myBatis框架之入门(四)

Mybatis多表管理查询

多表关联关系分析:

多表关联:至少两个表关联.分析多表关系的经验技巧:从一条记录出发,不要从表整体去分析,比如分析A表和B表关系,A表中的一条记录对应B表中的几条记录,如果是一条,那么A到B就是一对一;如果是多条,那么A到B就是一对多.

　　一对一:例如从订单表出发,到用户表,是一对一的关系.

　　一对多,例如从用户表出发,到订单表,一条用户数据可以在订单表中存在多条记录,这就是一对多.通过主外键来体现一对多的表结构.一的一方是主表,多的一方是从表,从表中有一个字段是外键,指向了主表的主键.

　　多对多,用户和角色表,一个用户可以有很多角色,一个角色有很多用户,多对多通过中间表来表现.

多表关联的sql语句表达分析

　　笛卡尔积
　　　　SELECT *FROM USER,orders

　　关联查询

　　内关联 inner join on

　　　　SELECT * FROM USER u,orders o WHERE u.id=o.user_id

　　　　SELECT * FROM USER u INNER JOIN orders o ON u.id=o.user_id;

　　左外连接

　　　　SELECT * FROM USER u LEFT JOIN orders o ON u.id=o.user_id

　   右外连接　

　　　　SELECT * FROM USER u RIGHT JOIN orders o ON u.id=o.user_id

一对一查询

需求：查询订单表全部数据，关联查询出订单对应的用户数据（username address）。

Sql语句自己来写，Mybatis只是帮我们执行sql语句同时封装结果集。

SQL语句：

SELECT o.id,o.user_id,o.number,o.note,u.address,u.username 
FROM orders

o LEFT JOIN USER u ON o.user_id=u.id

对于查询结果集，没有对应的pojo对象，因此MyBatis框架也不能封装结果集，但是可以将User对象放在Orders对象中，因为一对一关系，一个Orders对象可以对应一个User对象。需要手动映射方式，实现查询结果集封装。
pojo对象

public class Orders {
private Integer id;
private Integer userId;
private String number;
private Date createtime;
private String note;
private User user;
public User getUser() {
return user;
}
public void setUser(User user) {
this.user = user;
}
public Integer getId() {
return id;
}
public void setId(Integer id) {
this.id = id;
}
public Integer getUserId() {
return userId;
}
public void setUserId(Integer userId) {
this.userId = userId;
}
public String getNumber() {
return number;
}
public void setNumber(String number) {
this.number = number;
}
public Date getCreatetime() {
return createtime;
}
public void setCreatetime(Date createtime) {
this.createtime = createtime;
}
public String getNote() {
return note;
}
public void setNote(String note) {
this.note = note;
}
@Override
public String toString() {
return "Orders{" +
"id=" + id +
", userId=" + userId +
", number='" + number + '\'' +
", createtime=" + createtime +
", note='" + note + '\'' +
", user=" + user +
'}';
}
}

association（联合）标签，实现手动映射

　　propery属性：封装的pojo对象
　　javaType属性：封装的pojo对象类型

 <mapper namespace="com.itheima.mapper.OrdersMapper">
        <select id="queryOrdersUser" resultMap="ordersUserResultType" >
            SELECT o.id,o.user_id,o.number,o.note,u.address,u.username FROM
            orders o LEFT JOIN USER u
            ON o.user_id=u.id
        </select>
        <resultMap id="ordersUserResultType" type="orders">
            <id column="id" property="id"></id>
            <result column="user_id" property="userId"></result>
            <result column="number" property="number"></result>
            <result column="createtime" property="createtime"></result>
            <result column="note" property="note"></result>
            <!-- 手动映射，配置User对象-->
            <association property="user" javaType="user">
                <id column="user_id" property="id"></id>
                <result column="address" property="address"></result>
                <result column="username" property="username"></result>
            </association>
        </resultMap>
    </mapper>

@Test
public void queryOrdersUser(){
    SqlSession sqlSession = sqlSessionFactory.openSession();
    OrdersMapper mapper = sqlSession.getMapper(OrdersMapper.class);
    List<Orders> list = mapper.queryOrdersUser();
    if(list != null && list.size() > 0){
        for (Orders orders : list){
            System.out.println(orders);
        }
    }
    sqlSession.close();
}

往往还有另外一种方式：新创建一个pojo，这个pojo里面增加两个字段username和address，然后使用resultType自动映射即可，但是这种方式不推荐，因为pojo多了之后很混乱，不容易做技术管理，而且这点方式也有点low
一对多查询
查询全部用户数据，关联查询出订单数据。

SQL语句：SELECT
u.id,u.username,u.sex,u.birthday,u.address,o.user_id,o.id,o.number,o.createtime,o.note
FROM USER u LEFT JOIN orders o ON u.id=o.user_id

