SQL Server - find nth occurrence in a string - sql-server

I have a table column that contains values such as abc_1_2_3_4.gif or zzz_12_3_3_45.gif etc.
I want to find the index of each underscore _ in the above values. There will only ever be four underscores but given that they can be in any position in the string, how can I achieve this?
I've tried the substring and charindex function, but I can only reliably get hold of the first one. Any ideas?

You can use the same function inside for the position +1
charindex('_', [TEXT], (charindex('_', [TEXT], 1))+1)
in where +1 is the nth time you will want to find.

One way (2k8);
select 'abc_1_2_3_4.gif  ' as img into #T
insert #T values ('zzz_12_3_3_45.gif')
;with T as (
select 0 as row, charindex('_', img) pos, img from #T
union all
select pos + 1, charindex('_', img, pos + 1), img
from T
where pos > 0
)
select
img, pos
from T
where pos > 0
order by img, pos
>>>>
img pos
abc_1_2_3_4.gif 4
abc_1_2_3_4.gif 6
abc_1_2_3_4.gif 8
abc_1_2_3_4.gif 10
zzz_12_3_3_45.gif 4
zzz_12_3_3_45.gif 7
zzz_12_3_3_45.gif 9
zzz_12_3_3_45.gif 11
Update
;with T(img, starts, pos) as (
select img, 1, charindex('_', img) from #t
union all
select img, pos + 1, charindex('_', img, pos + 1)
from t
where pos > 0
)
select
*, substring(img, starts, case when pos > 0 then pos - starts else len(img) end) token
from T
order by img, starts
>>>
img starts pos token
abc_1_2_3_4.gif   1 4 abc
abc_1_2_3_4.gif   5 6 1
abc_1_2_3_4.gif   7 8 2
abc_1_2_3_4.gif   9 10 3
abc_1_2_3_4.gif   11 0 4.gif  
zzz_12_3_3_45.gif 1 4 zzz
zzz_12_3_3_45.gif 5 7 12
zzz_12_3_3_45.gif 8 9 3
zzz_12_3_3_45.gif 10 11 3
zzz_12_3_3_45.gif 12 0 45.gif

You can use the CHARINDEX and specify the starting location:
DECLARE #x VARCHAR(32) = 'MS-SQL-Server';
SELECT
STUFF(STUFF(#x,3 , 0, '/'), 8, 0, '/') InsertString
,CHARINDEX('-',LTRIM(RTRIM(#x))) FirstIndexOf
,CHARINDEX('-',LTRIM(RTRIM(#x)), (CHARINDEX('-', LTRIM(RTRIM(#x)) )+1)) SecondIndexOf
,CHARINDEX('-',#x,CHARINDEX('-',#x, (CHARINDEX('-',#x)+1))+1) ThirdIndexOf
,CHARINDEX('-',REVERSE(LTRIM(RTRIM(#x)))) LastIndexOf;
GO

You can use the following function to split the values by a delimiter. It'll return a table and to find the nth occurrence just make a select on it! Or change it a little for it to return what you need instead of the table.
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
Set #Cnt = 1
While (Charindex(#SplitOn,#RowData)>0)
Begin
Insert Into #RtnValue (data)
Select
Data = 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

DECLARE #str AS VARCHAR(100)
SET #str='1,2 , 3, 4, 5,6'
SELECT COALESCE(LTRIM(CAST(('<X>'+REPLACE(#str,',' ,'</X><X>')+'</X>') AS XML).value('(/X)[1]', 'varchar(128)')), ''),
COALESCE(LTRIM(CAST(('<X>'+REPLACE(#str,',' ,'</X><X>')+'</X>') AS XML).value('(/X)[2]', 'varchar(128)')), ''),
COALESCE(LTRIM(CAST(('<X>'+REPLACE(#str,',' ,'</X><X>')+'</X>') AS XML).value('(/X)[3]', 'varchar(128)')), ''),
COALESCE(LTRIM(CAST(('<X>'+REPLACE(#str,',' ,'</X><X>')+'</X>') AS XML).value('(/X)[4]', 'varchar(128)')), ''),
COALESCE(LTRIM(CAST(('<X>'+REPLACE(#str,',' ,'</X><X>')+'</X>') AS XML).value('(/X)[5]', 'varchar(128)')), ''),
COALESCE(LTRIM(CAST(('<X>'+REPLACE(#str,',' ,'</X><X>')+'</X>') AS XML).value('(/X)[6]', 'varchar(128)')), ''),
COALESCE(LTRIM(CAST(('<X>'+REPLACE(#str,',' ,'</X><X>')+'</X>') AS XML).value('(/X)[7]', 'varchar(128)')), ''),
COALESCE(LTRIM(CAST(('<X>'+REPLACE(#str,',' ,'</X><X>')+'</X>') AS XML).value('(/X)[8]', 'varchar(128)')), ''),
COALESCE(LTRIM(CAST(('<X>'+REPLACE(#str,',' ,'</X><X>')+'</X>') AS XML).value('(/X)[9]', 'varchar(128)')), '')

You can look for the four underscore in this way:
create table #test
( t varchar(50) );
insert into #test values
( 'abc_1_2_3_4.gif'),
('zzz_12_3_3_45.gif');
declare #t varchar(50);
declare #t_aux varchar(50);
declare #t1 int;
declare #t2 int;
declare #t3 int;
declare #t4 int;
DECLARE t_cursor CURSOR
FOR SELECT t FROM #test
OPEN t_cursor
FETCH NEXT FROM t_cursor into #t;​
set #t1 = charindex( '_', #t )
set #t2 = charindex( '_', #t , #t1+1)
set #t3 = charindex( '_', #t , #t2+1)
set #t4 = charindex( '_', #t , #t3+1)
select #t1, #t2, t3, t4
--do a loop to iterate over all table
you can test it here.
Or in this simple way:
select
charindex( '_', t ) as first,
charindex( '_', t, charindex( '_', t ) + 1 ) as second,
...
from
#test

You can try peeling the variable/array, assuming distinctness in your list
declare #array table ----table of values
(
id int identity(1,1)
,value nvarchar(max)
)
DECLARE #VALUE NVARCHAR(MAX)='val1_val2_val3_val4_val5_val6_val7'----string array
DECLARE #CURVAL NVARCHAR(MAX) ---current value
DECLARE #DELIM NVARCHAR(1)='_' ---delimiter
DECLARE #BREAKPT INT ---current index of the delimiter
WHILE EXISTS (SELECT #VALUE)
BEGIN
SET #BREAKPT=CHARINDEX(#DELIM,#VALUE) ---set the current index
---
If #BREAKPT<> 0 ---index at 0 breaks the loop
begin
SET #CURVAL=SUBSTRING(#VALUE,1,#BREAKPT-1) ---current value
set #VALUE=REPLACE(#VALUE,SUBSTRING(#VALUE,1,#BREAKPT),'') ---current value and delimiter, replace
insert into #array(value) ---insert data
select #CURVAL
end
else
begin
SET #CURVAL=#VALUE ---current value now last value
insert into #array(value) ---insert data
select #CURVAL
break ---break loop
end
end
select * from #array ---find nth occurance given the id

DECLARE #LEN INT
DECLARE #VAR VARCHAR(20)
SET #VAR = 'HELLO WORLD'
SET #LEN = LEN(#VAR)
--SELECT #LEN
SELECT PATINDEX('%O%',SUBSTRING(#VAR,PATINDEX('%O%' ,#VAR) + 1 ,PATINDEX('%O%',#VAR) + 1)) + PATINDEX('%O%',#VAR)

My SQL supports the function of a substring_Index where it will return the postion of a value in a string for the n occurance. A similar User defined function could be written to achieve this. Example in the link
Alternatively you could use charindex function call it x times to report the location of each _ given a starting postion +1 of the previously found instance. until a 0 is found
Edit: NM Charindex is the correct function

A simple sample to do this with xml conversion:
SELECT 'A|B|C'
, concat('<x>', REPLACE('A|B|C', '|', '</x><x>'), '</x>')
, cast(concat('<x>', REPLACE('A|B|C', '|', '</x><x>'), '</x>') as xml).query('/x[2]')
, cast(concat('<x>', REPLACE('A|B|C', '|', '</x><x>'), '</x>') as xml).value('/x[2]',
'varchar');
And here a translation for your sample:
SELECT gifname
,cast(concat('<x>', REPLACE(gifname, '_', '</x><x>'), '</x>') as xml).query('/x[2]') as xmlelement
, cast(concat('<x>', REPLACE(gifname, '_', '</x><x>'), '</x>') as xml).value('/x[2]', 'varchar(10)') as result
FROM (
SELECT 'abc_1_2_3_4.gif' as gifname
UNION ALL
SELECT 'zzz_12_3_3_45.gif'
) tmp

I did this creating several separate custom functions, one for each position of the searched character i.e. 2nd, 3rd:
CREATE FUNCTION [dbo].[fnCHARPOS2]
(#SEARCHCHAR VARCHAR(255),
#SEARCHSTRING VARCHAR(255))
RETURNS INT
AS
BEGIN
RETURN CHARINDEX(#SEARCHCHAR,#SEARCHSTRING(CHARINDEX(#SEARCHCHAR,#SEARCHSTRING,0)+1));
CREATE FUNCTION [dbo].[fnCHARPOS3]
(#SEARCHCHAR VARCHAR(255),
#SEARCHSTRING VARCHAR(255))
RETURNS INT
AS
BEGIN
RETURN CHARINDEX(#SEARCHCHAR,#SEARCHSTRING, (CHARINDEX(#SEARCHCHAR,#SEARCHSTRING, (CHARINDEX(#SEARCHCHAR,#SEARCHSTRING,0)+1)))+1);
You can then pass in as a parameter the character you are searching for and the string you are searching in:
So if you were searching for 'f' and wanted to know position of 1st 3 occurences:
select
database.dbo.fnCHARPOS2('f',tablename.columnname),
database.dbo.fnCHARPOS3('f',tablename.columnname)
from tablename
It worked for me!

