mysql必知必会笔记_《MySql必知必会》笔记整理
数据库基础
关键词：
数据库
表(表名唯一，取决多个因素，如不同数据库的表可以同名)
模式(关于数据库和表的布局及特性的信息)
列(表中的字段)
行[行(raw)和记录(record)很大程度可以等同，但行才是正确的术语]
数据类型(限制数据种类，帮助正确排序，磁盘优化方面的作用)
主键(primary key): 一列，其值可以唯一区分表中的行。
SQL [(Structured Query Language)：结构化查询语言。]
主键条件：
每行都应有一个主键，所以其值不为null。
任意两行间的主键值不同。
主键通常是一列，但也可多列共同构成主键。
主键设置建议：
不更新主键列中的值；
不重用主键列的值；
不在主键列中使用可能会更改的值。
SQL命令执行：
命令在mysql>之后输入；
命令用 ；或\g结束，仅按Enter不执行命令；
输入help或\h获取帮助；
输入quit或exit退出程序。
基本语句：
myslq -u root -p;
use Database;
#SHOW相关
SHOW databases;
SHOW tables;
SHOW columns FROM tables; -- 等于 describe "tables";
SHOW CREATE DATABASE db_name; 显示完整的建库语句
SHOW CREATE TABLE tbl_name;
SHOW [STORAGE] ENGINES
#SELECT相关
SELECT column_name1,column_name2 FROM table;
SELECT *FROM tables;
-- Distinct
-- 不能部分使用DISTINCT，其应用于所有列而不是其前置列
SELECT DISTINCT column_namw FROM table;
-- Limit 从第零个开始后的5个 取的时候排列顺序是从零开始的。
SELECT column_name FROM table_name LIMIT 5;
-- 从第二个开始后的5个
SELECT column_name FROM table_name LIMIT 2,5;
-- OFFSET 限制两个，从第三为开始取
SELECT column_name FROM table_name LIMIT 2 OFFSET 3;
-- 使用全限定的表名 库：manxc 表：tags
SELECT tags.tid FROM manxc.tags;
排序检索数据
关键字：ORDER BY
SELECT column_name FROM table_name ORDER BY column_name;
默认升序，字母按A-Z排，数字从小到大；
注：排序中文时出现问题。
解决：https://www.jianshu.com/p/8a9135f9cd47?utm_campaign
升序(默认)：ASC 降序：DESC
过滤数据
关键字：WHERE(同时可与其它关键字组合)
SELECT * FROM manxc.tags WHERE tags.tid BETWEEN 2 AND 9 ORDER BY tid DESC,tagname;
操作符
说明
=
等于
<>
不等于
!=
不等于
<
小于
<=
小于等于
>
大于
>=
大于等于
BETWEEN
在指定的两个值之间
eg:
mysql> SELECT 2 BETWEEN 1 AND 3, 2 BETWEEN 3 and 1;
-> 1, 0
mysql> SELECT 1 BETWEEN 2 AND 3;
-> 0
mysql> SELECT 'b' BETWEEN 'a' AND 'c';
-> 1
mysql> SELECT 2 BETWEEN 2 AND '3';
-> 1
mysql> SELECT 2 BETWEEN 2 AND 'x-3';
-> 0
WHERE 匹配字符加‘’；且其在执行匹配时默认不区分大小写；
mysql> SELECT uid,username,state FROM manxc.user WHERE username='FLY';
+-----+----------+-------+
| uid | username | state |
+-----+----------+-------+
| 3 | fly | 0 |
+-----+----------+-------+
空值检查：IS NULL (空值是无值和0和空格不同)
mysql> SELECT uid,username,state FROM user WHERE password IS NULL LIMIT 3;
+-----+----------+-------+
| uid | username | state |
+-----+----------+-------+
| 8 | dfdg | NULL |
| 9 | dgdg | NULL |
| 10 | gdg | NULL |
+-----+----------+-------+
数据过滤
关键词
操作符(operator) :用来联结或改变where子句的关键字。
AND 操作符
mysql> SELECT uid,username,state FROM USER WHERE state IS NULL AND uid <= 13;
+-----+-----------+-------+
| uid | username | state |
+-----+-----------+-------+
| 8 | dfdg | NULL |
| 9 | dgdg | NULL |
| 10 | gdg | NULL |
| 11 | dgdgh | NULL |
| 12 | dgklds | NULL |
| 13 | dgkljdlkg | NULL |
+-----+-----------+-------+
OR操作符：
mysql> SELECT uid,username,state FROM USER WHERE state IS NULL OR uid <= 13;
+-----+--------------+-------+
| uid | username | state |
+-----+--------------+-------+
| 4 | test1 | 0 |
| 3 | fly | 0 |
| 5 | test2 | 0 |
| 6 | test3 | 1 |
| 7 | 1 | 1 |
| 8 | dfdg | NULL |
| 9 | dgdg | NULL |
| 10 | gdg | NULL |
| 11 | dgdgh | NULL |
| 12 | dgklds | NULL |
| 13 | dgkljdlkg | NULL |
| 14 | fdjwe | NULL |
