Comma-separated values (CSV) parameter filtering - sql

Question

Need help on how to improve my SQL script for better performance. dbo.Products table has a million rows. I'm hesitant to rewrite it using dynamic SQL. Thanks!
DECLARE
#Brand varchar(MAX) = 'Brand 1, Brand 2, Brand 3',
#ItemCategory varchar(MAX) = 'IC1, IC2, IC3, IC4, IC5'
--will return all records if params where set to #Brand = NULL, #ItemCategory = NULL
SELECT
[Brand],
SUM([Amount]) AS [Amount]
FROM dbo.Products (NOLOCK)
LEFT JOIN [dbo].[Split](#Brand, ',') FilterBrand ON Brand = [FilterBrand].[Items]
LEFT JOIN [dbo].[Split](#ItemCategory, ',') FilterItemCategory ON ItemCategory = [FilterItemCategory].[Items]
WHERE
(#Brand IS NULL OR (#Brand IS NOT NULL AND [FilterBrand].[Items] IS NOT NULL)) AND
(#ItemCategory IS NULL OR (#ItemCategory IS NOT NULL AND [FilterItemCategory].[Items] IS NOT NULL))
GROUP BY
[Brand]
Below is the split table-valued function that I found on the web:
CREATE function [dbo].[Split]
(
#String varchar(8000),
#Delimiter char(1)
)
RETURNS #Results TABLE (Items varchar(4000))
AS
BEGIN
IF (#String IS NULL OR #String = '') RETURN
DECLARE #i int, #j int
SELECT #i = 1
WHILE #i <= LEN(#String)
BEGIN
SELECT #j = CHARINDEX(#Delimiter, #String, #i)
IF #j = 0
BEGIN
SELECT #j = len(#String) + 1
END
INSERT #Results SELECT RTRIM(SUBSTRING(#String, #i, #j - #i))
SELECT #i = #j + LEN(#Delimiter)
END
RETURN
END

Answer 1

Following solution are with out using functions
Declare #IDs Varchar(100)
SET #IDs = '2,4,6'
Select IsNull(STUFF((Select ', '+ CAST([Name] As Varchar(100)) From [TableName]
Where CharIndex(','+Convert(Varchar,[ID])+',', ','+#IDs+',')> 0
For XML Path('')),1,1,''),'') As [ColumnName]

Answer 2

Here is the function I use. I also have another that wraps this to return numeric values which I find helpful as well.
Edit: Sorry, as for how to improve the performance of the query, I usually split the values into table variables and perform my joins to that but that probably won't change your performance, just your readability. The only thing I can see in terms of performance is your double checking whether your joins produce anything. You really can't get much better performance with two conditional left joins on two tables. It basically boils down to indexes at that point.
(#Brand IS NULL OR [FilterBrand].[Items] IS NOT NULL)
Function:
ALTER FUNCTION [dbo].[fn_SplitDelimittedList]
(
#DelimittedList varchar(8000),
#Delimitter varchar(20)
)
RETURNS
#List TABLE
(
Item varchar(100)
)
AS
BEGIN
DECLARE #DelimitterLength INT
SET #DelimitterLength = LEN(#Delimitter)
-- Tack on another delimitter so we get the last item properly
set #DelimittedList = #DelimittedList + #Delimitter
declare #Position int
declare #Item varchar(500)
set #Position = patindex('%' + #Delimitter + '%' , #DelimittedList)
while (#Position <> 0)
begin
set #Position = #Position - 1
set #Item = LTRIM(RTRIM(left(#DelimittedList, #Position)))
INSERT INTO #List (Item) VALUES (#Item)
set #DelimittedList = stuff(#DelimittedList, 1, #Position + #DelimitterLength, '')
set #Position = patindex('%' + #Delimitter + '%' , #DelimittedList)
end
RETURN
END

Answer 3

Hey just try the split function I have created without using any while loops here.And just use this in place of your split function and use col to match in LEFT join.
ALTER function dbo.SplitString(#inputStr varchar(1000),#del varchar(5))
RETURNS #table TABLE(col varchar(100))
As
BEGIN
DECLARE #t table(col1 varchar(100))
INSERT INTO #t
select #inputStr
if CHARINDEX(#del,#inputStr,1) > 0
BEGIN
;WITH CTE as(select ROW_NUMBER() over (order by (select 0)) as id,* from #t)
,CTE1 as (
select id,ltrim(rtrim(LEFT(col1,CHARINDEX(#del,col1,1)-1))) as col,RIGHT(col1,LEN(col1)-CHARINDEX(#del,col1,1)) as rem from CTE
union all
select c.id,ltrim(rtrim(LEFT(rem,CHARINDEX(#del,rem,1)-1))) as col,RIGHT(rem,LEN(rem)-CHARINDEX(#del,rem,1))
from CTE1 c
where CHARINDEX(#del,rem,1)>0
)
INSERT INTO #table
select col from CTE1
union all
select rem from CTE1 where CHARINDEX(#del,rem,1)=0
END
ELSE
BEGIN
INSERT INTO #table
select col1 from #t
END
RETURN
END
DECLARE #Brand varchar(MAX) = 'Brand 1,Brand 2,Brand 3',
#ItemCategory varchar(MAX) = ' IC1 A ,IC2 B , IC3 C, IC4 D' --'IC1, IC2, IC3, IC4, IC5'
select * from dbo.SplitString(#ItemCategory,',')