I decided to use a recursive function because for me it was easier to follow the logic. Note that SQL Server has a default function recursion limit of 32, so this is only good for smaller workloads.
create function dbo._charindex_nth (
#FindThis varchar(8000),
#InThis varchar(max),
#StartFrom int,
#NthOccurence tinyint
)
returns bigint
as
begin
/*
Recursive helper used by dbo.charindex_nth to return the position of the nth occurance of #FindThis in #InThis
Who When What
PJR 160421 Initial
*/
declare #Pos bigint
if isnull(#NthOccurence, 0) <= 0 or isnull(#StartFrom, 0) <= 0
begin
select #Pos = 0
end else begin
if #NthOccurence = 1
begin
select #Pos = charindex(#FindThis, #InThis, #StartFrom)
end else begin
select #Pos = dbo._charindex_nth(#FindThis, #InThis, nullif(charindex(#FindThis, #InThis, #StartFrom), 0) + 1, #NthOccurence - 1)
end
end
return #Pos
end
create function dbo.charindex_nth (
#FindThis varchar(8000),
#InThis varchar(max),
#NthOccurence tinyint
)
returns bigint
as
begin
/*
Returns the position of the nth occurance of #FindThis in #InThis
Who When What
PJR 160421 Initial
*/
return dbo._charindex_nth(#FindThis, #InThis, 1, #NthOccurence)
end
declare #val varchar(max) = 'zzz_12_3_3_45.gif'
select dbo.charindex_nth('_', #val, 1) Underscore1
, dbo.charindex_nth('_', #val, 2) Underscore2
, dbo.charindex_nth('_', #val, 3) Underscore3
, dbo.charindex_nth('_', #val, 4) Underscore4

DECLARE #T AS TABLE(pic_name VARCHAR(100));
INSERT INTO #T VALUES ('abc_1_2_3_4.gif'),('zzz_12_3_3_45.gif');
SELECT A.pic_name, P1.D, P2.D, P3.D, P4.D
FROM #T A
CROSS APPLY (SELECT NULLIF(CHARINDEX('_', A.pic_name),0) AS D) P1
CROSS APPLY (SELECT NULLIF(CHARINDEX('_', A.pic_name, P1.D+1), 0) AS D) P2
CROSS APPLY (SELECT NULLIF(CHARINDEX('_', A.pic_name, P2.D+1),0) AS D) P3
CROSS APPLY (SELECT NULLIF(CHARINDEX('_', A.pic_name, P3.D+1),0) AS D) P4

I've used a function to grab the "nth" element from a delimited string field with great success. Like mentioned above, it's not a "fast" way of dealing with things but it sure as heck is convenient.
create function GetArrayIndex(#delimited nvarchar(max), #index int, #delimiter nvarchar(100) = ',') returns nvarchar(max)
as
begin
declare #xml xml, #result nvarchar(max)
set #xml = N'<root><r>' + replace(#delimited, #delimiter,'</r><r>') + '</r></root>'
select #result = r.value('.','varchar(max)')
from #xml.nodes('//root/r[sql:variable("#index")]') as records(r)
return #result
end

I was toying with a faster way to do this than simply iterating through the string.
CREATE FUNCTION [ssf_GetNthSeparatorPosition] ( #TargetString VARCHAR(MAX)
, #Sep VARCHAR(25)
, #n INTEGER )
RETURNS INTEGER
/****************************************************************************************
--#############################################################################
-- Returns the position of the Nth Charactor sequence
-- 1234567890123456789
-- Declare #thatString varchar(max) = 'hi,there,jay,yo'
Select dbo.ssf_GetNthSeparatorPosition(#thatString, ',', 3) --would return 13
--############################################################################
****************************************************************************************/
AS
BEGIN
DECLARE #Retval INTEGER = 0
DECLARE #CurPos INTEGER = 0
DECLARE #LenSep INTEGER = LEN(#Sep)
SELECT #CurPos = CHARINDEX(#Sep, #TargetString)
IF ISNULL(#LenSep, 0) > 0
AND #CurPos > 0
BEGIN
SELECT #CurPos = 0
;with lv0 AS (SELECT 0 g UNION ALL SELECT 0)
,lv1 AS (SELECT 0 g FROM lv0 a CROSS JOIN lv0 b) -- 4
,lv2 AS (SELECT 0 g FROM lv1 a CROSS JOIN lv1 b) -- 16
,lv3 AS (SELECT 0 g FROM lv2 a CROSS JOIN lv2 b) -- 256
,lv4 AS (SELECT 0 g FROM lv3 a CROSS JOIN lv3 b) -- 65,536
,lv5 AS (SELECT 0 g FROM lv4 a CROSS JOIN lv4 b) -- 4,294,967,296
,Tally (n) AS (SELECT ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) FROM lv5),
results
AS ( SELECT n - LEN(#Sep) AS Nth
, row_number() OVER ( ORDER BY n ) - 1 AS Position
FROM Tally t
WHERE n BETWEEN 1
AND DATALENGTH(#TargetString) + DATALENGTH(#Sep)
AND SUBSTRING(#Sep + #TargetString, n, LEN(#Sep)) = #Sep)
SELECT #CurPos = Nth
FROM results
WHERE results.Position = #n
END
RETURN #CurPos
END
GO

declare #a nvarchar(50)='Enter Your string '
declare #character char='e'
declare #nthoccurence int = 2
declare #i int = 1
declare #j int =0
declare #count int = len(#a)-len(replace(#a,#character,''))
if(#count >= #nthoccurence)
begin
while (#I <= #nthoccurence)
begin
set #j= CHARINDEX(#character,#a,#j+1)
set #i= #i+1
end
print #j
end
else
Print 'you have only '+convert(nvarchar ,#count)+' occurrences of '+#character
end

DECLARE #x VARCHAR(32) = 'MS-SQL-Server';
SELECT
SUBSTRING(#x,0,CHARINDEX('-',LTRIM(RTRIM(#x)))) A,
SUBSTRING(#x,CHARINDEX('-',LTRIM(RTRIM(#x)))+1,CHARINDEX('-'
,LTRIM(RTRIM(#x)))) B,
SUBSTRING(#x,CHARINDEX('-',REVERSE(LTRIM(RTRIM(#x))))+1,LEN(#x)-1) C
A B C
MS SQL Server

Inspired by Alex K's reply One way (2k8), I have created a script for a Token Function for the SQL Server for returning a specific token from a string.
I needed this for refacturing a SSIS-package to T-SQL without having to implement Alex' solution a number of times manually.
My function has one disadvantage: It returns the token value as a table (one column, one row) instead of as a varchar value. If anyone has a solution for this, please let me know.
DROP FUNCTION [RDW].[token]
GO
create function [RDW].[token] (#string varchar(8000), #split varchar(50), #returnIndex int)
returns table
as
return with T(img, starts, pos, [index]) as (
select #string, 1, charindex(#split, #string), 0
union all
select #string, pos + 1, charindex(#split, #string, pos + 1), [index]+1
from t
where pos > 0
)
select substring(img, starts, case when pos > 0 then pos - starts else len(img) end) token
from T
where [index] = #returnIndex
GO

Try this
CREATE FUNCTION [dbo].[CHARINDEX2] (
#expressionToFind VARCHAR(MAX),
#expressionToSearch VARCHAR(MAX),
#occurrenceIndex INT,
#startLocation INT = 0
)
RETURNS INT
AS BEGIN
IF #occurrenceIndex < 1 BEGIN
RETURN CAST('The argument #occurrenceIndex must be a positive integer.' AS INT)
END
IF #startLocation < 0 BEGIN
RETURN CAST('The argument #startLocation must be a non negative integer.' AS INT)
END
DECLARE #returnIndex INT
SET #returnIndex = CHARINDEX(#expressionToFind, #expressionToSearch, #startLocation)
IF (#occurrenceIndex = 1) BEGIN
RETURN #returnIndex
END
DECLARE #target_length INT
SET #target_length = LEN(#expressionToFind)
SET #occurrenceIndex += -1
WHILE (#occurrenceIndex > 0 AND #returnIndex > 0) BEGIN
SET #returnIndex = CHARINDEX(#expressionToFind, #expressionToSearch, #returnIndex + #target_length);
SET #occurrenceIndex += -1
END
RETURN #returnIndex
END
GO

I use this function:
CREATE FUNCTION [pub].[SplitString]
( #StringToSplit nvarchar(MAX),
#Delimiter as nvarchar(10)
)
-- Example of use:
-- select * from [pub].[SplitString] ('one;two;three;;for & six;', ';')
RETURNS
#returnList TABLE ([Item] [nvarchar] (4000), [ID] [int] IDENTITY(1,1))
AS
BEGIN
DECLARE #xml as xml
SET #xml =
cast
( ('<X>' + replace(replace(#StringToSplit, #Delimiter ,'</X><X>'),'&','&') + '</X>'
) as xml
)
INSERT INTO #returnList([Item])
SELECT
N.value('.', 'nvarchar(max)') as [Item]
FROM
#xml.nodes('X') as T(N)
RETURN
END