| 15 | gkdlkg | NULL |
| 16 | dgdlkjg | NULL |
| 17 | fdglkdjg | NULL |
| 18 | gkldssjgdsas | NULL |
| 19 | dgjkljg | NULL |
| 20 | djglkdg | NULL |
| 21 | kgdlksgj | NULL |
+-----+--------------+-------+
混合使用时的顺序：
在有多个or和and同时使用时，优先处理and，可以使用()提高优先级。
mysql> SELECT uid,username,state FROM USER WHERE (state IS NULL OR state =1) AND uid <=10;
+-----+----------+-------+
| uid | username | state |
+-----+----------+-------+
| 6 | test3 | 1 |
| 7 | 1 | 1 |
| 8 | dfdg | NULL |
| 9 | dgdg | NULL |
| 10 | gdg | NULL |
+-----+----------+-------+
5 rows in set (0.00 sec)
mysql> SELECT uid,username,state FROM USER WHERE state IS NULL OR state =1 AND uid <=10;
+-----+--------------+-------+
| uid | username | state |
+-----+--------------+-------+
| 6 | test3 | 1 |
| 7 | 1 | 1 |
| 8 | dfdg | NULL |
| 9 | dgdg | NULL |
| 10 | gdg | NULL |
| 11 | dgdgh | NULL |
| 12 | dgklds | NULL |
| 13 | dgkljdlkg | NULL |
| 14 | fdjwe | NULL |
| 15 | gkdlkg | NULL |
| 16 | dgdlkjg | NULL |
| 17 | fdglkdjg | NULL |
| 18 | gkldssjgdsas | NULL |
| 19 | dgjkljg | NULL |
| 20 | djglkdg | NULL |
| 21 | kgdlksgj | NULL |
+-----+--------------+-------+
16 rows in set (0.00 sec)
建议：使用具有AND和OR操作符的WHERE子句，都应该使用圆括号明确的分组，不用过分依赖计算次序，使用括号没有坏处且能消除歧义。
IN 操作符：
where子句使用in操作符
mysql> SELECT uid,username,state FROM user WHERE state IN (0,1);
+-----+----------+-------+
| uid | username | state |
+-----+----------+-------+
| 4 | test1 | 0 |
| 3 | fly | 0 |
| 5 | test2 | 0 |
| 6 | test3 | 1 |
| 7 | 1 | 1 |
+-----+----------+-------+
5 rows in set (0.00 sec)
IN 和 OR 有类似作用，此句表示查询state是0或1的。
IN操作符的优点：
使用IN时，计算次序更容易管理(操作符少了，没那么多or)
IN操作符的语法更清楚且直观；
IN一般比OR的执行更快；
IN的最大优点时可以包含其它SELECT语句，使得能更动态的建立WHERE子句。
NOT操作符：
where子句中，not用来否定之后跟的条件。
mysql> select * from tags where tid NOT IN (1,2,3,4,5,6,7,8,9,10);
+-----+---------+
| tid | tagname |
+-----+---------+
| 11 | 猎奇 |
| 12 | 少女 |
| 13 | 魔法 |
| 14 | 历史 |
| 15 | 机战 |
| 16 | 神魔 |
| 17 | 运动 |
| 18 | 励志 |
| 19 | 音乐 |
| 20 | 推理 |
| 21 | 美食 |
| 22 | 催泪 |
| 23 | 职场 |
| 26 | 搞笑 |
+-----+---------+
注：MySQL支持使用NOT对IN，BETWEEN，EXISTS子句取反。
统配符过滤
关键字：like
统配符匹配：
百分号(%)通配符：表示任何字符出现任意次数，(任意字数的任意字符)
注：
A.由配置方式，搜索时可以区分大小写的；
B.注意尾空格，尾空格会干扰匹配，可在其前后都家%,或者使用函数。
下划线(_): 任意的单个字符；
注：统配符的搜索处理比之前操作符的效率更低使用时：
1，不要过度使用，能用其他操作符的，尽量用。
2，除非有绝对必要，不要把通配符放在搜索模式的开始处，这样是最慢的。
3，注意统配符放的位置。
SELECT prod_id,prod_name FROM products
WHERE prod_name LIKE '%ton anvil';
正则表达式搜索
关键字：REGEXP 其后跟正则表达式
与LIKE与REGEXP的区别：
LIKE匹配整个列，如果匹配的文本在列值中出现，LIKE不会找到它，相应行也不会返回(除非使用统配符)，而REGEXP可以在列值中进行匹配：
mysql> SELECT uid,username,password,state FROM user WHERE username LIKE "test";
Empty set (0.00 sec)
mysql> SELECT uid,username,password,state FROM user WHERE username REGEXP "test";
+-----+----------+----------+-------+
| uid | username | password | state |
+-----+----------+----------+-------+
| 4 | test1 | 123 | 0 |
| 5 | test2 | 123 | 0 |
| 6 | test3 | 123 | 1 |
+-----+----------+----------+-------+
3 rows in set (0.05 sec)
注：MYSQL中正则匹配不区分大小写，如需区分可使用BINARY关键字，如WHERE prod-name REGEXP BINARY 'JetPack'.
