In this post, I am sharing a custom function which will replace the String in all the column of a PostgreSQL Table.

Postgres provides the string replace function, but at a time you can replace string for a particular column.

Recently, I got the requirement like replace same string value in multiple columns.
So, I prepared a user-defined function which I am sharing here.

Please check code, and try it yourself:

Create a sample table with data:

create table tbl_str (t1 character varying,t2 character varying);

insert into tbl_str values 
('AA','BB'),('BB','DD')
,('CC','BB'),('FF','GG');

Create a function for replacing string in all columns:

create or replace function fn_replace_string_for_all_columns(tablename text,oldvalue text,newvalue text) 
returns void as 
$$
declare 
eachrw record;
begin
    for eachrw in 
        select 'UPDATE '||$1||' SET '||C.COLUMN_NAME||' = REPLACE ('||C.COLUMN_NAME||','''||$2||''','''||$3||'''); ' SQLQuery
        FROM (select column_name from information_schema.columns where table_schema='public' and table_name =$1)c
    loop
        EXECUTE eachrw.SQLQuery;
    end loop;
end;
$$language plpgsql;

Call the function:

select *from fn_replace_string_for_all_columns('tbl_str', 'BB', 'ZZ');

Check the result: ‘BB’ replaced by ‘ZZ’

select *from tbl_str

  t1 | t2 
-----------
  AA | ZZ
  ZZ | DD
  CC | ZZ
  FF | GG

Check the below input string and expected output for counting the total Lowercase, Uppercase and Other Characters (Including SPACE) in the given string.

Input String:

'I am Anvesh Patel, owner @str DBRND.COM'

Expected Output:

UpperCase_Count LowerCase_Count OtherCharacter_Count
--------------- --------------- --------------------
11              19              9

Solution:

DECLARE @str AS VARCHAR(100) = 'I am Anvesh Patel, owner @str DBRND.COM'
 
SELECT 
        SUM(CASE WHEN ASCII(string) BETWEEN 65 AND 90 THEN 1 ELSE 0 END) as UpperCase_Count
       ,SUM(CASE WHEN ASCII(string) BETWEEN 97 AND 122 THEN 1 ELSE 0 END) as LowerCase_Count
       ,SUM(CASE WHEN (( ASCII(string) NOT BETWEEN 97 AND 122 ) AND ( ASCII(string) NOT BETWEEN 65 AND 90 )) THEN 1 ELSE 0 END) as OtherCharacter_Count 
FROM
(
    SELECT 
		DISTINCT Number 
		,SUBSTRING(@str,Number,1) as string
    FROM Master..spt_Values 
    WHERE Number > 0 and number <= LEN(@str)
)as t

Please try the different solution for this puzzle and share it via comment...

Check the below input data and expected output to split the string and generate a count of each string. Split the string by semi colon.

Input Data:

str1
--------------------
anvesh;dbrnd;blog
dbrnd;blog
dbrnd
blog;anvesh

Expected Output:

StringName           StringCounts
-------------------- ------------
anvesh               2
blog                 3
dbrnd                3

Create a table with data:

CREATE TABLE tbl_SplitData
(
    str1 VARCHAR(20)
)
GO
 
INSERT INTO tbl_SplitData (str1)
VALUES ('anvesh;dbrnd;blog'),('dbrnd;blog'),('dbrnd'),('blog;anvesh')
GO

Solution 1:

SELECT 
	StringName
	,COUNT(*) StringCounts 
FROM
(
    SELECT  CAST(CONCAT('<t>' , REPLACE(str1, ';' , '</t><t>' ) , '</t>' ) AS XML) Xmlcol FROM tbl_SplitData
) s
CROSS apply
(
    SELECT ProjectData.D.value('.', 'VARCHAR(20)') as StringName
    FROM s.xmlcol.nodes('t') as ProjectData(D)
)t
GROUP BY t.StringName

Please try the different solution for this puzzle and share it via comment...

Create a user defined function to count the total number of a word from the given string data.

