MYSQL - 合并具有多个重复值的行,然后删除重复项

MYSQL - Combine rows with multiple duplicate values and delete duplicates afterwards

所以我将我的数据库设置为单个 table。在那个 table 中,我收集了来源 URL 和描述(我从许多页面中抓取了产品描述)。不幸的是,如果有一个以上的段落,我最终在数据库中有多个行用于 URL/source 页面。

我想做的是,如果有多行具有相同的 URL,合并每行的描述,然后删除该 URL 的重复行。

我的 table 字面结构如下:

table             
+----+----------------------------+-------------+
| id | url                        | description |
+----+----------------------------+-------------+
|  1 | http://example.com/page-a  | paragraph 1 |
|  2 | http://example.com/page-a  | paragraph 2 |
|  3 | http://example.com/page-a  | paragraph 3 |
|  4 | http://example.com/page-b  | paragraph 1 |
|  5 | http://example.com/page-b  | paragraph 2 |
+----+----------------------------+-------------+

我想要的是这样的:

table             
+----+----------------------------+-------------------------------------+
| id | url                        | description                         |
+----+----------------------------+-------------------------------------+
|  1 | http://example.com/page-a  | paragraph 1 paragraph 2 paragraph 3 |
|  2 | http://example.com/page-b  | paragraph 1 paragraph 2             |
+----+----------------------------+-------------------------------------+

我不太在意更新 ID 是否正确,我只是希望能够合并段落应该在同一字段中的行,因为它们是相同的 URLs , 然后删除重复项。

如有任何帮助,我们将不胜感激!

创建一个新的临时文件table,截断原来的,然后重新插入数据:

create temporary table tempt as
    select (@rn := @rn + 1) as id, url,
           group_concat(description order by id separator ' ') as description
    from t cross join (select @rn := 0) params
    group by url 
    order by min(id);

-- Do lots of testing and checking here to be sure you have the data you want.

truncate table t;

insert into t(id, url, description)
    select id, url, description
    from tempt;

如果 id 已在 table 中自动递增,则您无需为其提供值。

过滤 table 很容易,只需将结果插入新的 table:

SELECT url, GROUP_CONCAT(description ORDER BY description SEPARATOR ' ') AS description
FROM `table`
GROUP BY url

SQL

SELECT MIN(id) as [ID],url, description= STUFF((SELECT '; ' 
+ ic.description FROM dbo.My_Table AS ic
WHERE ic.url= c.url
FOR XML PATH(''), TYPE).value('.','nvarchar(max)'), 1, 2, '')
FROM dbo.My_Table AS c
GROUP BY url
ORDER BY url;