进行or匹配 使用“|”
mysql> SELECT uid,username,password,state FROM user WHERE username REGEXP "1|lk";
+-----+-----------+----------+-------+
| uid | username | password | state |
+-----+-----------+----------+-------+
| 4 | test1 | 123 | 0 |
| 7 | 1 | 1 | 1 |
| 13 | dgkljdlkg | NULL | NULL |
| 15 | gkdlkg | NULL | NULL |
| 16 | dgdlkjg | NULL | NULL |
| 17 | fdglkdjg | NULL | NULL |
| 20 | djglkdg | NULL | NULL |
| 21 | kgdlksgj | NULL | NULL |
+-----+-----------+----------+-------+
8 rows in set (0.00 sec)
匹配字符：
使用“[]”匹配，相当于另一种形式的or；匹配其中的任意字符
mysql> SELECT uid,username,password,state FROM user WHERE username REGEXP "test";
+-----+----------+----------+-------+
| uid | username | password | state |
+-----+----------+----------+-------+
| 4 | test1 | 123 | 0 |
| 5 | test2 | 123 | 0 |
| 6 | test3 | 123 | 1 |
| 22 | test4 | NULL | NULL |
| 23 | test5 | NULL | NULL |
| 24 | test6 | NULL | NULL |
+-----+----------+----------+-------+
6 rows in set (0.00 sec)
mysql> SELECT uid,username,password,state FROM user WHERE username REGEXP "test[123]";
+-----+----------+----------+-------+
| uid | username | password | state |
+-----+----------+----------+-------+
| 4 | test1 | 123 | 0 |
| 5 | test2 | 123 | 0 |
| 6 | test3 | 123 | 1 |
+-----+----------+----------+-------+
3 rows in set (0.00 sec)
mysql> SELECT uid,username,password,state FROM user WHERE username REGEXP "[123]test";
Empty set (0.00 sec)
mysql> SELECT uid,username,password,state FROM user WHERE username REGEXP "[test]";
+-----+--------------+----------+-------+
| uid | username | password | state |
+-----+--------------+----------+-------+
| 4 | test1 | 123 | 0 |
| 5 | test2 | 123 | 0 |
| 6 | test3 | 123 | 1 |
| 12 | dgklds | NULL | NULL |
| 14 | fdjwe | NULL | NULL |
| 18 | gkldssjgdsas | NULL | NULL |
| 21 | kgdlksgj | NULL | NULL |
| 22 | test4 | NULL | NULL |
| 23 | test5 | NULL | NULL |
| 24 | test6 | NULL | NULL |
+-----+--------------+----------+-------+
10 rows in set (0.00 sec)
加上“^”表非
mysql> SELECT uid,username,password,state FROM user WHERE username REGEXP "test[^123]";
+-----+----------+----------+-------+
| uid | username | password | state |
+-----+----------+----------+-------+
| 22 | test4 | NULL | NULL |
| 23 | test5 | NULL | NULL |
| 24 | test6 | NULL | NULL |
+-----+----------+----------+-------+
3 rows in set (0.00 sec)
空格
mysql> SELECT uid,username,password,state FROM user WHERE username REGEXP "1|2 test";
+-----+----------+----------+-------+
| uid | username | password | state |
+-----+----------+----------+-------+
| 4 | test1 | 123 | 0 |
| 7 | 1 | 1 | 1 |
+-----+----------+----------+-------+
2 rows in set (0.00 sec)
匹配范围:如[1-3],[a-z];
mysql> SELECT uid,username,password,state FROM user WHERE username REGEXP "test[1-3]";
+-----+----------+----------+-------+
| uid | username | password | state |
+-----+----------+----------+-------+
| 4 | test1 | 123 | 0 |
| 5 | test2 | 123 | 0 |
| 6 | test3 | 123 | 1 |
+-----+----------+----------+-------+
3 rows in set (0.00 sec)
匹配特殊字符： 使用 “\\特殊字符”，即转义
\\.能够匹配.
