Get Occurrences of Character in String
Applicability:
SQL Server 2000: Tested
SQL Server 2005: Tested
SQL Server 2008: Tested
SQL Server 2008R2: Tested
SQL Server 2012: Not Tested
Credits:
Author: ChillyDBA
Date: 17 Jul 2012
Description
Returns the ordinal position of all occurrences of the supplied character in the supplied string
Takes 3 parameters:
@Character - the character to be searched for
@String - the string to be searched
@IsCaseSensitive - 1 indicates that case sensitivity should be enforced during the search. Default 0.
NOTE1: For case sensitivity (CS), this will enforce CS in a non-CS database but will not relax CS in a CS database
NOTE2: The function utilises the function udf_NumbersTable. This has several versions, depending on SQL Server version.
Code
Function:
DROP FUNCTION dbo.[udf_GetOccurrencesOfCharacterInString]
GO
CREATE FUNCTION [dbo].[udf_GetOccurrencesOfCharacterInString]
(
@Character CHAR(1), -- the character to search for
@String VARCHAR(MAX), -- the string to search
@IsCaseSensitive BIT = 0 -- is the search honor case sensitive?
)
RETURNS @Results TABLE
(
Position INT,
CHARacter CHAR(1) -- this is only returned as a confidence check. Not really necessary
)
AS
/**********************************************************
Purpose: Returns a list of character positions that the
supplied character occurs at in the supplied string.
NOTE: The case sensitive flag will only cause case sensitivity (CS) to
be applied in a non-CS DB. If will not apply non-CS in a CS DB
Utilises udf_NumbersTable to provide the required 'tally' table used
when parsing the string for matches
Author: ChillyDBA
History 17 Jul 2012 - Initial Issue
***********************************************************/
BEGIN
IF @IsCaseSensitive = 0
BEGIN
INSERT @Results (Position, CHARacter)
SELECT i, SUBSTRING(@String,i,1)
FROM dbo.udf_NumbersTable(1, DATALENGTH(@String), 1)
WHERE i <= LEN(@String)
AND SUBSTRING(@String,i,1) = @Character
ORDER BY i
END
ELSE
BEGIN
INSERT @Results (Position, CHARacter)
SELECT i, SUBSTRING(@String,i,1)
FROM dbo.udf_NumbersTable(1, DATALENGTH(@String), 1)
WHERE i <= LEN(@String)
AND ASCII(SUBSTRING(@String,i,1)) = ASCII(@Character) -- by converting the character to ASCII value, case-sensitivity is enforced
ORDER BY i
END
RETURN
END
Test Code:
SELECT * FROM dbo.udf_GetOccurrencesOfCharacterInString('e', ',Element01,Element02,Element03,Element04,Element05,', 0)
SELECT * FROM dbo.udf_GetOccurrencesOfCharacterInString('e', ',Element01,Element02,Element03,Element04,Element05,', 1)