This does not directly answer the question, but as I have found this while looking for a solution for extracting multiple values from a single string, with a specific delimiter, I post it anyway in case it could help someone.
SQL Server ships with the function PARSENAME that can extract up to 4 elements in a dot separated string, from the right :
SELECT PARSENAME('1234.5437.43278.532', 2) as third_element
--43278
You could use this function to retrieve the 2nd to the 5th element of the file name, by selecting from the second element to the end of the filename, without the extension.
Note that the code to remove the file extension (the 4th position from the right) is hardcoded here, so it's better to change it if you have some mix with .jpg and .jpeg file extensions for instance.
DECLARE #t TABLE (
c_filename VARCHAR(1000)
) INSERT #t
values
('abc_1_2_3_4.gif'),
('abcdefgh_1231_78432_398389_12144.png')
SELECT
LEFT(
c_filename,
CHARINDEX('_', c_filename) -1
) as first_element,
PARSENAME(
REPLACE(
/* PARSENAME only works with 4 elements */
/* we remove :
- the 1st element
- and the file extension */
SUBSTRING(
c_filename,
CHARINDEX('_', c_filename) + 1,
LEN(c_filename) - CHARINDEX('_', c_filename) -4
),
'_',
'.'
),
4 -- change this to get the nth element
) as second_element,
PARSENAME(
REPLACE(
SUBSTRING(
c_filename,
CHARINDEX('_', c_filename) + 1,
LEN(c_filename) - CHARINDEX('_', c_filename) -4
),
'_',
'.'
),
3
) as third_element,
PARSENAME(
REPLACE(
SUBSTRING(
c_filename,
CHARINDEX('_', c_filename) + 1,
LEN(c_filename) - CHARINDEX('_', c_filename) -4
),
'_',
'.'
),
2
) as fourth_element,
PARSENAME(
REPLACE(
SUBSTRING(
c_filename,
CHARINDEX('_', c_filename) + 1,
LEN(c_filename) - CHARINDEX('_', c_filename) -4
),
'_',
'.'
),
1
) as fifth_element
FROM
#t
+---------------+----------------+---------------+----------------+---------------+
| first_element | second_element | third_element | fourth_element | fifth_element |
+---------------+----------------+---------------+----------------+---------------+
| abc | 1 | 2 | 3 | 4 |
| abcdefghijkl | 12qwerty31 | 78891432 | 398977389 | 1212345344 |
+---------------+----------------+---------------+----------------+---------------+

Related

PIVOT the data from single column in sql server [duplicate]

