Mybatis高级应用

Mybatis是一个半自动的框架。相对于hibernate全自动模式，mybatis为开发人员提供了更加灵活的对sql语句操作的控制能力，有利于dba对相关的sql操作进行优化，同时也方便开发者构建复杂的sql操作。以下是Mybatis的相关高级应用，以供参考。

Mybatis的官方文档：http://mybatis.github.io/mybatis-3/zh/index.html

• 通过配置类进行构建SqlSessionFactory

Mybatis允许通过xml或配置类构建SqlSessionFactory

        DataSource dataSource = BlogDataSourceFactory.getBlogDataSource();
        TransactionFactory transactionFactory = new JdbcTransactionFactory();
        Environment environment = new Environment("development", transactionFactory, dataSource);
        Configuration configuration = new Configuration(environment);
        configuration.addMapper(BlogMapper.class);
        SqlSessionFactory sqlSessionFactory = new SqlSessionFactoryBuilder().build(configuration);

通过SqlSessionFactory获取SqlSession

        SqlSession session = sqlSessionFactory.openSession();
        try {
          BlogMapper mapper = session.getMapper(BlogMapper.class);
          Blog blog = mapper.selectBlog(101);
        } finally {
          session.close();
        }

• forEach应用

forEach按照官方支持对单一数组和集合进行遍历，以下是对单一数组进行遍历的示例，集合参数其实是类似的

	<delete id="deleteFile" parameterType="HashMap">
		delete
		From tbl_File
		Where themeKey = #{themeKey}
		<trim suffixOverrides="and">
		  <if test="arrFileUrl != null and arrFileUrl != '' ">
		      And fileUrl in
			  <foreach item="item" index="index" collection="arrFileUrl"
			      open="(" separator="," close=")">
			        #{item}
			  </foreach>
		  </if>
		</trim>
	</delete>

调用方法：

       HashMap<String, Object> argMap = new HashMap<String, Object>();
       argMap.put("themeKey", themeKey);
       /*arrFileUrl 是一个字符串数组*/
       argMap.put("arrFileUrl", arrFileUrl);

       deleteFile(HashMap<String, Object> argMap);

foreach支持遍历list,map,array 三种类型，因此利用foreach以上特性实现批量插入数据功能。

    <insert id="addTrainRecordBatch" useGeneratedKeys="true" parameterType="java.util.List">  
        <selectKey resultType="long" keyProperty="id" order="AFTER">  
            SELECT  
            LAST_INSERT_ID()  
        </selectKey>  

        insert into t_train_record (add_time,emp_id,activity_id,flag)   
        values  

        <foreach collection="list" item="item" index="index" separator="," >  
            (#{item.addTime},#{item.empId},#{item.activityId},#{item.flag})  
        </foreach>  
    </insert>  

• collection应用

	<resultMap id="Guide" type="Guide">
		<id column="spaguideKey" 		property="guideKey"/>
		<result column="GUIDENAME" 		property="guideName"/>
		<result column="GRADE" 			property="grade"/>
		<result column="gradeName" 		property="gradeName"/>
		<result column="SUBJECT" 		property="subject"/>
		<result column="subjectName" 	        property="subjectName"/>
		<result column="CREATECODE" 	        property="createCode"/>
		<result column="realName" 		property="realName"/>
		<result column="CREATETIME" 	        property="createTime"/>
		<result column="ISVALID" 		property="isValid"/>
		<result column="ISREVIEW" 		property="isReview"/>
		<result column="CONTENT" 		property="content"/>

                <collection  property="guideVideoList"  ofType="GuideVideo">
			<id column="spavideokey" 	property="videoKey"/>
			<result column="videoName" 	property="videoName"/>
			<result column="videoFileUrl" 	property="fileUrl"/>
                </collection>
                <collection  property="guideArchiveList" ofType="GuideArchive">
			<id column="SPAARCHIVESKEY" 	property="archiveKey"/>
			<result column="archiveName" 	property="archiveName"/>
			<result column="archiveFileUrl" property="fileUrl"/>
                </collection>
	</resultMap>

    <select id="getGuide" parameterType="Guide" resultMap="Guide">
        Select
            sg.SPAGUIDEKEY,
            GUIDENAME,
            CREATETIME,
            CONTENT,
            grade,
            gradeName,
            subject,
            subjectName,
            CREATECODE,
            realName,        
            spavideokey,
            videoName,
            videoFileName,
            videoFileUrl,          
            SPAARCHIVESKEY,
            ARCHIVENAME,
            archiveFileName,
            archiveFileUrl
        from vw_spa_guide sg
        left join vw_spa_guide_video sgv on sgv.spaguidekey = sg.spaguidekey
        left join vw_spa_guide_archive sga on sga.spaguidekey = sg.spaguidekey
    </select>

• Mybatis 嵌套查询

    <resultMap type="User" id="userResultMap">
        <id property="id" column="user_id"  />
        <result property="userName" column="user_userName"  />
        <result property="userAge" column="user_userAge"  />
        <result property="userAddress" column="user_userAddress"  />
    </resultMap>

    <resultMap id="articleResultMap" type="Article">
        <id property="id" column="article_id" />
        <result property="title" column="article_title" />
        <result property="content" column="article_content" />
        <association property="user" javaType="User" resultMap="userResultMap"/>
    </resultMap>

