sql 如何在查询时移除某一行数据以及这一行的上一行数据

小何开发2024-08-24 15:46

oracel 有个rownum 函数, 它可以为某一行数据增加一个序列比如查出的序列值便是

json 复制代码

[
{"name": "法外狂徒-张三",option":"申请团长", rownum:1},
{"name": "法外狂徒-张三","option":"申请人撤销", rownum:2},
{"name": "法外狂徒-张三","option":"申请旅长", rownum:3},
{"name": "法外狂徒-张三","option":"司令通过", rownum:4},
{"name": "法外狂徒-张三","option":"申请师长", rownum:5},
{"name": "法外狂徒-张三","option":"申请人撤销", rownum:6},
{"name": "法外狂徒-张三","option":"申请军长", rownum:7},
{"name": "法外狂徒-张三","option":"司令通过", rownum:8},
]

以上是表内的结果集,由于申请人撤销了操作,所以它之前的操作是无效的, 在页面也不显示, 申请人撤销这个字段是固定的,它之前的操作不固定, 所以如果想要过滤掉它之前的数据,我们需要先过去到申请人撤销那条数据的下标,再去-1.

想法, 我们可以先查询上面的结果集, 并且通过rownum为每个结果集定义一个固定的序号, 查询结果集select * from 结果集 where rownum not in ((select rownum from (结果集) where option = "申请人撤销" unine all select rownum -1 from (结果集) where option = "申请人撤销"))

这样我们就得到如下结果

json 复制代码

[

{"name": "法外狂徒-张三","option":"申请旅长", rownum:3},
{"name": "法外狂徒-张三","option":"司令通过", rownum:4},
{"name": "法外狂徒-张三","option":"申请军长", rownum:7},
{"name": "法外狂徒-张三","option":"司令通过", rownum:8},
]

由于结果集被反复使用到,我们在mybatis中可以将结果集sql 提出到sql片段中,便于反复调用

Mybaits.xml

xml 复制代码

<sql id= "dataSource">
  select t*,rownum as id from 表名 where t.name=#{name} order by operTime
</sql>

<sql id="removeRowIndex">
	select id from (<include refid="dataSource"/>) where option="申请人撤销"
	union all 
	select id-1 from (<include refid="dataSource"/>) where option="申请人撤销"
</sql>

<select id="getFolw" resultMap="java.util.HashMap" parameterType="java.util.HashMap">
  select * from(<include refid="dataSource"/>) q
  where q.id not in (
    (<include refid="removeRowIndex"/>)
    )
</select>