Mybatis聯(lián)合查詢(xún)XML與注解對(duì)比分析

更新時(shí)間：2026年03月12日 08:55:38 作者：很倔也很天真

文章對(duì)比了MyBatis中XML配置方式和注解方式查詢(xún)多表關(guān)聯(lián)的效率和實(shí)現(xiàn)方式,XML配置方式執(zhí)行聯(lián)合查詢(xún)效率較高,但注解方式雖然可以獨(dú)立使用,但在處理復(fù)雜映射時(shí)不如XML靈活

測(cè)試類(lèi)容

XML配置轉(zhuǎn)注解方式

實(shí)體類(lèi)

為了測(cè)試請(qǐng)忽略設(shè)計(jì)是否合理…

User.java

@Alias("User")
public class User {
    private Integer id;
    private String username;
    private String password;
    private Timestamp birthday;
    private Boolean sex;
}

UserInfo.java

@Alias("UserInfo")
public class UserInfo {
    private Integer id;
    private User user;
    private String name;
}

UserMessage.java

@Alias("UserMessage")
public class UserMessage {
    private Integer userId;
    private UserInfo userInfo;
    private List<Message> messages;
}

Message.java

@Alias("Message")
public class Message {
    private Integer id;
    private Integer userId;
    private String content;
    private Timestamp time;
}

三表之間的關(guān)聯(lián)為 userId

XML配置方式

XML配置

<select id="selectUserInfoAndMessages" resultMap="UserMessages">
    select
    ui.id as uiid,
    ui.name as name,
    ui.user_id as info_user_id,
    u.id as user_id,
    u.username
    as username,
    u.password as password,
    u.birthday as birthday,
    u.sex as
    sex,
    um.id as umid,
    um.user_id as um_user_id,
    um.content as content,
    um.time as um_time
    from user_info ui
    left outer join user u on u.id = ui.user_id
    left outer join user_message um
    on u.id = um.user_id
    where u.id = #{id}
</select>
<resultMap type="UserMessage" id="UserMessages">
    <id property="userId" column="user_id" />
    <association property="userInfo" column="info_user_id"
        javaType="UserInfo">
        <id property="id" column="uiid" />
        <result property="name" column="name" />
        <association property="user" column="user_id" javaType="User">
            <id property="id" column="user_id" />
            <result property="username" column="username" />
            <result property="password" column="password" />
            <result property="birthday" column="birthday" />
            <result property="sex" column="sex" />
        </association>
    </association>
    <collection property="messages" javaType="ArrayList"
        ofType="Message">
        <id property="id" column="umid" />
        <result property="userId" column="user_id" />
        <result property="content" column="content" />
        <result property="time" column="um_time" />
    </collection>
</resultMap>

查詢(xún)方法

public List<UserMessage> testSelectUserInfoAndMessages(Integer userId) {
    try (SqlSession session = sessionFactory.openSession()) {
        //映射器方式調(diào)用
        //return session.getMapper(UserMapper.class).selectUserInfoAndMessage(userId);
        return session.selectList("mybatis.dao.mappers.UserMapper.selectUserInfoAndMessages", userId);
    } catch (Exception e) {
        e.printStackTrace();
    }
    return null;
}

執(zhí)行過(guò)程

==> Preparing: select ui.id as uiid, ui.name as name, ui.user_id as info_user_id, u.id as user_id, u.username as username, u.password as password, u.birthday as birthday, u.sex as sex, um.id as umid, um.user_id as um_user_id, um.content as content, um.time as um_time from user_info ui left outer join user u on u.id = ui.user_id left outer join user_message um on u.id = um.user_id where u.id = ?
==> Parameters: 9(Integer)
<== Columns: uiid, name, info_user_id, user_id, username, password, birthday, sex, umid, um_user_id, content, um_time
<== Row: 5, 江流, 9, 9, test中文2, 1234567, 2018-07-17 17:04:09.0, 0, 1, 9, 信息1, 2018-07-18 11:36:58.0
<== Row: 5, 江流, 9, 9, test中文2, 1234567, 2018-07-17 17:04:09.0, 0, 2, 9, 信息2, 2018-07-18 11:37:05.0
<== Total: 2

結(jié)果

[UserMessage [userId=9, userInfo=UserInfo [id=5, user=User [id=9, username=test中文2, password=1234567, birthday=2018-07-17 17:04:09.0, sex=false], name=江流], messages=[Message [id=1, userId=9, content=信息1, time=2018-07-18 11:36:58.0], Message [id=2, userId=9, content=信息2, time=2018-07-18 11:37:05.0]]]]

此種方式只查詢(xún)了一次，結(jié)果集被過(guò)濾了重復(fù)項(xiàng)目，最終集合中只有一個(gè)UserMessage對(duì)象。

注解方式

注解方法

@Select("select * from user where id = #{userId}")
@Results(value = {
        @Result(property="userId" , column="id"),
        @Result(property="userInfo" , column="id", one = @One(select="mybatis.dao.mappers.UserMapper.getUserInfoByAnnotation")),
        @Result(property="messages" , column = "id" , many= @Many(select="mybatis.dao.mappers.UserMapper.getUserMessagesByAnnotation"))
})
public List<UserMessage> selectUserInfoAndMessagesByAnnotation(Integer userId);//查詢(xún)用戶信息和用戶消息

@Select("select * from user_info where user_id = #{userId}")
@Results(value={
        @Result(property="id" , column="id"),
        @Result(property="name" , column = "name"),
        @Result(property="user" , column = "user_id" , one = @One(select="mybatis.dao.mappers.UserMapper.getUserByAnnotation"))
})
public UserInfo getUserInfoByAnnotation(Integer userId);//查詢(xún)用戶信息

