8种常见的SQL错误用法
<h2 id="2">常见SQL错误用法</h2><h3 id="3">1. LIMIT 语句</h3>
<p>分页查询是最常用的场景之一，但也通常也是最容易出问题的地方。比如对于下面简朴的语句，一样寻常DBA想到的办法是在type, name, create_time字段上加组合索引。如许条件排序都能有效的利用到索引，性能迅速提升。</p>
<code >SELECT *
FROM operation
WHEREtype = 'SQLStats'
AND name = 'SlowLog'
ORDERBY create_time
LIMIT1000, 10;
</code>
<p>好吧，大概90%以上的DBA解决该问题就到此为止。但当 LIMIT 子句变成 “LIMIT 1000000,10” 时，程序员仍旧会诉苦：我只取10条记载为什么照旧慢？</p>
<p>要知道数据库也并不知道第1000000条记载从什么地方开始，纵然有索引也需要重新盘算一次。出现这种性能问题，多数情形下是程序员偷懒了。在前端数据浏览翻页，大概大数据分批导出等场景下，是可以将上一页的最大值当成参数作为查询条件的。SQL重新设计如下：</p>
<code >SELECT *
FROM operation
WHERE type = 'SQLStats'
AND name = 'SlowLog'
AND create_time > '2019-10-19 14:00:00'
ORDER BY create_time limit 10;
</code>
<p>在新设计下查询时间基本固定，不会随着数据量的增长而发生厘革。</p>
<h3 id="4">2. 隐式转换</h3>
<p>SQL语句中查询变量和字段界说类型不匹配是另一个常见的错误。比如下面的语句：</p>
<code >mysql> explain extended SELECT *
> FROM my_balance b
> WHEREb.bpn = 14000000123
> AND b.isverified IS NULL ;
mysql> show warnings;
| Warning | 1739 | Cannot use ref access on index 'bpn' due to type or collation conversion on field 'bpn'
</code>
<p>其中字段bpn的界说为varchar(20)，MySQL的策略是将字符串转换为数字之后再比较。函数作用于表字段，索引失效。</p>
<p>上述情况大概是应用程序框架主动填入的参数，而不是程序员的原意。现在应用框架很多很繁杂，利用方便的同时也警惕它大概给自己挖坑。</p>
<h3 id="5">3. 关联更新、删除</h3>
<p>虽然MySQL5.6引入了物化特性，但需要特别注意它现在仅仅针对查询语句的优化。对于更新或删除需要手工重写成JOIN。</p>
<p>比如下面UPDATE语句，MySQL实际实验的是循环/嵌套子查询（DEPENDENT SUBQUERY)，其实验时间可想而知。</p>
<code >UPDATE operation o
SET status = 'applying'
WHEREo.id IN (SELECT id
FROM (SELECT o.id,
o.status
FROM operation o
WHEREo.group = 123
AND o.status NOT IN ( 'done' )
ORDERBY o.parent,
o.id
LIMIT1) t);
</code>
<p>实验计划：</p>
<code >+----+--------------------+-------+-------+---------------+---------+---------+-------+------+-----------------------------------------------------+
| id | select_type | table | type| possible_keys | key | key_len | ref | rows | Extra |
+----+--------------------+-------+-------+---------------+---------+---------+-------+------+-----------------------------------------------------+
| 1| PRIMARY | o | index | | PRIMARY | 8 | | 24 | Using where; Using temporary |
| 2| DEPENDENT SUBQUERY | | | | | | | | Impossible WHERE noticed after reading const tables |
| 3| DERIVED | o | ref | idx_2,idx_5 | idx_5 | 8 | const | 1 | Using where; Using filesort |
+----+--------------------+-------+-------+---------------+---------+---------+-------+------+-----------------------------------------------------+
</code>
<p>重写为JOIN之后，子查询的选择模式从DEPENDENT SUBQUERY变成DERIVED,实验速率大大加速，从7秒低落到2毫秒。</p>
<code >UPDATE operation o
JOIN(SELECT o.id,
o.status
FROM operation o
WHEREo.group = 123
AND o.status NOT IN ( 'done' )
ORDERBY o.parent,
o.id
LIMIT1) t
ON o.id = t.id
SET status = 'applying'
</code>
<p>实验计划简化为：</p>
<code >+----+-------------+-------+------+---------------+-------+---------+-------+------+-----------------------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+------+---------------+-------+---------+-------+------+-----------------------------------------------------+
| 1| PRIMARY | | | | | | | | Impossible WHERE noticed after reading const tables |
| 2| DERIVED | o | ref| idx_2,idx_5 | idx_5 | 8 | const | 1 | Using where; Using filesort |
