oracle 简单备注

1. 建立数据库

备注：

1) oracle 不同于mysql 可以直接create database

2) oracle 创建schema时对应一个用户，即该schema的访问用户，与用户一一对应；但可以存在多个访问用户（带权限控制）

1.1 创建数据库文件

CREATE TABLESPACE XX LOGGING DATAFILE ‘D:\app\XX\oradata\orcl\XX.dbf‘ SIZE 1000M;

create temporary tablespace XX tempfile ‘D:\app\XX\oradata\orcl\XX.dbf‘ size 1000m;

1.2 创建用户

CREATE USER XX IDENTIFIED BY XX DEFAULT TABLESPACE XX TEMPORARY TABLESPACE XX;

1.3 授权

grant connect, resource to XX;
grant create session to XX;

 

2. 数据库操作(默认sccot用户)：

2.1 create

create table persons(
person_id NUMBER PRIMARY KEY,
first_name VARCHAR2(50) NOT NULL,
last_name VARCHAR2(50) NOT NULL,
score NUMBER,
type VARCHAR2(20)
);

备注：

1) ORA-02000: missing ALWAYS keyword : 在11g版本里不用用GENERATED BY DEFAULT AS IDENTITY ，要用PRIMARY KEY
2) oracle 本来只有number类型，用作number(19,2)，即有小数位；后为了兼容其他数据库，新增int，只能是整形

2.2 insert

insert into persons values(1, ‘fred‘, ‘xu‘, 0, ‘a‘);

insert into persons values(SEQ_PERSON_ID.NEXTVAL, ‘fred3‘, ‘xu3‘, 0, ‘b‘)

备注：

1）oracle下设置自增没有mysql那么简单，步骤如下：

1.1） CREATE SEQUENCE SEQ_PERSON_ID start with 100; #创建一个序列
1.2) INSERT INTO persons VALUES(SEQ_PERSON_ID.NEXTVAL, ‘fred1‘, ‘xu1‘, 0, ‘a‘);

#此时插入persons表记录的person_id被设置成了100

1.3）也可以采用触发器的形式
CREATE TRIGGER persons_trigger
BEFORE INSERT ON persons
FOR EACH ROW
WHEN (new.person_id is null)
begin
select SEQ_PERSON_ID.nextval into :new.person_id from sys.dual;
end;

备注：sys.dual 是个虚拟表，oracle保证里面只有一条记录
:new— 触发器执行过程中触发表作操作的当前行的新纪录
:old— 触发器执行过程中触发表作操作的当前行的旧纪录

在有触发器之后，插入数据时不需要填写主键 INSERT INTO persons(first_name, last_name, score, person_type) VALUES(‘fred2‘, ‘xu2‘, 0, ‘a‘);

2.3 select

select * from persons where first_name = ‘fred‘ or last_name = ‘xu1‘
select * from persons where first_name like ‘%fred%‘
select * from persons where CONCAT(first_name, last_name) like ‘%1%‘

select * from users where rownum BETWEEN 0 AND 5 #利用rownum关键字分页

2.4 group by

select AVG(score), person_type from persons group by person_type
select AVG(score), person_type from persons group by person_type having person_type = ‘b‘

备注：

1）ORA-00979: not a GROUP BY expression group by 后的列要能被处理

2.5 in

select * from persons where person_type in (‘a‘,‘b‘);

2.6 insert all

insert all
into persons(first_name, last_name, score, person_type) values(‘test‘, ‘test‘, 1, ‘c‘)
into persons(first_name, last_name, score, person_type) values(‘test1‘, ‘test1‘, 1, ‘c1‘)
into persons(first_name, last_name, score, person_type) values(‘test2‘, ‘test2‘, 1, ‘c2‘)

SELECT 1 FROM dual;

备注：

SELECT 1 FROM dual; 最后一次select必须有

2.7 视图

create view person_view as select * from persons where first_name like ‘%f%‘

select * from person_view;

备注：

视图是需表，只是逻辑定义；除非是一种物化视图，那个才是有物理占用

2.8 索引

create index person_index on persons (first_name);