I have a table like this
Value String
-------------------
1 Cleo, Smith
I want to separate the comma delimited string into two columns
Value Name Surname
-------------------
1 Cleo Smith
I need only two fixed extra columns
Your purpose can be solved using following query -
Select Value , Substring(FullName, 1,Charindex(',', FullName)-1) as Name,
Substring(FullName, Charindex(',', FullName)+1, LEN(FullName)) as Surname
from Table1
There is no readymade Split function in sql server, so we need to create user defined function.
CREATE FUNCTION Split (
#InputString VARCHAR(8000),
#Delimiter VARCHAR(50)
)
RETURNS #Items TABLE (
Item VARCHAR(8000)
)
AS
BEGIN
IF #Delimiter = ' '
BEGIN
SET #Delimiter = ','
SET #InputString = REPLACE(#InputString, ' ', #Delimiter)
END
IF (#Delimiter IS NULL OR #Delimiter = '')
SET #Delimiter = ','
--INSERT INTO #Items VALUES (#Delimiter) -- Diagnostic
--INSERT INTO #Items VALUES (#InputString) -- Diagnostic
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 = #ItemList minus one less item
SET #ItemList = SUBSTRING(#ItemList, #DelimIndex+1, LEN(#ItemList)-#DelimIndex)
SET #DelimIndex = CHARINDEX(#Delimiter, #ItemList, 0)
END -- End WHILE
IF #Item IS NOT NULL -- At least one delimiter was encountered in #InputString
BEGIN
SET #Item = #ItemList
INSERT INTO #Items VALUES (#Item)
END
-- No delimiters were encountered in #InputString, so just return #InputString
ELSE INSERT INTO #Items VALUES (#InputString)
RETURN
END -- End Function
GO
---- Set Permissions
--GRANT SELECT ON Split TO UserRole1
--GRANT SELECT ON Split TO UserRole2
--GO
;WITH Split_Names (Value,Name, xmlname)
AS
(
SELECT Value,
Name,
CONVERT(XML,'<Names><name>'
+ REPLACE(Name,',', '</name><name>') + '</name></Names>') AS xmlname
FROM tblnames
)
SELECT Value,
xmlname.value('/Names[1]/name[1]','varchar(100)') AS Name,
xmlname.value('/Names[1]/name[2]','varchar(100)') AS Surname
FROM Split_Names
and also check the link below for reference
http://jahaines.blogspot.in/2009/06/converting-delimited-string-of-values.html
xml based answer is simple and clean
refer this
DECLARE #S varchar(max),
#Split char(1),
#X xml
SELECT #S = 'ab,cd,ef,gh,ij',
#Split = ','
SELECT #X = CONVERT(xml,' <root> <myvalue>' +
REPLACE(#S,#Split,'</myvalue> <myvalue>') + '</myvalue> </root> ')
SELECT T.c.value('.','varchar(20)'), --retrieve ALL values at once
T.c.value('(/root/myvalue)[1]','VARCHAR(20)') , --retrieve index 1 only, which is the 'ab'
T.c.value('(/root/myvalue)[2]','VARCHAR(20)')
FROM #X.nodes('/root/myvalue') T(c)
I think this is cool
SELECT value,
PARSENAME(REPLACE(String,',','.'),2) 'Name' ,
PARSENAME(REPLACE(String,',','.'),1) 'Surname'
FROM table WITH (NOLOCK)
With CROSS APPLY
select ParsedData.*
from MyTable mt
cross apply ( select str = mt.String + ',,' ) f1
cross apply ( select p1 = charindex( ',', str ) ) ap1
cross apply ( select p2 = charindex( ',', str, p1 + 1 ) ) ap2
cross apply ( select Nmame = substring( str, 1, p1-1 )
, Surname = substring( str, p1+1, p2-p1-1 )
) ParsedData
There are multiple ways to solve this and many different ways have been proposed already. Simplest would be to use LEFT / SUBSTRING and other string functions to achieve the desired result.
Sample Data
DECLARE #tbl1 TABLE (Value INT,String VARCHAR(MAX))
INSERT INTO #tbl1 VALUES(1,'Cleo, Smith');
INSERT INTO #tbl1 VALUES(2,'John, Mathew');
Using String Functions like LEFT
SELECT
Value,
LEFT(String,CHARINDEX(',',String)-1) as Fname,
LTRIM(RIGHT(String,LEN(String) - CHARINDEX(',',String) )) AS Lname
FROM #tbl1
This approach fails if there are more 2 items in a String.
In such a scenario, we can use a splitter and then use PIVOT or convert the string into an XML and use .nodes to get string items. XML based solution have been detailed out by aads and bvr in their solution.
The answers for this question which use splitter, all use WHILE which is inefficient for splitting. Check this performance comparison. One of the best splitters around is DelimitedSplit8K, created by Jeff Moden. You can read more about it here
Splitter with PIVOT
DECLARE #tbl1 TABLE (Value INT,String VARCHAR(MAX))
INSERT INTO #tbl1 VALUES(1,'Cleo, Smith');
INSERT INTO #tbl1 VALUES(2,'John, Mathew');
SELECT t3.Value,[1] as Fname,[2] as Lname
FROM #tbl1 as t1
CROSS APPLY [dbo].[DelimitedSplit8K](String,',') as t2
PIVOT(MAX(Item) FOR ItemNumber IN ([1],[2])) as t3
Output
Value Fname Lname
1 Cleo Smith
2 John Mathew
DelimitedSplit8K by Jeff Moden
CREATE FUNCTION [dbo].[DelimitedSplit8K]
/**********************************************************************************************************************
Purpose:
Split a given string at a given delimiter and return a list of the split elements (items).
Notes:
1. Leading a trailing delimiters are treated as if an empty string element were present.
2. Consecutive delimiters are treated as if an empty string element were present between them.
3. Except when spaces are used as a delimiter, all spaces present in each element are preserved.
Returns:
iTVF containing the following:
ItemNumber = Element position of Item as a BIGINT (not converted to INT to eliminate a CAST)
Item = Element value as a VARCHAR(8000)
Statistics on this function may be found at the following URL:
http://www.sqlservercentral.com/Forums/Topic1101315-203-4.aspx
CROSS APPLY Usage Examples and Tests:
--=====================================================================================================================
-- TEST 1:
-- This tests for various possible conditions in a string using a comma as the delimiter. The expected results are
-- laid out in the comments
--=====================================================================================================================
--===== Conditionally drop the test tables to make reruns easier for testing.
-- (this is NOT a part of the solution)
IF OBJECT_ID('tempdb..#JBMTest') IS NOT NULL DROP TABLE #JBMTest
;
--===== Create and populate a test table on the fly (this is NOT a part of the solution).
-- In the following comments, "b" is a blank and "E" is an element in the left to right order.
-- Double Quotes are used to encapsulate the output of "Item" so that you can see that all blanks
-- are preserved no matter where they may appear.
SELECT *
INTO #JBMTest
FROM ( --# & type of Return Row(s)
SELECT 0, NULL UNION ALL --1 NULL
SELECT 1, SPACE(0) UNION ALL --1 b (Empty String)
SELECT 2, SPACE(1) UNION ALL --1 b (1 space)
SELECT 3, SPACE(5) UNION ALL --1 b (5 spaces)
SELECT 4, ',' UNION ALL --2 b b (both are empty strings)
SELECT 5, '55555' UNION ALL --1 E
SELECT 6, ',55555' UNION ALL --2 b E
SELECT 7, ',55555,' UNION ALL --3 b E b
SELECT 8, '55555,' UNION ALL --2 b B
SELECT 9, '55555,1' UNION ALL --2 E E
SELECT 10, '1,55555' UNION ALL --2 E E
SELECT 11, '55555,4444,333,22,1' UNION ALL --5 E E E E E
SELECT 12, '55555,4444,,333,22,1' UNION ALL --6 E E b E E E
SELECT 13, ',55555,4444,,333,22,1,' UNION ALL --8 b E E b E E E b
SELECT 14, ',55555,4444,,,333,22,1,' UNION ALL --9 b E E b b E E E b
SELECT 15, ' 4444,55555 ' UNION ALL --2 E (w/Leading Space) E (w/Trailing Space)
SELECT 16, 'This,is,a,test.' --E E E E
) d (SomeID, SomeValue)
;
--===== Split the CSV column for the whole table using CROSS APPLY (this is the solution)
SELECT test.SomeID, test.SomeValue, split.ItemNumber, Item = QUOTENAME(split.Item,'"')
FROM #JBMTest test
CROSS APPLY dbo.DelimitedSplit8K(test.SomeValue,',') split
;
--=====================================================================================================================
-- TEST 2:
-- This tests for various "alpha" splits and COLLATION using all ASCII characters from 0 to 255 as a delimiter against
-- a given string. Note that not all of the delimiters will be visible and some will show up as tiny squares because
-- they are "control" characters. More specifically, this test will show you what happens to various non-accented
-- letters for your given collation depending on the delimiter you chose.
--=====================================================================================================================
WITH
cteBuildAllCharacters (String,Delimiter) AS
(
SELECT TOP 256
'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789',
CHAR(ROW_NUMBER() OVER (ORDER BY (SELECT NULL))-1)
FROM master.sys.all_columns
)
SELECT ASCII_Value = ASCII(c.Delimiter), c.Delimiter, split.ItemNumber, Item = QUOTENAME(split.Item,'"')
FROM cteBuildAllCharacters c
CROSS APPLY dbo.DelimitedSplit8K(c.String,c.Delimiter) split
ORDER BY ASCII_Value, split.ItemNumber
;
-----------------------------------------------------------------------------------------------------------------------
Other Notes:
1. Optimized for VARCHAR(8000) or less. No testing or error reporting for truncation at 8000 characters is done.
2. Optimized for single character delimiter. Multi-character delimiters should be resolvedexternally from this
function.
3. Optimized for use with CROSS APPLY.
4. Does not "trim" elements just in case leading or trailing blanks are intended.
5. If you don't know how a Tally table can be used to replace loops, please see the following...
http://www.sqlservercentral.com/articles/T-SQL/62867/
6. Changing this function to use NVARCHAR(MAX) will cause it to run twice as slow. It's just the nature of
VARCHAR(MAX) whether it fits in-row or not.
7. Multi-machine testing for the method of using UNPIVOT instead of 10 SELECT/UNION ALLs shows that the UNPIVOT method
is quite machine dependent and can slow things down quite a bit.
-----------------------------------------------------------------------------------------------------------------------
Credits:
This code is the product of many people's efforts including but not limited to the following:
cteTally concept originally by Iztek Ben Gan and "decimalized" by Lynn Pettis (and others) for a bit of extra speed
and finally redacted by Jeff Moden for a different slant on readability and compactness. Hat's off to Paul White for
his simple explanations of CROSS APPLY and for his detailed testing efforts. Last but not least, thanks to
