Split string into two columns - sql

I have the following string to split into two columns:
Given:
DECLARE #String VARCHAR(MAX) = 'Mak^1,Jak^2,Smith^3,Lee^4,Joseph^5'
I want to split it into two columns:
column1 column2
-----------------
Mak 1
Jak 2
Smith 3
Lee 4
Joseph 5
My try:
Table-valued Function:
CREATE FUNCTION [dbo].[udf_Split]
(
#InputString VARCHAR(8000),
#Delimiter VARCHAR(50)
)
RETURNS #Items TABLE (ID INTEGER IDENTITY(1,1), Item VARCHAR(8000))
AS
BEGIN
IF #Delimiter = ' '
BEGIN
SET #Delimiter = ','
SET #InputString = REPLACE(#InputString, ' ', #Delimiter)
END
IF (#Delimiter IS NULL OR #Delimiter = '')
SET #Delimiter = ','
DECLARE #Item VARCHAR(8000)
DECLARE #ItemList VARCHAR(8000)
DECLARE #DelimIndex INT
SET #ItemList = #InputString
SET #DelimIndex = CHARINDEX(#Delimiter, #ItemList, 0)
WHILE (#DelimIndex != 0)
BEGIN
SET #Item = SUBSTRING(#ItemList, 0, #DelimIndex)
INSERT INTO #Items VALUES (#Item)
SET #ItemList = SUBSTRING(#ItemList, #DelimIndex+1, LEN(#ItemList)-#DelimIndex)
SET #DelimIndex = CHARINDEX(#Delimiter, #ItemList, 0)
END -- End WHILE
IF #Item IS NOT NULL
BEGIN
SET #Item = #ItemList
INSERT INTO #Items VALUES (#Item)
END
ELSE INSERT INTO #Items VALUES (#InputString)
RETURN
END
Function calling:
SELECT Item FROM [dbo].[udf_Split](#String ,',');
Output:
Item
--------------
Mak^1
Jak^2
Smith^3
Lee^4
Joseph^5

First, Please note that SQL Server 2008 r2 is out of extended support. It's high time to upgrade to a newer version.
For a single string, I would probably use a little dynamic SQL magic trick:
DECLARE #String VARCHAR(MAX) = 'Mak^1,Jak^2,Smith^3,Lee^4,Joseph^5'
DECLARE #Sql VARCHAR(MAX) = 'SELECT Name,Id FROM (VALUES (''' + REPLACE(REPLACE(REPLACE(#String,'''',''''''), ',', '),('''), '^', ''',') + ')) V(Name, Id)';
-- #Sql now contains this:
-- SELECT Name,Id FROM (VALUES ('Mak',1),('Jak',2),('Smith',3),('Lee',4),('Joseph',5)) V(Name, Id)
EXEC(#Sql)
Results:
Name Id
Mak 1
Jak 2
Smith 3
Lee 4
Joseph 5

In the most recent versions of SQL Server, you can use string_split():
select left(s.value, charindex('^', value) - 1) as column1,
stuff(s.value, 1, charindex('^', value), '') as column2
from string_split(#string, ',') s ;
You might find it most convenient to download a split function to handle this.
Otherwise, I think a recursive CTE is a simple enough approach:
with cte as (
select convert(varchar(max), null) as row,
#string as str
union all
select convert(varchar(max), left(str, charindex(',', str + ',') - 1)),
convert(varchar(max), stuff(str, 1, charindex(',', str + ','), ''))
from cte
where str <> ''
)
select left(cte.row, charindex('^', cte.row) - 1) as column1,
stuff(cte.row, 1, charindex('^', cte.row), '')
from cte
where row is not null;
Here is a db<>fiddle.

I feel a much better approach to this would be to get rid of that awful WHILE and use a set based approach; we'll be using delimitedsplit8K here (if you are on 2012+ use delimitedsplit8k_lead or on 2016+ you can STRING_SPLIT).
With that in mind, the above becomes quite trivial:
DECLARE #String varchar(MAX) = 'Mak^1,Jak^2,Smith^3,Lee^4,Joseph^5';
SELECT LEFT(DS.Item,CHARINDEX('^',DS.Item)-1) AS Col1,
STUFF(DS.Item,1, CHARINDEX('^',DS.Item),'') AS Col2
FROM dbo.DelimitedSplit8K(#String, ',') DS;

Try This Script below
DECLARE #String VARCHAR(MAX) = 'Mak^1,Jak^2,Smith^3,Lee^4,Joseph^5';
DECLARE #TempTable AS TABLE(data VARCHAR(MAX))
INSERT INTO #TempTable
SELECT #String
;WITH CTE
AS
(
SELECT Split.A.value('.','nvarchar(1000)') AS data
FROM
(
SELECT CAST('<S>'+REPLACE(data,',','</S><S>')+'</S>' AS XML ) AS data
FROM #TempTable
)AS A
CROSS APPLY data.nodes('S') AS Split(A)
)
SELECT LTRIM(RTRIM(SUBSTRING(data,0,CHARINDEX('^',data)))) AS column1,
LTRIM(RTRIM(SUBSTRING(data,CHARINDEX('^',data)+1,LEN (data)))) AS column2
FROM CTE
Result
column1 column2
-------------------
Mak 1
Jak 2
Smith 3
Lee 4
Joseph 5
Use the above script create table valued parameter function
CREATE FUNCTION [dbo].[udf_SplitFun](#InputData VARCHAR(MAX))
RETURNS #Return TABLE ( column1 VARCHAR(200),column2 INT)
AS
BEGIN
DECLARE #TempTable AS TABLE
(
data VARCHAR(MAX)
)
INSERT INTO #TempTable
SELECT #InputData
;WITH CTE
AS
(
SELECT Split.A.value('.','nvarchar(1000)') AS data
FROM
(
SELECT CAST('<S>'+REPLACE(data,',','</S><S>')+'</S>' AS XML ) AS data
FROM #TempTable
)AS A
CROSS APPLY data.nodes('S') AS Split(A)
)
INSERT INTO #Return(column1,column2)
SELECT LTRIM(RTRIM(SUBSTRING(data,0,CHARINDEX('^',data)))) AS column1,
LTRIM(RTRIM(SUBSTRING(data,CHARINDEX('^',data)+1,LEN (data)))) AS column2
FROM CTE
RETURN;
END
Execute the Function like below
DECLARE #InputData VARCHAR(MAX) = 'Mak^1,Jak^2,Smith^3,Lee^4,Joseph^5';
SELECT * FROM [dbo].[udf_SplitFun] (#InputData)
GO

You may use that split function another time to split each line by caret. Like:
SELECT SplitByCaret1.Item, SplitByCaret2.Item
FROM [dbo].[udf_Split](#String ,',') SplitByComma
CROSS APPLY (SELECT * FROM [dbo].[udf_Split](SplitByComma.Item ,'^') Splitted WHERE Splitted.ID=1) SplitByCaret1
CROSS APPLY (SELECT * FROM [dbo].[udf_Split](SplitByComma.Item ,'^') Splitted WHERE Splitted.ID=2) SplitByCaret2

Related

Split data from colon into new column

I have got a scenario where I want to split my data from colon into new column.
E.g YR:136;YR:50;JN:275;YM:138;IN:477;WO:150;G1:10;F2:10.
What I am looking for is this:
YR 136
YR 50
JN 275
YM 138
Can anybody tell me how to achieve this? Any help would be much appreciated.
for SQL 2016 and later you can use STRING_SPLIT. If not, search for CSV Spliter
declare #str varchar(100) = 'YR:136;YR:50;JN:275;YM:138;IN:477;WO:150;G1:10;F2:10'
select parsename(replace(value, ':', '.'), 2),
parsename(replace(value, ':', '.'), 1)
from string_split(#str, ';')
Hope the string will in the specified format every time. What I did was, first copied the same string to a new variable and added an extra ; to the end. Then used a WHILE loop to split each part before each ;. Then used LEFT and RIGHT function to take the parts before and after : from the splitted part. And inserted those values to a table variable #t.
Code
declare #str as varchar(max) ='YR:136;YR:50;JN:275;YM:138;IN:477;WO:150;G1:10;F2:10';
declare #rows as int;
set #rows = len(#str) - len(replace(#str, ';',''));
declare #t as table([col1] varchar(100), [col2] int);
declare #i as int;
set #i = 0;
declare #str2 as varchar(max);
set #str2 = #str + ';';
while(#i <= #rows)
begin
declare #col as varchar(100);
set #col = left(#str2, charindex(';', #str2, 1) - 1);
set #str2 = right(#str2, len(#str2) - charindex(';', #str2, 1));
insert into #t([col1], [col2])
select
left(#col, charindex(':', #col, 1) - 1),
right(#col, charindex(':', reverse(#col), 1) - 1);
set #i += 1;
end
select * from #t;
Find a demo here
First CREATE the function:
CREATE FUNCTION dbo.fnSplit(
#sInputList VARCHAR(8000) -- List of delimited items
, #sDelimiter VARCHAR(8000) = ',' -- delimiter that separates items
) RETURNS #List TABLE (item VARCHAR(8000))
BEGIN
DECLARE #sItem VARCHAR(8000)
WHILE CHARINDEX(#sDelimiter,#sInputList,0) <> 0
BEGIN
SELECT
#sItem=RTRIM(LTRIM(SUBSTRING(#sInputList,1,CHARINDEX(#sDelimiter,#sInputList,0)-1))),
#sInputList=RTRIM(LTRIM(SUBSTRING(#sInputList,CHARINDEX(#sDelimiter,#sInputList,0)+LEN(#sDelimiter),LEN(#sInputList))))
IF LEN(#sItem) > 0
INSERT INTO #List SELECT #sItem
END
IF LEN(#sInputList) > 0
INSERT INTO #List SELECT #sInputList -- Put the last item in
RETURN
END
GO
Then try this:
DECLARE #X VARCHAR(100), #Val VARCHAR(250) = 'YR:136;YR:50;JN:275;YM:138;IN:477;WO:150;G1:10;F2:10'
SELECT LEFT(value,CHARINDEX(':',value) - 1)A
,RIGHT(value,CHARINDEX(':',REVERSE(value)) - 1)B
FROM dbo.fn_Split(#Val,';')
Output:
A B
YR 136
YR 50
JN 275
YM 138
IN 477
WO 150
G1 10
F2 10
In SQL Server you can do that using STRING_SPLIT.
For more information read the documentation:
https://learn.microsoft.com/en-us/sql/t-sql/functions/string-split-transact-sql?view=sql-server-2017
--FOR SQL SERVER 2016+
DECLARE #str VARCHAR(100) = 'YR:136;YR:50;JN:275;YM:138;IN:477;WO:150;G1:10;F2:10'
SELECT SUBSTRING(value,0,CHARINDEX(':',value,0)) Id
, SUBSTRING(value,CHARINDEX(':',value,0)+1,100) Vals
FROM STRING_SPLIT(#str, ';')
--FOR Older SQL SERVER
DECLARE #x AS XML=''
SET #x = CAST('<A>'+ REPLACE(#str,';','</A><A>')+ '</A>' AS XML)
SELECT SUBSTRING(t.value('.', 'VARCHAR(MAX)') ,0,CHARINDEX(':',t.value('.', 'VARCHAR(MAX)') ,0)) Id
, SUBSTRING(t.value('.', 'VARCHAR(MAX)') ,CHARINDEX(':',t.value('.', 'VARCHAR(MAX)') ,0)+1,100) Vals
FROM #x.nodes('/A') AS x(t)
Try this below function
CREATE FUNCTION [dbo].[udf_GetUnsplitGivenData]
(
#string nvarchar(max)
)
RETURNS #OutTable TABLE
(
COl1 nvarchar(max),
COl2 nvarchar(max)
)
AS
BEGIN
DECLARE #Temp AS TABLE
(
DATA nvarchar(max)
)
INSERT INTO #Temp
SELECT #string
INSERT INTO #OutTable(COl1,COl2)
SELECT SUBSTRING(Data,1,CHARINDEX(':',Data )-1) AS COl1,
SUBSTRING(Data,CHARINDEX(':',Data )+1,LEN(Data)) AS COl2
FROM
(
SELECT Split.a.value('.','nvarchar(100)') AS Data
FROM
(
SELECT
CAST( '<S>'+REPLACE(Data,';','</S><S>')+'</S>' AS XML ) AS Data
FROM #Temp
) AS A
CROSS APPLY data.nodes('S') AS Split(a)
)dt
RETURN
END
SELECT * FROM [dbo].[udf_GetUnsplitGivenData] (
'YR:136;YR:50;JN:275;YM:138;IN:477;WO:150;G1:10;F2:10')
GO
Result
COl1 COl2
------------
YR 136
YR 50
JN 275
YM 138
IN 477
WO 150
G1 10
F2 10

