若数据量非常大，可以把List拆成多份，每份1000条数据。

import cn.hutool.core.collection.ListUtil;
import cn.hutool.core.util.StrUtil;
import cn.路劲.NetPointDTO;
import lombok.extern.slf4j.Slf4j;
import org.springframework.stereotype.Service;import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;@Slf4j
@Service
public class NetPointServiceImpl implements NetPointService {/*** 批量新增*/@Overridepublic boolean add(List<NetPointDTO> dataList) {//数据里的 编码 查重String netPointCode = dataList.stream().collect(Collectors.groupingBy(NetPointDTO::getNetPointCode, Collectors.counting())).entrySet().stream().filter(e -> e.getValue() > 1L).map(Map.Entry::getKey).collect(Collectors.joining("、"));if (StrUtil.isNotBlank(netPointCode)){throw new RuntimeException("网点编码 "+ netPointCode + " 重复");}//切分List， 此方法返回的是原List的视图，也就是说原List有变更，切分后的结果也会变更List<List<NetPointDTO>> partition = ListUtil.partition(dataList, 1000);Integer n = 0;for (List<NetPointDTO> list : partition){//Mapper 很简单，就不赘述了n = 执行批量新增的Mapper.addBatch(list);}return n > 0;}
}

NetPointDTO

import io.swagger.annotations.ApiModelProperty;
import lombok.Data;import javax.validation.constraints.NotBlank;
import javax.validation.constraints.Size;@Data
public class NetPointDTO {@ApiModelProperty(value = "网点信息编码", example = "企业网点编码", required = true)@NotBlank(message = "网点信息编码, 值不能为空")@Size(max = 50, message = "网点信息编码，最大长度 50位")private String netPointCode;;@ApiModelProperty(value = "网点名称", example = "网点名称", required = true)@NotBlank(message = "网点名称,值不能为空")@Size(max = 50, message = "网点名称，值长度 不能超过 50位")private String netPointName;@ApiModelProperty(value = "点坐标", example = "39.625934 118.232805", required = true)@NotBlank(message = "点坐标, 不能为空")private String point;@ApiModelProperty(value = "办公地点", example = "办公地址", required = true)@NotBlank(message = "办公地址, 不能为空")@Size(max = 100, message = "办公地点，值长度 不能超过 100位")private String officeAddress;
}

批量新增SQL.xml

useGeneratedKeys=“true” keyProperty=“id”
useGeneratedKeys设置为 true 时，
表示如果插入的表id以自增列为主键，则允许 JDBC 支持自动生成主键，并可将自动生成的主键id返回。
下边这种写法，实际上是通过循环，数据也是一条一条的插入的，好处是每条数据都有判断是否为null。

<insert id="addBatch" parameterType="java.util.List" keyProperty="id" useGeneratedKeys="true"><foreach collection="list" item="e" index="index" separator=";">insert into NET_POINT<trim prefix="(" suffix=")" suffixOverrides=","><if test="e.netPointCode != null"> net_point_code, </if><if test="e.netPointName != null"> NET_POINT_NAME, </if><if test="e.point != null"> POINT, </if><if test="e.officeAddress != null"> OFFICE_ADDRESS, </if>CREATE_TIME</trim>values<trim prefix="(" suffix=")" suffixOverrides=","><if test="e.netPointCode != null"> #{e.netPointCode}, </if><if test="e.netPointName != null"> #{e.netPointName}, </if><if test="e.point != null"> dmgeo.ST_PointFromText(CONCAT('point(', #{e.point}, ')'), 0), </if><if test="e.officeAddress != null"> #{e.officeAddress}, </if>now()</trim></foreach>
</insert>

sql中point点坐标，前边的函数 是达梦数据库的，
MySQL数据库的函数为 ST_GEOMFROMTEXT(CONCAT(‘POINT(’, #{point}, ‘)’))
若直接批量新增，可以这样写

<insert id="saveBatch">insert into NET_POINT(net_point_code, NET_POINT_NAME, POINT, CREATE_TIME)values<foreach collection="list" item="e" separator=",">( 	#{e.netPointCode}, #{e.netPointName},dmgeo.ST_PointFromText(CONCAT('point(', #{e.point}, ')'), 0),NOW())</foreach>
</insert>

批量修改SQL.xml

<update id="updateBatch" parameterType="java.util.List"><foreach collection="list" item="e" index="index" open="" close="" separator=";">UPDATE NET_POINT<set>update_time = NOW(),<if test="e.netPointName != null and e.netPointName != ''">NET_POINT_NAME = #{e.netPointName},</if><if test="e.point != null and e.point != ''">point = dmgeo.ST_PointFromText(CONCAT('point(', #{e.point}, ')'), 0),</if><if test="e.officeAddress != null and e.officeAddress != ''">OFFICE_ADDRESS = #{e.officeAddress},</if></set>WHERE NET_POINT_CODE = #{e.netPointCode}</foreach>
</update>