oracle里的int类型
在 ANSI SQL 标准 中，INTEGER 和 SMALLINT 是定义好的精确数值类型，但它们的 “长度”或“大小”并不是通过 (N) 括号来指定的（如 INT(4)），这一点与 MySQL 等数据库的非标准扩展完全不同。
- SMALLINT
语义：短整型整数
实际实现：通常是 16 位有符号整数，2个字节
取值范围：-32,768 ≤ SMALLINT ≤ 32,767 - INTEGER 或 INT
语义：标准整数
实际实现：通常是 32 位有符号整数，4个字节
取值范围：-2,147,483,648 ≤ INTEGER ≤ 2,147,483,647
oracle没有内置int类型，但是兼容int类型，在oracle里int和smallint认为是number(38)，元信息是这样的，但数据受约束不一样，number(38)受38位有效数字限制，而int只受number类型的最大限制限制，即NUMBER [ (p [, s]) ]，Number having precision p and scale s. The precision p can range from 1 to 38. The scale s can range from -84 to 127. Both precision and scale are in decimal digits. A NUMBER value requires from 1 to 22 bytes.
所以在oracle中的int类型约束等同于number类型，但元信息为number(38)，其实际存储数据的限制会远大于sql标准中的int和smallint
Data Types
| ANSI SQL Data Type | Oracle Data Type |
|---|---|
CHARACTER(n)CHAR(n) | CHAR(n) |
CHARACTER VARYING(n)CHAR VARYING(n) | VARCHAR2(n) |
NATIONAL CHARACTER(n)NATIONAL CHAR(n)NCHAR(n) | NCHAR(n) |
NATIONAL CHARACTER VARYING(n)NATIONAL CHAR VARYING(n)NCHAR VARYING(n) | NVARCHAR2(n) |
NUMERIC[(p,s)]DECIMAL[(p,s)] (Note 1) | NUMBER(p,s) |
INTEGERINTSMALLINT | NUMBER(38) |
FLOAT (Note 2)DOUBLE PRECISION (Note 3)REAL (Note 4) | FLOAT(126)FLOAT(126)FLOAT(63) |
以下测试版本：oracle 19.18.0.0.0
SQL> create table t1(c1 int,c2 smallint);Table created.SQL> desc t1Name Null? Type----------------------------------------------------------------- -------- --------------------------------------------C1 NUMBER(38)C2 NUMBER(38) --但是建表时，不支持int(4)这种限制类型长度，也不支持int4这种写法
SQL> create table t2(id int(4));
create table t2(id int(4))*
ERROR at line 1:
ORA-00907: missing right parenthesisSQL> create table t2(id int4);
create table t2(id int4)*
ERROR at line 1:
ORA-00902: invalid datatype--int类型与number(38)长度的不同
SQL> create table t1(c1 number(38));Table created.
SQL> desc t1Name Null? Type----------------------------------------------------------------- -------- --------------------------------------------C1 NUMBER(38)SQL> insert into t1 values (99999999999999999999999999999999999999); --38个91 row created.SQL> insert into t1 values (999999999999999999999999999999999999999); --39个9
insert into t1 values (999999999999999999999999999999999999999)*
ERROR at line 1:
ORA-01438: value larger than specified precision allowed for this columnSQL> select c1,to_char(c1),length(c1) from t1;C1 TO_CHAR(C1) LENGTH(C1)
---------- -------------------------------------------------------------------------------- ----------
1.0000E+38 99999999999999999999999999999999999999 38SQL> create table t3(c1 int);Table created.
SQL> desc t3Name Null? Type----------------------------------------------------------------- -------- --------------------------------------------C1 NUMBER(38)SQL> insert into t3 values (99999999999999999999999999999999999999); --38个91 row created.SQL> insert into t3 values (999999999999999999999999999999999999999); --39个91 row created.SQL> insert into t3 values (99999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999); --125个91 row created.
SQL> insert into t3 values (999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999); --126个9
insert into t3 values (999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999)*
ERROR at line 1:
ORA-01426: numeric overflowSQL> select c1,to_char(c1),length(c1) from t3;C1 TO_CHAR(C1) LENGTH(C1)
---------- -------------------------------------------------------------------------------- ----------
1.0000E+38 99999999999999999999999999999999999999 38
1.0000E+39 999999999999999999999999999999999999999 39
1.000E+125 1.000000000000000000000000000000000E+125 40--没有长度的number类型
SQL> create table t2(c1 number);Table created.SQL> desc t2Name Null? Type----------------------------------------------------------------- -------- --------------------------------------------C1 NUMBERSQL> insert into t2 values (99999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999);1 row created.SQL> insert into t2 values (999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999);
insert into t2 values (999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999)*
ERROR at line 1:
ORA-01426: numeric overflow
但是oracle在plsql对int类型的用法与在sql中有所区别，在plsql中可以使用int或者int(1) 写法，且要做精度检查，其中int等于number(38),int(1)等于number(1)
SQL> -- 开启输出
SET SERVEROUTPUT ON;SQL>--int(1)
SQL> DECLAREv1 INT(1);
BEGINv1 := 1;DBMS_OUTPUT.PUT_LINE('v1=' || v1);
END;
/ 2 3 4 5 6 7
v1=1PL/SQL procedure successfully completed.SQL> DECLAREv1 INT(1);
BEGINv1 := 11;DBMS_OUTPUT.PUT_LINE('v1=' || v1);
END;
/ 2 3 4 5 6 7
DECLARE
*
ERROR at line 1:
ORA-06502: PL/SQL: numeric or value error: number precision too large
ORA-06512: at line 4----int
SQL> DECLAREv1 INT;
BEGINv1 := 99999999999999999999999999999999999999; --38个9DBMS_OUTPUT.PUT_LINE('v1=' || v1);
END;
/ 2 3 4 5 6 7
v1=99999999999999999999999999999999999999PL/SQL procedure successfully completed.SQL> DECLAREv1 INT;
BEGINv1 := 999999999999999999999999999999999999999; --39个9DBMS_OUTPUT.PUT_LINE('v1=' || v1);
END;
/2 3 4 5 6 7 DECLARE
*
ERROR at line 1:
ORA-06502: PL/SQL: numeric or value error: number precision too large
ORA-06512: at line 4
所以，在oracle里，在sql里int类型不做精度检查，存储限制上等同于number，元信息上为number(38),在plsql里，要做精度检查，int完全等同于number(38),int(1)等同于number(1)
因此在oracle开发规范中，还是尽量使用number类型，避免使用int类型，不过如果是其他支持标准 int类型的数据库中，如果数据符合int类型特点，还是建议用int类型而非numeric类型，因为int类型是定长存储，一般性能会更好
)
)