How to extract words from a sentence in SQL Server? [duplicate]

I have a SQL Server 2008 R2 column containing a string which I need to split by a comma. I have seen many answers on StackOverflow but none of them works in R2. I have made sure I have select permissions on any split function examples. Any help greatly appreciated.
I've used this SQL before which may work for you:-
CREATE FUNCTION dbo.splitstring ( #stringToSplit VARCHAR(MAX) )
RETURNS
#returnList TABLE ([Name] [nvarchar] (500))
AS
BEGIN
DECLARE #name NVARCHAR(255)
DECLARE #pos INT
WHILE CHARINDEX(',', #stringToSplit) > 0
BEGIN
SELECT #pos = CHARINDEX(',', #stringToSplit)
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos-1)
INSERT INTO #returnList
SELECT #name
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, LEN(#stringToSplit)-#pos)
END
INSERT INTO #returnList
SELECT #stringToSplit
RETURN
END
and to use it:-
SELECT * FROM dbo.splitstring('91,12,65,78,56,789')
Instead of recursive CTEs and while loops, has anyone considered a more set-based approach? Note that this function was written for the question, which was based on SQL Server 2008 and comma as the delimiter. In SQL Server 2016 and above (and in compatibility level 130 and above), STRING_SPLIT() is a better option.
CREATE FUNCTION dbo.SplitString
(
#List nvarchar(max),
#Delim nvarchar(255)
)
RETURNS TABLE
AS
RETURN ( SELECT [Value] FROM
(
SELECT [Value] = LTRIM(RTRIM(SUBSTRING(#List, [Number],
CHARINDEX(#Delim, #List + #Delim, [Number]) - [Number])))
FROM (SELECT Number = ROW_NUMBER() OVER (ORDER BY name)
FROM sys.all_columns) AS x WHERE Number <= LEN(#List)
AND SUBSTRING(#Delim + #List, [Number], DATALENGTH(#Delim)/2) = #Delim
) AS y
);
GO
If you want to avoid the limitation of the length of the string being <= the number of rows in sys.all_columns (9,980 in model in SQL Server 2017; much higher in your own user databases), you can use other approaches for deriving the numbers, such as building your own table of numbers. You could also use a recursive CTE in cases where you can't use system tables or create your own:
CREATE FUNCTION dbo.SplitString
(
#List nvarchar(max),
#Delim nvarchar(255)
)
RETURNS TABLE WITH SCHEMABINDING
AS
RETURN ( WITH n(n) AS (SELECT 1 UNION ALL SELECT n+1
FROM n WHERE n <= LEN(#List))
SELECT [Value] = SUBSTRING(#List, n,
CHARINDEX(#Delim, #List + #Delim, n) - n)
FROM n WHERE n <= LEN(#List)
AND SUBSTRING(#Delim + #List, n, DATALENGTH(#Delim)/2) = #Delim
);
GO
But you'll have to append OPTION (MAXRECURSION 0) (or MAXRECURSION <longest possible string length if < 32768>) to the outer query in order to avoid errors with recursion for strings > 100 characters. If that is also not a good alternative then see this answer as pointed out in the comments, or this answer if you need an ordered split string function.
(Also, the delimiter will have to be NCHAR(<=1228). Still researching why.)
More on split functions, why (and proof that) while loops and recursive CTEs don't scale, and better alternatives, if you're splitting strings coming from the application layer:
Splitting strings
Finally the wait is over in SQL Server 2016 they have introduced Split string function : STRING_SPLIT
select * From STRING_SPLIT ('a,b', ',') cs
All the other methods to split string like XML, Tally table, while loop, etc.. has been blown away by this STRING_SPLIT function.
Here is an excellent article with performance comparison : Performance Surprises and Assumptions : STRING_SPLIT
The easiest way to do this is by using XML format.
1. Converting string to rows without table
QUERY
DECLARE #String varchar(100) = 'String1,String2,String3'
-- To change ',' to any other delimeter, just change ',' to your desired one
DECLARE #Delimiter CHAR = ','
SELECT LTRIM(RTRIM(Split.a.value('.', 'VARCHAR(100)'))) 'Value'
FROM
(
SELECT CAST ('<M>' + REPLACE(#String, #Delimiter, '</M><M>') + '</M>' AS XML) AS Data
) AS A
CROSS APPLY Data.nodes ('/M') AS Split(a)
RESULT
x---------x
| Value |
x---------x
| String1 |
| String2 |
| String3 |
x---------x
2. Converting to rows from a table which have an ID for each CSV row
SOURCE TABLE
x-----x--------------------------x
| Id | Value |
x-----x--------------------------x
| 1 | String1,String2,String3 |
| 2 | String4,String5,String6 |
x-----x--------------------------x
QUERY
-- To change ',' to any other delimeter, just change ',' before '</M><M>' to your desired one
DECLARE #Delimiter CHAR = ','
SELECT ID,LTRIM(RTRIM(Split.a.value('.', 'VARCHAR(100)'))) 'Value'
FROM
(
SELECT ID,CAST ('<M>' + REPLACE(VALUE, #Delimiter, '</M><M>') + '</M>' AS XML) AS Data
FROM TABLENAME
) AS A
CROSS APPLY Data.nodes ('/M') AS Split(a)
RESULT
x-----x----------x
| Id | Value |
x-----x----------x
| 1 | String1 |
| 1 | String2 |
| 1 | String3 |
| 2 | String4 |
| 2 | String5 |
| 2 | String6 |
x-----x----------x
I needed a quick way to get rid of the +4 from a zip code.
UPDATE #Emails
SET ZIPCode = SUBSTRING(ZIPCode, 1, (CHARINDEX('-', ZIPCODE)-1))
WHERE ZIPCode LIKE '%-%'
No proc... no UDF... just one tight little inline command that does what it must. Not fancy, not elegant.
Change the delimiter as needed, etc, and it will work for anything.
if you replace
WHILE CHARINDEX(',', #stringToSplit) > 0
with
WHILE LEN(#stringToSplit) > 0
you can eliminate that last insert after the while loop!
CREATE FUNCTION dbo.splitstring ( #stringToSplit VARCHAR(MAX) )
RETURNS
#returnList TABLE ([Name] [nvarchar] (500))
AS
BEGIN
DECLARE #name NVARCHAR(255)
DECLARE #pos INT
WHILE LEN(#stringToSplit) > 0
BEGIN
SELECT #pos = CHARINDEX(',', #stringToSplit)
if #pos = 0
SELECT #pos = LEN(#stringToSplit)
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos-1)
INSERT INTO #returnList
SELECT #name
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, LEN(#stringToSplit)-#pos)
END
RETURN
END
The often used approach with XML elements breaks in case of forbidden characters. This is an approach to use this method with any kind of character, even with the semicolon as delimiter.
The trick is, first to use SELECT SomeString AS [*] FOR XML PATH('') to get all forbidden characters properly escaped. That's the reason, why I replace the delimiter to a magic value to avoid troubles with ; as delimiter.
DECLARE #Dummy TABLE (ID INT, SomeTextToSplit NVARCHAR(MAX))
INSERT INTO #Dummy VALUES
(1,N'A&B;C;D;E, F')
,(2,N'"C" & ''D'';<C>;D;E, F');
DECLARE #Delimiter NVARCHAR(10)=';'; --special effort needed (due to entities coding with "&code;")!
WITH Casted AS
(
SELECT *
,CAST(N'<x>' + REPLACE((SELECT REPLACE(SomeTextToSplit,#Delimiter,N'§§Split$me$here§§') AS [*] FOR XML PATH('')),N'§§Split$me$here§§',N'</x><x>') + N'</x>' AS XML) AS SplitMe
FROM #Dummy
)
SELECT Casted.ID
,x.value(N'.',N'nvarchar(max)') AS Part
FROM Casted
CROSS APPLY SplitMe.nodes(N'/x') AS A(x)
The result
ID Part
1 A&B
1 C
1 D
1 E, F
2 "C" & 'D'
2 <C>
2 D
2 E, F
All the functions for string splitting that use some kind of Loop-ing (iterations) have bad performance. They should be replaced with set-based solution.
This code executes excellent.
CREATE FUNCTION dbo.SplitStrings
(
#List NVARCHAR(MAX),
#Delimiter NVARCHAR(255)
)
RETURNS TABLE
WITH SCHEMABINDING
AS
RETURN
(
SELECT Item = y.i.value('(./text())[1]', 'nvarchar(4000)')
FROM
(
SELECT x = CONVERT(XML, '<i>'
+ REPLACE(#List, #Delimiter, '</i><i>')
+ '</i>').query('.')
) AS a CROSS APPLY x.nodes('i') AS y(i)
);
GO
I had to write something like this recently. Here's the solution I came up with. It's generalized for any delimiter string and I think it would perform slightly better:
CREATE FUNCTION [dbo].[SplitString]
( #string nvarchar(4000)
, #delim nvarchar(100) )
RETURNS
#result TABLE
( [Value] nvarchar(4000) NOT NULL
, [Index] int NOT NULL )
AS
BEGIN
DECLARE #str nvarchar(4000)
, #pos int
, #prv int = 1
SELECT #pos = CHARINDEX(#delim, #string)
WHILE #pos > 0
BEGIN
SELECT #str = SUBSTRING(#string, #prv, #pos - #prv)
INSERT INTO #result SELECT #str, #prv
SELECT #prv = #pos + LEN(#delim)
, #pos = CHARINDEX(#delim, #string, #pos + 1)
END
INSERT INTO #result SELECT SUBSTRING(#string, #prv, 4000), #prv
RETURN
END
If you need a quick ad-hoc solution for common cases with minimum code, then this recursive CTE two-liner will do it:
DECLARE #s VARCHAR(200) = ',1,2,,3,,,4,,,,5,'
;WITH
a AS (SELECT i=-1, j=0 UNION ALL SELECT j, CHARINDEX(',', #s, j + 1) FROM a WHERE j > i),
b AS (SELECT SUBSTRING(#s, i+1, IIF(j>0, j, LEN(#s)+1)-i-1) s FROM a WHERE i >= 0)
SELECT * FROM b
Either use this as a stand-alone statement or just add the above CTEs to any of your queries and you will be able to join the resulting table b with others for use in any further expressions.
edit (by Shnugo)
If you add a counter, you will get a position index together with the List:
DECLARE #s VARCHAR(200) = '1,2333,344,4'
;WITH
a AS (SELECT n=0, i=-1, j=0 UNION ALL SELECT n+1, j, CHARINDEX(',', #s, j+1) FROM a WHERE j > i),
b AS (SELECT n, SUBSTRING(#s, i+1, IIF(j>0, j, LEN(#s)+1)-i-1) s FROM a WHERE i >= 0)
SELECT * FROM b;
The result:
n s
1 1
2 2333
3 344
4 4
I take the xml route by wrapping the values into elements (M but anything works):
declare #v nvarchar(max) = '100,201,abcde'
select
a.value('.', 'varchar(max)')
from
(select cast('<M>' + REPLACE(#v, ',', '</M><M>') + '</M>' AS XML) as col) as A
CROSS APPLY A.col.nodes ('/M') AS Split(a)
A solution using a CTE, if anyone should need that (apart from me, who obviously did, that is why I wrote it).
declare #StringToSplit varchar(100) = 'Test1,Test2,Test3';
declare #SplitChar varchar(10) = ',';
with StringToSplit as (
select
ltrim( rtrim( substring( #StringToSplit, 1, charindex( #SplitChar, #StringToSplit ) - 1 ) ) ) Head
, substring( #StringToSplit, charindex( #SplitChar, #StringToSplit ) + 1, len( #StringToSplit ) ) Tail
union all
select
ltrim( rtrim( substring( Tail, 1, charindex( #SplitChar, Tail ) - 1 ) ) ) Head
, substring( Tail, charindex( #SplitChar, Tail ) + 1, len( Tail ) ) Tail
from StringToSplit
where charindex( #SplitChar, Tail ) > 0
union all
select
ltrim( rtrim( Tail ) ) Head
, '' Tail
from StringToSplit
where charindex( #SplitChar, Tail ) = 0
and len( Tail ) > 0
)
select Head from StringToSplit
This is more narrowly-tailored. When I do this I usually have a comma-delimited list of unique ids (INT or BIGINT), which I want to cast as a table to use as an inner join to another table that has a primary key of INT or BIGINT. I want an in-line table-valued function returned so that I have the most efficient join possible.
Sample usage would be:
DECLARE #IDs VARCHAR(1000);
SET #IDs = ',99,206,124,8967,1,7,3,45234,2,889,987979,';
SELECT me.Value
FROM dbo.MyEnum me
INNER JOIN dbo.GetIntIdsTableFromDelimitedString(#IDs) ids ON me.PrimaryKey = ids.ID
I stole the idea from http://sqlrecords.blogspot.com/2012/11/converting-delimited-list-to-table.html, changing it to be in-line table-valued and cast as INT.
create function dbo.GetIntIDTableFromDelimitedString
(
#IDs VARCHAR(1000) --this parameter must start and end with a comma, eg ',123,456,'
--all items in list must be perfectly formatted or function will error
)
RETURNS TABLE AS
RETURN
SELECT
CAST(SUBSTRING(#IDs,Nums.number + 1,CHARINDEX(',',#IDs,(Nums.number+2)) - Nums.number - 1) AS INT) AS ID
FROM
[master].[dbo].[spt_values] Nums
WHERE Nums.Type = 'P'
AND Nums.number BETWEEN 1 AND DATALENGTH(#IDs)
AND SUBSTRING(#IDs,Nums.number,1) = ','
AND CHARINDEX(',',#IDs,(Nums.number+1)) > Nums.number;
GO
There is a correct version on here but I thought it would be nice to add a little fault tolerance in case they have a trailing comma as well as make it so you could use it not as a function but as part of a larger piece of code. Just in case you're only using it once time and don't need a function. This is also for integers (which is what I needed it for) so you might have to change your data types.
DECLARE #StringToSeperate VARCHAR(10)
SET #StringToSeperate = '1,2,5'
--SELECT #StringToSeperate IDs INTO #Test
DROP TABLE #IDs
CREATE TABLE #IDs (ID int)
DECLARE #CommaSeperatedValue NVARCHAR(255) = ''
DECLARE #Position INT = LEN(#StringToSeperate)
--Add Each Value
WHILE CHARINDEX(',', #StringToSeperate) > 0
BEGIN
SELECT #Position = CHARINDEX(',', #StringToSeperate)