Ron "BitBucket" McCullough and Wayne Sheffield for their extreme performance testing across multiple machines and
versions of SQL Server. The latest improvement brought an additional 15-20% improvement over Rev 05. Special thanks
to "Nadrek" and "peter-757102" (aka Peter de Heer) for bringing such improvements to light. Nadrek's original
improvement brought about a 10% performance gain and Peter followed that up with the content of Rev 07.
I also thank whoever wrote the first article I ever saw on "numbers tables" which is located at the following URL
and to Adam Machanic for leading me to it many years ago.
http://sqlserver2000.databases.aspfaq.com/why-should-i-consider-using-an-auxiliary-numbers-table.html
-----------------------------------------------------------------------------------------------------------------------
Revision History:
Rev 00 - 20 Jan 2010 - Concept for inline cteTally: Lynn Pettis and others.
Redaction/Implementation: Jeff Moden
- Base 10 redaction and reduction for CTE. (Total rewrite)
Rev 01 - 13 Mar 2010 - Jeff Moden
- Removed one additional concatenation and one subtraction from the SUBSTRING in the SELECT List for that tiny
bit of extra speed.
Rev 02 - 14 Apr 2010 - Jeff Moden
- No code changes. Added CROSS APPLY usage example to the header, some additional credits, and extra
documentation.
Rev 03 - 18 Apr 2010 - Jeff Moden
- No code changes. Added notes 7, 8, and 9 about certain "optimizations" that don't actually work for this
type of function.
Rev 04 - 29 Jun 2010 - Jeff Moden
- Added WITH SCHEMABINDING thanks to a note by Paul White. This prevents an unnecessary "Table Spool" when the
function is used in an UPDATE statement even though the function makes no external references.
Rev 05 - 02 Apr 2011 - Jeff Moden
- Rewritten for extreme performance improvement especially for larger strings approaching the 8K boundary and
for strings that have wider elements. The redaction of this code involved removing ALL concatenation of
delimiters, optimization of the maximum "N" value by using TOP instead of including it in the WHERE clause,
and the reduction of all previous calculations (thanks to the switch to a "zero based" cteTally) to just one
instance of one add and one instance of a subtract. The length calculation for the final element (not
followed by a delimiter) in the string to be split has been greatly simplified by using the ISNULL/NULLIF
combination to determine when the CHARINDEX returned a 0 which indicates there are no more delimiters to be
had or to start with. Depending on the width of the elements, this code is between 4 and 8 times faster on a
single CPU box than the original code especially near the 8K boundary.
- Modified comments to include more sanity checks on the usage example, etc.
- Removed "other" notes 8 and 9 as they were no longer applicable.
Rev 06 - 12 Apr 2011 - Jeff Moden
- Based on a suggestion by Ron "Bitbucket" McCullough, additional test rows were added to the sample code and
the code was changed to encapsulate the output in pipes so that spaces and empty strings could be perceived
in the output. The first "Notes" section was added. Finally, an extra test was added to the comments above.
Rev 07 - 06 May 2011 - Peter de Heer, a further 15-20% performance enhancement has been discovered and incorporated
into this code which also eliminated the need for a "zero" position in the cteTally table.
**********************************************************************************************************************/
--===== Define I/O parameters
(#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
;
GO
With SQL Server 2016 we can use string_split to accomplish this:
create table commasep (
id int identity(1,1)
,string nvarchar(100) )
insert into commasep (string) values ('John, Adam'), ('test1,test2,test3')
select id, [value] as String from commasep
cross apply string_split(string,',')
CREATE FUNCTION [dbo].[fn_split_string_to_column] (
#string NVARCHAR(MAX),
#delimiter CHAR(1)
)
RETURNS #out_put TABLE (
[column_id] INT IDENTITY(1, 1) NOT NULL,
[value] NVARCHAR(MAX)
)
AS
BEGIN
DECLARE #value NVARCHAR(MAX),
#pos INT = 0,
#len INT = 0
SET #string = CASE
WHEN RIGHT(#string, 1) != #delimiter
THEN #string + #delimiter
ELSE #string
END
WHILE CHARINDEX(#delimiter, #string, #pos + 1) > 0
BEGIN
SET #len = CHARINDEX(#delimiter, #string, #pos + 1) - #pos
SET #value = SUBSTRING(#string, #pos, #len)
INSERT INTO #out_put ([value])
SELECT LTRIM(RTRIM(#value)) AS [column]
SET #pos = CHARINDEX(#delimiter, #string, #pos + #len) + 1
END
RETURN
END
SELECT id,
Substring(NAME, 0, Charindex(',', NAME)) AS firstname,
Substring(NAME, Charindex(',', NAME), Len(NAME) + 1) AS lastname
FROM spilt
Try this (change instances of ' ' to ',' or whatever delimiter you want to use)
CREATE FUNCTION dbo.Wordparser
(
#multiwordstring VARCHAR(255),
#wordnumber NUMERIC
)
returns VARCHAR(255)
AS
BEGIN
DECLARE #remainingstring VARCHAR(255)
SET #remainingstring=#multiwordstring
DECLARE #numberofwords NUMERIC
SET #numberofwords=(LEN(#remainingstring) - LEN(REPLACE(#remainingstring, ' ', '')) + 1)
DECLARE #word VARCHAR(50)
DECLARE #parsedwords TABLE
(
line NUMERIC IDENTITY(1, 1),
word VARCHAR(255)
)
WHILE #numberofwords > 1
BEGIN
SET #word=LEFT(#remainingstring, CHARINDEX(' ', #remainingstring) - 1)
INSERT INTO #parsedwords(word)
SELECT #word
SET #remainingstring= REPLACE(#remainingstring, Concat(#word, ' '), '')
SET #numberofwords=(LEN(#remainingstring) - LEN(REPLACE(#remainingstring, ' ', '')) + 1)
IF #numberofwords = 1
BREAK
ELSE
CONTINUE
END
IF #numberofwords = 1
SELECT #word = #remainingstring
INSERT INTO #parsedwords(word)
SELECT #word
RETURN
(SELECT word
FROM #parsedwords
WHERE line = #wordnumber)
END
Example usage:
SELECT dbo.Wordparser(COLUMN, 1),
dbo.Wordparser(COLUMN, 2),
dbo.Wordparser(COLUMN, 3)
FROM TABLE
I think PARSENAME is the neat function to use for this example, as described in this article: http://www.sqlshack.com/parsing-and-rotating-delimited-data-in-sql-server-2012/
The PARSENAME function is logically designed to parse four-part object names. The nice thing about PARSENAME is that it’s not limited to parsing just SQL Server four-part object names – it will parse any function or string data that is delimited by dots.
The first parameter is the object to parse, and the second is the integer value of the object piece to return. The article is discussing parsing and rotating delimited data - company phone numbers, but it can be used to parse name/surname data also.
Example:
USE COMPANY;
SELECT PARSENAME('Whatever.you.want.parsed',3) AS 'ReturnValue';
The article also describes using a Common Table Expression (CTE) called ‘replaceChars’, to run PARSENAME against the delimiter-replaced values. A CTE is useful for returning a temporary view or result set.
After that, the UNPIVOT function has been used to convert some columns into rows; SUBSTRING and CHARINDEX functions have been used for cleaning up the inconsistencies in the data, and the LAG function (new for SQL Server 2012) has been used in the end, as it allows referencing of previous records.
We can create a function as this
CREATE Function [dbo].[fn_CSVToTable]
(
#CSVList Varchar(max)
)
RETURNS #Table TABLE (ColumnData VARCHAR(100))
AS
BEGIN
IF RIGHT(#CSVList, 1) <> ','
SELECT #CSVList = #CSVList + ','
DECLARE #Pos BIGINT,
#OldPos BIGINT
SELECT #Pos = 1,
#OldPos = 1
WHILE #Pos < LEN(#CSVList)
BEGIN
SELECT #Pos = CHARINDEX(',', #CSVList, #OldPos)
INSERT INTO #Table
SELECT LTRIM(RTRIM(SUBSTRING(#CSVList, #OldPos, #Pos - #OldPos))) Col001
SELECT #OldPos = #Pos + 1
END
RETURN
END
We can then seperate the CSV values into our respective columns using a SELECT statement
You can use a table-valued function STRING_SPLIT, which is available only under compatibility level 130. If your database compatibility level is lower than 130, SQL Server will not be able to find and execute the STRING_SPLIT function. You can change a compatibility level of the database using the following command:
ALTER DATABASE DatabaseName SET COMPATIBILITY_LEVEL = 130
Syntax
SELECT * FROM STRING_SPLIT ( string, separator )
see documentation here
I think following function will work for you:
You have to create a function in SQL first. Like this
CREATE FUNCTION [dbo].[fn_split](
#str VARCHAR(MAX),
#delimiter CHAR(1)
)
RETURNS #returnTable TABLE (idx INT PRIMARY KEY IDENTITY, item VARCHAR(8000))
AS
BEGIN
DECLARE #pos INT
SELECT #str = #str + #delimiter
WHILE LEN(#str) > 0
BEGIN
SELECT #pos = CHARINDEX(#delimiter,#str)
IF #pos = 1
INSERT #returnTable (item)
VALUES (NULL)
ELSE
INSERT #returnTable (item)
VALUES (SUBSTRING(#str, 1, #pos-1))
SELECT #str = SUBSTRING(#str, #pos+1, LEN(#str)-#pos)
END
RETURN
END
You can call this function, like this:
select * from fn_split('1,24,5',',')
Implementation:
Declare #test TABLE (
ID VARCHAR(200),
Data VARCHAR(200)
)
insert into #test
(ID, Data)
Values
('1','Cleo,Smith')
insert into #test
(ID, Data)
Values
('2','Paul,Grim')
select ID,
(select item from fn_split(Data,',') where idx in (1)) as Name ,
(select item from fn_split(Data,',') where idx in (2)) as Surname
from #test
Result will like this:
Use Parsename() function
with cte as(
select 'Aria,Karimi' as FullName
Union
select 'Joe,Karimi' as FullName
Union
select 'Bab,Karimi' as FullName
)
SELECT PARSENAME(REPLACE(FullName,',','.'),2) as Name,
PARSENAME(REPLACE(FullName,',','.'),1) as Family
FROM cte
Result
Name Family
----- ------
Aria Karimi
Bab Karimi
Joe Karimi
Try this:
declare #csv varchar(100) ='aaa,bb,csda,daass';
set #csv = #csv+',';
with cte as
(
select SUBSTRING(#csv,1,charindex(',',#csv,1)-1) as val, SUBSTRING(#csv,charindex(',',#csv,1)+1,len(#csv)) as rem
UNION ALL
select SUBSTRING(a.rem,1,charindex(',',a.rem,1)-1)as val, SUBSTRING(a.rem,charindex(',',a.rem,1)+1,len(A.rem))
from cte a where LEN(a.rem)>=1
) select val from cte
This function is most fast:
CREATE FUNCTION dbo.F_ExtractSubString
(
#String VARCHAR(MAX),
#NroSubString INT,
#Separator VARCHAR(5)
)
RETURNS VARCHAR(MAX) AS
BEGIN
DECLARE #St INT = 0, #End INT = 0, #Ret VARCHAR(MAX)
SET #String = #String + #Separator
WHILE CHARINDEX(#Separator, #String, #End + 1) > 0 AND #NroSubString > 0
