分享好友 数据库首页 频道列表

MYSQL-实现ORACLE- row_number() over(partition by ) 分组排序功能

数据库其他  2023-02-10 02:570

MYSQL-实现ORACLE- row_number() over(partition by ) 分组排序功能

 

　　由于MYSQL没有提供类似ORACLE中OVER()这样丰富的分析函数. 所以在MYSQL里需要实现这样的功能,我们只能用一些灵活的办法:

1.首先我们来创建实例数据:

drop table if exists heyf_t10;
create table heyf_t10 (empid int ,deptid int ,salary decimal(10,2) );

insert into heyf_t10 values
(1,10,5500.00),
(2,10,4500.00),
(3,20,1900.00),
(4,20,4800.00),
(5,40,6500.00),
(6,40,14500.00),
(7,40,44500.00),
(8,50,6500.00),
(9,50,7500.00);

 

2. 确定需求: 根据部门来分组,显示各员工在部门里按薪水排名名次.

显示结果预期如下:

+-------+--------+----------+------+
| empid | deptid | salary   | rank |
+-------+--------+----------+------+
|     1 |     10 |  5500.00 |    1 |
|     2 |     10 |  4500.00 |    2 |
|     4 |     20 |  4800.00 |    1 |
|     3 |     20 |  1900.00 |    2 |
|     7 |     40 | 44500.00 |    1 |
|     6 |     40 | 14500.00 |    2 |
|     5 |     40 |  6500.00 |    3 |
|     9 |     50 |  7500.00 |    1 |
|     8 |     50 |  6500.00 |    2 |
+-------+--------+----------+------+
9 rows in set (0.00 sec)

 

3. SQL 实现

SELECT
    empid,
    deptid,
    salary,
    rank
FROM
    (
        SELECT
            empid,
            deptid,
            salary,

        IF (
            @pdept = src.deptid ,@rank := @rank + 1 ,@rank := 1
        ) AS rank,
        @pdept := src.deptid AS g
    FROM
        (
            SELECT
                empid,
                deptid,
                salary
            FROM
                heyf_t10
            ORDER BY
                deptid ASC,
                salary DESC
        ) src,
        (
            SELECT
                @pdept := NULL ,@rank := 0
        ) var
    ) z;

 

4. 结果演示

mysql> SELECT
    ->  empid,
    ->  deptid,
    ->  salary,
    ->  rank
    -> FROM
    ->  (
    ->          SELECT
    ->                  empid,
    ->                  deptid,
    ->                  salary,
    ->
    ->          IF (
    ->                  @pdept = src.deptid ,@rank := @rank + 1 ,@rank := 1
    ->          ) AS rank,
    ->          @pdept := src.deptid AS g
    ->  FROM
    ->          (
    ->                  SELECT
    ->                          empid,
    ->                          deptid,
    ->                          salary
    ->                  FROM
    ->                          heyf_t10
    ->                  ORDER BY
    ->                          deptid ASC,
    ->                          salary DESC
    ->          ) src,
    ->          (
    ->                  SELECT
    ->                          @pdept := NULL ,@rank := 0
    ->          ) var
    ->  ) z;
+-------+--------+----------+------+
| empid | deptid | salary   | rank |
+-------+--------+----------+------+
|     1 |     10 |  5500.00 |    1 |
|     2 |     10 |  4500.00 |    2 |
|     4 |     20 |  4800.00 |    1 |
|     3 |     20 |  1900.00 |    2 |
|     7 |     40 | 44500.00 |    1 |
|     6 |     40 | 14500.00 |    2 |
|     5 |     40 |  6500.00 |    3 |
|     9 |     50 |  7500.00 |    1 |
|     8 |     50 |  6500.00 |    2 |
+-------+--------+----------+------+
9 rows in set (0.00 sec)

 

 

 我的SQL:

SELECT
    MESSAGE_ID,
    GET_USER_ID,
    SEND_USER_ID,
    MESSAGE_CONTEXT,
    CREATE_TIME
FROM
    (
        SELECT
            SRC.*,
        IF (
            @V_USER_ID = USER_ID ,@V_RANK := @V_RANK + 1 ,@V_RANK := 1
        ) AS RANK ,@V_USER_ID := USER_ID AS G_USER_ID
        FROM
            (
                SELECT
                    MESSAGE_ID,
                    USER_ID,
                    CREATE_TIME,
                    MESSAGE_CONTEXT,
                    GET_USER_ID,
                    SEND_USER_ID
                FROM
                    (
                        SELECT
                            MESSAGE_ID,
                            GET_USER_ID,
                            SEND_USER_ID,
                            GET_USER_ID AS USER_ID,
                            CREATE_TIME,
                            MESSAGE_CONTEXT
                        FROM
                            T_SD_MESSAGE
                        WHERE
                            GET_USER_ID != '1234'
                        UNION ALL
                            SELECT
                                MESSAGE_ID,
                                GET_USER_ID,
                                SEND_USER_ID,
                                SEND_USER_ID AS USER_ID,
                                CREATE_TIME,
                                MESSAGE_CONTEXT
                            FROM
                                T_SD_MESSAGE
                            WHERE
                                SEND_USER_ID != '1234'
                    ) METADATA
                ORDER BY
                    USER_ID ASC,
                    CREATE_TIME DESC
            ) SRC,
            (
                SELECT
                    @V_RANK = 0,
                    @V_USER_ID := NULL
            ) VARS
    ) SRC
WHERE
    RANK = 1
ORDER BY
    CREATE_TIME DESC

 

一个过程;

DROP PROCEDURE
IF EXISTS PROCE_USER_NEW_MSG;

DROP TEMPORARY TABLE
IF EXISTS TEM_USER_NEW_MSG;
DELIMITER || 


CREATE PROCEDURE PROCE_USER_NEW_MSG (
    IN FRIST_RESULT INT,
    IN FETCH_SIZE INT
)
BEGIN
    SELECT
        M.MESSAGE_ID,
        M.GET_USER_ID,
        M.SEND_USER_ID,
        M.MESSAGE_CONTEXT,
        M.CREATE_TIME,
        G_U.USER_NAME AS G_USER_NAME,
        S_U.USER_NAME AS S_USER_NAME,
        G_H.GENERAL_PIC_THUMBNAIL_URL AS G_HEADER,
        S_H.GENERAL_PIC_THUMBNAIL_URL AS S_HEADER
    FROM
        T_SD_MESSAGE M
    LEFT JOIN T_SD_USER G_U ON M.GET_USER_ID = G_U.USER_ID
    LEFT JOIN T_SD_USER S_U ON M.SEND_USER_ID = S_U.USER_ID
    LEFT JOIN T_SD_GENERAL_PICTURE G_H ON G_H.GENERAL_PICTURE_ID = G_U.USER_HEADER_PIC_ID
    LEFT JOIN T_SD_GENERAL_PICTURE S_H ON S_H.GENERAL_PICTURE_ID = S_U.USER_HEADER_PIC_ID
    WHERE
        M.MESSAGE_ID IN (
            SELECT
                MESSAGE_ID
            FROM
                (
                    SELECT
                        MESSAGE_ID,
                        RANK,
                        MESSAGE_CONTEXT,
                        CREATE_TIME
                    FROM
                        (
                            SELECT
                                SRC.*,
                            IF (
                                @V_USER_ID = USER_ID ,@V_RANK := @V_RANK + 1 ,@V_RANK := 1
                            ) AS RANK ,@V_USER_ID := USER_ID AS G_USER_ID
                            FROM
                                (
                                    SELECT
                                        MESSAGE_ID,
                                        USER_ID,
                                        CREATE_TIME,
                                        MESSAGE_CONTEXT,
                                        GET_USER_ID,
                                        SEND_USER_ID
                                    FROM
                                        (
                                            SELECT
                                                MESSAGE_ID,
                                                GET_USER_ID,
                                                SEND_USER_ID,
                                                GET_USER_ID AS USER_ID,
                                                CREATE_TIME,
                                                MESSAGE_CONTEXT
                                            FROM
                                                T_SD_MESSAGE
                                            WHERE
                                                GET_USER_ID != '3000'
                                            UNION ALL
                                                SELECT
                                                    MESSAGE_ID,
                                                    GET_USER_ID,
                                                    SEND_USER_ID,
                                                    SEND_USER_ID AS USER_ID,
                                                    CREATE_TIME,
                                                    MESSAGE_CONTEXT
                                                FROM
                                                    T_SD_MESSAGE
                                                WHERE
                                                    SEND_USER_ID != '3000'
                                        ) METADATA
                                    ORDER BY
                                        USER_ID ASC,
                                        CREATE_TIME DESC
                                ) SRC,
                                (
                                    SELECT
                                        @V_RANK = 0,
                                        @V_USER_ID := NULL
                                ) VARS
                        ) SRC
                    WHERE
                        RANK = 1
                    ORDER BY
                        CREATE_TIME DESC
                ) SRC
        )
    ORDER BY
        M.CREATE_TIME DESC
    LIMIT FRIST_RESULT,
    FETCH_SIZE ;
END|| 
DELIMITER ;


-- LIMIT  FRIST_RESULT ,FETCH_SIZE
CALL PROCE_USER_NEW_MSG (0, 2);

 

 

 

 

转自:　　http://ace105.blog.51cto.com/639741/724411

 

查看更多关于【数据库其他】的文章

展开全文
相关推荐
反对 0
举报 0
评论 0
图文资讯
热门推荐
优选好物
更多热点专题
更多推荐文章
去重复的sql(Oracle) 去重复的英文
1.利用group by 去重复2.可以利用下面的sql去重复，如下　　1) select id,name,sex from (select a.*,row_number() over(partition by a.id,a.set order by name) su from test a ) where su=1　　2)select id,name,sex from (select a.*,row_number() over(p

