基于Java如何用Mybatis实现oracle批量插入及分页查询

寻技术 JAVA编程 / Oracle 2023年09月10日 81

今天小编给大家分享一下基于Java如何用Mybatis实现oracle批量插入及分页查询的相关知识点,内容详细,逻辑清晰,相信大部分人都还太了解这方面的知识,所以分享这篇文章给大家参考一下,希望大家阅读完这篇文章后有所收获,下面我们一起来了解一下吧。

1、单条数据insert

<!--简单SQL-->
insert into userinfo (USERID, USERNAME, AGE) values(1001,'小明',20);

<!--Mybatis写法1,有序列,主键是自增ID,主键是序列-->
<insert id="insert" parameterType="com.zznode.modules.bean.UserInfo">
    <selectKey resultType="java.lang.Integer" order="BEFORE" keyProperty="userid">
      SELECT userinfo_userid_seq.nextval as userid from dual
    </selectKey>
    insert into EPG_ALARM_INFO (USERID, USERNAME, AGE)
    values (#{userid}, #{username}, #{age})
</insert>

<!--Mybatis写法2,无序列,主键是uuid,字符串-->
<insert id="insert" parameterType="com.zznode.modules.bean.UserInfo">
    insert into EPG_ALARM_INFO (USERID, USERNAME, AGE, TIME)
    values (#{userid}, #{username}, #{age}, sysdate)
</insert>

2、批量数据批量insert

insert all into 的方式返回值由最后的select 决定:

<!--简单SQL, 方法1-->
INSERT ALL 
INTO userinfo (USERID, USERNAME, AGE) values(1001,'小明',20)
INTO userinfo (USERID, USERNAME, AGE) values(1002,'小红',18)
INTO userinfo (USERID, USERNAME, AGE) values(1003,'张三',23)
select 3 from dual;
<!--简单SQL, 方法2-->
begin
    insert into userinfo (USERID, USERNAME, AGE) values(1001,'小明',20);
    insert into userinfo (USERID, USERNAME, AGE) values(1001,'小红',18);
    insert into userinfo (USERID, USERNAME, AGE) values(1001,'张三',23);
end;
<!--简单SQL, 方法3-->
insert into userinfo (USERID, USERNAME, AGE) 
select 1001, '小明', 20 from dual union all
select 1002, '小红', 18 from dual union all
select 1003, '张三', 23 from dual
<!--Mybatis写法1,无序列-->
<insert id="insertBatch" parameterType="java.util.List">
    INSERT ALL 
    <foreach collection="list" index="index" item="item">
        INTO userinfo (USERID, USERNAME, AGE)
        VALUES (#{item.userid}, #{item.username}, #{item.age})
    </foreach>
    select list.size from dual
</insert>
<!--Mybatis写法2,无序列-->
<insert id="insertBatch">
    insert into EPG_ALARM_INFO (USERID, USERNAME, AGE)
    <foreach collection="list" item="item" index="index" separator="union all">
        <!-- <foreach collection="list" item="item" index="index" separator="union all" open="(" close=")"> -->
        <!-- (select #{item.userid}, #{item.username}, #{item.age} from dual) -->
        
        <!-- 上面带括号,下面不带括号,都可以,少量数据不带括号效率高 -->
        select #{item.userid}, #{item.username}, #{item.age} from dual
    </foreach>
</insert>    
<!--Mybatis写法3,有序列-->
<insert id="insertBatch">
    insert into EPG_ALARM_INFO (USERID, USERNAME, AGE)
    SELECT userinfo_userid_seq.nextval, m.* FROM (
    <foreach collection="list" item="item" index="index" separator="union all">
        select #{item.username}, #{item.age} from dual
    </foreach>
    ) m
</insert>

3、创建序列

  • minvalue n (/nominvalue):最小值为n

  • maxvalue n (/nomaxvalue):最大值为n

  • start with n:从n开始计数

  • increment by n:每次增加n

  • cache n (/nocache):缓存n个sequence值 / 不缓存,如果缓存,则会有跳号的危险

  • noorder (/order):不保证序列号按顺序生成请求

  • cycle n (/nocycle):如果到达最大值n后,再次从start with n开始

  • currval:序列的当前值,新序列必须使用一次nextval 才能获取到值,否则会报错

  • nextval:表示序列的下一个值。新序列首次使用时获取的是该序列的初始值,从第二次使用时开始按照设置的步进递增

删除序列语法: drop sequence seq_表名

<!--
create sequence 序列名     
       increment by 1 	--每次增加几个,我这里是每次增加1
       start with 1   	--从1开始计数
       nomaxvalue      	--不设置最大值
       nocycle         	--一直累加,不循环
       nocache;        	--不建缓冲区
在插入语句中调用:序列名.nextval  生成自增主键。
-->
<!--创建序列-->
create sequence SEQ_USERINFO
minvalue 1
maxvalue 9999999999
start with 1
increment by 1
nocache;