BEGIN
SET #St = #End + 1
SET #End = CHARINDEX(#Separator, #String, #End + 1)
SET #NroSubString = #NroSubString - 1
END
IF #NroSubString > 0
SET #Ret = ''
ELSE
SET #Ret = SUBSTRING(#String, #St, #End - #St)
RETURN #Ret
END
GO
Example usage:
SELECT dbo.F_ExtractSubString(COLUMN, 1, ', '),
dbo.F_ExtractSubString(COLUMN, 2, ', '),
dbo.F_ExtractSubString(COLUMN, 3, ', ')
FROM TABLE
I encountered a similar problem but a complex one and since this is the first thread i found regarding that issue i decided to post my finding. i know it is complex solution to a simple problem but i hope that i could help other people who go to this thread looking for a more complex solution. i had to split a string containing 5 numbers (column name: levelsFeed) and to show each number in a separate column.
for example: 8,1,2,2,2
should be shown as :
1 2 3 4 5
-------------
8 1 2 2 2
Solution 1: using XML functions:
this solution for the slowest solution by far
SELECT Distinct FeedbackID,
, S.a.value('(/H/r)[1]', 'INT') AS level1
, S.a.value('(/H/r)[2]', 'INT') AS level2
, S.a.value('(/H/r)[3]', 'INT') AS level3
, S.a.value('(/H/r)[4]', 'INT') AS level4
, S.a.value('(/H/r)[5]', 'INT') AS level5
FROM (
SELECT *,CAST (N'<H><r>' + REPLACE(levelsFeed, ',', '</r><r>') + '</r> </H>' AS XML) AS [vals]
FROM Feedbacks
) as d
CROSS APPLY d.[vals].nodes('/H/r') S(a)
Solution 2: using Split function and pivot. (the split function split a string to rows with the column name Data)
SELECT FeedbackID, [1],[2],[3],[4],[5]
FROM (
SELECT *, ROW_NUMBER() OVER (PARTITION BY feedbackID ORDER BY (SELECT null)) as rn
FROM (
SELECT FeedbackID, levelsFeed
FROM Feedbacks
) as a
CROSS APPLY dbo.Split(levelsFeed, ',')
) as SourceTable
PIVOT
(
MAX(data)
FOR rn IN ([1],[2],[3],[4],[5])
)as pivotTable
Solution 3: using string manipulations functions - fastest by small margin over solution 2
SELECT FeedbackID,
SUBSTRING(levelsFeed,0,CHARINDEX(',',levelsFeed)) AS level1,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),4) AS level2,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),3) AS level3,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),2) AS level4,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),1) AS level5
FROM Feedbacks
since the levelsFeed contains 5 string values i needed to use the substring function for the first string.
i hope that my solution will help other that got to this thread looking for a more complex split to columns methods
Using instring function :)
select Value,
substring(String,1,instr(String," ") -1) Fname,
substring(String,instr(String,",") +1) Sname
from tablename;
Used two functions,
1. substring(string, position, length) ==> returns string from positon to length
2. instr(string,pattern) ==> returns position of pattern.
If we don’t provide length argument in substring it returns until end of string
This worked for me
CREATE FUNCTION [dbo].[SplitString](
#delimited NVARCHAR(MAX),
#delimiter NVARCHAR(100)
) RETURNS #t TABLE ( 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
mytable:
Value ColOne
--------------------
1 Cleo, Smith
The following should work if there aren't too many columns
ALTER TABLE mytable ADD ColTwo nvarchar(256);
UPDATE mytable SET ColTwo = LEFT(ColOne, Charindex(',', ColOne) - 1);
--'Cleo' = LEFT('Cleo, Smith', Charindex(',', 'Cleo, Smith') - 1)
UPDATE mytable SET ColTwo = REPLACE(ColOne, ColTwo + ',', '');
--' Smith' = REPLACE('Cleo, Smith', 'Cleo' + ',')
UPDATE mytable SET ColOne = REPLACE(ColOne, ',' + ColTwo, ''), ColTwo = LTRIM(ColTwo);
--'Cleo' = REPLACE('Cleo, Smith', ',' + ' Smith', '')
Result:
Value ColOne ColTwo
--------------------
1 Cleo Smith
DECLARE #INPUT VARCHAR (MAX)='N,A,R,E,N,D,R,A'
DECLARE #ELIMINATE_CHAR CHAR (1)=','
DECLARE #L_START INT=1
DECLARE #L_END INT=(SELECT LEN (#INPUT))
DECLARE #OUTPUT CHAR (1)
WHILE #L_START <=#L_END
BEGIN
SET #OUTPUT=(SUBSTRING (#INPUT,#L_START,1))
IF #OUTPUT!=#ELIMINATE_CHAR
BEGIN
PRINT #OUTPUT
END
SET #L_START=#L_START+1
END
You may find the solution in SQL User Defined Function to Parse a Delimited String helpful (from The Code Project).
This is the code part from this page:
CREATE FUNCTION [fn_ParseText2Table]
(#p_SourceText VARCHAR(MAX)
,#p_Delimeter VARCHAR(100)=',' --default to comma delimited.
)
RETURNS #retTable
TABLE([Position] INT IDENTITY(1,1)
,[Int_Value] INT
,[Num_Value] NUMERIC(18,3)
,[Txt_Value] VARCHAR(MAX)
,[Date_value] DATETIME
)
AS
/*
********************************************************************************
Purpose: Parse values from a delimited string
& return the result as an indexed table
Copyright 1996, 1997, 2000, 2003 Clayton Groom (Clayton_Groom#hotmail.com)
Posted to the public domain Aug, 2004
2003-06-17 Rewritten as SQL 2000 function.
Reworked to allow for delimiters > 1 character in length
and to convert Text values to numbers
2016-04-05 Added logic for date values based on "new" ISDATE() function, Updated to use XML approach, which is more efficient.
********************************************************************************
*/
BEGIN
DECLARE #w_xml xml;
SET #w_xml = N'<root><i>' + replace(#p_SourceText, #p_Delimeter,'</i><i>') + '</i></root>';
INSERT INTO #retTable
([Int_Value]
, [Num_Value]
, [Txt_Value]
, [Date_value]
)
SELECT CASE
WHEN ISNUMERIC([i].value('.', 'VARCHAR(MAX)')) = 1
THEN CAST(CAST([i].value('.', 'VARCHAR(MAX)') AS NUMERIC) AS INT)
END AS [Int_Value]
, CASE
WHEN ISNUMERIC([i].value('.', 'VARCHAR(MAX)')) = 1
THEN CAST([i].value('.', 'VARCHAR(MAX)') AS NUMERIC(18, 3))
END AS [Num_Value]
, [i].value('.', 'VARCHAR(MAX)') AS [txt_Value]
, CASE
WHEN ISDATE([i].value('.', 'VARCHAR(MAX)')) = 1
THEN CAST([i].value('.', 'VARCHAR(MAX)') AS DATETIME)
END AS [Num_Value]
FROM #w_xml.nodes('//root/i') AS [Items]([i]);
RETURN;
END;
GO
ALTER function get_occurance_index(#delimiter varchar(1),#occurence int,#String varchar(100))
returns int
AS Begin
--Declare #delimiter varchar(1)=',',#occurence int=2,#String varchar(100)='a,b,c'
Declare #result int
;with T as (
select 1 Rno,0 as row, charindex(#delimiter, #String) pos,#String st
union all
select Rno+1,pos + 1, charindex(#delimiter, #String, pos + 1), #String
from T
where pos > 0
)
select #result=pos
from T
where pos > 0 and rno = #occurence
return isnull(#result,0)
ENd
declare #data as table (data varchar(100))
insert into #data values('1,2,3')
insert into #data values('aaa,bbbbb,cccc')
select top 3 Substring (data,0,dbo.get_occurance_index( ',',1,data)) ,--First Record always starts with 0
Substring (data,dbo.get_occurance_index( ',',1,data)+1,dbo.get_occurance_index( ',',2,data)-dbo.get_occurance_index( ',',1,data)-1) ,
Substring (data,dbo.get_occurance_index( ',',2,data)+1,len(data)) , -- Last record cant be more than len of actual data
data
From #data
I found that using PARSENAME as above caused any name with a period to get nulled.
So if there was an initial or a title in the name followed by a dot they return NULL.
I found this worked for me:
SELECT
REPLACE(SUBSTRING(FullName, 1,CHARINDEX(',', FullName)), ',','') as Name,
REPLACE(SUBSTRING(FullName, CHARINDEX(',', FullName), LEN(FullName)), ',', '') as Surname
FROM Table1
it is so easy, you can take it by below query:
DECLARE #str NVARCHAR(MAX)='ControlID_05436b78-04ba-9667-fa01-9ff8c1b7c235,3'
SELECT LEFT(#str, CHARINDEX(',',#str)-1),RIGHT(#str,LEN(#str)-(CHARINDEX(',',#str)))
select distinct modelFileId,F4.*
from contract
cross apply (select XmlList=convert(xml, '<x>'+replace(modelFileId,';','</x><x>')+'</x>').query('.')) F2
cross apply (select mfid1=XmlNode.value('/x[1]','varchar(512)')
,mfid2=XmlNode.value('/x[2]','varchar(512)')
,mfid3=XmlNode.value('/x[3]','varchar(512)')
,mfid4=XmlNode.value('/x[4]','varchar(512)') from XmlList.nodes('x') F3(XmlNode)) F4
where modelFileId like '%;%'
order by modelFileId
Select distinct PROJ_UID,PROJ_NAME,RES_UID from E2E_ProjectWiseTimesheetActuals
where CHARINDEX(','+cast(PROJ_UID as varchar(8000))+',', #params) > 0 and CHARINDEX(','+cast(RES_UID as varchar(8000))+',', #res) > 0
I re-wrote an answer above and made it better:
CREATE FUNCTION [dbo].[CSVParser]
(
#s VARCHAR(255),
#idx NUMERIC
)
RETURNS VARCHAR(12)
BEGIN
DECLARE #comma int
SET #comma = CHARINDEX(',', #s)
WHILE 1=1
BEGIN
IF #comma=0
IF #idx=1
RETURN #s
ELSE
RETURN ''
IF #idx=1
BEGIN
DECLARE #word VARCHAR(12)
SET #word=LEFT(#s, #comma - 1)
RETURN #word
END
SET #s = RIGHT(#s,LEN(#s)-#comma)
SET #comma = CHARINDEX(',', #s)
SET #idx = #idx - 1
END
RETURN 'not used'
END
Example usage:
SELECT dbo.CSVParser(COLUMN, 1),
dbo.CSVParser(COLUMN, 2),
dbo.CSVParser(COLUMN, 3)
FROM TABLE
question is simple, but problem is hot :)
So I create some wrapper for string_split() which pivot result in more generic way. It's table function which returns values (nn, value1, value2, ... , value50) - enough for most CSV lines. If there are more values, they will wrap to next line - nn indicate line number. Set third parameter #columnCnt = [yourNumber] to wrap at specific position:
alter FUNCTION fn_Split50
(
#str varchar(max),
#delim char(1),
#columnCnt int = 50
)
RETURNS TABLE
AS
RETURN
(
SELECT *
FROM (SELECT
nn = (nn - 1) / #columnCnt + 1,
nnn = 'value' + cast(((nn - 1) % #columnCnt) + 1 as varchar(10)),
value
FROM (SELECT
nn = ROW_NUMBER() over (order by (select null)),
value
FROM string_split(#str, #delim) aa
) aa
where nn > 0
) bb
PIVOT
(
max(value)
FOR nnn IN (
value1, value2, value3, value4, value5, value6, value7, value8, value9, value10,
value11, value12, value13, value14, value15, value16, value17, value18, value19, value20,
value21, value22, value23, value24, value25, value26, value27, value28, value29, value30,
value31, value32, value33, value34, value35, value36, value37, value38, value39, value40,
value41, value42, value43, value44, value45, value46, value47, value48, value49, value50
)
) AS PivotTable
)
Example of using:
select * from dbo.fn_split50('zz1,aa2,ss3,dd4,ff5', ',', DEFAULT)
select * from dbo.fn_split50('zz1,aa2,ss3,dd4,ff5,gg6,hh7,jj8,ww9,qq10', ',', 3)
select * from dbo.fn_split50('zz1,11,aa2,22,ss3,33,dd4,44,ff5,55,gg6,66,hh7,77,jj8,88,ww9,99,qq10,1010', ',',2)
Hope, it will helps :)

How to make this IsSimilar(varchar,varchar) function more performant

I need to implement a function to look for similar names. It's for the development of a new system that has migrated the data of a previous system.
One of the features would be that on account creation, it would try and look for person records that are there with a similar name and propose them.
Examples for similar names for "John Johnson" could be:
John Johnson
Jon Jonsen
John & Jane Johnson-Peters
Fam. Johnsen
J. Johnson
To achieve this I've created some SQL functions and functional wise they work:
[dbo].[Levenshtein]: A copy of the top rated answer from this question
[dbo].[SplitString]: Which is to split a name string based on '/', '', '&', '+' and '-'
CREATE FUNCTION [dbo].[SplitString](
#s nvarchar(4000)
)
RETURNS #table TABLE ([value] VARCHAR(4000))
WITH SCHEMABINDING
AS
BEGIN
DECLARE #repl VARCHAR(4000) = REPLACE(REPLACE(REPLACE(REPLACE(#s, '/', '-'),'&', '-'),'+', '-'),'\', '-');
INSERT INTO #table
SELECT TRIM(value) FROM STRING_SPLIT(#repl, '-', NULL)
WHERE RTRIM(value) <> '';
RETURN
END
[dbo].[IsSimilar]: Takes 2 strings, calls the split function and checks if any part of the splits are similar, meaning within Levenshtein distance 5, an initial, or 'Fam'
CREATE FUNCTION [dbo].[IsSimilar](
#s nvarchar(4000)
, #t nvarchar(4000)
)
RETURNS BIT
WITH SCHEMABINDING
AS
BEGIN
DECLARE #sT TABLE (idx int IDENTITY(1,1), [value] VARCHAR(4000))
DECLARE #tT TABLE (idx int IDENTITY(1,1), [value] VARCHAR(4000))
DECLARE #sNrOfWords INT,
#tNrOfWords INT,
#sCount INT = 1,
#tCount INT = 1,
#sVal VARCHAR(4000),
#tVal VARCHAR(4000)
IF (#s = 'fam' OR #s = 'fam.' OR #t = 'fam' OR #t = 'fam.')
return 1
INSERT INTO #sT SELECT [value] FROM [dbo].[SplitString](#s)
INSERT INTO #tT SELECT [value] FROM [dbo].[SplitString](#t)
SET #sNrOfWords = (SELECT COUNT([value]) FROM #sT)
SET #tNrOfWords = (SELECT COUNT([value]) FROM #tT)
IF (#sNrOfWords > 0 AND #tNrOfWords > 0)
BEGIN
WHILE (#sCount <= (SELECT MAX(idx) FROM #sT))
BEGIN
SET #sVal = (SELECT [value] FROM #sT WHERE idx = #sCount)
WHILE (#tCount <= (SELECT MAX(idx) FROM #tT))
BEGIN
SET #tVal = (SELECT [value] FROM #tT WHERE idx = #tCount)
IF (((LEN(#sVal) = 1 OR LEN(#tVal) = 1 OR SUBSTRING(#sVal, 2, 1) IN ('.', ' ') OR SUBSTRING(#tVal, 2, 1) IN ('.', ' ')) AND SUBSTRING(#sVal, 1, 1) = SUBSTRING(#tVal, 1, 1)) OR ((SELECT [dbo].[Levenshtein](#sVal,#tVal, 5)) IS NOT NULL))
RETURN 1
SET #tCount = #tCount + 1
END
SET #sCount = #sCount + 1
SET #tCount = 1
END
END
RETURN 0
END
Sadly this solution isn't the most performant :(
It takes 12-13s to find this record based on my mispelled name, which off course is too long.
The full table is only 512 records at the moment.
Any help on getting this more performant?
I know looping isn't recomended in SQL, so probably something to gain there.
I'm not a DBA or SQL specialist, no idea how to write that differently without the loops.
Didn't think I could use a join as there's no equality.
After implementing the suggestions in the comments on the OP, I managed to cut down the same SELECT statement from 12-13s to about 1s, which is a lot more acceptable.
The SplitString has been changed to an inline function:
Create FUNCTION [dbo].[SplitString](
#s nvarchar(4000)
)
RETURNS TABLE
WITH SCHEMABINDING
AS
RETURN(
SELECT TRIM(value) AS value FROM STRING_SPLIT(REPLACE(REPLACE(REPLACE(REPLACE(#s, '/', '-'),'&', '-'),'+', '-'),'\', '-'), '-', NULL)
WHERE RTRIM(value) <> ''
);
Cutting down on the variables and using a Join statement for the IsSimilar function gives a boost as well:
CREATE FUNCTION [dbo].[IsSimilar](
#s nvarchar(4000)
, #t nvarchar(4000)
)
RETURNS BIT
AS
BEGIN
IF (#s = 'fam' OR #s = 'fam.' OR #t = 'fam' OR #t = 'fam.')
return 1
IF (LEN(TRIM(#s)) > 0 AND LEN(TRIM(#t)) > 0)
BEGIN
RETURN (SELECT IIF (EXISTS(SELECT [sT].[value] FROM (SELECT [value] FROM [dbo].[SplitString](#s)) AS sT INNER JOIN (SELECT [value] FROM [dbo].[SplitString](#t)) AS tT ON (((LEN([sT].[value]) = 1 OR LEN([tT].[value]) = 1 OR SUBSTRING([sT].[value], 2, 1) IN ('.', ' ') OR SUBSTRING([tT].[value], 2, 1) IN ('.', ' ')) AND SUBSTRING([sT].[value], 1, 1) = SUBSTRING([tT].[value], 1, 1)) OR (NOT(SUBSTRING([sT].[value], 2, 1) IN ('.', ' ') OR SUBSTRING([tT].[value], 2, 1) IN ('.', ' ')) AND (SELECT [dbo].[Levenshtein]([sT].[value],[tT].[value], 5)) IS NOT NULL)) ), 1, 0))
END
RETURN 0
END
I don't know how much this boost will hold up to real big data, but in this case that'll not be the case as Person records get linked to Account records with every new account creation and only Person records with AccountID = NULL will be considered.

