抄自:https://www.cnblogs.com/qiu18359243869/p/9474515.html
提示:dual是一个虚拟表,用来构成select的语法规则,oracle保证dual里面永远只有一条记录。
一、基本操作
--新建表: create table table1( id varchar(300) primary key, name varchar(200) not null); --插入数据 insert into table1 (id,name) values ('aa','bb'); --更新数据 update table1 set id = 'bb' where id='cc'; --删除数据 delete from table1 where id ='cc'; --清空表数据 --是DDL操作,不产生rollback,速度快,delete操作在删除大数据量时会很慢 truncate table table1; --删除表 drop table table1; --修改表名: alter table table1 rename to table2; --添加新字段 --alter table 表名 add(字段名 字段类型 默认值 是否为空); alter table user add(age number(6)); alter table user add (course varchar2(30) default '空' not null); --修改字段 --alter table 表名 modify (字段名 字段类型 默认值 是否为空); alter table user modify((age number(8)); --修改字段名 --alter table 表名 rename column 列名 to 新列名; alter table user rename column course to newcourse; -- 删除字段 --alter table 表名 drop column 字段名; alter table user drop column course; --表数据复制: insert into table1 (select * from table2); --复制表结构: create table table1 select * from table2 where 1>1; --复制表结构和数据: create table table1 select * from table2; --复制指定字段: create table table1 as select id, name from table2 where 1>1; --条件查询: select id,name (case gender when 0 then '男' when 1 then ‘女’ end ) gender from table1
--添加主键 ALTER TABLE TABLE_NAME ADD CONSTRAINT PK_NAME PRIMARY KEY(COLUMN_NAME); --添加外键 --有三种形式的外键约束: --1.普通外键约束(如果存在子表引用父表主键,则无法删除父表记录) --2.级联外键约束(可删除存在引用的父表记录,而且同时把所有有引用的子表记录也删除) --3.置空外键约束(可删除存在引用的父表记录,同时将子表中引用该父表主键的外键字段自动设为NULL,但该字段应允许空值) --普通外键 ALTER TABLE T_INVOICE_DETAIL ADD CONSTRAINT FK_INVOICE_ID FOREIGN KEY(INVOICE_ID ) REFERENCES T_INVOICE(ID); --级联外键 ALTER TABLE T_INVOICE_DETAIL ADD CONSTRAINT FK_INVOICE_ID FOREIGN KEY(INVOICE_ID ) REFERENCES T_INVOICE(ID) ON DELETE CASCADE; --置空外键 ALTER TABLE T_INVOICE_DETAIL ADD CONSTRAINT FK_INVOICE_ID FOREIGN KEY(INVOICE_ID ) REFERENCES T_INVOICE(ID) ON DELETE SET NULL; --删除主键/外键 ALTER TABLE TABLE_NAME DROP CONSTRAINT PK_NAME ; --创建序列 create sequence sequence_name minvalue 1 --最小值 nomaxvalue --不设置最大值(由机器决定),或 根据表字段的值范围设置 maxvalue maxvalue 999 -- 最大值 start with 1 --从1开始计数,数值可变 increment by 1 --每次加1,数值可变 nocycle --一直累加,不循环;cycle:达到最大值后,将从头开始累加 nocache; --不建缓冲区。 如果建立cache那么系统将自动读取cache值个seq,这样会加快运行速度;如果在单机中使用cache,或者oracle死了,那么下次读取的seq值将不连贯,所以不建议使用cache。 --修改序列 alter sequence sequence_name-- 序列名 也可以更改 minvalue 1 maxvalue 99999999 start with 10 increment by 1 cycle -- 到10000后,从头开始 nocache; --删除序列 drop sequence sequence_name --查询数据库的所有序列 select * from all_sequences; --查询序列的下一个值 SELECT SEQ_NAME.NEXTVAL FROM dual; --修改序列的下个值 ----X为正数负数都可以 alter sequence SEQ_NAME increment BY X; --重新将序列的值的增长值设为1 alter sequence SEQ_NAME increment BY 1; --创建索引 create index 索引名 on 表名(列名1,列名2,......); --删除索引 drop index 索引名; --在数据库中查找表名 select * from user_tables where table_name like 'tablename%'; --查看表的所有索引 select * from all_indexes where table_name = 'tablename'; --查看表的所有索引列 select* from all_ind_columns where table_name = 'tablename';
二、数学函数