\\f换页
\\n换行
\\r回车
\\t制表
\\纵向制表
注意：
a)为了匹配 \ 本身，需要使用 \\\
b)在一般情况下正则表达式的转义加一个“\”就可以了，在MySQL中需要加两个。
匹配字符类：
[:alnum:]=[a-zA-Z0-9]
[:alpha:]=[a-zA-Z]
[:digit:]=[0-9]
[:lower:]=[a-z]
[:upper:]=[A-Z]
[:xdigit:]=[a-fA-F0-9]
重复次数匹配将其加在之后：
元字符
说明
*
0个或多个匹配
+
1个或多个匹配(等于{1，})
？
0个或1个匹配(等于{0，1})
{n}
指定数目匹配
{n,}
不少于指定数目匹配
{n,m}
匹配数目的范围
匹配任意三个连续数字：
问：把其【【】】是一种更好习惯？还是有什么区别
答：所以这里的区别是[[:digit:]],[:digit:]有的
mysql> SELECT uid,username,password,state FROM user WHERE password REGEXP "[[:digit:]]{3}";
+-----+----------+----------+-------+
| uid | username | password | state |
+-----+----------+----------+-------+
| 4 | test1 | 123 | 0 |
| 3 | fly | 123 | 0 |
| 5 | test2 | 123 | 0 |
| 6 | test3 | 123 | 1 |
| 8 | dfdg | d124 | NULL |
| 9 | dgdg | 123r | NULL |
+-----+----------+----------+-------+
6 rows in set (0.00 sec)
mysql> SELECT uid,username,password,state FROM user WHERE password REGEXP "[:digit:]{3}";
+-----+----------+----------+-------+
| uid | username | password | state |
+-----+----------+----------+-------+
| 4 | test1 | 123 | 0 |
| 3 | fly | 123 | 0 |
| 5 | test2 | 123 | 0 |
| 6 | test3 | 123 | 1 |
| 8 | dfdg | d124 | NULL |
| 9 | dgdg | 123r | NULL |
+-----+----------+----------+-------+
6 rows in set (0.00 sec)
定位元字符：
元字符
说明
^
文本的开始
$
文本的结尾
[[:<:>
词的开始
[[:>:]]
词的结尾
示例:找出密码中以数字开头的记录：
mysql> SELECT uid,username,password,state FROM user WHERE password REGEXP "^[[:digit:]\\.]";
+-----+----------+----------+-------+
| uid | username | password | state |
+-----+----------+----------+-------+
| 4 | test1 | 123 | 0 |
| 3 | fly | 123 | 0 |
| 5 | test2 | 123 | 0 |
| 6 | test3 | 123 | 1 |
| 7 | 1 | 1 | 1 |
| 9 | dgdg | 123r | NULL |
| 10 | gdg | 1d23 | NULL |
| 11 | dgdgh | 1.23s | NULL |
| 12 | dgklds | 2.31t | NULL |
+-----+----------+----------+-------+
9 rows in set (0.00 sec)
// 注意区别
mysql> SELECT uid,username,password,state FROM user WHERE password REGEXP "^[:digit:]\\.";
+-----+----------+----------+-------+
| uid | username | password | state |
+-----+----------+----------+-------+
| 11 | dgdgh | 1.23s | NULL |
| 12 | dgklds | 2.31t | NULL |
+-----+----------+----------+-------+
2 rows in set (0.00 sec)
创建计算字段
Concat()函数拼接字段
Concat()拼接串，把多个串连接起来形成一个较长的串。各串之间用“，”分隔。
注:多数DBMS使用+或||来实现拼接,而mysql使用Concat()函数
将用户名按 用户名(用户id)拼接出来：
mysql> SELECT Concat(username,'(',uid,')') FROM user ORDER BY uid DESC LIMIT 4;
+------------------------------+
| Concat(username,'(',uid,')') |
+------------------------------+
| test6(24) |
| test5(23) |
| test4(22) |
| kgdlksgj(21) |
+------------------------------+
4 rows in set (0.00 sec)
RTrim( )函数：删除值右边的所有空格
mysql> SELECT Concat(Rtrim(username),'(',uid,')') FROM user ORDER BY uid DESC LIMIT 4;
+-------------------------------------+
| Concat(Rtrim(username),'(',uid,')') |
+-------------------------------------+
| test6(24) |
| test5(23) |
| test4(22) |
| kgdlksgj(21) |
+-------------------------------------+
4 rows in set (0.00 sec)
LTrim( )函数去掉串左边的空格，Trim( )函数去掉串两边的空格；
使用别名：
别名(alias)用AS关键字赋予，使用别名能让客户机更好的使用数据，别名有时也叫导出列；
mysql> SELECT Concat(username,'(',uid,')')
-> AS uinfo
-> FROM user LIMIT 4;
+----------+
| uinfo |
+----------+
| test1(4) |
| fly(3) |
| test2(5) |
| test3(6) |
+----------+
4 rows in set (0.00 sec)