MySql-如何查询删除数据表重复记录

查询 SELECT name, email, COUNT (*)FROM usersGROUP BY name, emailHAVING COUNT(*) 1 重点来了，查询容易，那应该如何删除重复记录呢？ 演示数据 表结构： mysql desc demo;+-------+------------------+------+-----+---------+----------------+ | Field

查询

<code class=" hljs oxygene"><span class="hljs-keyword">SELECT</span>
    name, email, COUNT<span class="hljs-comment">(*)
FROM
    users
GROUP BY
    name, email
HAVING 
    COUNT(*)</span> > <span class="hljs-number">1</span></code>

重点来了，查询容易，那应该如何删除重复记录呢？

演示数据
表结构：

<code class=" hljs asciidoc"><span class="hljs-header">mysql> desc demo;
+-------+------------------+------+-----+---------+----------------+</span>
<span class="hljs-header">| Field | Type             | Null | Key | Default | Extra          |
+-------+------------------+------+-----+---------+----------------+</span>
| id    | int(11) unsigned | NO   | PRI | NULL    | auto<span class="hljs-emphasis">_increment |
| site  | varchar(100)     | NO   | MUL |         |                |
+-------+------------------+------+-----+---------+----------------+
2 rows in set (0.00 sec)</span></code>

数据：

<code class=" hljs asciidoc"><span class="hljs-header">mysql> select * from demo order by id;
+----+------------------------+</span>
<span class="hljs-header">| id | site                   |
+----+------------------------+</span>
|  1 | http://www.CodeBit.cn  |
|  2 | http://YITU.org        |
|  3 | http://www.ShuoWen.org |
|  4 | http://www.CodeBit.cn  |
<span class="hljs-header">|  5 | http://www.ShuoWen.org |
+----+------------------------+</span>
5 rows in set (0.00 sec)</code>

当没有创建表或创建索引权限的时候，可以用下面的方法：

如果你要删除较旧的重复记录，可以使用下面的语句：

<code class=" hljs haml">mysql> delete from a 
    -<span class="ruby">> using demo as a, demo as b
</span>    -<span class="ruby">> where (a.id > b.id)
</span>    -<span class="ruby">> <span class="hljs-keyword">and</span> (a.site = b.site);
</span>Query OK, 2 rows affected (0.12 sec)</code>

<code class=" hljs asciidoc"><span class="hljs-header">mysql> select * from demo order by id;
+----+------------------------+</span>
<span class="hljs-header">| id | site                   |
+----+------------------------+</span>
|  1 | http://www.CodeBit.cn  |
|  2 | http://YITU.org        |
<span class="hljs-header">|  3 | http://www.ShuoWen.org |
+----+------------------------+</span>
3 rows in set (0.00 sec)</code>

如果你要删除较新的重复记录，可以使用下面的语句：

<code class=" hljs haml">mysql> delete from a 
    -<span class="ruby">> using demo as a, demo as b
</span>    -<span class="ruby">> where (a.id < b.id)
</span>    -<span class="ruby">> <span class="hljs-keyword">and</span> (a.site = b.site);
</span>Query OK, 2 rows affected (0.12 sec)</code>

<code class=" hljs asciidoc"><span class="hljs-header">mysql> select * from demo order by id;
+----+------------------------+</span>
<span class="hljs-header">| id | site                   |
+----+------------------------+</span>
|  2 | http://YITU.org        |
|  4 | http://www.CodeBit.cn  |
<span class="hljs-header">|  5 | http://www.ShuoWen.org |
+----+------------------------+</span>
3 rows in set (0.00 sec)</code>

你可以用下面的语句先确认将被删除的重复记录：

<code class=" hljs asciidoc">mysql> SELECT a.* 
<span class="hljs-code">    -> FROM demo a, demo b</span>
<span class="hljs-code">    -> WHERE a.id > b.id</span>
<span class="hljs-header">    -> AND (a.site = b.site);
+----+------------------------+</span>
<span class="hljs-header">| id | site                   |
+----+------------------------+</span>
|  1 | http://www.CodeBit.cn  |
<span class="hljs-header">|  3 | http://www.ShuoWen.org |
+----+------------------------+</span>
2 rows in set (0.00 sec)</code>

如果有创建索引的权限，可以用下面的方法：

在表上创建唯一键索引：

<code class=" hljs avrasm">mysql> alter ignore table demo <span class="hljs-keyword">add</span> unique index ukey (site)<span class="hljs-comment">;</span>
Query OK, <span class="hljs-number">5</span> rows affected (<span class="hljs-number">0.46</span> <span class="hljs-keyword">sec</span>)
<span class="hljs-label">Records:</span> <span class="hljs-number">5</span>  Duplicates: <span class="hljs-number">2</span>  Warnings: <span class="hljs-number">0</span></code>