SELECT #CommaSeperatedValue = SUBSTRING(#StringToSeperate, 1, #Position-1)
INSERT INTO #IDs
SELECT #CommaSeperatedValue
SELECT #StringToSeperate = SUBSTRING(#StringToSeperate, #Position+1, LEN(#StringToSeperate)-#Position)
END
--Add Last Value
IF (LEN(LTRIM(RTRIM(#StringToSeperate)))>0)
BEGIN
INSERT INTO #IDs
SELECT SUBSTRING(#StringToSeperate, 1, #Position)
END
SELECT * FROM #IDs
I modified +Andy Robinson's function a little bit. Now you can select only required part from returning table:
CREATE FUNCTION dbo.splitstring ( #stringToSplit VARCHAR(MAX) )
RETURNS
#returnList TABLE ([numOrder] [tinyint] , [Name] [nvarchar] (500)) AS
BEGIN
DECLARE #name NVARCHAR(255)
DECLARE #pos INT
DECLARE #orderNum INT
SET #orderNum=0
WHILE CHARINDEX('.', #stringToSplit) > 0
BEGIN
SELECT #orderNum=#orderNum+1;
SELECT #pos = CHARINDEX('.', #stringToSplit)
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos-1)
INSERT INTO #returnList
SELECT #orderNum,#name
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, LEN(#stringToSplit)-#pos)
END
SELECT #orderNum=#orderNum+1;
INSERT INTO #returnList
SELECT #orderNum, #stringToSplit
RETURN
END
Usage:
SELECT Name FROM dbo.splitstring('ELIS.YD.CRP1.1.CBA.MDSP.T389.BT') WHERE numOrder=5
Simples
DECLARE #String varchar(100) = '11,21,84,85,87'
SELECT * FROM TB_PAPEL WHERE CD_PAPEL IN (SELECT value FROM STRING_SPLIT(#String, ','))
-- EQUIVALENTE
SELECT * FROM TB_PAPEL WHERE CD_PAPEL IN (11,21,84,85,87)
here is a version that can split on a pattern using patindex, a simple adaptation of the post above. I had a case where I needed to split a string that contained multiple separator chars.
alter FUNCTION dbo.splitstring ( #stringToSplit VARCHAR(1000), #splitPattern varchar(10) )
RETURNS
#returnList TABLE ([Name] [nvarchar] (500))
AS
BEGIN
DECLARE #name NVARCHAR(255)
DECLARE #pos INT
WHILE PATINDEX(#splitPattern, #stringToSplit) > 0
BEGIN
SELECT #pos = PATINDEX(#splitPattern, #stringToSplit)
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos-1)
INSERT INTO #returnList
SELECT #name
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, LEN(#stringToSplit)-#pos)
END
INSERT INTO #returnList
SELECT #stringToSplit
RETURN
END
select * from dbo.splitstring('stringa/stringb/x,y,z','%[/,]%');
result looks like this
stringa
stringb
x
y
z
Personnaly I use this function :
ALTER FUNCTION [dbo].[CUST_SplitString]
(
#String NVARCHAR(4000),
#Delimiter NCHAR(1)
)
RETURNS TABLE
AS
RETURN
(
WITH Split(stpos,endpos)
AS(
SELECT 0 AS stpos, CHARINDEX(#Delimiter,#String) AS endpos
UNION ALL
SELECT endpos+1, CHARINDEX(#Delimiter,#String,endpos+1)
FROM Split
WHERE endpos > 0
)
SELECT 'Id' = ROW_NUMBER() OVER (ORDER BY (SELECT 1)),
'Data' = SUBSTRING(#String,stpos,COALESCE(NULLIF(endpos,0),LEN(#String)+1)-stpos)
FROM Split
)
I have developed a double Splitter (Takes two split characters) as requested Here. Could be of some value in this thread seeing its the most referenced for queries relating to string splitting.
CREATE FUNCTION uft_DoubleSplitter
(
-- Add the parameters for the function here
#String VARCHAR(4000),
#Splitter1 CHAR,
#Splitter2 CHAR
)
RETURNS #Result TABLE (Id INT,MId INT,SValue VARCHAR(4000))
AS
BEGIN
DECLARE #FResult TABLE(Id INT IDENTITY(1, 1),
SValue VARCHAR(4000))
DECLARE #SResult TABLE(Id INT IDENTITY(1, 1),
MId INT,
SValue VARCHAR(4000))
SET #String = #String+#Splitter1
WHILE CHARINDEX(#Splitter1, #String) > 0
BEGIN
DECLARE #WorkingString VARCHAR(4000) = NULL
SET #WorkingString = SUBSTRING(#String, 1, CHARINDEX(#Splitter1, #String) - 1)
--Print #workingString
INSERT INTO #FResult
SELECT CASE
WHEN #WorkingString = '' THEN NULL
ELSE #WorkingString
END
SET #String = SUBSTRING(#String, LEN(#WorkingString) + 2, LEN(#String))
END
IF ISNULL(#Splitter2, '') != ''
BEGIN
DECLARE #OStartLoop INT
DECLARE #OEndLoop INT
SELECT #OStartLoop = MIN(Id),
#OEndLoop = MAX(Id)
FROM #FResult
WHILE #OStartLoop <= #OEndLoop
BEGIN
DECLARE #iString VARCHAR(4000)
DECLARE #iMId INT
SELECT #iString = SValue+#Splitter2,
#iMId = Id
FROM #FResult
WHERE Id = #OStartLoop
WHILE CHARINDEX(#Splitter2, #iString) > 0
BEGIN
DECLARE #iWorkingString VARCHAR(4000) = NULL
SET #IWorkingString = SUBSTRING(#iString, 1, CHARINDEX(#Splitter2, #iString) - 1)
INSERT INTO #SResult
SELECT #iMId,
CASE
WHEN #iWorkingString = '' THEN NULL
ELSE #iWorkingString
END
SET #iString = SUBSTRING(#iString, LEN(#iWorkingString) + 2, LEN(#iString))
END
SET #OStartLoop = #OStartLoop + 1
END
INSERT INTO #Result
SELECT MId AS PrimarySplitID,
ROW_NUMBER() OVER (PARTITION BY MId ORDER BY Mid, Id) AS SecondarySplitID ,
SValue
FROM #SResult
END
ELSE
BEGIN
INSERT INTO #Result
SELECT Id AS PrimarySplitID,
NULL AS SecondarySplitID,
SValue
FROM #FResult
END
RETURN
Usage:
--FirstSplit
SELECT * FROM uft_DoubleSplitter('ValueA=ValueB=ValueC=ValueD==ValueE&ValueA=ValueB=ValueC===ValueE&ValueA=ValueB==ValueD===','&',NULL)
--Second Split
SELECT * FROM uft_DoubleSplitter('ValueA=ValueB=ValueC=ValueD==ValueE&ValueA=ValueB=ValueC===ValueE&ValueA=ValueB==ValueD===','&','=')
Possible Usage (Get second value of each split):
SELECT fn.SValue
FROM uft_DoubleSplitter('ValueA=ValueB=ValueC=ValueD==ValueE&ValueA=ValueB=ValueC===ValueE&ValueA=ValueB==ValueD===', '&', '=')AS fn
WHERE fn.mid = 2
A recursive cte based solution
declare #T table (iden int identity, col1 varchar(100));
insert into #T(col1) values
('ROOT/South America/Lima/Test/Test2')
, ('ROOT/South America/Peru/Test/Test2')
, ('ROOT//South America/Venuzuala ')
, ('RtT/South America / ')
, ('ROOT/South Americas// ');
declare #split char(1) = '/';
select #split as split;
with cte as
( select t.iden, case when SUBSTRING(REVERSE(rtrim(t.col1)), 1, 1) = #split then LTRIM(RTRIM(t.col1)) else LTRIM(RTRIM(t.col1)) + #split end as col1, 0 as pos , 1 as cnt
from #T t
union all
select t.iden, t.col1 , charindex(#split, t.col1, t.pos + 1), cnt + 1
from cte t
where charindex(#split, t.col1, t.pos + 1) > 0
)
select t1.*, t2.pos, t2.cnt
, ltrim(rtrim(SUBSTRING(t1.col1, t1.pos+1, t2.pos-t1.pos-1))) as bingo
from cte t1
join cte t2
on t2.iden = t1.iden
and t2.cnt = t1.cnt+1
and t2.pos > t1.pos
order by t1.iden, t1.cnt;
With all due respect to #AviG this is the bug free version of function deviced by him to return all the tokens in full.
IF EXISTS (SELECT * FROM sys.objects WHERE type = 'TF' AND name = 'TF_SplitString')
DROP FUNCTION [dbo].[TF_SplitString]
GO
-- =============================================
-- Author: AviG
-- Amendments: Parameterize the delimeter and included the missing chars in last token - Gemunu Wickremasinghe
-- Description: Tabel valued function that Breaks the delimeted string by given delimeter and returns a tabel having split results
-- Usage
-- select * from [dbo].[TF_SplitString]('token1,token2,,,,,,,,token969',',')
-- 969 items should be returned
-- select * from [dbo].[TF_SplitString]('4672978261,4672978255',',')
-- 2 items should be returned
-- =============================================
CREATE FUNCTION dbo.TF_SplitString
( #stringToSplit VARCHAR(MAX) ,
#delimeter char = ','
)
RETURNS
#returnList TABLE ([Name] [nvarchar] (500))
AS
BEGIN
DECLARE #name NVARCHAR(255)
DECLARE #pos INT
WHILE LEN(#stringToSplit) > 0
BEGIN
SELECT #pos = CHARINDEX(#delimeter, #stringToSplit)
if #pos = 0
BEGIN
SELECT #pos = LEN(#stringToSplit)
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos)
END
else
BEGIN
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos-1)
END
INSERT INTO #returnList
SELECT #name
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, LEN(#stringToSplit)-#pos)
END
RETURN
END
This is based on Andy Robertson's answer, I needed a delimiter other than comma.
CREATE FUNCTION dbo.splitstring ( #stringToSplit nvarchar(MAX), #delim nvarchar(max))
RETURNS
#returnList TABLE ([value] [nvarchar] (MAX))
AS
BEGIN
DECLARE #value NVARCHAR(max)
DECLARE #pos INT
WHILE CHARINDEX(#delim, #stringToSplit) > 0
BEGIN
SELECT #pos = CHARINDEX(#delim, #stringToSplit)
SELECT #value = SUBSTRING(#stringToSplit, 1, #pos - 1)
INSERT INTO #returnList
SELECT #value
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos + LEN(#delim), LEN(#stringToSplit) - #pos)
END
INSERT INTO #returnList
SELECT #stringToSplit
RETURN
END
GO
And to use it:
SELECT * FROM dbo.splitstring('test1 test2 test3', ' ');
(Tested on SQL Server 2008 R2)
EDIT: correct test code
ALTER FUNCTION [dbo].func_split_string
(
#input as varchar(max),
#delimiter as varchar(10) = ";"
)
RETURNS #result TABLE
(
id smallint identity(1,1),
csv_value varchar(max) not null
)
AS
BEGIN
DECLARE #pos AS INT;
DECLARE #string AS VARCHAR(MAX) = '';
WHILE LEN(#input) > 0
BEGIN
SELECT #pos = CHARINDEX(#delimiter,#input);
IF(#pos<=0)
select #pos = len(#input)
IF(#pos <> LEN(#input))
SELECT #string = SUBSTRING(#input, 1, #pos-1);
ELSE
SELECT #string = SUBSTRING(#input, 1, #pos);
INSERT INTO #result SELECT #string
SELECT #input = SUBSTRING(#input, #pos+len(#delimiter), LEN(#input)-#pos)
END
RETURN
END
You can Use this function:
CREATE FUNCTION SplitString
(
#Input NVARCHAR(MAX),
#Character CHAR(1)
)
RETURNS #Output TABLE (
Item NVARCHAR(1000)
)
AS
BEGIN
DECLARE #StartIndex INT, #EndIndex INT
SET #StartIndex = 1
IF SUBSTRING(#Input, LEN(#Input) - 1, LEN(#Input)) <> #Character
BEGIN
SET #Input = #Input + #Character
END
WHILE CHARINDEX(#Character, #Input) > 0
BEGIN
SET #EndIndex = CHARINDEX(#Character, #Input)
INSERT INTO #Output(Item)
SELECT SUBSTRING(#Input, #StartIndex, #EndIndex - 1)
SET #Input = SUBSTRING(#Input, #EndIndex + 1, LEN(#Input))
END
RETURN
END
GO
Here is an example that you can use as function or also you can put the same logic in procedure.
--SELECT * from [dbo].fn_SplitString ;
CREATE FUNCTION [dbo].[fn_SplitString]
(#CSV VARCHAR(MAX), #Delimeter VARCHAR(100) = ',')
RETURNS #retTable TABLE
(
[value] VARCHAR(MAX) NULL
)AS
BEGIN
DECLARE
#vCSV VARCHAR (MAX) = #CSV,
#vDelimeter VARCHAR (100) = #Delimeter;
IF #vDelimeter = ';'
BEGIN
SET #vCSV = REPLACE(#vCSV, ';', '~!~#~');
SET #vDelimeter = REPLACE(#vDelimeter, ';', '~!~#~');
END;
SET #vCSV = REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(#vCSV, '&', '&'), '<', '<'), '>', '>'), '''', '&apos;'), '"', '"');
DECLARE #xml XML;
SET #xml = '<i>' + REPLACE(#vCSV, #vDelimeter, '</i><i>') + '</i>';
INSERT INTO #retTable
SELECT
x.i.value('.', 'varchar(max)') AS COLUMNNAME
FROM #xml.nodes('//i')AS x(i);
RETURN;
END;
/*
Answer to T-SQL split string
Based on answers from Andy Robinson and AviG
Enhanced functionality ref: LEN function not including trailing spaces in SQL Server
This 'file' should be valid as both a markdown file and an SQL file
*/
CREATE FUNCTION dbo.splitstring ( --CREATE OR ALTER
#stringToSplit NVARCHAR(MAX)
) RETURNS #returnList TABLE ([Item] NVARCHAR (MAX))
AS BEGIN
DECLARE #name NVARCHAR(MAX)
DECLARE #pos BIGINT
SET #stringToSplit = #stringToSplit + ',' -- this should allow entries that end with a `,` to have a blank value in that "column"
WHILE ((LEN(#stringToSplit+'_') > 1)) BEGIN -- `+'_'` gets around LEN trimming terminal spaces. See URL referenced above
SET #pos = COALESCE(NULLIF(CHARINDEX(',', #stringToSplit),0),LEN(#stringToSplit+'_')) -- COALESCE grabs first non-null value
SET #name = SUBSTRING(#stringToSplit, 1, #pos-1) --MAX size of string of type nvarchar is 4000
SET #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, 4000) -- With SUBSTRING fn (MS web): "If start is greater than the number of characters in the value expression, a zero-length expression is returned."
INSERT INTO #returnList SELECT #name --additional debugging parameters below can be added
-- + ' pos:' + CAST(#pos as nvarchar) + ' remain:''' + #stringToSplit + '''(' + CAST(LEN(#stringToSplit+'_')-1 as nvarchar) + ')'
END
RETURN
END
GO
/*
Test cases: see URL referenced as "enhanced functionality" above
SELECT *,LEN(Item+'_')-1 'L' from splitstring('a,,b')
Item | L
--- | ---
a | 1
| 0
b | 1
SELECT *,LEN(Item+'_')-1 'L' from splitstring('a,,')
Item | L
--- | ---
a | 1
| 0
| 0
SELECT *,LEN(Item+'_')-1 'L' from splitstring('a,, ')
Item | L
--- | ---
a | 1
| 0
| 1
SELECT *,LEN(Item+'_')-1 'L' from splitstring('a,, c ')
Item | L
--- | ---
a | 1
| 0
c | 3
*/
The easiest way:
Install SQL Server 2016
Use STRING_SPLIT https://msdn.microsoft.com/en-us/library/mt684588.aspx
It works even in express edition :).

T-SQL split string

I have a SQL Server 2008 R2 column containing a string which I need to split by a comma. I have seen many answers on StackOverflow but none of them works in R2. I have made sure I have select permissions on any split function examples. Any help greatly appreciated.
I've used this SQL before which may work for you:-
CREATE FUNCTION dbo.splitstring ( #stringToSplit VARCHAR(MAX) )
RETURNS
#returnList TABLE ([Name] [nvarchar] (500))
AS
BEGIN
DECLARE #name NVARCHAR(255)
DECLARE #pos INT
WHILE CHARINDEX(',', #stringToSplit) > 0
BEGIN
SELECT #pos = CHARINDEX(',', #stringToSplit)
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos-1)
INSERT INTO #returnList
SELECT #name
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, LEN(#stringToSplit)-#pos)
END
INSERT INTO #returnList
SELECT #stringToSplit
RETURN
END
and to use it:-
SELECT * FROM dbo.splitstring('91,12,65,78,56,789')
Instead of recursive CTEs and while loops, has anyone considered a more set-based approach? Note that this function was written for the question, which was based on SQL Server 2008 and comma as the delimiter. In SQL Server 2016 and above (and in compatibility level 130 and above), STRING_SPLIT() is a better option.
CREATE FUNCTION dbo.SplitString
(
#List nvarchar(max),
#Delim nvarchar(255)
)
RETURNS TABLE
AS
RETURN ( SELECT [Value] FROM
(
SELECT [Value] = LTRIM(RTRIM(SUBSTRING(#List, [Number],
CHARINDEX(#Delim, #List + #Delim, [Number]) - [Number])))
FROM (SELECT Number = ROW_NUMBER() OVER (ORDER BY name)
FROM sys.all_columns) AS x WHERE Number <= LEN(#List)
AND SUBSTRING(#Delim + #List, [Number], DATALENGTH(#Delim)/2) = #Delim
) AS y
);
GO
If you want to avoid the limitation of the length of the string being <= the number of rows in sys.all_columns (9,980 in model in SQL Server 2017; much higher in your own user databases), you can use other approaches for deriving the numbers, such as building your own table of numbers. You could also use a recursive CTE in cases where you can't use system tables or create your own:
CREATE FUNCTION dbo.SplitString
(
#List nvarchar(max),
#Delim nvarchar(255)
)
RETURNS TABLE WITH SCHEMABINDING
AS
RETURN ( WITH n(n) AS (SELECT 1 UNION ALL SELECT n+1
FROM n WHERE n <= LEN(#List))
SELECT [Value] = SUBSTRING(#List, n,
CHARINDEX(#Delim, #List + #Delim, n) - n)
FROM n WHERE n <= LEN(#List)
AND SUBSTRING(#Delim + #List, n, DATALENGTH(#Delim)/2) = #Delim
);
GO
But you'll have to append OPTION (MAXRECURSION 0) (or MAXRECURSION <longest possible string length if < 32768>) to the outer query in order to avoid errors with recursion for strings > 100 characters. If that is also not a good alternative then see this answer as pointed out in the comments, or this answer if you need an ordered split string function.
(Also, the delimiter will have to be NCHAR(<=1228). Still researching why.)
More on split functions, why (and proof that) while loops and recursive CTEs don't scale, and better alternatives, if you're splitting strings coming from the application layer:
Splitting strings
Finally the wait is over in SQL Server 2016 they have introduced Split string function : STRING_SPLIT
select * From STRING_SPLIT ('a,b', ',') cs
All the other methods to split string like XML, Tally table, while loop, etc.. has been blown away by this STRING_SPLIT function.
Here is an excellent article with performance comparison : Performance Surprises and Assumptions : STRING_SPLIT
The easiest way to do this is by using XML format.
1. Converting string to rows without table
QUERY
DECLARE #String varchar(100) = 'String1,String2,String3'
-- To change ',' to any other delimeter, just change ',' to your desired one
DECLARE #Delimiter CHAR = ','
SELECT LTRIM(RTRIM(Split.a.value('.', 'VARCHAR(100)'))) 'Value'
FROM
(
SELECT CAST ('<M>' + REPLACE(#String, #Delimiter, '</M><M>') + '</M>' AS XML) AS Data
) AS A
CROSS APPLY Data.nodes ('/M') AS Split(a)
RESULT
x---------x
| Value |
x---------x
| String1 |
| String2 |
| String3 |
x---------x
2. Converting to rows from a table which have an ID for each CSV row
SOURCE TABLE
x-----x--------------------------x
| Id | Value |
x-----x--------------------------x
| 1 | String1,String2,String3 |
| 2 | String4,String5,String6 |
x-----x--------------------------x
QUERY
-- To change ',' to any other delimeter, just change ',' before '</M><M>' to your desired one
DECLARE #Delimiter CHAR = ','
SELECT ID,LTRIM(RTRIM(Split.a.value('.', 'VARCHAR(100)'))) 'Value'
FROM
(
SELECT ID,CAST ('<M>' + REPLACE(VALUE, #Delimiter, '</M><M>') + '</M>' AS XML) AS Data
FROM TABLENAME
) AS A
CROSS APPLY Data.nodes ('/M') AS Split(a)
RESULT
x-----x----------x
| Id | Value |
x-----x----------x
| 1 | String1 |
| 1 | String2 |
| 1 | String3 |
| 2 | String4 |
| 2 | String5 |
| 2 | String6 |
x-----x----------x
I needed a quick way to get rid of the +4 from a zip code.
UPDATE #Emails
SET ZIPCode = SUBSTRING(ZIPCode, 1, (CHARINDEX('-', ZIPCODE)-1))
WHERE ZIPCode LIKE '%-%'
No proc... no UDF... just one tight little inline command that does what it must. Not fancy, not elegant.
Change the delimiter as needed, etc, and it will work for anything.
if you replace
WHILE CHARINDEX(',', #stringToSplit) > 0
with
WHILE LEN(#stringToSplit) > 0
you can eliminate that last insert after the while loop!
CREATE FUNCTION dbo.splitstring ( #stringToSplit VARCHAR(MAX) )
RETURNS
#returnList TABLE ([Name] [nvarchar] (500))
AS
BEGIN
DECLARE #name NVARCHAR(255)
DECLARE #pos INT
WHILE LEN(#stringToSplit) > 0
BEGIN
SELECT #pos = CHARINDEX(',', #stringToSplit)
if #pos = 0
SELECT #pos = LEN(#stringToSplit)
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos-1)
INSERT INTO #returnList
SELECT #name
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, LEN(#stringToSplit)-#pos)
END
RETURN
END
The often used approach with XML elements breaks in case of forbidden characters. This is an approach to use this method with any kind of character, even with the semicolon as delimiter.
The trick is, first to use SELECT SomeString AS [*] FOR XML PATH('') to get all forbidden characters properly escaped. That's the reason, why I replace the delimiter to a magic value to avoid troubles with ; as delimiter.
DECLARE #Dummy TABLE (ID INT, SomeTextToSplit NVARCHAR(MAX))
INSERT INTO #Dummy VALUES
(1,N'A&B;C;D;E, F')
,(2,N'"C" & ''D'';<C>;D;E, F');
DECLARE #Delimiter NVARCHAR(10)=';'; --special effort needed (due to entities coding with "&code;")!
WITH Casted AS
(
SELECT *
,CAST(N'<x>' + REPLACE((SELECT REPLACE(SomeTextToSplit,#Delimiter,N'§§Split$me$here§§') AS [*] FOR XML PATH('')),N'§§Split$me$here§§',N'</x><x>') + N'</x>' AS XML) AS SplitMe
FROM #Dummy
)
SELECT Casted.ID
,x.value(N'.',N'nvarchar(max)') AS Part
FROM Casted
CROSS APPLY SplitMe.nodes(N'/x') AS A(x)
The result
ID Part
1 A&B
1 C
1 D
1 E, F
2 "C" & 'D'
2 <C>
2 D
2 E, F
All the functions for string splitting that use some kind of Loop-ing (iterations) have bad performance. They should be replaced with set-based solution.
This code executes excellent.
CREATE FUNCTION dbo.SplitStrings
(
#List NVARCHAR(MAX),
#Delimiter NVARCHAR(255)
)
RETURNS TABLE
WITH SCHEMABINDING
AS
RETURN
(
SELECT Item = y.i.value('(./text())[1]', 'nvarchar(4000)')
FROM
(
SELECT x = CONVERT(XML, '<i>'
+ REPLACE(#List, #Delimiter, '</i><i>')
+ '</i>').query('.')
) AS a CROSS APPLY x.nodes('i') AS y(i)
);
GO
I had to write something like this recently. Here's the solution I came up with. It's generalized for any delimiter string and I think it would perform slightly better:
CREATE FUNCTION [dbo].[SplitString]
( #string nvarchar(4000)
, #delim nvarchar(100) )
RETURNS
#result TABLE
( [Value] nvarchar(4000) NOT NULL
, [Index] int NOT NULL )
AS
BEGIN
DECLARE #str nvarchar(4000)
, #pos int
, #prv int = 1
SELECT #pos = CHARINDEX(#delim, #string)
WHILE #pos > 0
BEGIN
SELECT #str = SUBSTRING(#string, #prv, #pos - #prv)
INSERT INTO #result SELECT #str, #prv
SELECT #prv = #pos + LEN(#delim)
, #pos = CHARINDEX(#delim, #string, #pos + 1)
END
INSERT INTO #result SELECT SUBSTRING(#string, #prv, 4000), #prv
RETURN
END
If you need a quick ad-hoc solution for common cases with minimum code, then this recursive CTE two-liner will do it:
DECLARE #s VARCHAR(200) = ',1,2,,3,,,4,,,,5,'
;WITH
a AS (SELECT i=-1, j=0 UNION ALL SELECT j, CHARINDEX(',', #s, j + 1) FROM a WHERE j > i),
b AS (SELECT SUBSTRING(#s, i+1, IIF(j>0, j, LEN(#s)+1)-i-1) s FROM a WHERE i >= 0)
SELECT * FROM b
Either use this as a stand-alone statement or just add the above CTEs to any of your queries and you will be able to join the resulting table b with others for use in any further expressions.
edit (by Shnugo)
If you add a counter, you will get a position index together with the List:
DECLARE #s VARCHAR(200) = '1,2333,344,4'
;WITH
a AS (SELECT n=0, i=-1, j=0 UNION ALL SELECT n+1, j, CHARINDEX(',', #s, j+1) FROM a WHERE j > i),
b AS (SELECT n, SUBSTRING(#s, i+1, IIF(j>0, j, LEN(#s)+1)-i-1) s FROM a WHERE i >= 0)
SELECT * FROM b;
The result:
n s
1 1
2 2333
3 344
4 4
I take the xml route by wrapping the values into elements (M but anything works):
declare #v nvarchar(max) = '100,201,abcde'
select
a.value('.', 'varchar(max)')
from
(select cast('<M>' + REPLACE(#v, ',', '</M><M>') + '</M>' AS XML) as col) as A
CROSS APPLY A.col.nodes ('/M') AS Split(a)
A solution using a CTE, if anyone should need that (apart from me, who obviously did, that is why I wrote it).
declare #StringToSplit varchar(100) = 'Test1,Test2,Test3';
declare #SplitChar varchar(10) = ',';
with StringToSplit as (
select
ltrim( rtrim( substring( #StringToSplit, 1, charindex( #SplitChar, #StringToSplit ) - 1 ) ) ) Head
, substring( #StringToSplit, charindex( #SplitChar, #StringToSplit ) + 1, len( #StringToSplit ) ) Tail
union all
select
ltrim( rtrim( substring( Tail, 1, charindex( #SplitChar, Tail ) - 1 ) ) ) Head
, substring( Tail, charindex( #SplitChar, Tail ) + 1, len( Tail ) ) Tail
from StringToSplit
where charindex( #SplitChar, Tail ) > 0
union all
select
ltrim( rtrim( Tail ) ) Head
, '' Tail
from StringToSplit
where charindex( #SplitChar, Tail ) = 0
and len( Tail ) > 0
)
select Head from StringToSplit
This is more narrowly-tailored. When I do this I usually have a comma-delimited list of unique ids (INT or BIGINT), which I want to cast as a table to use as an inner join to another table that has a primary key of INT or BIGINT. I want an in-line table-valued function returned so that I have the most efficient join possible.
Sample usage would be:
DECLARE #IDs VARCHAR(1000);
SET #IDs = ',99,206,124,8967,1,7,3,45234,2,889,987979,';
SELECT me.Value
FROM dbo.MyEnum me
INNER JOIN dbo.GetIntIdsTableFromDelimitedString(#IDs) ids ON me.PrimaryKey = ids.ID
I stole the idea from http://sqlrecords.blogspot.com/2012/11/converting-delimited-list-to-table.html, changing it to be in-line table-valued and cast as INT.
create function dbo.GetIntIDTableFromDelimitedString
(
#IDs VARCHAR(1000) --this parameter must start and end with a comma, eg ',123,456,'
--all items in list must be perfectly formatted or function will error
)
RETURNS TABLE AS
RETURN
SELECT
CAST(SUBSTRING(#IDs,Nums.number + 1,CHARINDEX(',',#IDs,(Nums.number+2)) - Nums.number - 1) AS INT) AS ID
FROM
[master].[dbo].[spt_values] Nums
WHERE Nums.Type = 'P'
AND Nums.number BETWEEN 1 AND DATALENGTH(#IDs)
AND SUBSTRING(#IDs,Nums.number,1) = ','
AND CHARINDEX(',',#IDs,(Nums.number+1)) > Nums.number;
GO
There is a correct version on here but I thought it would be nice to add a little fault tolerance in case they have a trailing comma as well as make it so you could use it not as a function but as part of a larger piece of code. Just in case you're only using it once time and don't need a function. This is also for integers (which is what I needed it for) so you might have to change your data types.
DECLARE #StringToSeperate VARCHAR(10)
SET #StringToSeperate = '1,2,5'
--SELECT #StringToSeperate IDs INTO #Test
DROP TABLE #IDs
CREATE TABLE #IDs (ID int)
DECLARE #CommaSeperatedValue NVARCHAR(255) = ''
DECLARE #Position INT = LEN(#StringToSeperate)
--Add Each Value
WHILE CHARINDEX(',', #StringToSeperate) > 0
BEGIN
SELECT #Position = CHARINDEX(',', #StringToSeperate)
SELECT #CommaSeperatedValue = SUBSTRING(#StringToSeperate, 1, #Position-1)
INSERT INTO #IDs
SELECT #CommaSeperatedValue
SELECT #StringToSeperate = SUBSTRING(#StringToSeperate, #Position+1, LEN(#StringToSeperate)-#Position)
END
--Add Last Value
IF (LEN(LTRIM(RTRIM(#StringToSeperate)))>0)
BEGIN
INSERT INTO #IDs
SELECT SUBSTRING(#StringToSeperate, 1, #Position)
END
SELECT * FROM #IDs
I modified +Andy Robinson's function a little bit. Now you can select only required part from returning table:
CREATE FUNCTION dbo.splitstring ( #stringToSplit VARCHAR(MAX) )
RETURNS
#returnList TABLE ([numOrder] [tinyint] , [Name] [nvarchar] (500)) AS
BEGIN
DECLARE #name NVARCHAR(255)
DECLARE #pos INT
DECLARE #orderNum INT
SET #orderNum=0
WHILE CHARINDEX('.', #stringToSplit) > 0
BEGIN
SELECT #orderNum=#orderNum+1;
SELECT #pos = CHARINDEX('.', #stringToSplit)
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos-1)
INSERT INTO #returnList
SELECT #orderNum,#name
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, LEN(#stringToSplit)-#pos)
END
SELECT #orderNum=#orderNum+1;
INSERT INTO #returnList
SELECT #orderNum, #stringToSplit
RETURN
END
Usage:
SELECT Name FROM dbo.splitstring('ELIS.YD.CRP1.1.CBA.MDSP.T389.BT') WHERE numOrder=5
Simples
DECLARE #String varchar(100) = '11,21,84,85,87'
SELECT * FROM TB_PAPEL WHERE CD_PAPEL IN (SELECT value FROM STRING_SPLIT(#String, ','))
-- EQUIVALENTE
SELECT * FROM TB_PAPEL WHERE CD_PAPEL IN (11,21,84,85,87)
here is a version that can split on a pattern using patindex, a simple adaptation of the post above. I had a case where I needed to split a string that contained multiple separator chars.
alter FUNCTION dbo.splitstring ( #stringToSplit VARCHAR(1000), #splitPattern varchar(10) )
RETURNS
#returnList TABLE ([Name] [nvarchar] (500))
AS
BEGIN
DECLARE #name NVARCHAR(255)
DECLARE #pos INT
WHILE PATINDEX(#splitPattern, #stringToSplit) > 0
BEGIN
SELECT #pos = PATINDEX(#splitPattern, #stringToSplit)
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos-1)
INSERT INTO #returnList
SELECT #name
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, LEN(#stringToSplit)-#pos)
END
INSERT INTO #returnList
SELECT #stringToSplit
RETURN
END
select * from dbo.splitstring('stringa/stringb/x,y,z','%[/,]%');
result looks like this
stringa
stringb
x
y
z
Personnaly I use this function :
ALTER FUNCTION [dbo].[CUST_SplitString]
(
#String NVARCHAR(4000),
#Delimiter NCHAR(1)
)
RETURNS TABLE
AS
RETURN
(
WITH Split(stpos,endpos)
AS(
SELECT 0 AS stpos, CHARINDEX(#Delimiter,#String) AS endpos
UNION ALL
SELECT endpos+1, CHARINDEX(#Delimiter,#String,endpos+1)
FROM Split
WHERE endpos > 0
)
SELECT 'Id' = ROW_NUMBER() OVER (ORDER BY (SELECT 1)),
'Data' = SUBSTRING(#String,stpos,COALESCE(NULLIF(endpos,0),LEN(#String)+1)-stpos)
FROM Split
)
I have developed a double Splitter (Takes two split characters) as requested Here. Could be of some value in this thread seeing its the most referenced for queries relating to string splitting.
CREATE FUNCTION uft_DoubleSplitter
(
-- Add the parameters for the function here
#String VARCHAR(4000),
#Splitter1 CHAR,
#Splitter2 CHAR
)
RETURNS #Result TABLE (Id INT,MId INT,SValue VARCHAR(4000))
AS
BEGIN
DECLARE #FResult TABLE(Id INT IDENTITY(1, 1),
SValue VARCHAR(4000))
DECLARE #SResult TABLE(Id INT IDENTITY(1, 1),
MId INT,
SValue VARCHAR(4000))
SET #String = #String+#Splitter1
WHILE CHARINDEX(#Splitter1, #String) > 0
BEGIN
DECLARE #WorkingString VARCHAR(4000) = NULL
SET #WorkingString = SUBSTRING(#String, 1, CHARINDEX(#Splitter1, #String) - 1)
--Print #workingString
INSERT INTO #FResult
SELECT CASE
WHEN #WorkingString = '' THEN NULL
ELSE #WorkingString
END
SET #String = SUBSTRING(#String, LEN(#WorkingString) + 2, LEN(#String))
END
IF ISNULL(#Splitter2, '') != ''
BEGIN
DECLARE #OStartLoop INT
DECLARE #OEndLoop INT
SELECT #OStartLoop = MIN(Id),
#OEndLoop = MAX(Id)
FROM #FResult
WHILE #OStartLoop <= #OEndLoop
BEGIN
DECLARE #iString VARCHAR(4000)
DECLARE #iMId INT
SELECT #iString = SValue+#Splitter2,
#iMId = Id
FROM #FResult
WHERE Id = #OStartLoop
WHILE CHARINDEX(#Splitter2, #iString) > 0
BEGIN
DECLARE #iWorkingString VARCHAR(4000) = NULL
SET #IWorkingString = SUBSTRING(#iString, 1, CHARINDEX(#Splitter2, #iString) - 1)
INSERT INTO #SResult
SELECT #iMId,
CASE
WHEN #iWorkingString = '' THEN NULL
ELSE #iWorkingString
END
SET #iString = SUBSTRING(#iString, LEN(#iWorkingString) + 2, LEN(#iString))
END
SET #OStartLoop = #OStartLoop + 1
END
INSERT INTO #Result
SELECT MId AS PrimarySplitID,
ROW_NUMBER() OVER (PARTITION BY MId ORDER BY Mid, Id) AS SecondarySplitID ,
SValue
FROM #SResult
END
ELSE
BEGIN
INSERT INTO #Result
SELECT Id AS PrimarySplitID,
NULL AS SecondarySplitID,
SValue
FROM #FResult
END
RETURN
Usage:
--FirstSplit
SELECT * FROM uft_DoubleSplitter('ValueA=ValueB=ValueC=ValueD==ValueE&ValueA=ValueB=ValueC===ValueE&ValueA=ValueB==ValueD===','&',NULL)
--Second Split
SELECT * FROM uft_DoubleSplitter('ValueA=ValueB=ValueC=ValueD==ValueE&ValueA=ValueB=ValueC===ValueE&ValueA=ValueB==ValueD===','&','=')
Possible Usage (Get second value of each split):
SELECT fn.SValue
FROM uft_DoubleSplitter('ValueA=ValueB=ValueC=ValueD==ValueE&ValueA=ValueB=ValueC===ValueE&ValueA=ValueB==ValueD===', '&', '=')AS fn
WHERE fn.mid = 2
A recursive cte based solution
declare #T table (iden int identity, col1 varchar(100));
insert into #T(col1) values
('ROOT/South America/Lima/Test/Test2')
, ('ROOT/South America/Peru/Test/Test2')
, ('ROOT//South America/Venuzuala ')
, ('RtT/South America / ')
, ('ROOT/South Americas// ');
declare #split char(1) = '/';
select #split as split;
with cte as
( select t.iden, case when SUBSTRING(REVERSE(rtrim(t.col1)), 1, 1) = #split then LTRIM(RTRIM(t.col1)) else LTRIM(RTRIM(t.col1)) + #split end as col1, 0 as pos , 1 as cnt
from #T t
union all
select t.iden, t.col1 , charindex(#split, t.col1, t.pos + 1), cnt + 1
from cte t
where charindex(#split, t.col1, t.pos + 1) > 0
)
select t1.*, t2.pos, t2.cnt
, ltrim(rtrim(SUBSTRING(t1.col1, t1.pos+1, t2.pos-t1.pos-1))) as bingo
from cte t1
join cte t2
on t2.iden = t1.iden
and t2.cnt = t1.cnt+1
and t2.pos > t1.pos
order by t1.iden, t1.cnt;
With all due respect to #AviG this is the bug free version of function deviced by him to return all the tokens in full.
IF EXISTS (SELECT * FROM sys.objects WHERE type = 'TF' AND name = 'TF_SplitString')
DROP FUNCTION [dbo].[TF_SplitString]
GO
-- =============================================
-- Author: AviG
-- Amendments: Parameterize the delimeter and included the missing chars in last token - Gemunu Wickremasinghe
-- Description: Tabel valued function that Breaks the delimeted string by given delimeter and returns a tabel having split results
-- Usage
-- select * from [dbo].[TF_SplitString]('token1,token2,,,,,,,,token969',',')
-- 969 items should be returned
-- select * from [dbo].[TF_SplitString]('4672978261,4672978255',',')
-- 2 items should be returned
-- =============================================
CREATE FUNCTION dbo.TF_SplitString
( #stringToSplit VARCHAR(MAX) ,
#delimeter char = ','
)
RETURNS
#returnList TABLE ([Name] [nvarchar] (500))
AS
BEGIN
DECLARE #name NVARCHAR(255)
DECLARE #pos INT
WHILE LEN(#stringToSplit) > 0
BEGIN
SELECT #pos = CHARINDEX(#delimeter, #stringToSplit)
if #pos = 0
BEGIN
SELECT #pos = LEN(#stringToSplit)
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos)
END
else
BEGIN
SELECT #name = SUBSTRING(#stringToSplit, 1, #pos-1)
END
INSERT INTO #returnList
SELECT #name
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, LEN(#stringToSplit)-#pos)
END
RETURN
END
This is based on Andy Robertson's answer, I needed a delimiter other than comma.
CREATE FUNCTION dbo.splitstring ( #stringToSplit nvarchar(MAX), #delim nvarchar(max))
RETURNS
#returnList TABLE ([value] [nvarchar] (MAX))
AS
BEGIN
DECLARE #value NVARCHAR(max)
DECLARE #pos INT
WHILE CHARINDEX(#delim, #stringToSplit) > 0
BEGIN
SELECT #pos = CHARINDEX(#delim, #stringToSplit)
SELECT #value = SUBSTRING(#stringToSplit, 1, #pos - 1)
INSERT INTO #returnList
SELECT #value
SELECT #stringToSplit = SUBSTRING(#stringToSplit, #pos + LEN(#delim), LEN(#stringToSplit) - #pos)
END
INSERT INTO #returnList
SELECT #stringToSplit
RETURN
END
GO
And to use it:
SELECT * FROM dbo.splitstring('test1 test2 test3', ' ');
(Tested on SQL Server 2008 R2)
EDIT: correct test code
ALTER FUNCTION [dbo].func_split_string
(
#input as varchar(max),
#delimiter as varchar(10) = ";"
)
RETURNS #result TABLE
(
id smallint identity(1,1),
csv_value varchar(max) not null
)
AS
BEGIN
DECLARE #pos AS INT;
DECLARE #string AS VARCHAR(MAX) = '';
WHILE LEN(#input) > 0
BEGIN
SELECT #pos = CHARINDEX(#delimiter,#input);
IF(#pos<=0)
select #pos = len(#input)
IF(#pos <> LEN(#input))
SELECT #string = SUBSTRING(#input, 1, #pos-1);
ELSE
SELECT #string = SUBSTRING(#input, 1, #pos);
INSERT INTO #result SELECT #string
SELECT #input = SUBSTRING(#input, #pos+len(#delimiter), LEN(#input)-#pos)
END
RETURN
END
You can Use this function:
CREATE FUNCTION SplitString
(
#Input NVARCHAR(MAX),
#Character CHAR(1)
)
RETURNS #Output TABLE (
Item NVARCHAR(1000)
)
AS
BEGIN
DECLARE #StartIndex INT, #EndIndex INT
SET #StartIndex = 1
IF SUBSTRING(#Input, LEN(#Input) - 1, LEN(#Input)) <> #Character
BEGIN
SET #Input = #Input + #Character
END
WHILE CHARINDEX(#Character, #Input) > 0
BEGIN
SET #EndIndex = CHARINDEX(#Character, #Input)
INSERT INTO #Output(Item)
SELECT SUBSTRING(#Input, #StartIndex, #EndIndex - 1)
SET #Input = SUBSTRING(#Input, #EndIndex + 1, LEN(#Input))
END
RETURN
END
GO
Here is an example that you can use as function or also you can put the same logic in procedure.
--SELECT * from [dbo].fn_SplitString ;
CREATE FUNCTION [dbo].[fn_SplitString]
(#CSV VARCHAR(MAX), #Delimeter VARCHAR(100) = ',')
RETURNS #retTable TABLE
(
[value] VARCHAR(MAX) NULL
)AS
BEGIN
DECLARE
#vCSV VARCHAR (MAX) = #CSV,
#vDelimeter VARCHAR (100) = #Delimeter;
IF #vDelimeter = ';'
BEGIN
SET #vCSV = REPLACE(#vCSV, ';', '~!~#~');
SET #vDelimeter = REPLACE(#vDelimeter, ';', '~!~#~');
END;
SET #vCSV = REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(#vCSV, '&', '&'), '<', '<'), '>', '>'), '''', '&apos;'), '"', '"');
DECLARE #xml XML;
SET #xml = '<i>' + REPLACE(#vCSV, #vDelimeter, '</i><i>') + '</i>';
INSERT INTO #retTable
SELECT
x.i.value('.', 'varchar(max)') AS COLUMNNAME
FROM #xml.nodes('//i')AS x(i);
RETURN;
END;
/*
Answer to T-SQL split string
Based on answers from Andy Robinson and AviG
Enhanced functionality ref: LEN function not including trailing spaces in SQL Server
This 'file' should be valid as both a markdown file and an SQL file
*/
CREATE FUNCTION dbo.splitstring ( --CREATE OR ALTER
#stringToSplit NVARCHAR(MAX)
) RETURNS #returnList TABLE ([Item] NVARCHAR (MAX))
AS BEGIN
DECLARE #name NVARCHAR(MAX)
DECLARE #pos BIGINT
SET #stringToSplit = #stringToSplit + ',' -- this should allow entries that end with a `,` to have a blank value in that "column"
WHILE ((LEN(#stringToSplit+'_') > 1)) BEGIN -- `+'_'` gets around LEN trimming terminal spaces. See URL referenced above
SET #pos = COALESCE(NULLIF(CHARINDEX(',', #stringToSplit),0),LEN(#stringToSplit+'_')) -- COALESCE grabs first non-null value
SET #name = SUBSTRING(#stringToSplit, 1, #pos-1) --MAX size of string of type nvarchar is 4000
SET #stringToSplit = SUBSTRING(#stringToSplit, #pos+1, 4000) -- With SUBSTRING fn (MS web): "If start is greater than the number of characters in the value expression, a zero-length expression is returned."
INSERT INTO #returnList SELECT #name --additional debugging parameters below can be added
-- + ' pos:' + CAST(#pos as nvarchar) + ' remain:''' + #stringToSplit + '''(' + CAST(LEN(#stringToSplit+'_')-1 as nvarchar) + ')'
END
RETURN
END
GO
/*
Test cases: see URL referenced as "enhanced functionality" above
SELECT *,LEN(Item+'_')-1 'L' from splitstring('a,,b')
Item | L
--- | ---
a | 1
| 0
b | 1
SELECT *,LEN(Item+'_')-1 'L' from splitstring('a,,')
Item | L
--- | ---
a | 1
| 0
| 0
SELECT *,LEN(Item+'_')-1 'L' from splitstring('a,, ')
Item | L
--- | ---
a | 1
| 0
| 1
SELECT *,LEN(Item+'_')-1 'L' from splitstring('a,, c ')
Item | L
--- | ---
a | 1
| 0
c | 3
*/
The easiest way:
Install SQL Server 2016
Use STRING_SPLIT https://msdn.microsoft.com/en-us/library/mt684588.aspx
It works even in express edition :).

String concatenation in SQL server

Consider a situation we have two variables in SQL Server 2005's SP as below,
#string1 = 'a,b,c,d'
#string2 = 'c,d,e,f,g'
Is there a solution to get a new string out of that like (#string1 U #string2) without using any loops. i.e the final string should be like,
#string3 = 'a,b,c,d,e,f,g'
In case you need to do this as a set and not one row at a time. Given the following split function:
USE tempdb;
GO
CREATE FUNCTION dbo.SplitStrings(#List nvarchar(max))
RETURNS TABLE
AS
RETURN ( SELECT Item FROM
( SELECT Item = x.i.value(N'./text()[1]', N'nvarchar(max)')
FROM ( SELECT [XML] = CONVERT(xml, '<i>'
+ REPLACE(#List,',', '</i><i>') + '</i>').query('.')
) AS a CROSS APPLY [XML].nodes('i') AS x(i) ) AS y
WHERE Item IS NOT NULL
);
GO
Then with the following table and sample data, and string variable, you can get all of the results this way:
DECLARE #foo TABLE(ID INT IDENTITY(1,1), col NVARCHAR(MAX));
INSERT #foo(col) SELECT N'c,d,e,f,g';
INSERT #foo(col) SELECT N'c,e,b';
INSERT #foo(col) SELECT N'd,e,f,x,a,e';
DECLARE #string NVARCHAR(MAX) = N'a,b,c,d';
;WITH x AS
(
SELECT f.ID, c.Item FROM #foo AS f
CROSS APPLY dbo.SplitStrings(f.col) AS c
), y AS
(
SELECT ID, Item FROM x
UNION
SELECT x.ID, s.Item
FROM dbo.SplitStrings(#string) AS s
CROSS JOIN x
)
SELECT ID, Items = STUFF((SELECT ',' + Item
FROM y AS y2 WHERE y2.ID = y.ID
FOR XML PATH(''), TYPE).value(N'./text()[1]', N'nvarchar(max)'), 1, 1, N'')
FROM y
GROUP BY ID;
Results:
ID Items
-- ----------
1 a,b,c,d,e,f,g
2 a,b,c,d,e
3 a,b,c,d,e,f,x
On newer versions (SQL Server 2017+), the query is much simpler, and you don't need to create your own custom string-splitting function:
;WITH x AS
(
SELECT f.ID, c.value FROM #foo AS f
CROSS APPLY STRING_SPLIT
(
CONCAT(f.col, N',', #string), N','
) AS c GROUP BY f.ID, c.value
)
SELECT ID, STRING_AGG(value, N',')
WITHIN GROUP (ORDER BY value)
FROM x GROUP BY ID;
Example db<>fiddle
Now that all said, what you really should do is follow the previous advice and store these things in a related table in the first place. You can use the same type of splitting methodology to store the strings separately whenever an insert or update happens, instead of just dumping the CSV into a single column, and your applications shouldn't really have to change the way they're passing data into your procedures. But it sure will be easier to get the data out!
EDIT
Adding a potential solution for SQL Server 2008 that is a bit more convoluted but gets things done with one less loop (using a massive table scan and replace instead). I don't think this is any better than the solution above, and it is certainly less maintainable, but it is an option to test out should you find you are able to upgrade to 2008 or better (and also for any 2008+ users who come across this question).
SET NOCOUNT ON;
-- let's pretend this is our static table:
CREATE TABLE #x
(
ID int IDENTITY(1,1),
col nvarchar(max)
);
INSERT #x(col) VALUES(N'c,d,e,f,g'), (N'c,e,b'), (N'd,e,f,x,a,e');
-- and here is our parameter:
DECLARE #string nvarchar(max) = N'a,b,c,d';
The code:
DECLARE #sql nvarchar(max) = N'DECLARE #src TABLE(ID INT, col NVARCHAR(32));
DECLARE #dest TABLE(ID int, col nvarchar(32));';
SELECT #sql += '
INSERT #src VALUES(' + RTRIM(ID) + ','''
+ REPLACE(col, ',', '''),(' + RTRIM(ID) + ',''') + ''');'
FROM #x;
SELECT #sql += '
INSERT #dest VALUES(' + RTRIM(ID) + ','''
+ REPLACE(#string, ',', '''),(' + RTRIM(ID) + ',''') + ''');'
FROM #x;
SELECT #sql += '
WITH x AS (SELECT ID, col FROM #src UNION SELECT ID, col FROM #dest)
SELECT DISTINCT ID, Items = STUFF((SELECT '','' + col
FROM x AS x2 WHERE x2.ID = x.ID FOR XML PATH('''')), 1, 1, N'''')
FROM x;'
EXEC sys.sp_executesql #sql;
GO
DROP TABLE #x;
This is much trickier to do in 2005 (though not impossible) because you need to change the VALUES() clauses to UNION ALL...
Two ways you can do that:
Build a CLR function to do the job for you. Move the logic back to .NET code which is much easier platform for string manipulation.
If you have to use SQL Server, then you will need to:
"explode" the two strings into two tables, this function might help: http://blog.logiclabz.com/sql-server/split-function-in-sql-server-to-break-comma-separated-strings-into-table.aspx
Get a unique list of strings from the two tables. (simple query)
"implode" the two string tables into a variable (http://stackoverflow.com/questions/194852/concatenate-many-rows-into-a-single-text-string)
Found this function dbo.Split in a related answer, which you can use like this:
declare #string1 nvarchar(50) = 'a,b,c,d'
declare #string2 nvarchar(50) = 'c,d,e,f,g'
select * from dbo.split(#string1, ',')
select * from dbo.split(#string2, ',')
declare #data nvarchar(100) = ''
select #data = #data + ',' + Data from (
select Data from dbo.split(#string1, ',')
union
select Data from dbo.split(#string2, ',')
) as d
select substring(#data, 2, LEN(#data))
The last SELECT returns
a,b,c,d,e,f,g
How about
set #string3 = #string1+','+#string2
Sorry, wasn't clear you wanted only unique occurrences. What version of SQL server are you using? String manipulation functions vary per version.
If you don't mind a UDF to split the string, try this:
CREATE FUNCTION dbo.Split
(
#RowData nvarchar(2000),
#SplitOn nvarchar(5)
)
RETURNS #RtnValue table
(
Id int identity(1,1),
Data nvarchar(100)
)
AS
BEGIN
Declare #Cnt int
declare #data varchar(100)
Set #Cnt = 1
While (Charindex(#SplitOn,#RowData)>0)
Begin
Insert Into #RtnValue (data)
Select ltrim(rtrim(Substring(#RowData,1,Charindex(#SplitOn,#RowData)-1)))
Set #RowData = Substring(#RowData,Charindex(#SplitOn,#RowData)+1,len(#RowData))
Set #Cnt = #Cnt + 1
End
Insert Into #RtnValue (data)
Select Data = ltrim(rtrim(#RowData))
Return
END
and the code to use the UDF
go
#string1 = 'a,b,c,d'
#string2 = 'c,d,e,f,g'
declare #string3 varchar(200)
set #string3 = ''
select #string3 = #string3+data+','
from ( select data,min(id) as Id from dbo.split(#string1+','+#string2,',')
group by data ) xx
order by xx.id
print left(#string3,len(#string3)-1)
The following SQL function will convert a comma separated list to a table variable...
CREATE FUNCTION [dbo].[udfCsvToTable]( #CsvString VARCHAR( 8000))
-- Converts a comman separated value into a table variable
RETURNS #tbl TABLE( [Value] VARCHAR( 100) COLLATE DATABASE_DEFAULT NOT NULL)
AS BEGIN
DECLARE #Text VARCHAR( 100)
SET #CsvString = RTRIM( LTRIM( #CsvString))
SET #CsvString = REPLACE( #CsvString, CHAR( 9), '')
SET #CsvString = REPLACE( #CsvString, CHAR( 10), '')
SET #CsvString = REPLACE( #CsvString, CHAR( 13), '')
IF LEN( #CsvString) < 1 RETURN
WHILE LEN( #CsvString) > 0 BEGIN
IF CHARINDEX( ',', #CsvString) > 0 BEGIN
SET #Text = LEFT( #CsvString, CHARINDEX( ',', #CsvString) - 1)
SET #CsvString = LTRIM( RTRIM( RIGHT( #CsvString, LEN( #CsvString) - CHARINDEX( ',', #CsvString))))
END
ELSE BEGIN
SET #Text = #CsvString
SET #CsvString = ''
END
INSERT #tbl VALUES( LTRIM( RTRIM( #Text)))
END
RETURN
END
You can then union the two tables together, like so...
SELECT * FROM udfCsvToTable('a,b,c,d')
UNION
SELECT * FROM udfCsvToTable('c,d,e,f,g')
Which will give you a result set of:
a
b
c
d
e
f
g

Split function equivalent in T-SQL?

I’m looking to split '1,2,3,4,5,6,7,8,9,10,11,12,13,14,15...' (comma delimited) into a table or table variable.
Does anyone have a function that returns each one in a row?
Try this
DECLARE #xml xml, #str varchar(100), #delimiter varchar(10)
SET #str = '1,2,3,4,5,6,7,8,9,10,11,12,13,14,15'
SET #delimiter = ','
SET #xml = cast(('<X>'+replace(#str, #delimiter, '</X><X>')+'</X>') as xml)
SELECT C.value('.', 'varchar(10)') as value FROM #xml.nodes('X') as X(C)
OR
DECLARE #str varchar(100), #delimiter varchar(10)
SET #str = '1,2,3,4,5,6,7,8,9,10,11,12,13,14,15'
SET #delimiter = ','
;WITH cte AS
(
SELECT 0 a, 1 b
UNION ALL
SELECT b, CHARINDEX(#delimiter, #str, b) + LEN(#delimiter)
FROM CTE
WHERE b > a
)
SELECT SUBSTRING(#str, a,
CASE WHEN b > LEN(#delimiter)
THEN b - a - LEN(#delimiter)
ELSE LEN(#str) - a + 1 END) value
FROM cte WHERE a > 0
Many more ways of doing the same is here How to split comma delimited string?
Here is somewhat old-fashioned solution:
/*
Splits string into parts delimitered with specified character.
*/
CREATE FUNCTION [dbo].[SDF_SplitString]
(
#sString nvarchar(2048),
#cDelimiter nchar(1)
)
RETURNS #tParts TABLE ( part nvarchar(2048) )
AS
BEGIN
if #sString is null return
declare #iStart int,
#iPos int
if substring( #sString, 1, 1 ) = #cDelimiter
begin
set #iStart = 2
insert into #tParts
values( null )
end
else
set #iStart = 1
while 1=1
begin
set #iPos = charindex( #cDelimiter, #sString, #iStart )
if #iPos = 0
set #iPos = len( #sString )+1
if #iPos - #iStart > 0
insert into #tParts
values ( substring( #sString, #iStart, #iPos-#iStart ))
else
insert into #tParts
values( null )
set #iStart = #iPos+1
if #iStart > len( #sString )
break
end
RETURN
END
In SQL Server 2008 you can achieve the same with .NET code. Maybe it would work faster, but definitely this approach is easier to manage.
You've tagged this SQL Server 2008 but future visitors to this question (using SQL Server 2016+) will likely want to know about STRING_SPLIT.
With this new builtin function you can now just use
SELECT TRY_CAST(value AS INT)
FROM STRING_SPLIT ('1,2,3,4,5,6,7,8,9,10,11,12,13,14,15', ',')
Some restrictions of this function and some promising results of performance testing are in this blog post by Aaron Bertrand.
This is most like .NET, for those of you who are familiar with that function:
CREATE FUNCTION dbo.[String.Split]
(
#Text VARCHAR(MAX),
#Delimiter VARCHAR(100),
#Index INT
)
RETURNS VARCHAR(MAX)
AS BEGIN
DECLARE #A TABLE (ID INT IDENTITY, V VARCHAR(MAX));
DECLARE #R VARCHAR(MAX);
WITH CTE AS
(
SELECT 0 A, 1 B
UNION ALL
SELECT B, CONVERT(INT,CHARINDEX(#Delimiter, #Text, B) + LEN(#Delimiter))
FROM CTE
WHERE B > A
)
INSERT #A(V)
SELECT SUBSTRING(#Text,A,CASE WHEN B > LEN(#Delimiter) THEN B-A-LEN(#Delimiter) ELSE LEN(#Text) - A + 1 END) VALUE
FROM CTE WHERE A >0
SELECT #R
= V
FROM #A
WHERE ID = #Index + 1
RETURN #R
END
SELECT dbo.[String.Split]('121,2,3,0',',',1) -- gives '2'
here is the split function that u asked
CREATE FUNCTION [dbo].[split](
#delimited NVARCHAR(MAX),
#delimiter NVARCHAR(100)
) RETURNS #t TABLE (id INT IDENTITY(1,1), val NVARCHAR(MAX))
AS
BEGIN
DECLARE #xml XML
SET #xml = N'<t>' + REPLACE(#delimited,#delimiter,'</t><t>') + '</t>'
INSERT INTO #t(val)
SELECT r.value('.','varchar(MAX)') as item
FROM #xml.nodes('/t') as records(r)
RETURN
END
execute the function like this
select * from dbo.split('1,2,3,4,5,6,7,8,9,10,11,12,13,14,15',',')
DECLARE
#InputString NVARCHAR(MAX) = 'token1,token2,token3,token4,token5'
, #delimiter varchar(10) = ','
DECLARE #xml AS XML = CAST(('<X>'+REPLACE(#InputString,#delimiter ,'</X><X>')+'</X>') AS XML)
SELECT C.value('.', 'varchar(10)') AS value
FROM #xml.nodes('X') as X(C)
Source of this response:
http://sqlhint.com/sqlserver/how-to/best-split-function-tsql-delimited
I am tempted to squeeze in my favourite solution. The resulting table will consist of 2 columns: PosIdx for position of the found integer; and Value in integer.
create function FnSplitToTableInt
(
#param nvarchar(4000)
)
returns table as
return
with Numbers(Number) as
(
select 1
union all
select Number + 1 from Numbers where Number < 4000
),
Found as
(
select
Number as PosIdx,
convert(int, ltrim(rtrim(convert(nvarchar(4000),
substring(#param, Number,
charindex(N',' collate Latin1_General_BIN,
#param + N',', Number) - Number))))) as Value
from
Numbers
where
Number <= len(#param)
and substring(N',' + #param, Number, 1) = N',' collate Latin1_General_BIN
)
select
PosIdx,
case when isnumeric(Value) = 1
then convert(int, Value)
else convert(int, null) end as Value
from
Found
It works by using recursive CTE as the list of positions, from 1 to 100 by default. If you need to work with string longer than 100, simply call this function using 'option (maxrecursion 4000)' like the following:
select * from FnSplitToTableInt
(
'9, 8, 7, 6, 5, 4, 3, 2, 1, 0, ' +
'9, 8, 7, 6, 5, 4, 3, 2, 1, 0, ' +
'9, 8, 7, 6, 5, 4, 3, 2, 1, 0, ' +
'9, 8, 7, 6, 5, 4, 3, 2, 1, 0, ' +
'9, 8, 7, 6, 5, 4, 3, 2, 1, 0'
)
option (maxrecursion 4000)
CREATE FUNCTION Split
(
#delimited nvarchar(max),
#delimiter nvarchar(100)
) RETURNS #t TABLE
(
-- Id column can be commented out, not required for sql splitting string
id int identity(1,1), -- I use this column for numbering splitted parts
val nvarchar(max)
)
AS
BEGIN
declare #xml xml
set #xml = N'<root><r>' + replace(#delimited,#delimiter,'</r><r>') + '</r></root>'
insert into #t(val)
select
r.value('.','varchar(max)') as item
from #xml.nodes('//root/r') as records(r)
RETURN
END
GO
usage
Select * from dbo.Split(N'1,2,3,4,6',',')
This simple CTE will give what's needed:
DECLARE #csv varchar(max) = '1,2,3,4,5,6,7,8,9,10,11,12,13,14,15';
--append comma to the list for CTE to work correctly
SET #csv = #csv + ',';
--remove double commas (empty entries)
SET #csv = replace(#csv, ',,', ',');
WITH CteCsv AS (
SELECT CHARINDEX(',', #csv) idx, SUBSTRING(#csv, 1, CHARINDEX(',', #csv) - 1) [Value]
UNION ALL
SELECT CHARINDEX(',', #csv, idx + 1), SUBSTRING(#csv, idx + 1, CHARINDEX(',', #csv, idx + 1) - idx - 1) FROM CteCsv
WHERE CHARINDEX(',', #csv, idx + 1) > 0
)
SELECT [Value] FROM CteCsv
This is another version which really does not have any restrictions (e.g.: special chars when using xml approach, number of records in CTE approach) and it runs much faster based on a test on 10M+ records with source string average length of 4000. Hope this could help.
Create function [dbo].[udf_split] (
#ListString nvarchar(max),
#Delimiter nvarchar(1000),
#IncludeEmpty bit)
Returns #ListTable TABLE (ID int, ListValue nvarchar(1000))
AS
BEGIN
Declare #CurrentPosition int, #NextPosition int, #Item nvarchar(max), #ID int, #L int
Select #ID = 1,
#L = len(replace(#Delimiter,' ','^')),
#ListString = #ListString + #Delimiter,
#CurrentPosition = 1
Select #NextPosition = Charindex(#Delimiter, #ListString, #CurrentPosition)
While #NextPosition > 0 Begin
Set #Item = LTRIM(RTRIM(SUBSTRING(#ListString, #CurrentPosition, #NextPosition-#CurrentPosition)))
If #IncludeEmpty=1 or LEN(#Item)>0 Begin
Insert Into #ListTable (ID, ListValue) Values (#ID, #Item)
Set #ID = #ID+1
End
Set #CurrentPosition = #NextPosition+#L
Set #NextPosition = Charindex(#Delimiter, #ListString, #CurrentPosition)
End
RETURN
END
/* *Object: UserDefinedFunction [dbo].[Split] Script Date: 10/04/2013 18:18:38* */
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
ALTER FUNCTION [dbo].[Split]
(#List varchar(8000),#SplitOn Nvarchar(5))
RETURNS #RtnValue table
(Id int identity(1,1),Value nvarchar(100))
AS
BEGIN
Set #List = Replace(#List,'''','')
While (Charindex(#SplitOn,#List)>0)
Begin
Insert Into #RtnValue (value)
Select
Value = ltrim(rtrim(Substring(#List,1,Charindex(#SplitOn,#List)-1)))
Set #List = Substring(#List,Charindex(#SplitOn,#List)+len(#SplitOn),len(#List))
End
Insert Into #RtnValue (Value)
Select Value = ltrim(rtrim(#List))
Return
END
go
Select *
From [Clv].[Split] ('1,2,3,3,3,3,',',')
GO
Using tally table here is one split string function(best possible approach) by Jeff Moden
CREATE FUNCTION [dbo].[DelimitedSplit8K]
(#pString VARCHAR(8000), #pDelimiter CHAR(1))
RETURNS TABLE WITH SCHEMABINDING AS
RETURN
--===== "Inline" CTE Driven "Tally Table" produces values from 0 up to 10,000...
-- enough to cover NVARCHAR(4000)
WITH E1(N) AS (
SELECT 1 UNION ALL SELECT 1 UNION ALL SELECT 1 UNION ALL
SELECT 1 UNION ALL SELECT 1 UNION ALL SELECT 1 UNION ALL
SELECT 1 UNION ALL SELECT 1 UNION ALL SELECT 1 UNION ALL SELECT 1
), --10E+1 or 10 rows
E2(N) AS (SELECT 1 FROM E1 a, E1 b), --10E+2 or 100 rows
E4(N) AS (SELECT 1 FROM E2 a, E2 b), --10E+4 or 10,000 rows max
cteTally(N) AS (--==== This provides the "base" CTE and limits the number of rows right up front
-- for both a performance gain and prevention of accidental "overruns"
SELECT TOP (ISNULL(DATALENGTH(#pString),0)) ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) FROM E4
),
cteStart(N1) AS (--==== This returns N+1 (starting position of each "element" just once for each delimiter)
SELECT 1 UNION ALL
SELECT t.N+1 FROM cteTally t WHERE SUBSTRING(#pString,t.N,1) = #pDelimiter
),
cteLen(N1,L1) AS(--==== Return start and length (for use in substring)
SELECT s.N1,
ISNULL(NULLIF(CHARINDEX(#pDelimiter,#pString,s.N1),0)-s.N1,8000)
FROM cteStart s
)
--===== Do the actual split. The ISNULL/NULLIF combo handles the length for the final element when no delimiter is found.
SELECT ItemNumber = ROW_NUMBER() OVER(ORDER BY l.N1),
Item = SUBSTRING(#pString, l.N1, l.L1)
FROM cteLen l
;
Referred from Tally OH! An Improved SQL 8K “CSV Splitter” Function
This blog came with a pretty good solution using XML in T-SQL.
This is the function I came up with based on that blog (change function name and result type cast per need):
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE FUNCTION [dbo].[SplitIntoBigints]
(#List varchar(MAX), #Splitter char)
RETURNS TABLE
AS
RETURN
(
WITH SplittedXML AS(
SELECT CAST('<v>' + REPLACE(#List, #Splitter, '</v><v>') + '</v>' AS XML) AS Splitted
)
SELECT x.v.value('.', 'bigint') AS Value
FROM SplittedXML
CROSS APPLY Splitted.nodes('//v') x(v)
)
GO
CREATE Function [dbo].[CsvToInt] ( #Array varchar(4000))
returns #IntTable table
(IntValue int)
AS
begin
declare #separator char(1)
set #separator = ','
declare #separator_position int
declare #array_value varchar(4000)
set #array = #array + ','
while patindex('%,%' , #array) <> 0
begin
select #separator_position = patindex('%,%' , #array)
select #array_value = left(#array, #separator_position - 1)
Insert #IntTable
Values (Cast(#array_value as int))
select #array = stuff(#array, 1, #separator_position, '')
end
This works great for me https://www.sqlshack.com/the-string-split-function-in-sql-server/
After two hours of resarching this topic this is the simplest solution (without using XML ect.).
You should only remember to use string_split after from.
DROP TABLE IF EXISTS #Countries
GO
DROP TABLE IF EXISTS #CityList
GO
CREATE TABLE #Countries
(Continent VARCHAR(100),
Country VARCHAR(100))
GO
CREATE TABLE #CityList
(Country VARCHAR(100),
City VARCHAR(5000))
GO
INSERT INTO #Countries
VALUES('Europe','France'),('Europe','Germany')
INSERT INTO #CityList
VALUES('France','Paris,Marsilya,Lyon,Lille,Nice'), ('Germany','Berlin,Hamburg,Munih,Frankfurt,Koln')
SELECT
CN.Continent,CN.Country,value
FROM #CityList CL CROSS APPLY string_split(CL.City,',') INNER JOIN
#Countries CN ON CL.Country = CN.Country
DROP TABLE IF EXISTS #Countries
GO
DROP TABLE IF EXISTS #CityList
You write this function in sql server after that problem will be solved.
http://csharpdotnetsol.blogspot.in/2013/12/csv-function-in-sql-server-for-divide.html