MYSQL查询 - 尝试从外部MySQL数据库构建HTML表

时间:2014-11-28 12:57:53

标签: mysql database syntax

我是PHP和MySQL连接的初学者,但我真的很愿意解决这个问题,我已经花了一些时间在它上面。我对我制作的数据库没有任何问题,但是这个数据库是来自我的一个客户端的外部数据库,他想从DB queires构建一个HTML表。数据库中的数据完全混合并添加到同一列中,这就是为什么理解SQL查询语法对我来说很复杂。

在数据库中,我有表中的所有数据:trr_postmeta

meta_id post_id meta_key         meta_value
3515    1479    company_sidlo    Praha
3516    1479    company_ico      02925885
3517    1479    company_date     04/23/2014
3518    1479    company_price    29400
3519    1479    company_url      https://or.justice.cz/ias/ui/rejstrik-firma.vysled...
3520    1479    company_state    zamluvená
3530    1481    _edit_last       7
3531    1481    _edit_lock       1413898849:7
3532    1481    company_typ      s.r.o.
3533    1481    company_sidlo    Praha
3534    1481    company_ico      02925605
3535    1481    company_date     04/23/2014
3536    1481    company_price    26900

对于我的HTML输出,我需要这样的东西:

--------------------
ID | ICO      | Typ
--------------------
 1 | 64654664 | sro
--------------------

我的SQL查询如下:

SELECT meta_id, meta_value FROM trr_postmeta WHERE meta_key =  "company_typ" AND meta_key = "company_ico" AND meta_key = "company_date" AND meta_key = "company_price" AND meta_key = "company_url" AND meta_key = "company_state";

但我认为我的语法WHERE仅适用于第一个条件。你能帮我解决我做错的事吗?我认为99%是SQL查询语法,但我无法弄清楚这一点。我需要在一列中获得company_typ和company_ico等的值。

由于

2 个答案:

答案 0 :(得分:0)

如果我理解你想要将行值作为列标题,我创建了一个小例子,你可以根据你的要求修改它

mysql> create table trr_postmeta (meta_key varchar(100),meta_value varchar(100));
Query OK, 0 rows affected (0.11 sec)

mysql> insert into trr_postmeta values ('company_typ','ABC'),('company_ico','pic.jpg'),('company_date','2014-11-28'),('company_state','LA');
Query OK, 4 rows affected (0.05 sec)
Records: 4  Duplicates: 0  Warnings: 0

mysql> select * from trr_postmeta;
+---------------+------------+
| meta_key      | meta_value |
+---------------+------------+
| company_typ   | ABC        |
| company_ico   | pic.jpg    |
| company_date  | 2014-11-28 |
| company_state | LA         |
+---------------+------------+
4 rows in set (0.00 sec)

所以查询看起来像

select 
group_concat(if(meta_key = 'company_typ',meta_value,null)) as 'Company Type', 
group_concat(if(meta_key = 'company_ico',meta_value,null)) as 'Company Icon',
group_concat(if(meta_key = 'company_state',meta_value,null)) as 'Company State' 
from trr_postmeta ;

输出看起来像

+--------------+--------------+---------------+
| Company Type | Company Icon | Company State |
+--------------+--------------+---------------+
| ABC          | pic.jpg      | LA            |
+--------------+--------------+---------------+

答案 1 :(得分:0)

一定是一年中的时间......

 DROP TABLE IF EXISTS eav_hell;

 CREATE TABLE eav_hell
 (meta_id INT NOT NULL AUTO_INCREMENT PRIMARY KEY
 ,post_id INT NOT NULL
 ,meta_key  VARCHAR(20) NOT NULL
 ,meta_value VARCHAR(100) NOT NULL
 ,UNIQUE(post_id,meta_key)
 );

 INSERT INTO eav_hell VALUES
 (3515    ,1479    ,'company_sidlo','Praha'),
 (3516    ,1479    ,'company_ico','02925885'),
 (3517    ,1479    ,'company_date','04/23/2014'),
 (3518    ,1479    ,'company_price','29400'),
 (3519    ,1479    ,'company_url','https://or.justice.cz/ias/ui/rejstrik-firma.vysled...'),
 (3520    ,1479    ,'company_state','zamluvená'),
 (3530    ,1481    ,'_edit_last','7'),
 (3531    ,1481    ,'_edit_lock','1413898849:7'),
 (3532    ,1481    ,'company_typ','s.r.o.'),
 (3533    ,1481    ,'company_sidlo','Praha'),
 (3534    ,1481    ,'company_ico','02925605'),
 (3535    ,1481    ,'company_date','04/23/2014'),
 (3536    ,1481    ,'company_price','26900');

 SELECT post_id
      , MAX(CASE WHEN meta_key = 'company_typ' THEN meta_value END) company_typ  
      , MAX(CASE WHEN meta_key = 'company_sidlo' THEN meta_value END) company_sidlo 
      , MAX(CASE WHEN meta_key = 'company_ico' THEN meta_value END) company_ico  
      , MAX(CASE WHEN meta_key = 'company_date' THEN meta_value END) company_date 
      , MAX(CASE WHEN meta_key = 'company_price' THEN meta_value END) company_price
      , MAX(CASE WHEN meta_key = 'company_url' THEN meta_value END) company_url  
      , MAX(CASE WHEN meta_key = 'company_state' THEN meta_value END) company_state
      , MAX(CASE WHEN meta_key = '_edit_last' THEN meta_value END) _edit_last   
      , MAX(CASE WHEN meta_key = '_edit_lock' THEN meta_value END) _edit_lock   
   FROM eav_hell
  GROUP  
     BY post_id;

 +---------+-------------+---------------+-------------+--------------+---------------+-------------------------------------------------------+---------------+------------+--------------+
 | post_id | company_typ | company_sidlo | company_ico | company_date | company_price | company_url                                           | company_state | _edit_last | _edit_lock   |
 +---------+-------------+---------------+-------------+--------------+---------------+-------------------------------------------------------+---------------+------------+--------------+
 |    1479 | NULL        | Praha         | 02925885    | 04/23/2014   | 29400         | https://or.justice.cz/ias/ui/rejstrik-firma.vysled... | zamluvená     | NULL       | NULL         |
 |    1481 | s.r.o.      | Praha         | 02925605    | 04/23/2014   | 26900         | NULL                                                  | NULL          | 7          | 1413898849:7 |
 +---------+-------------+---------------+-------------+--------------+---------------+-------------------------------------------------------+---------------+------------+--------------+
顺便说一句,如果使用像这样的EAV模型,那么根据数据类型将值分成不同的表(或至少列)仍然是一个好主意 - 所以日期在日期表中,数字在数字表,字符串到字符串表。