Sample function execution:

select [dbo].[fn_GetWordCount]('I am Database Engineer and I am the owner of dbrnd.com') AS TotalWordCount

Expected Output:

TotalWordCount
--------------
11

User defined function:

CREATE FUNCTION [dbo].[fn_GetWordCount] (@InputStringData VARCHAR(100)) 
RETURNS INT
AS
BEGIN

DECLARE @i INT
DECLARE @Char CHAR(1)
DECLARE @PrevChar CHAR(1)
DECLARE @Counter INT

SET @i = 1
SET @Counter = 0

WHILE @i <= LEN(@InputStringData)
BEGIN
    SET @Char = SUBSTRING(@InputStringData, @i, 1)
    SET @PrevChar = CASE WHEN @i = 1 THEN ' '
                         ELSE SUBSTRING(@InputStringData, @i - 1, 1)
                    END
    IF @PrevChar = ' ' AND @Char != ' '
    SET @Counter = @Counter + 1
    SET @i = @i + 1
END

RETURN @Counter 
END
GO

Please try the different solution for this puzzle and share it via comment...

Last week on my Facebook Page, I have received a couple message about what is a similar solution of SQL Server STUFF () in PostgreSQL.

Most of the Database Developers require to perform String Aggregation based on different group of records.

Since PostgreSQL 9.0, STRING_AGG(expression, delimiter) function is available to perform String Aggregation operation.
Using STRING_AGG(), We can concatenate strings using different type of delimiter symbols.

Example of STRING_AGG():

Create a sample Students table:

CREATE TABLE tbl_Students
(
	StudID INT 
	,StudName CHARACTER VARYING
	,StudGrades CHAR(1)
);

Insert few sample records:

INSERT INTO tbl_Students 
VALUES 
(1,'Anvesh','A'),(2,'Kimly','B')
,(3,'Jenny','C'),(4,'Ali','B')
,(5,'Mukesh','D'),(6,'Sofia','A')
,(7,'Roy','C'),(8,'Martin','C');

Concatenate Students Name per each Student Grade and arrange by Grade wise row (Using STRING_AGG()):

SELECT 	
	StudGrades
	,STRING_AGG(StudName,', ') AS StudPerGrade
FROM tbl_Students
GROUP BY StudGrades
ORDER BY 1 ;

The Result:

studgrades |    studpergrade
------------+--------------------
 A          | Anvesh, Sofia
 B          | Kimly, Ali
 C          | Jenny, Roy, Martin
 D          | Mukesh
(4 rows)

In this post, I am sharing a small demonstration on how to perform case-sensitive string comparison in MySQL.

For example, if you are searching string with LIKE ‘a%’, also you get all records of LIKE ‘A%’ – the reason is the case-insensitive comparison.
By default, string comparisons are case insensitive because strings are non-binary. For the case-sensitive comparison, we should use binary collation.

If you need case-sensitive string comparison, you should use this COLLATION latin1_bin.

You can visit this article about, How to change default COLLATION in MySQL.


The String Case Sensitive Comparison using LIKE BINARY:

You can also perform string case-sensitive comparison using LIKE BINARY operator.
Generally, we need for password comparison.

Below is a full demonstration on this:

Create a sample table and data:

CREATE TABLE tbl_Students
(
    StudentID INT AUTO_INCREMENT
    ,StudentName VARCHAR(255)
    ,CONSTRAINT pk_tbl_Students_StudentID PRIMARY KEY (StudentID)
);

INSERT INTO tbl_Students (StudentName) 
VALUES ('Anvesh'),('Roy'),('martin'),('Loris'),('Jenny');

Execute below three queries and check the result:
Using normal LIKE you will get Case insensitive result and using LIKE BINARY you will get the Case sensitive result.

SELECT 
    StudentID
    ,StudentName
FROM tbl_Students WHERE StudentName LIKE 'roy';
-- Return one record of Student 'Roy'

SELECT 
    StudentID
    ,StudentName
FROM tbl_Students WHERE StudentName LIKE BINARY 'roy';
-- No record for Student 'roy' because this is case sensitive check,'roy' and 'Roy' both are different. 

SELECT 
    StudentID
    ,StudentName
FROM tbl_Students WHERE StudentName LIKE BINARY 'Roy';
-- Return one record of Student 'Roy'