一个用户对应多个订单数据，因此在pojo中，订单对象是存储在集合中，并保存在User对象中

pojo对象

public class User {
  
    private int id;
  
    private String username;
  
    private String sex;
  
    private Date birthday;
  
    private String address;
  
    private List<Orders> ordersList;
  

    public List<Orders> getOrdersList() {
    return ordersList;
 }
  

    public void setOrdersList(List<Orders> ordersList) {
    this.ordersList = ordersList;
 }
  

    public int getId() {
    return id;
 }
  

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

    public String getUsername() {
    return username;
 }
  

    public void setUsername(String username) {
    this.username = username;
 }
  

    public String getSex() {
    return sex;
 }
  

    public void setSex(String sex) {
    this.sex = sex;
 }
  

    public Date getBirthday() {
    return birthday;
 }
  

    public void setBirthday(Date birthday) {
    this.birthday = birthday;
 }
  

    public String getAddress() {
    return address;
 }
  

    public void setAddress(String address) {
    this.address = address;
 }
  
    @Override
  

    public String toString() {
    return "User{" +
        "id=" + id +
        ", username='" + username + '\'' +
        ", sex='" + sex + '\'' +
        ", birthday=" + birthday +
        ", address='" + address + '\'' +
                collection标签：
        property属性：封装的对应的属性名
        ofType属性：已经指定为一个集合List，需要指定集合中的泛型
        ", ordersList=" + ordersList +
        '}';
 }
}　

    <select id="queryUserOrders" resultMap="userOrdersResultMap">
         SELECT u.id,u.username,u.sex,u.birthday,u.address,o.user_id,o.id,
         o.number,o.createtime,o.note
         FROM USER u LEFT JOIN orders o
         ON u.id=o.user_id
    </select>

    <resultMap id="userOrdersResultMap" type="user">
         <id column="id" property="id"></id>
         <result column="username" property="username"></result>
         <result column="sex" property="sex"></result>
         <result column="birthday" property="birthday"></result>
         <result column="address" property="address"></result>
         <collection property="ordersList" ofType="orders">
           <id column="id" property="id"></id>
           <result column="user_id" property="userId"></result>
           <result column="number" property="number"></result>
           <result column="createtime" property="createtime"></result>
           <result column="note" property="note"></result>
         </collection>
    </resultMap>

    @Test
    public void queryUserOrders(){
        SqlSession sqlSession = sqlSessionFactory.openSession();
        UserMapper mapper = sqlSession.getMapper(UserMapper.class);
        List<User> list = mapper.queryUserOrders();
        if(list != null && list.size() > 0){
            for(User user : list){
                System.out.println(user);
            }
        }
        sqlSession.close();
    }

多对多查询

多对多：双向的一对多，从A表到B表出发是一对多的关系，从B表出发到A表也是一对多的关系
SQL语句：

SELECT r.RID,r.RNAME,r.RDESC,u.id,u.username,u.sex,u.birthday,u.address
FROM role r LEFT JOIN user_role ur
ON r.RID=ur.RID LEFT JOIN USER u ON u.id=ur.UID

pojo对象

public class Role {
    private Integer rid;
    private String rname;
    private String rdesc;
    private List<User> userList;
    public Integer getRid() {
        return rid;
    }
    public void setRid(Integer rid) {
        this.rid = rid;
    }
    public String getRname() {
        return rname;
    }
    public void setRname(String rname) {
        this.rname = rname;
    }
    public String getRdesc() {
        return rdesc;
    }
    public void setRdesc(String rdesc) {
        this.rdesc = rdesc;
    }
    public List<User> getUserList() {
        return userList;
    }
    public void setUserList(List<User> userList) {
        this.userList = userList;
    }
    @Override
    public String toString() {
        return "Role{" +
                "rid=" + rid +
                ", rname='" + rname + '\'' +
                ", rdesc='" + rdesc + '\'' +
                ", userList=" + userList +
                '}';
    }
}

手动映射roleMapper.xml

  <mapper namespace="com.itheima.mapper.RoleMapper">
        <select id="queryRoleUser" resultMap="roleUserResultMap">
            SELECT
            r.RID,r.RNAME,r.RDESC,u.id,u.username,u.sex,u.birthday,u.address FROM
            role r LEFT JOIN user_role ur ON r.RID=ur.RID LEFT JOIN USER u ON
            u.id=ur.UID
        </select>
        <resultMap id="roleUserResultMap" type="role">
            <id column="rid" property="rid"></id>
            <result column="rname" property="rname"></result>
            <result column="rdesc" property="rdesc"></result>
            <result column="" property=""></result>
            <collection property="userList" ofType="user">
                <id column="id" property="id"></id>
                <result column="username" property="username"></result>
                <result column="sex" property="sex"></result>
                <result column="birthday" property="birthday"></result>
                <result column="address" property="address"></result>
            </collection>
        </resultMap>
    </mapper>