+----+-------------+-------+------+---------------+-------+---------+-------+------+-----------------------------------------------------+
</code>
<h3 id="6">4. 混淆排序</h3>
<p>MySQL不能利用索引进行混淆排序。但在某些场景，照旧有时机利用特别方法提升性能的。</p>
<code >SELECT *
FROM my_order o
INNER JOIN my_appraise a ON a.orderid = o.id
ORDERBY a.is_reply ASC,
a.appraise_time DESC
LIMIT0, 20
</code>
<p>实验计划显示为全表扫描：</p>
<code >+----+-------------+-------+--------+-------------+---------+---------+---------------+---------+-+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra
+----+-------------+-------+--------+-------------+---------+---------+---------------+---------+-+
|1 | SIMPLE | a | ALL | idx_orderid | NULL | NULL | NULL | 1967647 | Using filesort |
|1 | SIMPLE | o | eq_ref | PRIMARY | PRIMARY | 122 | a.orderid | 1 | NULL |
+----+-------------+-------+--------+---------+---------+---------+-----------------+---------+-+
</code>
<p>由于is_reply只有0和1两种状态，我们按照下面的方法重写后，实验时间从1.58秒低落到2毫秒。</p>
<code >SELECT *
FROM ((SELECT *
FROM my_order o
INNER JOIN my_appraise a
ON a.orderid = o.id
AND is_reply = 0
ORDERBY appraise_time DESC
LIMIT0, 20)
UNION ALL
(SELECT *
FROM my_order o
INNER JOIN my_appraise a
ON a.orderid = o.id
AND is_reply = 1
ORDERBY appraise_time DESC
LIMIT0, 20)) t
ORDERBYis_reply ASC,
appraisetime DESC
LIMIT20;
</code>
<h3 id="7">5. EXISTS语句</h3>
<p>MySQL对待EXISTS子句时，仍旧接纳嵌套子查询的实验方式。如下面的SQL语句：</p>
<code >SELECT *
FROM my_neighbor n
LEFT JOIN my_neighbor_apply sra
ON n.id = sra.neighbor_id
AND sra.user_id = 'xxx'
WHEREn.topic_status < 4
AND EXISTS(SELECT 1
FROM message_info m
WHEREn.id = m.neighbor_id
AND m.inuser = 'xxx')
AND n.topic_type <> 5
</code>
<p>实验计划为：</p>
<code >+----+--------------------+-------+------+-----+------------------------------------------+---------+-------+---------+ -----+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+--------------------+-------+------+ -----+------------------------------------------+---------+-------+---------+ -----+
|1 | PRIMARY | n | ALL|| NULL | NULL | NULL| 1086041 | Using where |
|1 | PRIMARY | sra | ref|| idx_user_id | 123 | const | 1 | Using where |
|2 | DEPENDENT SUBQUERY | m | ref|| idx_message_info | 122 | const | 1 | Using index condition; Using where |
+----+--------------------+-------+------+ -----+------------------------------------------+---------+-------+---------+ -----+
</code>
<p>去掉exists更改为join，能够制止嵌套子查询，将实验时间从1.93秒低落为1毫秒。</p>
<code >SELECT *
FROM my_neighbor n
INNER JOIN message_info m
ON n.id = m.neighbor_id
AND m.inuser = 'xxx'
LEFT JOIN my_neighbor_apply sra
ON n.id = sra.neighbor_id
AND sra.user_id = 'xxx'
WHEREn.topic_status < 4
AND n.topic_type <> 5
</code>
<p>新的实验计划：</p>
<code >+----+-------------+-------+--------+ -----+------------------------------------------+---------+ -----+------+ -----+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+--------+ -----+------------------------------------------+---------+ -----+------+ -----+
|1 | SIMPLE | m | ref | | idx_message_info | 122 | const | 1 | Using index condition |
