SQL_Standards_2 |
Please refer SQL_Standards1.htm :
Although the most of the database vendors adhere to ANSI - SQL
standards; also have their own flavors added to SQL language.
SQL Querying:
PostgreSQL and MySQL :
: :SELECT ename e, job empjob, deptno FROM emp
limit 5
Oracle 10. and 11.2 g
:SELECT ename e, job empjob, deptno, sal * 0.025
esal FROM emp where rownum < 6
This document introduces a basic know how of SQL Data Types. |
Oracle emp Table:

PostgreSQL: EMP Table

|
Data Type: The programming languages like c/c++ and java were
used to create the databases like oracle and PostgreSQL , and therefore
columns in a table also support derivatives of similar data types.
As you noticed in the above two images that data type
integer (o,d), where 'o' stands for over all number and 'd' stands
of decimal point , was declared as NUMBER(o,d) data type in
Oracle and 'numeric (o,d) in PostgreSQL.
The variable character data type, was declared as
VARCHAR2(n) in Oracle, and character varying (n) in PostgreSQL 9.2
and up.
|
Common Built in data-types in a Database: PGS (PostgreSQL)
Name/Uses |
Feature |
Special Notes: |
LOB /BLOB
/BFILE |
stores large binary objects up to 4gb |
LONG RAW,RAW(N) |
BOOLEAN |
True/False |
Oracle does not have boolean ** |
CHAR(n) |
Fixed length character data |
|
CLOB |
Long character string |
LONG VARCHAR |
DATE |
Y,M,D,HR,MIN,SEC |
No fractional second or time zone |
DOUBLE |
|
|
INTEGER |
Stores whole number/No fraction |
|
LONG |
stores value up to 2gb, only one column per table, backward
compatibility |
don't create tables with LONG, use BLOB, CLOB |
NUMBER |
stores up to 40 digits |
sub types are float |
NUMBER(o,d) |
unlimited number with exact precision (money matters) of
decimal points |
numeric(n) in postgresql |
REAL |
float with binary precision |
single for 32 bit 4 bytes, double for 64bit 8
bytes |
TIMESTAMP |
extension of DATE, stores Y,M,D hr:min:sec |
has fractional second, but no time zone |
TIMESTAMP with TIMEZONE |
extension of DATE, stores Y,M,D hr:min:sec |
has both time-zone and fractional second |
VARCHAR2(n) |
variable-length of a character string, with n=max |
character varying(n) in PGSQL |
|
|
|
|
|
|
**Serial types: Stores a sequential
INTEGER increment at the succeeding row insertions. |
**Ref: postgresql 9.3 pdf
data types real and double precision are inexact, which means
that some values could be stored with approximation. |
|
Oracle :
- In Oracle-SQL there is no data type boolean. It can, however, be
simulated by using either char(1) or number(1).
|
PostgreSQL : Please refer postgresql-9.3-A4.pdf Ch-8 |
Pseudo columns:
Oracle
SELECT ename e, job empjob, deptno, sal * 0.025
esal FROM emp where rownum < 6

PostgreSQL (9.2 and up)

MySQL:
test Scripts : SELECT ename e, job empjob, deptno, sal * 0.025 esal
FROM emp limit 5

|
|
|
|
|
|
|
|
|