<code class=" hljs asciidoc"><span class="hljs-header">mysql> select * from demo order by id;
+----+------------------------+</span>
<span class="hljs-header">| id | site                   |
+----+------------------------+</span>
|  1 | http://www.CodeBit.cn  |
|  2 | http://YITU.org        |
<span class="hljs-header">|  3 | http://www.ShuoWen.org |
+----+------------------------+</span>
3 rows in set (0.00 sec)</code>

重复记录被删除后，如果需要，可以删除索引：

<code class=" hljs avrasm">mysql> alter table demo drop index ukey<span class="hljs-comment">;</span>
Query OK, <span class="hljs-number">3</span> rows affected (<span class="hljs-number">0.37</span> <span class="hljs-keyword">sec</span>)
<span class="hljs-label">Records:</span> <span class="hljs-number">3</span>  Duplicates: <span class="hljs-number">0</span>  Warnings: <span class="hljs-number">0</span></code>

如果有创建表的权限，可以用下面的方法：

创建一个新表，然后将原表中不重复的数据插入新表：

<code class=" hljs oxygene">mysql> <span class="hljs-keyword">create</span> table demo_new <span class="hljs-keyword">as</span> <span class="hljs-keyword">select</span> * <span class="hljs-keyword">from</span> demo <span class="hljs-keyword">group</span> <span class="hljs-keyword">by</span> site;
Query OK, <span class="hljs-number">3</span> rows affected (<span class="hljs-number">0.19</span> sec)
Records: <span class="hljs-number">3</span>  Duplicates: <span class="hljs-number">0</span>  Warnings: <span class="hljs-number">0</span></code>

<code class=" hljs asciidoc"><span class="hljs-header">mysql> show tables;
+----------------+</span>
<span class="hljs-header">| Tables_in_test |
+----------------+</span>
| demo           |
<span class="hljs-header">| demo_new       |
+----------------+</span>
2 rows in set (0.00 sec)</code>

<code class=" hljs asciidoc"><span class="hljs-header">mysql> select * from demo order by id;
+----+------------------------+</span>
<span class="hljs-header">| id | site                   |
+----+------------------------+</span>
|  1 | http://www.CodeBit.cn  |
|  2 | http://YITU.org        |
|  3 | http://www.ShuoWen.org |
|  4 | http://www.CodeBit.cn  |
<span class="hljs-header">|  5 | http://www.ShuoWen.org |
+----+------------------------+</span>
5 rows in set (0.00 sec)</code>

<code class=" hljs asciidoc"><span class="hljs-header">mysql> select * from demo_new order by id;
+----+------------------------+</span>
<span class="hljs-header">| id | site                   |
+----+------------------------+</span>
|  1 | http://www.CodeBit.cn  |
|  2 | http://YITU.org        |
<span class="hljs-header">|  3 | http://www.ShuoWen.org |
+----+------------------------+</span>
3 rows in set (0.00 sec)</code>

然后将原表备份，将新表重命名为当前表：

<code class=" hljs asciidoc">mysql> rename table demo to demo<span class="hljs-emphasis">_old, demo_</span>new to demo;
Query OK, 0 rows affected (0.04 sec)

<span class="hljs-header">mysql> show tables;
+----------------+</span>
<span class="hljs-header">| Tables_in_test |
+----------------+</span>
| demo           |
<span class="hljs-header">| demo_old       |
+----------------+</span>
2 rows in set (0.00 sec)</code>

<code class=" hljs asciidoc"><span class="hljs-header">mysql> select * from demo order by id;
+----+------------------------+</span>
<span class="hljs-header">| id | site                   |
+----+------------------------+</span>
|  1 | http://www.CodeBit.cn  |
|  2 | http://YITU.org        |
<span class="hljs-header">|  3 | http://www.ShuoWen.org |
+----+------------------------+</span>
3 rows in set (0.00 sec)</code>

注意：使用这种方式创建的表会丢失原表的索引信息！

<code class=" hljs asciidoc"><span class="hljs-header">mysql> desc demo;
+-------+------------------+------+-----+---------+-------+</span>
<span class="hljs-header">| Field | Type             | Null | Key | Default | Extra |
+-------+------------------+------+-----+---------+-------+</span>
| id    | int(11) unsigned | NO   |     | 0       |       |
<span class="hljs-header">| site  | varchar(100)     | NO   |     |         |       |
+-------+------------------+------+-----+---------+-------+</span>
2 rows in set (0.00 sec)</code>

如果要保持和原表信息一致，你可以使用 show create table demo; 来查看原表的创建语句，然后使用原表的创建语句创建新表，接着使用 insert … select 语句插入数据，再重命名表即可。