mybatis oracle 批量新增

Posted 2022-08-09 lidada

 

 

假定场景：批量导入用户信息

一般批量新增使用 SELECT … INSERT INTO 和 INSERT INTO … SELECT

我们这次使用第二种

 

一、先建一张用户信息表模拟批量导入用户信息

create table u_info
    id           NUMBER not null,
    info_no      VARCHAR2(32) not null,
    name         VARCHAR2(32) not null,
    birthday     DATE,
    age          NUMBER,
    create_date  DATE not null

-- 自动按天数分区
-- tablespace TBS_DATA 
--partition by range (create_date) interval (numtodsinterval(1, ‘DAY‘))
--(partition P20190101 values less than (TO_DATE(‘2019-01-01 00:00:00‘, ‘SYYYY-MM-DD HH24:MI:SS‘, ‘NLS_CALENDAR=GREGORIAN‘)))
;

--create unique index IU_INFO_NO_DATE on u_info (info_no,create_date) tablespace TBS_IDX online local;

-- Add comments
comment on table u_info is ‘用户信息表‘;
comment on column u_info.id is ‘主键‘;
comment on column u_info.info_no is ‘用户编号‘;
comment on column u_info.name is ‘姓名‘;
comment on column u_info.birthday is ‘生日‘;
comment on column u_info.age is ‘年龄‘;
comment on column u_info.create_date is ‘创建时间‘;

 

二、mybatis xml（传入集合参数，建议批量数量控制在3000以内）

<insert id="insertBatch" parameterType="java.util.List">
    /**UserInfoMapper.insertBatch*/
    INSERT INTO u_info(
        id,
        info_no,
        name,
        birthday,
        age,
        create_date
    )
    SELECT
    S.*,
    SYSDATE
    FROM(
    <foreach item="bean" index="index" collection="list" separator="UNION ALL">
        SELECT
        #bean.id,              jdbcType=DECIMAL,
        #bean.infoNo,          jdbcType=VARCHAR,
        #bean.name,            jdbcType=VARCHAR,
        #bean.birthday,        jdbcType=TIMESTAMP,
        #bean.age,             jdbcType=DECIMAL
        FROM DUAL
    </foreach>
    ) S
</insert>