Audience> converter = new ParamsConverter<Audience>() {
            @Override
            public Object[] convert(Audience audience) {
                return new Object[]{audience.getId(), audience.getName(), audience.getPay()};
            }
        };

        String sql = "insert into audience(id,name,pay) values(?,?,?)";
        // 构造数据.
        List<Audience> list = new ArrayList<>();
        Audience audience1 = new Audience();
        audience1.setId(250);
        audience1.setName("ff");
        audience1.setPay(1000.00);

        Audience audience2 = new Audience();
        audience2.setId(251);
        audience2.setName("ff1");
        audience2.setPay(1000.00);

        list.add(audience1);
        list.add(audience2);

        dataAccessor.insert(sql, list, Audience.class, converter);
        
    }

}

Audience

package org.wit.ff.jdbc;

import org.apache.commons.lang3.builder.ReflectionToStringBuilder;
import org.apache.commons.lang3.builder.ToStringStyle;

/**
 * Created by F.Fang on 2015/2/16.
 * Version :2015/2/16
 */
public class Audience {
    private Integer id;

    private String name;

    private Double pay;

    public Integer getId() {
        return id;
    }

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

    public String getName() {
        return name;
    }

    public void setName(String name) {
        this.name = name;
    }

    public Double getPay() {
        return pay;
    }

    public void setPay(Double pay) {
        this.pay = pay;
    }

    @Override
    public String toString() {
        return ReflectionToStringBuilder.toString(this, ToStringStyle.SIMPLE_STYLE);
    }
}

配置

<!-- 数据库连接池 -->
    <bean id="dataSource"
          class="org.apache.commons.dbcp.BasicDataSource"
          destroy-method="close">
        <property name="driverClassName" value="${db.driverClass}"/>
        <property name="url" value="${db.jdbcUrl}"/>
        <property name="username" value="${db.user}"/>
        <property name="password" value="${db.password}"/>
    </bean>

    <bean id="dataAccessor" class="org.wit.ff.jdbc.access.dbutils.DefaultDataAccessor">
        <property name="dataSource" ref="dataSource"/>
    </bean>

备注：实际数据源配置还是需要优化的

看上去还凑合，调用比较简单，那么接下来看看具体的实现吧！

AbstractDataAccessor##

• QueryRunner无参构造对象
• 抽象获取连接和关闭连接的方法

package org.wit.ff.jdbc.access.dbutils;

import org.apache.commons.dbutils.QueryRunner;
import org.apache.commons.dbutils.handlers.*;
import org.apache.commons.lang3.StringUtils;
import org.wit.ff.jdbc.converter.ParamsConverter;
import org.wit.ff.jdbc.access.IDataAccessor;
import org.wit.ff.jdbc.exception.DbUtilsDataAccessException;
import org.wit.ff.jdbc.id.IdGenerator;

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

/**
 * Created by F.Fang on 2015/3/31.
 * Version :2015/3/31
 */
public abstract class AbstractDataAccessor implements IDataAccessor {

    /**
     * 获取连接的方法时抽象的,目的是为了和事务绑定,开放Connection来源.
     * @return
     * @throws SQLException
     */
    protected abstract Connection getConnection() throws SQLException;

    /**
     * 获取连接的方法时抽象的,目的是为了和事务绑定,因为事务执行前后包含的sql逻辑可能不只一条, 不可在一条sql执行逻辑完成后关闭.
     * @param conn
     */
    protected abstract void closeConn(Connection conn);

    @Override
    public <T> List<T> query(String sql, Class<T> resultType){
        return query(sql, null, resultType);
    }