@Select("select * from user where id = #{userId}")
public User getUserByAnnotation(Integer userId);//查詢(xún)用戶

@Select("select * from user_message where user_id = #{userId}")
public List<Message> getUserMessagesByAnnotation(Integer userId);//查詢(xún)用戶消息集合

查詢(xún)方法

public List<UserMessage> testSelectUserMessagesByAnnotation(Integer userId) {
    try (SqlSession session = sessionFactory.openSession()) {
        return session.getMapper(UserMapper.class).selectUserInfoAndMessagesByAnnotation(userId);
    } catch (Exception e) {
        e.printStackTrace();
    }
    return null;
}

執(zhí)行過(guò)程

==> Preparing: select * from user where id = ?
==> Parameters: 9(Integer)
<== Columns: id, username, password, sex, birthday
<== Row: 9, test中文2, 1234567, 0, 2018-07-17 17:04:09.0
====> Preparing: select * from user_info where user_id = ?
====> Parameters: 9(Integer)
<==== Columns: id, user_id, name
<==== Row: 5, 9, 江流
======> Preparing: select * from user where id = ?
======> Parameters: 9(Integer)
<====== Columns: id, username, password, sex, birthday
<====== Row: 9, test中文2, 1234567, 0, 2018-07-17 17:04:09.0
<====== Total: 1
<==== Total: 1
====> Preparing: select * from user_message where user_id = ?
====> Parameters: 9(Integer)
<==== Columns: id, user_id, content, time
<==== Row: 1, 9, 信息1, 2018-07-18 11:36:58.0
<==== Row: 2, 9, 信息2, 2018-07-18 11:37:05.0
<==== Total: 2
<== Total: 1

可以看出執(zhí)行了幾次查詢(xún)語(yǔ)句。

結(jié)果

[UserMessage [userId=9, userInfo=UserInfo [id=5, user=User [id=9, username=test中文2, password=1234567, birthday=2018-07-17 17:04:09.0, sex=false], name=江流], messages=[Message [id=1, userId=9, content=信息1, time=2018-07-18 11:36:58.0], Message [id=2, userId=9, content=信息2, time=2018-07-18 11:37:05.0]]]]

總結(jié)

XML執(zhí)行聯(lián)合查詢(xún)效率是要高一些的
注解方式方法倒是可以獨(dú)立使用，雖然XML也可以用select屬性
混合使用更好

套用官方的話

1、因?yàn)樽畛踉O(shè)計(jì)時(shí)，MyBatis 是一個(gè) XML 驅(qū)動(dòng)的框架。配置信息是基于 XML 的，而且映射語(yǔ)句也是定義在 XML 中的。而到了 MyBatis 3，就有新選擇了。MyBatis 3 構(gòu)建在全面且強(qiáng)大的基于 Java 語(yǔ)言的配置 API 之上。這個(gè)配置 API 是基于 XML 的 MyBatis 配置的基礎(chǔ)，也是新的基于注解配置的基礎(chǔ)。注解提供了一種簡(jiǎn)單的方式來(lái)實(shí)現(xiàn)簡(jiǎn)單映射語(yǔ)句，而不會(huì)引入大量的開(kāi)銷(xiāo)。

2、注意不幸的是，Java 注解的的表達(dá)力和靈活性十分有限。盡管很多時(shí)間都花在調(diào)查、設(shè)計(jì)和試驗(yàn)上，最強(qiáng)大的 MyBatis 映射并不能用注解來(lái)構(gòu)建——并不是在開(kāi)玩笑，的確是這樣。比方說(shuō)，C#屬性就沒(méi)有這些限制，因此 MyBatis.NET 將會(huì)比 XML 有更豐富的選擇。也就是說(shuō)，基于 Java 注解的配置離不開(kāi)它的特性。

附數(shù)據(jù)表SQL

CREATE TABLE `user` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `username` varchar(25) DEFAULT NULL,
  `password` varchar(25) DEFAULT NULL,
  `sex` bit(1) DEFAULT b'1',
  `birthday` datetime DEFAULT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=16 DEFAULT CHARSET=utf8;

CREATE TABLE `user_info` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `user_id` int(11) DEFAULT NULL,
  `name` varchar(25) DEFAULT NULL,
  PRIMARY KEY (`id`),
  KEY `user_id` (`user_id`),
  CONSTRAINT `user_info_ibfk_1` FOREIGN KEY (`user_id`) REFERENCES `user` (`id`) ON DELETE CASCADE ON UPDATE CASCADE
) ENGINE=InnoDB AUTO_INCREMENT=6 DEFAULT CHARSET=utf8;

CREATE TABLE `user_message` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `user_id` int(11) DEFAULT NULL,
  `content` varchar(255) DEFAULT NULL,
  `time` datetime DEFAULT NULL,
  PRIMARY KEY (`id`),
  KEY `message_ibfk_1` (`user_id`),
  CONSTRAINT `user_message_ibfk_1` FOREIGN KEY (`user_id`) REFERENCES `user` (`id`) ON DELETE CASCADE ON UPDATE CASCADE
) ENGINE=InnoDB AUTO_INCREMENT=4 DEFAULT CHARSET=utf8;

以上為個(gè)人經(jīng)驗(yàn)，希望能給大家一個(gè)參考，也希望大家多多支持腳本之家。

您可能感興趣的文章: