How to split a comma delimited string using ONLY a select statement? - sql-server

Question

Here's the situation:
I cannot implement a custom function. PLEASE let's not get into a debate about why. That's just the way the sitation is and I cannot change that
I would need to somehow split the comma delimited values up using a SELECT statement of some kind
I cannot use the built in STRING_SPLIT function because I'd need to set the database compatibly to 130, which I cannot do, due to permission issue
So with that all mentioned, how can I spit up something like 'This,Is,A,Sentence' using a select statement?
I am using SQL Server 2008.

Answer 1

Query
Declare #String nvarchar(500) = 'This,Is,A,Sentence';
SELECT Split.a.value('.', 'VARCHAR(100)') Words
FROM (
SELECT Cast ('<X>'
+ Replace(#String, ',', '</X><X>')
+ '</X>' AS XML) AS Data
) AS t CROSS APPLY Data.nodes ('/X') AS Split(a);
Result Set:
╔══════════╗
║ Words ║
╠══════════╣
║ This ║
║ Is ║
║ A ║
║ Sentence ║
╚══════════╝

Answer 2

A plus 1 to the guys suggesting cte's. You're probably going to find that their suggestions are going to be the best option. The code I use for this is usually wrapped in a user defined function, but I've stripped it out for an example here:
DECLARE #str NVARCHAR(MAX) = 'This,Is,A,Sentence'
, #Delimiter NCHAR(1) = ',';
WITH cte AS
(SELECT 1 AS ID
, CASE WHEN CHARINDEX(#Delimiter, #str, 1) = 0 THEN #str
ELSE LEFT(#str, CHARINDEX(#Delimiter, #str, 1) - 1)
END Words
, CASE WHEN CHARINDEX(#Delimiter, #str, 1) = 0 THEN ''
ELSE STUFF(#str, 1, CHARINDEX(#Delimiter, #str, 1), '')
END Remainder
UNION ALL
SELECT cte.ID + 1
, CASE WHEN CHARINDEX(#Delimiter, cte.Remainder, 1) = 0 THEN cte.Remainder
ELSE LEFT(cte.Remainder, CHARINDEX(#Delimiter, cte.Remainder, 1) - 1)
END
, CASE WHEN CHARINDEX(#Delimiter, cte.Remainder, 1) = 0 THEN ''
ELSE STUFF(cte.Remainder, 1, CHARINDEX(#Delimiter, cte.Remainder, 1), '')
END Remainder
FROM cte
WHERE Remainder <> '')
SELECT cte.ID [Index]
, cte.Words
FROM cte;
Result set:
You can of course strip out the id/index column if you're not going to need it
Incidentally, if you compare this to the built in Split function in the latest version of SQL Server, this is actually far more efficient. Running both together cte version 16% of load, built in function 84%. That's a big difference.

Answer 3

Well, not a single SELECT, but here's something you can use without calling a proc or UDF:
CREATE TABLE #OutputTable( SValues VARCHAR(100) )
DECLARE #StringInput VARCHAR(MAX)
,#StringTemp VARCHAR(100)
WHILE LEN(#StringInput) > 0
BEGIN
SET #StringTemp = LEFT(#StringInput,
ISNULL(NULLIF(CHARINDEX(',', #StringInput) - 1, -1),
LEN(#StringInput)))
SET #StringInput = SUBSTRING(#StringInput,
ISNULL(NULLIF(CHARINDEX(',', #StringInput), 0),
LEN(#StringInput)) + 1, LEN(#StringInput))
INSERT INTO #OutputTable ( SValues )
VALUES ( #StringTemp )
END
#OutputTable will hold each "word" as a record. You can then query it as needed.

Answer 4

If you can use a CTE .....
DECLARE #word VARCHAR(200) = 'This,Is,A,Sentence'
;WITH
a AS (SELECT i=-1, j=0 UNION ALL SELECT j, CHARINDEX(',', #word, j + 1) FROM a WHERE j > i),
b AS (SELECT SUBSTRING(#word, i+1, IIF(j>0, j, LEN(#word)+1)-i-1) word FROM a WHERE i >= 0)
SELECT * FROM b

Answer 5

With credit to question: Find nth Occurrence in a string
DECLARE #Val varchar(200) = 'this,is,a,string'
;with t as (
select #Val as val, 1 as starts, charindex(',', #Val) as pos
union all
select #Val, pos + 1, charindex(',', #Val, pos + 1)
from t
where pos > 0
)
select
*, substring(#Val, starts, case when pos > 0 then pos - starts else len(#Val) end) token
from T
order by starts

Answer 6

This is what I would do:
DECLARE #t TABLE
(
ID INT,
Words VARCHAR(500)
)
INSERT #t VALUES (1,'This,Is,A,Sentence')
SELECT
LTRIM(RTRIM(m.n.value('.[1]','varchar(8000)'))) AS Words
FROM
(
SELECT CAST('<XMLRoot><RowData>' + REPLACE(Words,',','</RowData><RowData>') + '</RowData></XMLRoot>' AS XML) AS x
FROM #t
)t
CROSS APPLY x.nodes('/XMLRoot/RowData')m(n)
Results:
|Words|
This
Is
A
Sentence