--绝对值:abs() select abs(-2) value from dual; --(2) --取整函数(大):ceil() select ceil(-2.001) value from dual; --(-2) --取整函数(小):floor() select floor(-2.001) value from dual; --(-3) --取整函数(截取):trunc() select trunc(-2.001) value from dual; -- (-2) --四舍五入:round() select round(1.234564,4) value from dual; --(1.2346) --取平方:Power(m,n) select power(4,2) value from dual; --(16) --取平方根:SQRT() select sqrt(16) value from dual; --(4) --取随机数:dbms_random(minvalue,maxvalue) select dbms_random.value() from dual; (默认是0到1之间) select dbms_random.value(2,4) value from dual; (2-4之间随机数) --取符号:Sign() select sign(-3) value from dual; --(-1) select sign(3) value from dual; --(1) --取集合的最大值:greatest(value) select greatest(-1,3,5,7,9) value from dual; --(9) --取集合的最小值:least(value) select least(-1,3,5,7,9) value from dual; --(-1) --处理Null值:nvl(空值,代替值) select nvl(null,10) value from dual; --(10) select nvl(score,10) score from student;
三、rownum相关
--rownum小于某个数时可以直接作为查询条件(注意oracle不支持select top) select * from student where rownum <3; --查询rownum大于某个数值,需要使用子查询,并且rownum需要有别名 select * from(select rownum rn ,id,name from student) where rn>2; select * from (select rownum rn, student.* from student) where rn >3; --区间查询 select * from (select rownum rn, student.* from student) where rn >3 and rn<6; --排序+前n条 select * from (select rownum rn, t.* from ( select d.* from DJDRUVER d order by drivernumber)t )p where p.rn<10; --排序+区间查询1 select * from (select rownum rn, t.* from ( select d.* from DJDRIVER d order by DJDRIVER_DRIVERTIMES)t )p where p.rn<9 and p.rn>6; --排序+区间查询2 select * from (select rownum rn, t.* from ( select d.* from DJDRIVER d order by DJDRIVER_DRIVERTIMES)t where rownum<9 )p where p.rn>6;--效率远高于方式一
四、分页查询
假设每页显示10条
1. 不包含排序
--效率低 select * from (select rownum rn, d.* from DJDRIVER d )p where p.rn<=20 and p.rn>=10; select * from (select rownum rn, d.* from DJDRIVER d )p where p.rn between 10 and 20; --效率高 select * from (select rownum rn, d.* from DJDRIVER d where rownum<=20 )p where p.rn>=10;
2. 包含排序
--排序+区间查询1(效率低) select * from (select rownum rn, t.* from ( select d.* from DJDRIVER d order by DJDRIVER_DRIVERTIMES)t )p where p.rn<=20 and p.rn>=10; select * from (select rownum rn, t.* from ( select d.* from DJDRIVER d order by DJDRIVER_DRIVERTIMES)t )p where p.rn between 10 and 20; --排序+区间查询2(效率高) select * from (select rownum rn, t.* from ( select d.* from DJDRIVER d order by DJDRIVER_DRIVERTIMES)t where rownum<=20 )p where p.rn>=10;
五、时间处理
1. to_char和to_date基本使用
--日期 --年 yyyy yyy yy year --月 month mm mon month --日+星期 dd ddd(一年中第几天) dy day --小时 hh hh24 --分 mi --秒 ss
select to_char(sysdate,'yyyy-mm-dd hh24:mi:ss')currenttime, to_char(sysdate,'yyyy') year, to_char(sysdate,'mm') month, to_char(sysdate,'dd') day, to_char(sysdate,'day') week, to_char(sysdate,'hh24')hour, to_char(sysdate,'mi') minute, to_char(sysdate,'ss') second from dual;
select to_date('2009-07-04 05:02:01','yyyy-mm-dd hh24:mi:ss')currenttime, to_char(to_date('2009-07-04 05:02:01','yyyy-mm-dd hh24:mi:ss'),'yyyy')year, to_char(to_date('2009-07-04 05:02:01','yyyy-mm-dd hh24:mi:ss'),'mm')month, to_char(to_date('2009-07-04 05:02:01','yyyy-mm-dd hh24:mi:ss'),'dd') day, to_char(to_date('2009-07-04 05:02:01','yyyy-mm-dd hh24:mi:ss'),'day') week, to_char(to_date('2009-07-04 05:02:01','yyyy-mm-dd hh24:mi:ss'),'day','NLS_DATE_LANGUAGE=American') week, --设置语言 to_char(to_date('2009-07-04 05:02:01','yyyy-mm-dd hh24:mi:ss'),'hh24')hour, to_char(to_date('2009-07-04 05:02:01','yyyy-mm-dd hh24:mi:ss'),'mi') minute, to_char(to_date('2009-07-04 05:02:01','yyyy-mm-dd hh24:mi:ss'),'ss') second from dual;
