文章目录
 mybatis工具类insert,delete,updateresultMap结果映射标签动态sqlmybatis的三种执行器cache与cache-ref缓存lazyLoading延迟加载proxy getMapper
 
 
mybatis工具类
 
public class MybatisUtils {
    private  static  SqlSessionFactory sqlSessionFactory
;
    static {
         try {
             
        String resource 
= "mybatis-config.xml";
        InputStream inputStream 
= Resources
.getResourceAsStream(resource
);
        sqlSessionFactory 
= new SqlSessionFactoryBuilder().build(inputStream
);
        } catch (IOException e
) {
            e
.printStackTrace();
        }
    }
    
    
    public  static SqlSession 
getSqlSession(){
        return  sqlSessionFactory
.openSession();
    }
}
 
insert,delete,update
 
对应的curd标签,没有resultType,接口返回值类型是int(Integer)insert标签可能需要得到新增记录id,useGeneratedKeys=“true” keyProperty=“uid”需要手动commit 
resultMap结果映射标签
 
对查询结果做映射处理 
<resultMap id
="aa" type
="com.javasm.entity.Goods">
        <id column
="gid" property
="gid"></id
> 
        <result column
="goodsColor" property
="gcolor"></result
>
</resultMap
>
 
对象关系映射 
  association与collection选择性的使用,自己写独立的查询方法也可以实现。association与collection两个使用方法: 
    拆分多次单表查询,通过select属性指定二次查询的位置。一次多表连接查询,通过做字段映射把不同的字段映射的不同的对象中。  1-1、1-N 
  情景:查询用户时,把该用户下的订单信息查询出来。  
public class Sysuser {
    private Integer uid
;
    private String uname
;
    private String uphone
;
    private String upwd
;
    private String createTime
;
    private List
<GoodsOrder> ordersList
; 
......
}
public Sysuser 
selectUserOrdersByKey(int key
);
public List
<GoodsOrder> selectOrderByUserId(int uid
);
<select id
="selectUserOrdersByKey" resultMap
="userRM" parameterType
="int">
  select 
* from sysuser where uid
=#
{uid
}
</select
>
<resultMap id
="userRM" type
="sysuser">
    <id column
="uid" property
="uid"></id
>
    <result column
="uname" property
="uname"></result
>
    <result column
="uphone" property
="uphone"></result
>
    <result column
="upwd" property
="upwd"></result
>
    <result column
="create_time" property
="createTime"></result
>
    <!--做
1-N映射
,N
-N
-->
    <collection property
="ordersList" ofType
="GoodsOrder" column
="uid" select
="com.javasm.mapper.GoodsOrderMapper.selectOrderByUserId"></collection
>
</resultMap
>
<select id
="selectOrderByUserId"  resultType
="goodsOrder">
    select 
* from goodsOrder where uid
=#
{uid
}
</select
>
public Sysuser 
selectUserOrdersByKey2(int key
);
<select id
="selectUserOrdersByKey2" resultMap
="userRM2" parameterType
="int">
  select u
.*
,o
.oid
,o
.otime
,o
.oprice from sysuser u
,goodsOrder o where o
.uid
=u
.uid and u
.uid
=#
{uid
}
</select
>
<resultMap id
="userRM2" type
="sysuser">
    <id column
="uid" property
="uid"></id
>
    <result column
="uname" property
="uname"></result
>
    <result column
="uphone" property
="uphone"></result
>
    <result column
="upwd" property
="upwd"></result
>
    <result column
="create_time" property
="createTime"></result
>
    <!--做
1-N映射
,N
-N
-->
    <collection property
="ordersList" ofType
="GoodsOrder">
        <id column
="oid" property
="oid"></id
>
        <result column
="otime" property
="otime"></result
>
        <result column
="oprice" property
="oprice"></result
>
    </collection
>
</resultMap
>
 