<!--删除序列-->
drop sequence SEQ_USERINFO

4、oracle分页查询

前端与后端交互,分页查询

service业务实现:

public List<TBadUserW> queryPageBadUserInfo(TbadUserQuery queryModel) {
    log.info("分页查询请求参数,{}", JSON.toJSONString(queryModel));
    int pageNum = queryModel.getPageNum(); // 开始页
    int pageSize = queryModel.getPageSize(); // 每页数量
    queryModel.setStart((pageNum - 1) * pageSize); // 开始行数 (+1后)
    queryModel.setEnd(pageNum * pageSize); // 结束行数
    List<TBadUserW> beans = badUserWDao.queryPageBadUserInfo(queryModel);
    log.info("最终查询数量:", beans.size());
    return beans;
}

mapper.xml文件:

<select id="queryPageInfo" parameterType="com.zznode.test.bean.TbadUserQuery"
        resultMap="BaseResultMap" >
    SELECT tt.*	FROM
    (
    	<!--前端分页需要 total总记录-->
        SELECT t.*, ROWNUM rown, COUNT (*) OVER () total FROM
        (
            select <include refid="Base_Column_List"/> from T_BAD_USER_W
            <where>
                <if test="city != null and city !=''">
                    and city = #{city}
                </if>
                <if test="county != null and county != ''">
                    and county = #{county}
                </if>
                <if test="startTime != null and startTime !=''">
                    and loadtime >= to_date(#{startTime} , 'yyyy-mm-dd hh44:mi:ss')
                </if>
                <if test="endTime != null and endTime !=''">
                    and loadtime <![CDATA[<=]]> to_date(#{endTime} , 'yyyy-mm-dd hh44:mi:ss')
                </if>
            </where>
        )t
    )tt
    where tt.rown > #{start} and tt.rown <![CDATA[<=]]> #{end}
</select>

后端海量数据导出,批量查询

service业务实现:

public List<TBadUserW> queryPageBadUserInfo(TbadUserQuery queryModel) {
    log.info("分页查询请求参数,{}", JSON.toJSONString(queryModel));
    List<TBadUserW> result = new ArrayList<>();
    int pageNum = queryModel.getPageNum(); // 开始页
    int pageSize = queryModel.getPageSize(); // 每页数量(可以每页设置为200/500/1000),每次查询的条数
    boolean searchAll = true;
    while (searchAll){
        queryModel.setStart((pageNum - 1) * pageSize); // 开始行数 (+1后)
        queryModel.setEnd(pageNum * pageSize); // 结束行数
        List<TBadUserW> beans = badUserWDao.queryPageBadUserInfo(queryModel);
        if (null == beans || beans.size() < pageSize) {
            searchAll = false;
        }
        if (CollectionUtils.isNotEmpty(beans)) {
            result.addAll(beans);
        }
        pageNum++;
    }
    log.info("最终查询数量:", result.size());
    return result;
}

mapper.xml文件编写

<!--这种写法是比较高效的分批查询方法,分批不需要查询total总量,不支持total-->
<select id="queryPageInfo" parameterType="com.zznode.test.bean.TbadUserQuery"
        resultMap="BaseResultMap" >
    SELECT tt.*	FROM
    (
        SELECT t.*, ROWNUM rown FROM
        (
            select <include refid="Base_Column_List"/> from T_BAD_USER_W
            <where>
                <if test="city != null and city !=''">
                    and city = #{city}
                </if>
                <if test="county != null and county != ''">
                    and county = #{county}
                </if>
                <if test="startTime != null and startTime !=''">
                    and loadtime >= to_date(#{startTime} , 'yyyy-mm-dd hh44:mi:ss')
                </if>
                <if test="endTime != null and endTime !=''">
                    and loadtime <![CDATA[<=]]> to_date(#{endTime} , 'yyyy-mm-dd hh44:mi:ss')
                </if>
            </where>
        )t where ROWNUM <![CDATA[<=]]> #{end}
    )tt
    where tt.rown > #{start}
</select>
关闭

用微信“扫一扫”