MySQL開發技巧
參考：
1、 MySQL開發技巧
2、 MySQL開發技巧2
常見的SQL語句型別
DDL:資料定義語言
TPL:事務處理語言
DCL:資料控制語言
DML:資料操作語言（CRUD）
如何正確的使用Join從句
| 內連線 | INNER |
| 全外連線(不支援) | FULL OUTER |
| 左外連線 | LEFT OUTER |
| 右外連線 | RIGHT OUTER |
| 交叉連線(又稱笛卡爾積) | CROSS |
全外連線(FULL OUTER)實現方式：
SELECT A.*,B.* FROM table A LEFT JOIN table B ON B.Key = A.Key
UNION ALL
SELECT A.*,B.* FROM table B RIGNT JOIN table A ON A.Key = B.Key
使用JOIN更新表
UPDATE table A
JOIN table B ON A.key = B.key
SET A.column=set_value;
使用JOIN優化子查詢
準備表：
CREATE TABLE `users` (
`id` smallint(5) unsigned NOT NULL AUTO_INCREMENT,
`username` varchar(20) NOT NULL,
`password` varchar(32) NOT NULL,
`age` tinyint(3) unsigned NOT NULL DEFAULT '10',
`sex` tinyint(1) DEFAULT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=33 DEFAULT CHARSET=utf8
CREATE TABLE `province` (
`id` smallint(5) unsigned NOT NULL AUTO_INCREMENT,
`pname` varchar(20) NOT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=5 DEFAULT CHARSET=utf8
第1種子查詢：
select A.id,A.username,(select pname from province B WHERE A.username=B.pname) as pname from users A；
show profiles;
第2種外連線查詢；
select A.id,A.username,B.pname from users A left join province B on A.username=B.pname;
show profiles;
所以第2種比較好。
使用join優化聚合子查詢
如何實現分組選擇
查詢每人打怪最多的兩條記錄；
方法1：
WITH tmp AS(
SELECT a.user_name,b.timestr,b.kills, ROW_NUMBER() over(partition by a.user_name order by b.kills) cnt
FROM user1 a
JOIN user_kills b ON a.id = b.user_id
)select * from tmp where cnt <=2;
方法2：
SELECT d.user_name,c.timestr,kills
FROM(
SELECT user_id,timestr,kills,( SELECT COUNT(*) FROM user_kills b WHERE b.user_id = a.user_id AND a.kills <=b.kills) AS cnt
FROM user kills a
GROUP BY user_id,timestr,kills
)c
JOIN user1 d ON c.user_id = d.id
WHERE cnt <=2;
檢視mysql語句執行時間
1. Show profiles是5.0.37之後新增的，要想使用此功能，要確保版本在5.0.37之後。
2.確定支援show profile 後，檢視profile是否開啟，資料庫預設是不開啟的。變數profiling是使用者變數，每次都得重新啟用。
檢視方法： show variables like "%pro%";
設定開啟方法： set profiling = 1;
再次檢視show variables like "%pro%"; 已經是開啟的狀態了。
3.可以開始執行一些想要分析的sql語句了，執行完後，show profiles; 即可檢視所有sql的總的執行時間。
show profile for query 1 即可檢視第1個sql語句的執行的各個操作的耗時詳情。
show profile cpu, block io, memory,swaps,context switches,source for query 6;可以查看出一條SQL語句執行的各種資源消耗情況，比如CPU，IO等
show profile all for query 6 檢視第6條語句的所有的執行資訊。
測試完畢後，關閉引數：
mysql> set profiling=0方法二： timestampdiff來檢視執行時間。
這種方法有一點要注意，就是三條sql語句要儘量連一起執行，不然誤差太大，根本不準
set @d=now();
select * from comment;
select timestampdiff(second,@d,now());
如果是用命令列來執行的話，有一點要注意，就是在select timestampdiff(second,@d,now());後面，一定要多copy一個空行，不然最後一個sql要你自己按回車執行，這樣就不準了。
行轉列
場景：
1、報表統計
2、彙總顯示
set names 'gbk';
mysql> select a.user_name,sum(kills)
-> from user1 a
-> join user_kills b on a.id =b.user_id
-> group by a.user_name;
+-----------+------------+
| user_name | sum(kills) |
+-----------+------------+
| 孫悟空 | 47 |
| 沙僧 | 9 |
| 豬八戒 | 24 |
+-----------+------------+
3 rows in set (0.01 sec)
--方法1：
select *
from(
select sum(kills) as '沙僧' from user1 a join user_kills b on a.id =b.user_id and a.user_name='沙僧'
) a
cross join(
select sum(kills) as '豬八戒' from user1 a join user_kills b on a.id =b.user_id and a.user_name='豬八戒'
) b
cross join(
select sum(kills) as '孫悟空' from user1 a join user_kills b on a.id =b.user_id and a.user_name='孫悟空'
) c
;
--方法2：
select sum(case when user_name='孫悟空' then kills end) as '孫悟空'
,sum(case when user_name='豬八戒' then kills end) as '豬八戒'
,sum(case when user_name='沙僧' then kills end) as '沙僧'
from user1 a
join user_kills b on a.id = b.user_id
;
mysql> select *
-> from(
-> select sum(kills) as '沙僧' from user1 a join user_kills b on a.id =b.user_id and a.user_n
ame='沙僧'
-> ) a
-> cross join(
-> select sum(kills) as '豬八戒' from user1 a join user_kills b on a.id =b.user_id and a.user
_name='豬八戒'
-> ) b
-> cross join(
-> select sum(kills) as '孫悟空' from user1 a join user_kills b on a.id =b.user_id and a.user
_name='孫悟空'
-> ) c
-> ;
+------+--------+--------+
| 沙僧 | 豬八戒 | 孫悟空 |
+------+--------+--------+
| 9 | 24 | 47 |
+------+--------+--------+
1 row in set (0.01 sec)
mysql> select sum(case when user_name='孫悟空' then kills end) as '孫悟空'
-> ,sum(case when user_name='豬八戒' then kills end) as '豬八戒'
-> ,sum(case when user_name='沙僧' then kills end) as '沙僧'
-> from user1 a
-> join user_kills b on a.id = b.user_id
-> ;
+--------+--------+------+
| 孫悟空 | 豬八戒 | 沙僧 |
+--------+--------+------+
| 47 | 24 | 9 |
+--------+--------+------+
1 row in set (0.00 sec)
列轉行