N-1、N-N 
  情景:查询订单信息以及订单的所属人信息  
public class GoodsOrder {
    private String oid
;
    private String otime
;
    private Double oprice
;
    private Sysuser suser
;
......
}
public GoodsOrder 
selectOrderAndUserByKey(String oid
);
public Sysuser 
selectUserByKey(int key
);
<!--把多表查询拆分两次的单表查询,推荐使用单表查询
-->
<select id
="selectOrderAndUserByKey" parameterType
="String" resultMap
="aa">
    select 
* from goodsOrder where oid
=#
{oid
}
</select
>
<resultMap id
="aa" type
="goodsOrder">
    <id column
="oid" property
="oid"></id
>
    <result column
="oprice" property
="oprice"></result
>
    <result column
="otime" property
="otime"></result
>
    <!--<result column
="uid" property
="suser.uid"></result
>-->
    <!--association
:做对象
1-1,N
-1关系映射 
-->
    <!--把uid列的值作为参数,传入select所指向的查询语句,返回结果是javaType,赋值给goodsOrder的suser属性
-->
    <association property
="suser" javaType
="Sysuser" column
="uid" select
="com.javasm.mapper.SysuserMapper.selectUserByKey"></association
>
</resultMap
>
<sql id
="userFields">
    uid
,uname
,uphone
,upwd
,create_time
</sql
>
<select id
="selectUserByKey" resultType
="sysuser" parameterType
="int" >
  select
   
<include refid
="userFields"></include
>
   from sysuser where uid
=#
{uid
}
</select
>
public GoodsOrder 
selectOrderAndUserByKey2(String oid
);
<!--做表连接查询,把需要的数据全部查询出来
-->
<select id
="selectOrderAndUserByKey2" parameterType
="String" resultMap
="aa2">
    select o
.*
,u
.uname
,u
.uphone
,u
.upwd
,u
.create_time from goodsOrder o
,Sysuser u where o
.uid
=u
.uid and o
.oid
=#
{oid
}
</select
>
<resultMap id
="aa2" type
="goodsOrder">
    <id column
="oid" property
="oid"></id
>
    <result column
="oprice" property
="oprice"></result
>
    <result column
="otime" property
="otime"></result
>
    <association property
="suser" javaType
="Sysuser">
        <id column
="uid" property
="uid"></id
>
        <result column
="uname" property
="uname"></result
>
        <result column
="uphone" property
="uphone"></result
>
        <result column
="upwd" property
="upwd"></result
>
        <result column
="create_time" property
="createTime"></result
>
    </association
>
</resultMap
>
 
 
 扩展资料:简述数据库设计中一对多和多对多的应用场景
 
 
动态sql
 
在映射文件中进行sql语句拼接。where: 中间有内容,自动生成where关键字,自动删掉紧跟其后的and或or。if:条件判断,有test属性写条件表达式,and或or拼接多条件 
<select id
="selectGoods" parameterType
="goods" resultMap
="aa">
    select 
* from goods
    
<where>
        <if test
="gid!=null">
            and gid
=#
{gid
}
        </if>
        <if test
="gname!=null and gname!=''">
            and gname like 
"%"#
{gname
}"%"
        </if>
        <if test
="gprice!=null and gprice > 0">
            and gprice
>=#
{gprice
}
        </if>
    </where
>
</select
>
 
set:生成set关键字,用在update标签中(实现字段的部分更新),自动删除最后的逗号. 
<update id
="updateGoods" parameterType
="goods">
  update goods
   
<set>
       <if test
="gname!=null">
          gname
=#
{gname
},
       </if>
       <if test
="gprice!=null">
           gprice
=#
{gprice
},
       </if>
       <if test
="gcolor!=null">
           goodsColor
=#
{gcolor
}
       </if>
   </set
>
   where gid
=#
{gid
}
</update
>
 
foreach:做批量删除与批量添加.批量添加注意按量提交。20条提交一次. 
<!--mybatis把所有参数都是封成map,数组的key:array
,集合key:list
-->
<delete id
="delGoodsByKeyArray">
    delete from goods where gid in
    
<foreach collection
="array" open
="(" close
=")" item
="goodsid" separator
=",">
      #
{goodsid
}
    </foreach
>
</delete
>
<delete id
="delGoodsByKeyList">
    delete from goods where gid in
    
<foreach collection
="list" open
="(" close
=")" item
="goodsid" separator
=",">
        #
{goodsid
}
    </foreach
>
</delete
>
<insert id
="insertGoodsArray">
    insert into 
goods(gid
,gname
,gprice
) value
    
<foreach collection
="array" item
="g" separator
=",">
        (#
{g
.gid
},#
{g
.gname
},#
{g
.gprice
})
    </foreach
>
</insert
>
 
mybatis的三种执行器
 
 SimpleExecutor:每执行一次update或select,就开启一个Statement对象,用完立刻关闭Statement对象。
  ReuseExecutor:执行update或select,以sql作为key查找Statement对象,存在就使用,不存在就创建,用完后,不关闭Statement对象,而是放置于Map内,供下一次使用。简言之,就是重复使用Statement对象。
  BatchExecutor:执行update(没有select,JDBC批处理不支持select),将所有sql都添加到批处理中(addBatch()),等待统一执行(executeBatch()),它缓存了多个Statement对象,每个Statement对象都是addBatch()完毕后,等待逐一执行executeBatch()批处理。与JDBC批处理相同。
  
 
 Mybatis 的三种执行器
 
 
cache与cache-ref缓存
 
 
所有的缓存都是基于内存,提高查询效率。标准套路:发起查询–>先去内存查,如果查到则返回–>如果查不到,则查询数据库,并把查询结果放缓存–>返回。使用缓存会导致的问题:数据同步、数据延迟、数据安全…session级别:默认生效。没有用 
  只能够在同一个会话中生效。 sessionFactory级别:全局唯一对象。 
  不同会话之间共享缓存。  
<!-- 启用二级缓存 
-->
<setting name
="cacheEnabled" value
="true"/>
 
lazyLoading延迟加载
 
让用户的体验度更加优秀。 
  分页、前端图片的延迟加载,树形组件,移动端的信息延迟加载。mybatis中的延迟加载只针对于使用了二次查询。对于第二次查询可延迟查询。仅支持association关联对象和collection关联集合对象的延迟加载,association指的就是一对一,collection指的就是一对多查询。它的原理是,使用CGLIB创建目标对象的代理对象,当调用目标方法时,进入拦截器方法,比如调用a.getB().getName(),拦截器invoke()方法发现a.getB()是null值,那么就会单独发送事先保存好的查询关联B对象的sql,把B查询上来,然后调用a.setB(b),于是a的对象b属性就有值了,接着完成a.getB().getName()方法的调用。这就是延迟加载的基本原理。 当然了,不光是Mybatis,几乎所有的包括Hibernate,支持延迟加载的原理都是一样的。  
<!-- 启用延迟加载 
-->
<setting name
="lazyLoadingEnabled" value
="true"/>
<!-- 积极加载改为消极加载 
-->
<setting name
="aggressiveLazyLoading" value
="false"/>
<!-- 调用toString,equals不触发延迟对象的加载 
-->
<setting name
="lazyLoadTriggerMethods" value
=""/>
 
proxy getMapper
 
 代理模式:适合于用在已经存在一个对象,而这个对象中的某些方法需要更新,没有办法改动源码(jar中),需要用代理模式。
  SqlSeccion.getMapper的动态代理实现
  
public class TestGetMapper {
    public static void main(String
[] args
) {
        SqlSessionFactory factory 
= SessionFactoryUtil
.get();
        SqlSession session 
= factory
.openSession();
        SysuserMapper mapper 
= getMapper(SysuserMapper
.class,session
);
        Sysuser sysuser 
= mapper
.selectUserByKey(1);
        System
.out
.println(sysuser
);
    }
    public static <T> T 
getMapper(Class
<T> clz
,SqlSession session
){
        T t 
= (T
)Proxy
.newProxyInstance(clz
.getClassLoader(), new Class[]{clz
}, new InvocationHandler() {
            @Override
            public Object 
invoke(Object proxy
, Method method
, Object
[] args
) throws Throwable 
{
                String methodName 
= method
.getName();
                Class
<?> declaringClass 
= method
.getDeclaringClass();
                String name 
= declaringClass
.getName();
                String id 
= name
+"."+methodName
;
                MappedStatement mappedStatement 
= session
.getConfiguration().getMappedStatement(id
);
                SqlCommandType sqlCommandType 
= mappedStatement
.getSqlCommandType();
                if(sqlCommandType
==SqlCommandType
.SELECT
){
                    Class
<?> returnType 
= method
.getReturnType();
                    if(Collection
.class.isAssignableFrom(returnType
)){
                        return session
.selectList(id
,args
[0]);
                    }else{
                        return session
.selectOne(id
,args
[0]);
                    }
                }
                return null
;
            }
        });
        return t
;
    }
}