    @Test
    public void queryRoleUser() {
  SqlSession sqlSession = sqlSessionFactory.openSession();
  RoleMapper mapper = sqlSession.getMapper(RoleMapper.class);
  List<Role> roleList = mapper.queryRoleUser();
  if (roleList != null && roleList.size() > 0) {
    for (Role role : roleList) {
      System.out.println(role);
   }
 }
  sqlSession.close();
    }

小案例:

用户主键查询用户信息

<!--
 主键查询用户信息
-->
<select id="queryUserById" resultType="user" parameterType="int">
 select * from user where id = #{id}
</select>

/**
* 主键查询用户信息
*/
@Test
public void queryUserById(){
  SqlSession sqlSession = sqlSessionFactory.openSession();
  UserMapper userMapper = sqlSession.getMapper(UserMapper.class);
  User user = userMapper.queryUserById(1);
  System.out.println(user);
  sqlSession.close();
}

订单查询用户信息 一对一

<!--
 订单，查询订单对应的用户
-->
<select id="queryOrdersUser" resultMap="queryOrdersUserResultMap">
 SELECT
o.id,o.user_id,o.number,o.createtime,o.note,u.username,u.address FROM
orders o LEFT JOIN USER u
 ON o.user_id = u.id
</select>
用户查询订单信息 一对多
<resultMap id="queryOrdersUserResultMap" type="orders">
  <id column="id" property="id"></id>
  <result column="user_id" property="userId"></result>
  <result column="number" property="number"></result>
  <result column="createtime" property="createtime"></result>
  <result column="note" property="note"></result>
  <association property="user" javaType="user">
    <id column="user_id" property="id"></id>
    <result column="username" property="username"></result>
    <result column="address" property="address"></result>
  </association>
</resultMap>

/**
* 订单查询用户信息，一对一
*/
@Test
public void queryOrdersUser(){
  System.out.println();
  SqlSession sqlSession = sqlSessionFactory.openSession();
  OrdersMapper ordersMapper =
sqlSession.getMapper(OrdersMapper.class);
  List<Orders> ordersList = ordersMapper.queryOrdersUser();
  if(ordersList != null && ordersList.size() > 0){
    for (Orders orders : ordersList){
      System.out.println(orders);
   }
 }
  sqlSession.close();
}

用户查询订单信息 一对多

<!--
 根据用户查询订单，一对多
-->
<select id="queryUserOrders" resultMap="queryUserOrdersResultMap">
 SELECT u.id,u.username,u.sex,u.birthday,u.address,o.id
oid,o.number,o.createtime FROM USER u LEFT JOIN orders o
一对一延迟加载
针对的是关联对象
用户和订单从面相对象的角度来说就是关联对象，当只需要订单数据，尚不需要用户数据的时
候，就不应该去查询用户表，啥时候用到用户数据，啥时候查询
 ON u.id = o.user_id
</select>
<resultMap id="queryUserOrdersResultMap" type="user">
  <id column="id" property="id"></id>
  <result column="username" property="username"></result>
  <result column="sex" property="sex"></result>
  <result column="birthday" property="birthday"></result>
  <result column="address" property="address"></result>
  <collection property="ordersList" ofType="orders">
    <id column="oid" property="id"></id>
    <result column="number" property="number"></result>
    <result column="createtime" property="createtime"></result>
  </collection>
</resultMap>

/**
* 用户查询订单，一对多
*/
@Test
public void queryUserOrders(){
  System.out.println();
  SqlSession sqlSession = sqlSessionFactory.openSession();
  UserMapper mapper = sqlSession.getMapper(UserMapper.class);
  List<User> list = mapper.queryUserOrders();
  if(list != null && list.size() > 0){
    for(User user : list){
      System.out.println(user);
   }
 }
  sqlSession.close();
}