2. months_between
select months_between(to_date('03-31-2014','MM-DD-YYYY'),to_date('12-31-2013','MM-DD-YYYY')) "MONTHS" FROM DUAL;
3. next_day
select sysdate today, next_day(sysdate,6) nextweek from dual;
4. 时间区间
-- borrow 为具体的表名 select cardid, borrowdate from borrow where to_date(borrowdate,'yyyy-mm-dd hh24:mi:ss') between to_date('2014-02-01 00:00:00','yyyy-mm-dd hh24:mi:ss') and to_date('2014-05-01 00:00:00','yyyy-mm-dd hh24:mi:ss');
5. interval
select to_char(sysdate,'yyyy-mm-dd hh24:mi:ss') currenttime, to_char(sysdate - interval '7' year,'yyyy-mm-dd hh24:mi:ss') intervalyear, to_char(sysdate - interval '7' month,'yyyy-mm-dd hh24:mi:ss') intervalMonth, to_char(sysdate - interval '7' day,'yyyy-mm-dd hh24:mi:ss') intervalday, to_char(sysdate - interval '7' hour,'yyyy-mm-dd hh24:mi:ss') intervalHour, to_char(sysdate - interval '7' minute,'yyyy-mm-dd hh24:mi:ss') intervalMinute, to_char(sysdate - interval '7' second,'yyyy-mm-dd hh24:mi:ss') intervalSecond from dual;
6. add_months
select add_months(sysdate,12) newtime from dual;
7. extract
select extract(month from sysdate) "This Month", extract(year from add_months(sysdate,36)) " Years" from dual;
六、字符函数
--字符函数 select substr('abcdefg',1,5)substr, --字符串截取 instr('abcdefg','bc') instr, --查找子串 'Hello'||'World' concat, --连接 trim(' wish ') trim, --去前后空格 rtrim('wish ') rtrim, --去后面空格 ltrim(' wish') ltrim, --去前面空格 trim(leading 'w' from 'wish') deleteprefix, --去前缀 trim(trailing 'h' from 'wish') deletetrailing, --去后缀 trim('w' from 'wish') trim1, ascii('A') A1, ascii('a') A2, --ascii(转换为对应的十进制数) chr(65) C1, chr(97) C2, --chr(十进制转对应字符) length('abcdefg') len, --length lower('WISH')lower, upper('wish')upper, initcap('wish')initcap, --大小写变换 replace('wish1','1','youhappy') replace, --替换 translate('wish1','1','y')translate, --转换,对应一位(前面的位数大于等于后面的位数) translate('wish1','sh1','hy')translate1, concat('11','22') concat --连接 from dual;
七、to_number
--to_number(expr) --to_number(expr,format) --to_number(expr,format,'nls-param') select to_number('0123')number1, --converts a string to number trunc(to_number('0123.123'),2) number2, to_number('120.11','999.99') number3, to_number('0a','xx') number4, --converts a hex number to decimal to_number(100000,'xxxxxx') number5 from dual;
八、聚合函数
student表如下:
1.count
--count (distinct|all) select count(1) as count from student;--效率最高 select count(*) as count from student; select count(distinct score) from student;
2. avg
--avg (distinct|all) select avg(score) score from student; select avg(distinct score) from student; select classno,avg(score) score from student group by classno;
3. max
--max (distinct|all) select max(score) from student; select classno, max(score) score from student group by classno;
4. min
--min (distinct|all) select min(score) from student; select classno, min(score) score from student group by classno;
5. stddev(standard deviation)标准差
--stddev select stddev(score) from student; select classno, stddev(score) score from student group by classno;
6. sum
--sum select sum(score) from student; select classno, sum(score) score from student group by classno;
7. median--中位数
--median select median(score) from student; select classno, median(score) score from student group by classno;