0评论2023-02-10893

Oracle SQL七次提速技巧
以下SQL执行时间按序号递减。1,动态SQL，没有绑定变量，每次执行都做硬解析操作，占用较大的共享池空间，若共享池空间不足，会导致其他SQL语句的解析信息被挤出共享池。create or replace procedure proc1as beginfor i in 1..100000 loop    execute imme

0评论2023-02-10755

SQL ORACLE case when函数用法
case when 用法(1)简单case函数：格式：　　case 列名 　　when 条件值1 then 选项1　　when 条件值1 then 选项2......　　else 默认值 end例如：　　select 　　case job_level　　when '1' then '1111'　　when '2' then '2222'　　　when '3' then '3333

0评论2023-02-10564

mysql下如何执行sql脚本 执行SQL脚本
1.编写sql脚本,假设内容如下:  create database dearabao;  use dearabao;  create table niuzi (name varchar(20));  保存脚本文件,假设我把它保存在F盘的hello world目录下,于是该文件的路径为:F:\hello world\niuzi.sql2.执行sql脚本,可以有2种方法: 

0评论2023-02-10699

MySQL 5.7版本sql_mode=only_full_group_by问题
用到GROUP BY 语句查询时com.mysql.jdbc.exceptions.jdbc4.MySQLSyntaxErrorException: Expression #2 of SELECT list is not in GROUP BY clause and contains nonaggregated column 'col_user_6.a.START_TIME' which is not functionally dependent on colu

0评论2023-02-10973

Oracle迁移到MySQL性能下降的注意点 oracle数据库迁移需要注意的问题
背景：最近有较多的客户系统由原来由Oracle改造到MySQL后出现了性能问题CPU 100%，或是后台的CRM系统复杂SQL在业务高峰的时候出现堆积导致业务故障。在我的记忆里面淘宝最初从Oracle迁移到MySQL期间也遇到了很多SQL的性能问题，记忆最为深刻的子查询，当初的

0评论2023-02-10580

ORACLE中通过SQL语句(alter table)来增加、删除、修改字段
1.添加字段：alter table  表名  add (字段  字段类型)  [ default  '输入默认值']  [null/not null]  ;2.添加备注：comment on column  库名.表名.字段名 is  '输入的备注';  如： 我要在ers_data库中  test表 document_type字段添加备注  comm

0评论2023-02-10584

MySQL与Oracle 差异比较之六触发器
触发器编号类别ORACLEMYSQL注释1创建触发器语句不同create or replace trigger TG_ES_FAC_UNIT  before insert or update or delete on ES_FAC_UNIT  for each rowcreate trigger `hs_esbs`.`TG_INSERT_ES_FAC_UNIT` BEFORE INSERT on `hs_esbs`.`es_fac_u

0评论2023-02-10914

mysql where条件：某时间字段为今天的sql语句
1.查询：注册时间为今天的所有用户数：select count(*) from customer where TO_DAYS(createtime) = TO_DAYS(NOW())2.获取当前时间到凌晨24点还有多长时间：（Java中可用于判断某时间是否为今天）final Calendar cal = Calendar.getInstance();    ca

0评论2023-02-10717

mysql中的sql
变量用户变量: 在用户变量前加@系统变量: 在系统变量前加@@运算符算术运算符有: +(加), -(减), * (乘), / (除) 和％ (求模) 五中运算位运算符有:(位于), | (位或), ^ (位异或), ~ (位取反),(位右移),(位左移)比较运算符有: = (等于),(大于),(小于), = (大

0评论2023-02-10936

Oracle的HINT可以强制指定SQL的执行计划，比如选择索引、表的连接顺序以及表的连接方式等等。（转）
在Oracle中查看所有的表: select * from tab/dba_tables/dba_objects/cat; 看用户建立的表 :  select table_name from user_tables;  //当前用户的表 select table_name from all_tables;  //所有用户的表 select table_name from dba_tables;  //包

0评论2023-02-10857

Oracle sql 子字符串长度判断
Oracle sql 子字符串长度判断 select t.* from d_table t WHEREsubstr(t.col,1,1)='8' and instr(t.col,'/')0 and length(substr(t.col,1,instr(t.col,'/')))5; 字符串的前两位都是数字：select * from d_table t WHERE regexp_like(substr(t.col,1,2), '^[

0评论2023-02-10759

Oracle、MySql、Sql Server比对
MySql:廉价(部分免费)：当前，MySQL採用双重授权(DualLicensed)，他们是GPL和MySQLAB制定的商业许可协议。假设你在一个遵循GPL的***（开源）项目中使用MySQL，那么你能够遵循GPL协议免费使用MySQL。否则，你须要购买MySQLAB制定的那个商业许可协议。Windows $

0评论2023-02-10441

Oracle 存储过程,临时表,动态SQL测试
--创建事务级别的结果临时表create global temporary table tmp_yshy( c1 varchar2(100), c2 varchar2(100))on commit delete rows;--创建事务级别的存储sql语句的临时表create global temporary table tmp_sql( c1 varchar2(4000))on commit delete rows;测

0评论2023-02-10508

更多推荐