    <select id="getUserArticles" parameterType="int" resultMap="articleResultMap">
       select user.id user_id,user.userName user_userName,user.userAddress user_userAddress,
       article.id article_id,article.title article_title,article.content article_content
       from user,article
       where user.id=article.userid and user.id=#{id}
    </select>

借用别人的例子来说明。

• mybatis 自定义主键

    <insert id="insertTest" parameterType="Test">
        <selectKey  keyProperty="testKey" order="BEFORE" resultType="String">
              SELECT  SEQ_TST_TEST.nextval
              FROM DUAL
         </selectKey>

        insert into TST_TEST(
            TESTKEY,
            TESTNAME
        )values(
            #{testKey},
            #{testName}
        )
    </insert>

mybatis 相对于oracle的自增长先进行查询读取下一个主键,oracle不支持useGeneratedKeys，然后再进行插入操作。相对于其他数据库可以采用如下方式

        <insert id="insertTest" parameterType="Test" useGeneratedKeys="true" keyProperty="testKey">
            insert into Test(testKey, testName)
            values(#{testKey},#{testName})
        </insert>  

• mybatis复用语句块

        <!--定义可重用的SQL代码段-->
        <sql id="multiplexSql">testKey, testName</sql>  

        <select id="getTest" parameterType="int" resultType="hashmap">
            select
                <include refid="multiplexSql"/>
            from Blog
            where id = #{testKey}
        </select>  

• Mybatis执行函数，返回结果集

	<select id="getSelfStatisticData" parameterType="HashMap" statementType="CALLABLE" >
 	    {#{result,mode=OUT,jdbcType=CURSOR, resultMap=SelfStatisticData} = call PKG_RRT_SelfStatics.Fn_GetSelfStatData(#{userCode,jdbcType=VARCHAR,mode=IN})}
	</select> 	

Java调用的代码

        public interface SelfStatisticDataDao {
	     public List<SelfStatisticData> getSelfStatisticData(Map<String, Object> statMap);
        }

statMap 中的键值对对应着Fn_GetSelfStatData()函数的参数，键名与参数名保持完全一致，区分大小写。
                 SelfStatisticData定义的实体保持与结果集的字段一致。

• Mybatis执行没有返回值的存储过程

    <select id="insertGuideIntegral" parameterType="HashMap" statementType="CALLABLE" >
         {
             call PKG_Center_Integral_guide.Pro_SyncGuideIntegral(
                 #{userCode,jdbcType=VARCHAR,mode=IN},
                 #{integralKey,jdbcType=VARCHAR,mode=IN},
                 #{gradeKey,jdbcType=VARCHAR,mode=IN},
                 #{subjectKey,jdbcType=VARCHAR,mode=IN}
             )
         }
    </select>

• Mybatis执行带有返回两个游标结果集和输出参数

     <resultMap type="IntegralResult" id="integralResult">
         <result column="integralKey" property="integralKey"/>
         <result column="integralName" property="integralName"/>
     </resultMap>  

     <resultMap type="GuideIntegralResult" id="guideIntegralResult">
         <result column="guideIntegralKey" property="guideIntegralKey"/>
         <result column="guideIntegralName" property="guideIntegralName"/>
     </resultMap>  

     <select id="get" parameterType="java.util.Map" statementType="CALLABLE"  resultMap="integralResult, guideIntegralResult">
         {
             call PKG_Center_Integral_guide.Pro_GuideIntegral(
                 #{userCode,jdbcType=VARCHAR,mode=IN},
                 #{subjectKey,jdbcType=VARCHAR,mode=IN},
                 #{userName, mode=OUT, jdbcType=String}
             )
         }
     </select> 

• Mybatis 支持结构体类型

      #{middleInitial, mode=OUT, jdbcType=STRUCT, jdbcTypeName=MY_TYPE, resultMap=departmentResultMap}

MY_TYPE为数据库中自定义的结构体

• 定义输入输出小数点

       #{height,javaType=double,jdbcType=NUMERIC,numericScale=2}  

• 使用association进行复杂映射

        <resultMap type="Blog" id="Blog_result">
            <id column="id" property="id" />
            <result column="title" property="title"/>  

            <!-- 映射关联的对象 -->
            <association property="author" javaType="Author">
                <id column="author_id" property="id"/>
                <result column="username" property="username"/>
                <result column="password" property="password"/>
                <result column="email" property="email"/>
                <result column="bio" property="bio"/>
            </association>
         </resultMap>  

        <select id="selectBlog_by_id" parameterType="int" resultMap="Blog_result">  
            select  
                b.id,  
                b.title,  
                b.author_id,  
                a.id,  
                a.username,  
                a.password,  
                a.email,  
                a.bio  
            from  Blog b
            left join Author a  on  b.author_id = a.id  
            where  b.id = #{id}  
        </select>

• 定义输入输出和指定游标结果集

    #{
        department,
        mode=OUT,
        jdbcType=CURSOR,
        javaType=ResultSet,
        resultMap=departmentResultMap
    }

mode属性允许你指定IN，OUT或INOUT参数。如果mode为OUT（或INOUT），而且jdbcType为CURSOR（也就是Oracle的REFCURSOR），你必须指定一个resultMap来映射结果集到参数类型。

• 参考资料

http://blog.csdn.net/rootsuper/article/details/8542236