    /**
     * 查询时BeanListHandler将结果集转换成对象列表.
     * @param sql 查询语句
     * @param params 查询参数
     * @param resultType 返回类型
     * @param <T>
     * @return
     */
    @Override
    public <T> List<T> query(String sql, Object[] params, Class<T> resultType) {
        if (resultType == null || StringUtils.isBlank(sql)) {
            throw new DbUtilsDataAccessException("resultType can't be null and sql can't be blank!");
        }
        QueryRunner runner = new QueryRunner();
        Connection conn = null;
        List<T> result = null;
        try {
            conn = getConnection();
            if (params != null) {
                result = (List<T>) runner.query(conn, sql, new BeanListHandler(resultType), params);
            } else {
                result = (List<T>) runner.query(conn, sql, new BeanListHandler(resultType));
            }
        } catch (Exception e) {
            throw new DbUtilsDataAccessException("query error, sql is:" + sql, e);
        } finally {
            closeConn(conn);
        }
        return result;
    }

    /**
     *
     * @param sql insert语句.
     * @param params 对象参数列表.
     * @param paramsType 参数类型.
     * @param converter 参数转换器.
     * @param <T>
     */
    @Override
    public <T> void insert(String sql, List<T> params, Class<T> paramsType, ParamsConverter<T> converter) {
        if (StringUtils.isBlank(sql)) {
            throw new DbUtilsDataAccessException("sql can't be blank!");
        }
        if (params == null || params.isEmpty()) {
            throw new DbUtilsDataAccessException("params can't be null or empty!");
        }
        if (paramsType == null || converter == null) {
            throw new DbUtilsDataAccessException("paramsType or converter can't be null!");
        }

        QueryRunner runner = new QueryRunner();
        Connection conn = null;
        ArrayListHandler handler = new ArrayListHandler();
        List<Object[]> keys = null;
        try {
            conn = getConnection();
            int len = params.size();
            Object[][] arr = new Object[len][];
            for (int i = 0; i < len; ++i) {
                // 将单个对象处理成单行记录,用数组表示.
                arr[i] = converter.convert(params.get(i));
            }
            // 获取主键.
            keys = (List<Object[]>) runner.insertBatch(conn, sql, handler, arr);
            if (keys != null && keys.size() == len) {
                // implements IdGenerator.
                // 如果当前对象实现了IdGenerator接口, 则执行主键赋值的逻辑, 赋值逻辑由用户自定义在具体的对象类型当中
                if (params.get(0) instanceof IdGenerator) {
                    for (int i = 0; i < len; ++i) {
                        // 对每一个对象执行主键赋值(解析)
                        ((IdGenerator) params.get(i)).parseGenKey(keys.get(i));
                    }
                }
            }
        } catch (Exception e) {
            throw new DbUtilsDataAccessException("batch insert error, sql is:" + sql, e);
        }finally {
            closeConn(conn);
        }
    }

    /**
     * 批量插入.
     * @param sql
     * @param params
     */
    @Override
    public void insert(String sql, Object[][] params) {
        if(params==null || StringUtils.isBlank(sql)){
            throw new DbUtilsDataAccessException("params can't be null and sql can't be empty!");
        }
        QueryRunner runner = new QueryRunner();
        Connection conn = null;
        ArrayListHandler handler = new ArrayListHandler();
        try {
            conn = getConnection();
            runner.insertBatch(conn, sql, handler, params);
        } catch (Exception e) {
            throw new DbUtilsDataAccessException("batch insert error, sql is:" + sql, e);
        }finally {
            closeConn(conn);
        }
    }

    /**
     * 单条插入.
     * @param sql
     * @param params
     */
    @Override
    public void insert(String sql, Object[] params) {
        if(params==null || StringUtils.isBlank(sql)){
            throw new DbUtilsDataAccessException("params can't be null and sql can't be empty!");
        }
        QueryRunner runner = new QueryRunner();
        Connection conn = null;
        ArrayHandler handler = new ArrayHandler();
        try {
            conn = getConnection();
            runner.insert(conn, sql, handler, params);
        } catch (Exception e) {
            throw new DbUtilsDataAccessException("insert error, sql is:" + sql, e);
        }finally {
            closeConn(conn);
        }
    }

