mybatis 使用DISTINCT 獲取需要列符合條件的列 做動態拼接

考慮到表信息量大 需要對一些列加入索引 數據量大 不建議使用 默認返回1024個字節長度，超過長度會被截取掉。

解決方法：

在mysql的配置文件,my.ini中mysqld下增加group_concat_max_len = 102400

mysql中執行：set global group_concat_max_len=102400;

                      或者set session group_concat_max_len=102400;

global對全局有效，session只對當前會話有效。需要管理員才能操作global！

重啟mysql服務器生效

linux下的mysql一樣的設置，只不過linux下的mysql配置文件是/etc/my.cnf設置！

<select id="selectAggrColumnValues" parameterType="org.ggj.nms.db.condition.WiErpStockCondition"
            resultType="string">

        select
        <choose>
            <when test="field  == 'orginal_value_id'">
                group_concat(DISTINCT orginal_value_id)
            </when>
            <when test="field  == 'aging_degree'">
                group_concat(DISTINCT aging_degree)
            </when>
            <when test="field  == 'province_id'">
                group_concat(DISTINCT province_id)
            </when>
            <when test="field  == 'city_id'">
                group_concat(DISTINCT city_id)
            </when>
            <when test="field  == 'district_id'">
                group_concat(DISTINCT district_id)
            </when>
            <otherwise></otherwise>
        </choose>
        as lst
        from wi_erp_stock
        <where>
            state = 20
            /*省id*/
            <choose>
                <when test="condition.provinceId != null || 0 == condition.provinceId">
                    and province_id = #{condition.provinceId,jdbcType=INTEGER}
                </when>
                <otherwise>
                    and province_id !=-1
                </otherwise>
            </choose>
            /*市id*/
            <if test="condition.cityId != null">
                and city_id=#{condition.cityId,jdbcType=INTEGER}
            </if>
            /*區id*/
            <if test="condition.districtId != null">
                and district_id=#{condition.districtId,jdbcType=INTEGER}
            </if>
            /*新舊程度*/
            <choose>
                <when test="condition.agingDegree == null || 0 == condition.agingDegree ">
                    and aging_degree !=0
                </when>
                <otherwise>
                    AND aging_degree = #{condition.agingDegree,jdbcType=INTEGER}
                </otherwise>
            </choose>
            /*原值下標*/
            <if test="condition.originalValueRange != null">
                and orginal_value_id=#{condition.originalValueRange,jdbcType=INTEGER}
            </if>

            <if test="condition.categoryLevel1 != null">
                AND category_level1 = #{condition.categoryLevel1,jdbcType=BIGINT}
            </if>

            <if test="condition.categoryLevel2 != null">
                AND category_level2 = #{condition.categoryLevel2,jdbcType=BIGINT}
            </if>

            <if test="condition.categoryLevel3 != null">
                AND category_level3 = #{condition.categoryLevel3,jdbcType=BIGINT}
            </if>

            <if test="condition.categoryLevel4 != null">
                AND category_level4 = #{condition.categoryLevel4,jdbcType=BIGINT}
            </if>

            <if test="condition.inputAssetNo != null and condition.inputAssetNo != ''">
                and asset_no = #{condition.inputAssetNo,jdbcType=VARCHAR}
            </if>

            <if test="condition.inputSkuName != null and condition.inputSkuName != ''">
                <![CDATA[    and sku_name like CONCAT('%',#{condition.inputSkuName,jdbcType=VARCHAR},'%')   ]]>
            </if>

            <if test="condition.inputDeptIds != null and condition.inputDeptIds.size() > 0">
                and dept_id incondition.
                <foreach item="item" collection="condition.inputDeptIds" index="index" open="(" separator="," close=")">
                    #{item,jdbcType=BIGINT}
                </foreach>
            </if>
        </where>

    </select>

 結果;

lst
1,23,34,45,35,45，,4,34,546，，344

 

可以使用 group_concat 函數，將不重復的其他字段拼接起來，通過分組獲取到ID列符合條件的id

select distinct name,group_concat(id) from table group by name

結果：

name id 
   a 1
   b 2,5
   c 3,4