Skip to main content

Useful String functions in SQL

Select length('String length') str_LENGTH from dual;

STR_LENGTH
----------------------
13


--Replace string with
Select REPLACE('REPLACE String ','REPLACE','Replaced') str_REPLACE from dual;

STR_REPLACE
----------------
Replaced String


--default replaced value will be null if param2 is not passed
Select REPLACE('REPLACE String ','REPLACE') str_REPLACE from dual;

STR_REPLACE
-----------
String

--Print a portion of string
Select SUBSTR('Sub string',1,3) str_LENGTH from dual;

STR_LENGTH
----------
Sub

--INSTR function
Select instr('Find first occurence of a char in the string','a') str_INSTR from dual;

STR_INSTR
----------------------
25

--TO_CHAR function is mostly used with dates to convert them into desired format
Select to_char(sysdate,'dd-Mon-yyyy hh:mi:ss') str_TOCHAR from dual;

STR_TOCHAR
--------------------
01-Feb-2012 10:11:45

Select to_char(sysdate,'dd-Mon-yy') str_TOCHAR from dual;

STR_TOCHAR
----------
01-Feb-12

Select to_char(sysdate,'hh:mi:ss') str_TOCHAR from dual;

STR_TOCHAR
----------
10:11:45

Select to_char(sysdate,'dd-Mon-yyyy hh:mi Day') str_TOCHAR from dual;


STR_TOCHAR
---------------------------
01-Feb-2012 10:11 Wednesday

Comments

Popular posts from this blog

Design and Operations ( Employee- Model )

-- DB design drop table emp_Salary; drop table emp_Designation; drop table emp; drop table Designations; drop table Departments; drop table Salary_Bands; drop table Addresses; Create table emp( emp_id number, emp_Name varchar2(20), emp_address_id varchar2(10), emp_created_date date, emp_created_by varchar2(20), constraint e_pk primary key(emp_id) ); Create table Departments( dept_id number, dept_name Varchar2(20), constraints dept_pk primary key(dept_id) ); Create table Designations( dept_id number references Departments(dept_id), desg_id number, desg_description varchar2(20), constraints desg_pk primary key(dept_id,desg_id) ); Create table emp_Designation( emp_id number references emp(emp_id), emp_des_dept_id number references departments(dept_...

How to remove duplicate rows without using DISTINCT or ROWID or GROUP BY methods in Oracle?

Make use of SET operators and CTAS (Create Table AS) method.. Example: Create table tmp_tb1 as Select * from table1 intersect Select * from table1; truncate table table1; insert into table1 select * from tmp_tb1; -- This will now have removed all duplicates in the table... commit; *** You can use UNION operator in place of INTERSECT

Identify current and prev details of a customer

  create table customer_rating( id char, name char, irrating char, jcrrating char, procDate date, flag char ); insert into customer_rating values('A','D','X','Y','28-Jul-21','M'); insert into customer_rating values('A','D','M','L','27-Jul-21','M'); Select cr.id, cr.name, cr_prev.irrating pcob_rating, cr.irrating cob_rating, cr_prev.jcrrating pcob_jcr_rating, cr.jcrrating cob_jcr_rating From Customer_Rating cr join Customer_Rating cr_Prev  ON cr.procdate = '28-Jul-21' AND cr_prev.id = cr.id AND cr_prev.procdate = (  Select max(procdate) from Customer_Rating  Where id=cr.id and procdate < cr.procdate ) ;