    /**
     * 批量更新.
     * @param sql
     * @param params
     * @return
     */
    @Override
    public int[] update(String sql, Object[][] params) {
        if(params==null || StringUtils.isBlank(sql)){
            throw new DbUtilsDataAccessException("params can't be null and sql can't be empty!");
        }
        QueryRunner runner = new QueryRunner();
        Connection conn = null;
        int[] result = null;
        try {
            conn = getConnection();
            result = runner.batch(conn, sql, params);
        } catch (Exception e) {
            throw new DbUtilsDataAccessException("batch update error, sql is:" + sql, e);
        }finally {
            closeConn(conn);
        }
        return result;
    }

    /**
     * 单条更新.
     * @param sql
     * @param params
     * @return
     */
    @Override
    public int update(String sql, Object[] params) {
        QueryRunner runner = new QueryRunner();
        Connection conn = null;
        int result = 0;
        try {
            conn = getConnection();
            if (params != null) {
                result = runner.update(conn, sql, params);
            } else {
                result = runner.update(conn, sql);
            }
        } catch (Exception e) {
            throw new DbUtilsDataAccessException("update error, sql is:" + sql, e);
        }finally {
            closeConn(conn);
        }
        return result;
    }

    /**
     * 删除,没有必要批量.
     * 即使是批量,也可以调用批量更新的方法.
     * 事实上此方法并无太大必要,只是为了避免歧义而已.
     * @param sql
     * @param params
     * @return
     */
    @Override
    public int delete(String sql, Object[] params) {
        return update(sql, params);
    }

}

DefaultDataAccessor##

package org.wit.ff.jdbc.access.dbutils;

import org.apache.commons.dbutils.DbUtils;

import javax.sql.DataSource;
import java.sql.Connection;
import java.sql.SQLException;

/**
 * Created by F.Fang on 2015/3/31.
 * 默认以配置数据源的方式获取连接.
 * Version :2015/3/31
 */
public class DefaultDataAccessor extends AbstractDataAccessor {

    private DataSource dataSource;

    protected Connection getConnection() throws SQLException{
        return dataSource.getConnection();
    }

    @Override
    protected void closeConn(Connection conn) {
        try {
            DbUtils.close(conn);
        } catch (SQLException e) {
            // do nothing!
        }
    }

    public void setDataSource(DataSource dataSource) {
        this.dataSource = dataSource;
    }
}

DefaultTransactionDataAccessor##

package org.wit.ff.jdbc.access.dbutils;

import org.springframework.jdbc.datasource.DataSourceUtils;

import javax.sql.DataSource;
import java.sql.Connection;
import java.sql.SQLException;

/**
 * Created by F.Fang on 2015/4/23.
 * Version :2015/4/23
 */
public class DefaultTransactionDataAccessor extends AbstractDataAccessor {

    private DataSource dataSource;

    protected Connection getConnection() throws SQLException {

        return DataSourceUtils.getConnection(dataSource);
        // 虽然可以将连接绑定到事务,但是当外部循dataAccessor的方法时,循环调用时会产生多个连接.
        //return DataSourceUtils.getConnection(dataSource);
        // 以下写法无法解决事务问题,无法将连接绑定到Spring 事务.
        // return dataSource.getConnection();
    }

    @Override
    protected void closeConn(Connection conn) {
        // do nothing!
    }

    public void setDataSource(DataSource dataSource) {
        this.dataSource = dataSource;
    }
}

采用Spring事务集成方案，将Connection交给Spring管理，实际上Spring将Connection绑定到当前执行线程当中(以ConnectionHolder包装Connection，ThreadLocal包装ConnectionHolder绑定线程)
详情参考DataSourceTransactionMananger的源代码，最好在doBeign和doCleanupAfterCompletion执行断点调试

备注：AbstractDataAccessor开放getConnection()和closeConnection方法的原因也是调用过程Connection的开启和关闭不能由自身维持，否则会导致事务处理时Spring持有的Connection和AbstractDataAccessor执行方法的Connection不一致或在执行时拿到了关闭了的Connection对象。

QA#