|1 | SIMPLE | n | eq_ref | | PRIMARY | 122 | ighbor_id | 1 | Using where |
|1 | SIMPLE | sra | ref | | idx_user_id | 123 | const | 1 | Using where |
+----+-------------+-------+--------+ -----+------------------------------------------+---------+ -----+------+ -----+
</code>
<h3 id="8">6. 条件下推</h3>
<p>外部查询条件不能够下推到复杂的视图或子查询的情况有：</p>
<ol>
<li>聚合子查询；</li>
<li>含有LIMIT的子查询；</li>
<li>UNION 或UNION ALL子查询；</li>
<li>输出字段中的子查询；</li>
</ol>
<p>如下面的语句，从实验计划可以看出其条件作用于聚合子查询之后：</p>
<code >SELECT *
FROM (SELECT target,
Count(*)
FROM operation
GROUPBY target) t
WHEREtarget = 'rm-xxxx'
</code>
<code >+----+-------------+------------+-------+---------------+-------------+---------+-------+------+-------------+
| id | select_type | table | type| possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+-------+---------------+-------------+---------+-------+------+-------------+
|1 | PRIMARY | <derived2> | ref | <auto_key0> | <auto_key0> | 514 | const | 2 | Using where |
|2 | DERIVED | operation| index | idx_4 | idx_4 | 519 | NULL| 20 | Using index |
+----+-------------+------------+-------+---------------+-------------+---------+-------+------+-------------+
</code>
<p>确定从语义上查询条件可以直接下推后，重写如下：</p>
<code >SELECT target,
Count(*)
FROM operation
WHEREtarget = 'rm-xxxx'
GROUPBY target
</code>
<p>实验计划变为：</p>
<code >+----+-------------+-----------+------+---------------+-------+---------+-------+------+--------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-----------+------+---------------+-------+---------+-------+------+--------------------+
| 1 | SIMPLE | operation | ref | idx_4 | idx_4 | 514 | const | 1 | Using where; Using index |
+----+-------------+-----------+------+---------------+-------+---------+-------+------+--------------------+
</code>
<p>关于MySQL外部条件不能下推的详细表明说明请参考从前文章：MySQL · 性能优化 · 条件下推到物化表</p>
<h3 id="9">7. 提前缩小范围</h3>
<p>先上初始SQL语句：</p>
<code >SELECT *
FROM my_order o
LEFT JOIN my_userinfo u
ON o.uid = u.uid
LEFT JOIN my_productinfo p
ON o.pid = p.pid
WHERE( o.display = 0 )
AND ( o.ostaus = 1 )
ORDERBY o.selltime DESC
LIMIT0, 15
</code>
<p>该SQL语句原意是：先做一系列的左连接，然后排序取前15条记载。从实验计划也可以看出，最后一步估算排序记载数为90万，时间斲丧为12秒。</p>
<code >+----+-------------+-------+--------+---------------+---------+---------+-----------------+--------+----------------------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+--------+---------------+---------+---------+-----------------+--------+----------------------------------------------------+
|1 | SIMPLE | o | ALL | NULL | NULL | NULL | NULL | 909119 | Using where; Using temporary; Using filesort |
|1 | SIMPLE | u | eq_ref | PRIMARY | PRIMARY | 4 | o.uid | 1 | NULL |
|1 | SIMPLE | p | ALL | PRIMARY | NULL | NULL | NULL | 6 | Using where; Using join buffer (Block Nested Loop) |
+----+-------------+-------+--------+---------------+---------+---------+-----------------+--------+----------------------------------------------------+
</code>
<p>由于最后WHERE条件以及排序均针对最左主表，因此可以先对my_order排序提前缩小数据量再做左连接。SQL重写后如下，实验时间缩小为1毫秒左右。</p>
<code >SELECT *
FROM (
SELECT *
FROM my_order o
WHERE( o.display = 0 )
AND ( o.ostaus = 1 )
ORDERBY o.selltime DESC
LIMIT0, 15
