Question

CREATE TABLE Table1
    (`id` int, `group_id` int, `data_id` int, `value` varchar(19))
;

INSERT INTO Table1
    (`id`, `group_id`, `data_id`, `value`)
VALUES
    (1, 20, 15, 'Supplier'),
    (2, 20, 1, 'Rahul'),
    (3, 20, 2, 'Sharma'),
    (4, 20, 3, '05/08/90'),
    (5, 20, 4, 'india'),
    (6, 21, 15, 'Consumer'),
    (7, 21, 1, 'Rajesh'),
    (8, 21, 2, 'Sharma'),
    (9, 21, 3, '05/08/88'),
    (10, 21, 4, 'india')
;

现在创建一个表我必须按供应商或消费者对值列中的数据进行排序，如下所示：消费者：

f-name l-name    dob     location
-----------------------------------
Rajesh Sharma 05/08/88   india
and so on ....

供应商

f-name l-name    dob     location
-----------------------------------
Rahul Sharma 05/08/90   india

Answer 1

首先，您需要从您拥有的entity-attribute-value存储创建记录。您可以通过各种方式执行此操作。以下是使用group by的方法：

select max(case when data_id = 1 then value end) as fname,
       max(case when data_id = 2 then value end) as lname,
       max(case when data_id = 3 then value end) as dob,
       max(case when data_id = 4 then value end) as location 
from table1
group by group_id
having max(case when data_id = 15 then value end) = 'Consumer';

having子句执行过滤，但您也可以添加指定记录类型的列。

Answer 2

您可以像这样创建简单的函数：

 CREATE FUNCTION GET_DATA(P_TYPE INT ,P_GROUP_ID INT) RETURNS VARCHAR(19) DETERMINISTIC
 BEGIN
    DECLARE var varchar(19);
    select value into var  from Table1 WHERE group_id = P_GROUP_ID and data_id = P_TYPE;
 return var;
 END//



 -- you can use like this
 select GET_DATA(1,20) FIRST_NAME,
 GET_DATA(2,20) LAST_NAME,
 GET_DATA(3,20) DOB,
 GET_DATA(4,20) LOCATION;


 -- or you can use like this

select 
GET_DATA(15,a.group_id) Type,
GET_DATA(1,a.group_id) FirstName,
GET_DATA(2,a.group_id) LastName,
GET_DATA(3,a.group_id) Dob,
GET_DATA(4,a.group_id) Location
from 
 (select distinct group_id from Table1) as a;

小提琴示例http://sqlfiddle.com/#!8/e9647/7

如何在my-sql中完成这项工作？

2 个答案: