Mysql 高级(进阶)SQL 语句二
MySQL高级SQL语句
1、EXISTS
用来测试内查询有没有产生任何结果,类似布尔值是否为真
如果有的话,系统就会执行外查询中的SQL语句,若是没有,那整个SQL语句就不会产生任何结果。
语法:SELECT 字段1 FROM 表1 WHERE EXISTS (SELECT * FROM 表2 WHERE 条件); 例: select region from REGION where exists (select * from FARE where money = '700');
2、inner join、left join、right join
inner join(等值相连)
只返回两个表中联接字段相等的行
语法:SELECT 字段 FROM 表1 INNER JOIN 表2 ON 表1.字段 = 表2.字段;
left join(左联接)
返回包括左表中所有记录和右表中联接字段相等的记录
语法:SELECT 字段 FROM 表1 LEFT JOIN 表2 ON 表1.字段 = 表2.字段;
right join(右联接)
返回包括右表中的所有记录和左表中联接字段相等的记录
语法:SELECT 字段 FROM 表1 RIGHT JOIN 表2 ON 表1.字段 = 表2.字段;
例: update FARE set site='xingjiang' where money='2200'; select * from REGION AS A inner join FARE AS B on A.site = B.site; select * from REGION AS A left join FARE AS B on A.site = B.site; select * from REGION AS A right join FARE AS B on A.site = B.site;
3、CREATE VIEW 视图
视图:可被当作虚拟表或存储查询。
视图跟表格不同的是,表格中有实际存储资料,而视图是建立在表格之上的一个架构,它本身并不实际存储资料。
临时表在用户退出或者同数据库的连接断开后就自动消失了,而视图不会。
视图不含有数据,只存储它的定义,它的用途一般可以简化复杂的查询。比如你要对几个表格进行连接查询,而且还要进行统计排序等操作,写SQL语句会很麻烦,用视图将几个表格连接起来,然后对这个视图进行查询操作,就和对一个表查询一样,很方便。
语法:CREATE VIEW 视图表名 AS SELECT语句; 例: select A.region,SUM(B.money) from REGION AS A inner join FARE AS B on A.site = B.site GROUP BY region; create view C AS select A.region,SUM(B.money) from REGION AS A inner join FARE AS B on A.site = B.site GROUP BY region; show tables; select * from C;
删除视图表
4、UNION 联集
将两个SQL语句的结果合并起来,两个SQL语句产生的字段需要是同样的资料种类
UNION:生成结果的资料值将没有重复,且按照字段的顺序进行排序。
语法:SELECT语句1 UNION SELECT语句2;
UNION ALL:将生成结果的资料值都列出来,无论有无重复
语法:SELECT语句1 UNION ALL SELECT语句2
例:
select region from REGION union select money from FARE;
select site from REGION union select site from FARE;
select site from REGION union all select site from FARE;
6、交集值
取两个SQL语句结果的交集
例: select * from FARE; insert into FARE values ('xingjiang',2200,'2021-09-05'); select * from FARE; select A.site from FARE AS A inner join REGION AS B on A.site = B.site; select A.site from FARE AS A inner join REGION AS B using(site); select site from (select site from FARE union all select site from REGION) A group by A.site HAVING count(*) > 1; select site from (select A.site from FARE A inner join REGION B on A.site = B.site) C group by C.site; select site from FARE where site in (select site from REGION);
7、无交集值
显示第一个SQL语句的结果,且与第二个SQL语句没有交集的结果,还不能重复
select distinct site from FARE where (site) not in (select site from REGION);
例: select A.*,B.* from FARE A left join REGION B using(site); select site from FARE A left join REGION B using(site) where B.site is null; select distinct site from FARE A left join REGION B using(site) where B.site is null;
8、CASE
是SQL用来作为IF-THEN-ELSE之类逻辑的关键字
语法:
SELECT CASE (字段名)
WHEN 条件1 THEN 结果1
WHEN 条件2 THEN 结果2
……
ELSE 结果N
END
FROM 表名
#条件可以是一个数值或是公式。ELSE子句不是必须的。
例: select * from FARE; select case site when 'changsha' then money + 500 when 'tianjin' then money - 500 else money * 2 end A,site from FARE;
1、算排名
表格自我连接(self join),然后将结果依序列出,算出每一行之前(包括那一行本身)有多少行数
例: create table SCORE (name char(10),score int(4)); insert into SCORE values('zhangsan',95); insert into SCORE values('lisi',96); insert into SCORE values('wangwu',75); insert into SCORE values('zhaoliu',80); insert into SCORE values('qianqi',90); select * from SCORE;
select A1.name,A1.score,count(A2.score) rank from SCORE A1,SCORE A2 where A1.score < A2.score OR (A1.score=A2.score and A1.name=A2.name) group by A1.name order by A1.score desc;
#当A1的score字段值小于A2的score字段值、或者两表score字段值相等并且name字段值相等时,从A1和A2表中 查询A1的name字段值、A1的score字段值、和A2的score字段的非空值 rank是别名 ,并为A1的name字段分组,A1的score字段降序排序
A1.score为75时,A2.score可为75、80、90、95、96 则 count(A2.score)为5
A1.score为80时,A2.score可为80、90、95、96 则 count(A2.score)为4
A1.score为90时,A2.score可为90、95、96 则 count(A2.score)为3
A1.score为95时,A2.score可为95、96 则 count(A2.score)为2
A1.score为96时,A2.score可为96 则 count(A2.score)为1
这样再显示A1的name字段分组,A1的score字段降序排序 ,rank字段
则为下表所示
+----------+-------+------+
| name | score | rank |
+----------+-------+------+
| lisi | 96 | 1 |
| zhangsan | 95 | 2 |
| qianqi | 90 | 3 |
| zhaoliu | 80 | 4 |
| wangwu | 75 | 5 |
+----------+-------+------+
2、算中位数
例: select * from (select A1.name,A1.score,count(A2.score) rank from SCORE A1,SCORE A2 where A1.score < A2.score OR (A1.score=A2.score and A1.name=A2.name) group by A1.name order by A1.score desc) A3 where A3.rank = (select (count(*)+1) DIV 2 from SCORE); select score mid from (select A1.name,A1.score,count(A2.score) rank from SCORE A1,SCORE A2 where A1.score < A2.score OR (A1.score=A2.score and A1.name=A2.name) group by A1.name order by A1.score desc) A3 where A3.rank = (select (count(*)+1) DIV 2 from SCORE);
3、算累积总计
select A1.*,sum(A2.score) sum_socore from SCORE A1,SCORE A2 where A1.score < A2.score or(A1.score=A2.score and A1.name=A2.name) group by A1.name order by A1.score desc;
4、算总合百分比
select A1.*,A1.score/(select sum(score) from SCORE) z_sum from SCORE A1,SCORE A2 where A1.score < A2.score or (A1.score=A2.score and A1.name=A2.name) group by A1.name;
select sum(socre) from SCORE 是为了算出字段值总合,然后每一行一一除以总合,算出每行的总合百分比。
5、算累计总合百分比
select A1.name,A1.score,sum(A2.score),sum(A2.score)/(select sum(score) from SCORE) Z from SCORE A1,SCORE A2 where A1.score < A2.score or (A1.score=A2.score and A1.name=A2.name) group by A1.name order by A1.score desc; select A1.name,A1.score,sum(A2.score),TRUNCATE(sum(A2.score)/(select sum(score) from SCORE),2) ||'%' Z from SCORE A1,SCORE A2 where A1.score < A2.score or (A1.score=A2.score and A1.name=A2.name) group by A1.name order by A1.score desc;
6、空值(null)和无值(’’)的区别
1、无值的长度为0,不占用空间;而空值null 的长度是null,是占用空间的
2、IS NULL或者IS NOT NULL,是用来判断字段是不是NULL或者不是NULL,是不能查出是不是无值的
3、无值的判断使用=’‘或者<>’'来处理。<>代表不等于
4、在通过count()指定字段统计又多少行数时,如果遇到NULL值会自动忽略掉,遇到空值会自动加入记录中进行计算
create table SITE(site varchar(20)); insert into SITE values('nanjing'); insert into SITE values('beijing'); insert into SITE values(''); insert into SITE values('taijin'); insert into SITE values(); insert into SITE values(''); select * from SITE;
select length(site) from SITE; #查询SITE表site字段值长度 select * from SITE where site is NULL; select * from SITE where site is not NULL; select * from SITE where site =''; select * from SITE where site <> '';
7、正则表达式(与Shell部分用法基本一样)
匹配模式 描述 实例 ^ 匹配文本的开始字符 ‘^bd’ 匹配以 bd 开头的字符串 $ 匹配文本的结束字符 ‘qn$’ 匹配以 qn 结尾的字符串 . 匹配任何单个字符 ‘s.t’ 匹配任何 s 和 t 之间有一个字符的字符串 * 匹配零个或多个在它前面的字符 ‘fo*t’ 匹配 t 前面有任意个 o + 匹配前面的字符 1 次或多次 ‘hom+’ 匹配以 ho 开头,后面至少一个m 的字符串 字符串 匹配包含指定的字符串 ‘clo’ 匹配含有 clo 的字符串 p1|p2 匹配 p1 或 p2 ‘bg|fg’ 匹配 bg 或者 fg [...] 匹配字符集合中的任意一个字符 ‘[abc]’ 匹配 a 或者 b 或者 c [^...] 匹配不在括号中的任何字符 ‘[^ab]’ 匹配不包含 a 或者 b 的字符串 {n} 匹配前面的字符串 n 次 ‘g{2}’ 匹配含有 2 个 g 的字符串 {n,m} 匹配前面的字符串至少 n 次,至多m 次 ‘f{1,3}’ 匹配 f 最少 1 次,最多 3 次
语法:SELECT 字段 FROM 表名 WHERE 字段 REGEXP 匹配模式
例: select * from FARE where site regexp '^[n]'; select * from FARE where site regexp '[n]'; select * from FARE where site regexp 'jin|jing';
8、存储过程(与Shell函数类似,代码的复用)
存储过程是一组为了完成特定功能的SQL语句集合
存储过程在使用过程中是将常用或者复杂的工作预先使用SQL语句写好并用一个指定的名称来进行储存,这个过程经编译和优化后存储在数据库服务器中,当需要使用该存储过程时,只需要调用它即可,存储过程在执行上比传统SQL速度更快,执行效率更高。
存储过程的优点:
- 执行一次后,会将生成的二进制代码驻留缓冲区,提高执行效率
- SQL语句加上控制语句的集合,灵活性高
- 在服务器端存储,客户端调用时,降低网络负载
- 可多次重复被调用,可随时修改,不影响客户端调用
- 可完成所有的数据库操作,也可控制数据库的信息访问权限
语法: DELIMITER !! #将语句的结束符号从分号;临时修改,以防出问题,可以自定义 CREATE PROCEDURE XXX() #创建存储过程,过程名自定义,()可带参数 BEGIN #过程体以关键字BEGIN开始 select * from xxx; #过程体语句 END!! #过程体以关键字END结尾 DELIMITER ; #将语句的结束符号恢复为分号 call XXX; #调用存储过程 ====查看存储过程==== show create procedure [数据库.]储存过程名; #查看某个储存过程的具体信息 show create procedure XXX; show procedure status [like '%XXX%'] \G
例: DELIMITER !! CREATE PROCEDURE KIND1() BEGIN select * from FARE; END!! DELIMITER ; CALL KIND1;
show create procedure train_ticket.KIND1\G show procedure status like '%KIND1%'\G
存储过程的参数
IN 输入参数,表示调用者向过程传入值(传入值可以是字面量或变量)
OUT 输出参数:表示过程向调用者传出值(可以返回多个值,传出值只能是变量)
例: DELIMITER !! CREATE PROCEDURE KIND2(IN place char(20)) BEGIN select * from FARE where site=place; END!! DELIMITER ; CALL KIND2('beijing');
9、存储过程的条件语句
DELIMITER !! CREATE PROCEDURE KIND3(IN num int(10)) BEGIN declare var int; set var=num*2; if var>=10 then update FARE set money=money+1; else update FARE set money=money-1; end if; END!!
DELIMITER ;
10、循环语句while
create table xxx(id int); DELIMITER !! CREATE PROCEDURE KIND4() BEGIN declare var int; set var=0; while var<5 do insert into xxx values(var); set var=var+1; end while; END!! DELIMITER ; CALL KIND4; select * from xxx;