一对一延迟加载

针对的是关联对象

用户和订单从面相对象的角度来说就是关联对象，当只需要订单数据，尚不需要用户数据的时候，就不应该去查询用户表，啥时候用到用户数据，啥时候查询
       关联的sql语句肯定要拆分了

association标签
column属性：关联查询条件的属性
select属性：指定另一个查询语句

<!--
 订单，查询订单对应的用户
 拆分SQL语句，实现延迟加载
-->
<select id="queryOrdersUser" resultMap="queryOrdersUserResultMap">
 SELECT o.id,o.user_id,o.number,o.createtime,o.note FROM orders o
</select>
<resultMap id="queryOrdersUserResultMap" type="orders">
  <id column="id" property="id"></id>
  <result column="user_id" property="userId"></result>
  <result column="number" property="number"></result>
  <result column="createtime" property="createtime"></result>
  <result column="note" property="note"></result>
  <!--
   column指定查询的参数
   select关联其他查询语句
  -->
  <association property="user" javaType="user" column="user_id"
select="queryUserByUserId">
   <!-- <id column="user_id" property="id"></id>
    <result column="username" property="username"></result>
    <result column="address" property="address"></result>-->
  </association>
</resultMap>
<select id="queryUserByUserId" parameterType="int" resultType="user">
 select id,username,sex,birthday,address from user where id=#
{user_id}
</select>

/**
* 订单查询用户信息，一对一
* 延迟加载
*/
@Test
public void queryOrdersUser(){
  System.out.println();
  SqlSession sqlSession = sqlSessionFactory.openSession();
  OrdersMapper ordersMapper =
sqlSession.getMapper(OrdersMapper.class);
  List<Orders> ordersList = ordersMapper.queryOrdersUser();
  if(ordersList != null && ordersList.size() > 0){
    for (Orders orders : ordersList){
      System.out.println(orders);
      System.out.println(orders.getUser());
   }
 }
  sqlSession.close();
}

开启延迟加载配置

<settings>
<!-- 开启延迟加载-->
<setting name="lazyLoadingEnabled" value="true" />
<!-- 关闭立即加载-->
<setting name="aggressiveLazyLoading" value="false" />
<!-- 设定tostring等方法延迟加载-->
<setting name="lazyLoadTriggerMethods" value="true" />
</settings>

一对多延迟加载

<!--
根据用户查询订单，一对多
延迟加载，拆分SQL语句
-->
<select id="queryUserOrders" resultMap="queryUserOrdersResultMap">
SELECT u.id,u.username,u.sex,u.birthday,u.address FROM USER u
</select>
<resultMap id="queryUserOrdersResultMap" type="user">
<id column="id" property="id"></id>
<result column="username" property="username"></result>
<result column="sex" property="sex"></result>
<result column="birthday" property="birthday"></result>
<result column="address" property="address"></result>
<collection property="ordersList" ofType="orders" column="id"
select="queryOrdersByUserId">
<!-- <id column="oid" property="id"></id>
<result column="number" property="number"></result>
<result column="createtime" property="createtime"></result>-->
</collection>
</resultMap>
<select id="queryOrdersByUserId" parameterType="int" resultType="orders">
select id,number,createtime from orders where user_id = #{user_id}
</select>

/**
* 用户查询订单，一对多
* 延迟加载
*/
@Test
public void queryUserOrders(){
System.out.println();
SqlSession sqlSession = sqlSessionFactory.openSession();
UserMapper mapper = sqlSession.getMapper(UserMapper.class);
List<User> list = mapper.queryUserOrders();
if(list != null && list.size() > 0){
for(User user : list){
System.out.println(user);
List<Orders> ordersList = user.getOrdersList();
System.out.println(ordersList);
}
}
sqlSession.close();
}

MyBatis一级缓存

一级缓存：是SqlSession级别的，也就是同一个SqlSession内执行相同select语句的时候，不
再去查询数据库，而是从Mybatis内部的缓存内存结构去直接拿到数据。
缓存失效时机：
sqlSession关闭
sqlSession提交事务（意味着可能是一个增删改的动作，需要更新缓存，那么这个时候Mybatis就会把已有的一级缓存给清理掉）
MyBatis二级缓存

开启二级缓存

<setting name="cacheEnabled" value="true" />

<!-- 具体Mapper.xml配置-->
<cache></cache>

执行结果

/**
* 主键查询用户信息
*/
@Test
public void queryUserById(){
SqlSession sqlSession = sqlSessionFactory.openSession();
UserMapper userMapper = sqlSession.getMapper(UserMapper.class);
User user = userMapper.queryUserById(1);
System.out.println(user);
sqlSession.close();
SqlSession sqlSession1 = sqlSessionFactory.openSession();
UserMapper userMapper1 = sqlSession1.getMapper(UserMapper.class);
User user1 = userMapper1.queryUserById(1);
System.out.println(user1);
sqlSession1.close();
}

ratio，两次都会去查询二级缓存，但是第一次的时候缓存中没有数据，第二次才有，所以命中率是0.5

注意：Mybatis的二级缓存在使用的时候有很多限制，在实际开发的时候Mybatis的二级缓存应用很少

比如在其他mapper.xml文件中有对user表的更新，那么UserMapper.xml的二级缓存就可能错误，我们很难要求对一张表的操作全部放到同一个mapper.xml中。