) o
LEFT JOIN my_userinfo u
ON o.uid = u.uid
LEFT JOIN my_productinfo p
ON o.pid = p.pid
ORDER BYo.selltime DESC
limit 0, 15
</code>
<p>再查抄实验计划：子查询物化后（select_type=DERIVED)参与JOIN。虽然估算行扫描仍旧为90万，但是利用了索引以及LIMIT 子句后，实际实验时间变得很小。</p>
<code >
+----+-------------+------------+--------+---------------+---------+---------+-------+--------+----------------------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+--------+---------------+---------+---------+-------+--------+----------------------------------------------------+
|1 | PRIMARY | <derived2> | ALL | NULL | NULL | NULL | NULL| 15 | Using temporary; Using filesort |
|1 | PRIMARY | u | eq_ref | PRIMARY | PRIMARY | 4 | o.uid | 1 | NULL |
|1 | PRIMARY | p | ALL | PRIMARY | NULL | NULL | NULL| 6 | Using where; Using join buffer (Block Nested Loop) |
|2 | DERIVED | o | index| NULL | idx_1 | 5 | NULL| 909112 | Using where |
+----+-------------+------------+--------+---------------+---------+---------+-------+--------+----------------------------------------------------+
</code>
<h3 id="10">8. 中心结果集下推</h3>
<p>再来看下面这个已经初步优化过的例子(左连接中的主表优先作用查询条件)：</p>
<code >SELECT a.*,
c.allocated
FROM (
SELECT resourceid
FROM my_distribute d
WHERE isdelete = 0
AND cusmanagercode = '1234567'
ORDER BY salecode limit 20) a
LEFT JOIN
(
SELECT resourcesid， sum(ifnull(allocation, 0) * 12345) allocated
FROM my_resources
GROUP BY resourcesid) c
ON a.resourceid = c.resourcesid
</code>
<p>那么该语句还存在别的问题吗？不丢脸出子查询 c 是全表聚合查询，在表数量特别大的情况下会导致整个语句的性能降落。</p>
<p>着实对于子查询 c，左连接最后结果集只关心能和主表resourceid能匹配的数据。因此我们可以重写语句如下，实验时间从原来的2秒降落到2毫秒。</p>
<code >SELECT a.*,
c.allocated
FROM (
SELECT resourceid
FROM my_distribute d
WHERE isdelete = 0
AND cusmanagercode = '1234567'
ORDER BY salecode limit 20) a
LEFT JOIN
(
SELECT resourcesid， sum(ifnull(allocation, 0) * 12345) allocated
FROM my_resources r,
(
SELECT resourceid
FROM my_distribute d
WHERE isdelete = 0
AND cusmanagercode = '1234567'
ORDER BY salecode limit 20) a
WHERE r.resourcesid = a.resourcesid
GROUP BY resourcesid) c
ON a.resourceid = c.resourcesid
</code>
<p>但是子查询 a 在我们的SQL语句中出现了多次。这种写法不光存在额外的开销，还使得整个语句显的繁杂。利用WITH语句再次重写：</p>
<code >WITH a AS
(
SELECT resourceid
FROM my_distribute d
WHERE isdelete = 0
AND cusmanagercode = '1234567'
ORDER BY salecode limit 20)
SELECT a.*,
c.allocated
FROM a
LEFT JOIN
(
SELECT resourcesid， sum(ifnull(allocation, 0) * 12345) allocated
FROM my_resources r,
a
WHERE r.resourcesid = a.resourcesid
GROUP BY resourcesid) c
ON a.resourceid = c.resourcesid
</code>
<h2 id="11">总结</h2>
<ol>
<li>数据库编译器产生实验计划，决定着SQL的实际实验方式。但是编译器只是尽力服务，全部数据库的编译器都不是精致绝伦的。上述提到的多数场景，在别的数据库中也存在性能问题。相识数据库编译器的特性，才能避规其短处，写出高性能的SQL语句。</li>
<li>程序员在设计数据模子以及编写SQL语句时，要把算法的头脑或意识带进来。</li>
<li>编写复杂SQL语句要养成利用WITH语句的习惯。简便且思绪清晰的SQL语句也能减小数据库的负担 ^^。</li>
</ol>
<p><br><br />
<hr><br />
<font color="Darkorange" size="4px"><b>关注微信公众号【程序员的梦想】，专注于Java，SpringBoot，SpringCloud，微服务，Docker以及前后端分离等全栈技术。</b></font></p>
<p><div align="center"></div></p><br><br/><br/><br/><br/><br/>来源：<a href="https://www.cnblogs.com/cxydmx/p/11728419.html" target="_blank">https://www.cnblogs.com/cxydmx/p/11728419.html</a>
页:
[1]