Split Delimited strings and insert them into table columns SQL 2014 [duplicate]

I have a table like this
Value String
-------------------
1 Cleo, Smith
I want to separate the comma delimited string into two columns
Value Name Surname
-------------------
1 Cleo Smith
I need only two fixed extra columns
Your purpose can be solved using following query -
Select Value , Substring(FullName, 1,Charindex(',', FullName)-1) as Name,
Substring(FullName, Charindex(',', FullName)+1, LEN(FullName)) as Surname
from Table1
There is no readymade Split function in sql server, so we need to create user defined function.
CREATE FUNCTION Split (
#InputString VARCHAR(8000),
#Delimiter VARCHAR(50)
)
RETURNS #Items TABLE (
Item VARCHAR(8000)
)
AS
BEGIN
IF #Delimiter = ' '
BEGIN
SET #Delimiter = ','
SET #InputString = REPLACE(#InputString, ' ', #Delimiter)
END
IF (#Delimiter IS NULL OR #Delimiter = '')
SET #Delimiter = ','
--INSERT INTO #Items VALUES (#Delimiter) -- Diagnostic
--INSERT INTO #Items VALUES (#InputString) -- Diagnostic
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 = #ItemList minus one less item
SET #ItemList = SUBSTRING(#ItemList, #DelimIndex+1, LEN(#ItemList)-#DelimIndex)
SET #DelimIndex = CHARINDEX(#Delimiter, #ItemList, 0)
END -- End WHILE
IF #Item IS NOT NULL -- At least one delimiter was encountered in #InputString
BEGIN
SET #Item = #ItemList
INSERT INTO #Items VALUES (#Item)
END
-- No delimiters were encountered in #InputString, so just return #InputString
ELSE INSERT INTO #Items VALUES (#InputString)
RETURN
END -- End Function
GO
---- Set Permissions
--GRANT SELECT ON Split TO UserRole1
--GRANT SELECT ON Split TO UserRole2
--GO
;WITH Split_Names (Value,Name, xmlname)
AS
(
SELECT Value,
Name,
CONVERT(XML,'<Names><name>'
+ REPLACE(Name,',', '</name><name>') + '</name></Names>') AS xmlname
FROM tblnames
)
SELECT Value,
xmlname.value('/Names[1]/name[1]','varchar(100)') AS Name,
xmlname.value('/Names[1]/name[2]','varchar(100)') AS Surname
FROM Split_Names
and also check the link below for reference
http://jahaines.blogspot.in/2009/06/converting-delimited-string-of-values.html
xml based answer is simple and clean
refer this
DECLARE #S varchar(max),
#Split char(1),
#X xml
SELECT #S = 'ab,cd,ef,gh,ij',
#Split = ','
SELECT #X = CONVERT(xml,' <root> <myvalue>' +
REPLACE(#S,#Split,'</myvalue> <myvalue>') + '</myvalue> </root> ')
SELECT T.c.value('.','varchar(20)'), --retrieve ALL values at once
T.c.value('(/root/myvalue)[1]','VARCHAR(20)') , --retrieve index 1 only, which is the 'ab'
T.c.value('(/root/myvalue)[2]','VARCHAR(20)')
FROM #X.nodes('/root/myvalue') T(c)
I think this is cool
SELECT value,
PARSENAME(REPLACE(String,',','.'),2) 'Name' ,
PARSENAME(REPLACE(String,',','.'),1) 'Surname'
FROM table WITH (NOLOCK)
With CROSS APPLY
select ParsedData.*
from MyTable mt
cross apply ( select str = mt.String + ',,' ) f1
cross apply ( select p1 = charindex( ',', str ) ) ap1
cross apply ( select p2 = charindex( ',', str, p1 + 1 ) ) ap2
cross apply ( select Nmame = substring( str, 1, p1-1 )
, Surname = substring( str, p1+1, p2-p1-1 )
) ParsedData
There are multiple ways to solve this and many different ways have been proposed already. Simplest would be to use LEFT / SUBSTRING and other string functions to achieve the desired result.
Sample Data
DECLARE #tbl1 TABLE (Value INT,String VARCHAR(MAX))
INSERT INTO #tbl1 VALUES(1,'Cleo, Smith');
INSERT INTO #tbl1 VALUES(2,'John, Mathew');
Using String Functions like LEFT
SELECT
Value,
LEFT(String,CHARINDEX(',',String)-1) as Fname,
LTRIM(RIGHT(String,LEN(String) - CHARINDEX(',',String) )) AS Lname
FROM #tbl1
This approach fails if there are more 2 items in a String.
In such a scenario, we can use a splitter and then use PIVOT or convert the string into an XML and use .nodes to get string items. XML based solution have been detailed out by aads and bvr in their solution.
The answers for this question which use splitter, all use WHILE which is inefficient for splitting. Check this performance comparison. One of the best splitters around is DelimitedSplit8K, created by Jeff Moden. You can read more about it here
Splitter with PIVOT
DECLARE #tbl1 TABLE (Value INT,String VARCHAR(MAX))
INSERT INTO #tbl1 VALUES(1,'Cleo, Smith');
INSERT INTO #tbl1 VALUES(2,'John, Mathew');
SELECT t3.Value,[1] as Fname,[2] as Lname
FROM #tbl1 as t1
CROSS APPLY [dbo].[DelimitedSplit8K](String,',') as t2
PIVOT(MAX(Item) FOR ItemNumber IN ([1],[2])) as t3
Output
Value Fname Lname
1 Cleo Smith
2 John Mathew
DelimitedSplit8K by Jeff Moden
CREATE FUNCTION [dbo].[DelimitedSplit8K]
/**********************************************************************************************************************
Purpose:
Split a given string at a given delimiter and return a list of the split elements (items).
Notes:
1. Leading a trailing delimiters are treated as if an empty string element were present.
2. Consecutive delimiters are treated as if an empty string element were present between them.
3. Except when spaces are used as a delimiter, all spaces present in each element are preserved.
Returns:
iTVF containing the following:
ItemNumber = Element position of Item as a BIGINT (not converted to INT to eliminate a CAST)
Item = Element value as a VARCHAR(8000)
Statistics on this function may be found at the following URL:
http://www.sqlservercentral.com/Forums/Topic1101315-203-4.aspx
CROSS APPLY Usage Examples and Tests:
--=====================================================================================================================
-- TEST 1:
-- This tests for various possible conditions in a string using a comma as the delimiter. The expected results are
-- laid out in the comments
--=====================================================================================================================
--===== Conditionally drop the test tables to make reruns easier for testing.
-- (this is NOT a part of the solution)
IF OBJECT_ID('tempdb..#JBMTest') IS NOT NULL DROP TABLE #JBMTest
;
--===== Create and populate a test table on the fly (this is NOT a part of the solution).
-- In the following comments, "b" is a blank and "E" is an element in the left to right order.
-- Double Quotes are used to encapsulate the output of "Item" so that you can see that all blanks
-- are preserved no matter where they may appear.
SELECT *
INTO #JBMTest
FROM ( --# & type of Return Row(s)
SELECT 0, NULL UNION ALL --1 NULL
SELECT 1, SPACE(0) UNION ALL --1 b (Empty String)
SELECT 2, SPACE(1) UNION ALL --1 b (1 space)
SELECT 3, SPACE(5) UNION ALL --1 b (5 spaces)
SELECT 4, ',' UNION ALL --2 b b (both are empty strings)
SELECT 5, '55555' UNION ALL --1 E
SELECT 6, ',55555' UNION ALL --2 b E
SELECT 7, ',55555,' UNION ALL --3 b E b
SELECT 8, '55555,' UNION ALL --2 b B
SELECT 9, '55555,1' UNION ALL --2 E E
SELECT 10, '1,55555' UNION ALL --2 E E
SELECT 11, '55555,4444,333,22,1' UNION ALL --5 E E E E E
SELECT 12, '55555,4444,,333,22,1' UNION ALL --6 E E b E E E
SELECT 13, ',55555,4444,,333,22,1,' UNION ALL --8 b E E b E E E b
SELECT 14, ',55555,4444,,,333,22,1,' UNION ALL --9 b E E b b E E E b
SELECT 15, ' 4444,55555 ' UNION ALL --2 E (w/Leading Space) E (w/Trailing Space)
SELECT 16, 'This,is,a,test.' --E E E E
) d (SomeID, SomeValue)
;
--===== Split the CSV column for the whole table using CROSS APPLY (this is the solution)
SELECT test.SomeID, test.SomeValue, split.ItemNumber, Item = QUOTENAME(split.Item,'"')
FROM #JBMTest test
CROSS APPLY dbo.DelimitedSplit8K(test.SomeValue,',') split
;
--=====================================================================================================================
-- TEST 2:
-- This tests for various "alpha" splits and COLLATION using all ASCII characters from 0 to 255 as a delimiter against
-- a given string. Note that not all of the delimiters will be visible and some will show up as tiny squares because
-- they are "control" characters. More specifically, this test will show you what happens to various non-accented
-- letters for your given collation depending on the delimiter you chose.
--=====================================================================================================================
WITH
cteBuildAllCharacters (String,Delimiter) AS
(
SELECT TOP 256
'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789',
CHAR(ROW_NUMBER() OVER (ORDER BY (SELECT NULL))-1)
FROM master.sys.all_columns
)
SELECT ASCII_Value = ASCII(c.Delimiter), c.Delimiter, split.ItemNumber, Item = QUOTENAME(split.Item,'"')
FROM cteBuildAllCharacters c
CROSS APPLY dbo.DelimitedSplit8K(c.String,c.Delimiter) split
ORDER BY ASCII_Value, split.ItemNumber
;
-----------------------------------------------------------------------------------------------------------------------
Other Notes:
1. Optimized for VARCHAR(8000) or less. No testing or error reporting for truncation at 8000 characters is done.
2. Optimized for single character delimiter. Multi-character delimiters should be resolvedexternally from this
function.
3. Optimized for use with CROSS APPLY.
4. Does not "trim" elements just in case leading or trailing blanks are intended.
5. If you don't know how a Tally table can be used to replace loops, please see the following...
http://www.sqlservercentral.com/articles/T-SQL/62867/
6. Changing this function to use NVARCHAR(MAX) will cause it to run twice as slow. It's just the nature of
VARCHAR(MAX) whether it fits in-row or not.
7. Multi-machine testing for the method of using UNPIVOT instead of 10 SELECT/UNION ALLs shows that the UNPIVOT method
is quite machine dependent and can slow things down quite a bit.
-----------------------------------------------------------------------------------------------------------------------
Credits:
This code is the product of many people's efforts including but not limited to the following:
cteTally concept originally by Iztek Ben Gan and "decimalized" by Lynn Pettis (and others) for a bit of extra speed
and finally redacted by Jeff Moden for a different slant on readability and compactness. Hat's off to Paul White for
his simple explanations of CROSS APPLY and for his detailed testing efforts. Last but not least, thanks to
Ron "BitBucket" McCullough and Wayne Sheffield for their extreme performance testing across multiple machines and
versions of SQL Server. The latest improvement brought an additional 15-20% improvement over Rev 05. Special thanks
to "Nadrek" and "peter-757102" (aka Peter de Heer) for bringing such improvements to light. Nadrek's original
improvement brought about a 10% performance gain and Peter followed that up with the content of Rev 07.
I also thank whoever wrote the first article I ever saw on "numbers tables" which is located at the following URL
and to Adam Machanic for leading me to it many years ago.
http://sqlserver2000.databases.aspfaq.com/why-should-i-consider-using-an-auxiliary-numbers-table.html
-----------------------------------------------------------------------------------------------------------------------
Revision History:
Rev 00 - 20 Jan 2010 - Concept for inline cteTally: Lynn Pettis and others.
Redaction/Implementation: Jeff Moden
- Base 10 redaction and reduction for CTE. (Total rewrite)
Rev 01 - 13 Mar 2010 - Jeff Moden
- Removed one additional concatenation and one subtraction from the SUBSTRING in the SELECT List for that tiny
bit of extra speed.
Rev 02 - 14 Apr 2010 - Jeff Moden
- No code changes. Added CROSS APPLY usage example to the header, some additional credits, and extra
documentation.
Rev 03 - 18 Apr 2010 - Jeff Moden
- No code changes. Added notes 7, 8, and 9 about certain "optimizations" that don't actually work for this
type of function.
Rev 04 - 29 Jun 2010 - Jeff Moden
- Added WITH SCHEMABINDING thanks to a note by Paul White. This prevents an unnecessary "Table Spool" when the
function is used in an UPDATE statement even though the function makes no external references.
Rev 05 - 02 Apr 2011 - Jeff Moden
- Rewritten for extreme performance improvement especially for larger strings approaching the 8K boundary and
for strings that have wider elements. The redaction of this code involved removing ALL concatenation of
delimiters, optimization of the maximum "N" value by using TOP instead of including it in the WHERE clause,
and the reduction of all previous calculations (thanks to the switch to a "zero based" cteTally) to just one
instance of one add and one instance of a subtract. The length calculation for the final element (not
followed by a delimiter) in the string to be split has been greatly simplified by using the ISNULL/NULLIF
combination to determine when the CHARINDEX returned a 0 which indicates there are no more delimiters to be
had or to start with. Depending on the width of the elements, this code is between 4 and 8 times faster on a
single CPU box than the original code especially near the 8K boundary.
- Modified comments to include more sanity checks on the usage example, etc.
- Removed "other" notes 8 and 9 as they were no longer applicable.
Rev 06 - 12 Apr 2011 - Jeff Moden
- Based on a suggestion by Ron "Bitbucket" McCullough, additional test rows were added to the sample code and
the code was changed to encapsulate the output in pipes so that spaces and empty strings could be perceived
in the output. The first "Notes" section was added. Finally, an extra test was added to the comments above.
Rev 07 - 06 May 2011 - Peter de Heer, a further 15-20% performance enhancement has been discovered and incorporated
into this code which also eliminated the need for a "zero" position in the cteTally table.
**********************************************************************************************************************/
--===== Define I/O parameters
(#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
;
GO
With SQL Server 2016 we can use string_split to accomplish this:
create table commasep (
id int identity(1,1)
,string nvarchar(100) )
insert into commasep (string) values ('John, Adam'), ('test1,test2,test3')
select id, [value] as String from commasep
cross apply string_split(string,',')
CREATE FUNCTION [dbo].[fn_split_string_to_column] (
#string NVARCHAR(MAX),
#delimiter CHAR(1)
)
RETURNS #out_put TABLE (
[column_id] INT IDENTITY(1, 1) NOT NULL,
[value] NVARCHAR(MAX)
)
AS
BEGIN
DECLARE #value NVARCHAR(MAX),
#pos INT = 0,
#len INT = 0
SET #string = CASE
WHEN RIGHT(#string, 1) != #delimiter
THEN #string + #delimiter
ELSE #string
END
WHILE CHARINDEX(#delimiter, #string, #pos + 1) > 0
BEGIN
SET #len = CHARINDEX(#delimiter, #string, #pos + 1) - #pos
SET #value = SUBSTRING(#string, #pos, #len)
INSERT INTO #out_put ([value])
SELECT LTRIM(RTRIM(#value)) AS [column]
SET #pos = CHARINDEX(#delimiter, #string, #pos + #len) + 1
END
RETURN
END
SELECT id,
Substring(NAME, 0, Charindex(',', NAME)) AS firstname,
Substring(NAME, Charindex(',', NAME), Len(NAME) + 1) AS lastname
FROM spilt
Try this (change instances of ' ' to ',' or whatever delimiter you want to use)
CREATE FUNCTION dbo.Wordparser
(
#multiwordstring VARCHAR(255),
#wordnumber NUMERIC
)
returns VARCHAR(255)
AS
BEGIN
DECLARE #remainingstring VARCHAR(255)
SET #remainingstring=#multiwordstring
DECLARE #numberofwords NUMERIC
SET #numberofwords=(LEN(#remainingstring) - LEN(REPLACE(#remainingstring, ' ', '')) + 1)
DECLARE #word VARCHAR(50)
DECLARE #parsedwords TABLE
(
line NUMERIC IDENTITY(1, 1),
word VARCHAR(255)
)
WHILE #numberofwords > 1
BEGIN
SET #word=LEFT(#remainingstring, CHARINDEX(' ', #remainingstring) - 1)
INSERT INTO #parsedwords(word)
SELECT #word
SET #remainingstring= REPLACE(#remainingstring, Concat(#word, ' '), '')
SET #numberofwords=(LEN(#remainingstring) - LEN(REPLACE(#remainingstring, ' ', '')) + 1)
IF #numberofwords = 1
BREAK
ELSE
CONTINUE
END
IF #numberofwords = 1
SELECT #word = #remainingstring
INSERT INTO #parsedwords(word)
SELECT #word
RETURN
(SELECT word
FROM #parsedwords
WHERE line = #wordnumber)
END
Example usage:
SELECT dbo.Wordparser(COLUMN, 1),
dbo.Wordparser(COLUMN, 2),
dbo.Wordparser(COLUMN, 3)
FROM TABLE
I think PARSENAME is the neat function to use for this example, as described in this article: http://www.sqlshack.com/parsing-and-rotating-delimited-data-in-sql-server-2012/
The PARSENAME function is logically designed to parse four-part object names. The nice thing about PARSENAME is that it’s not limited to parsing just SQL Server four-part object names – it will parse any function or string data that is delimited by dots.
The first parameter is the object to parse, and the second is the integer value of the object piece to return. The article is discussing parsing and rotating delimited data - company phone numbers, but it can be used to parse name/surname data also.
Example:
USE COMPANY;
SELECT PARSENAME('Whatever.you.want.parsed',3) AS 'ReturnValue';
The article also describes using a Common Table Expression (CTE) called ‘replaceChars’, to run PARSENAME against the delimiter-replaced values. A CTE is useful for returning a temporary view or result set.
After that, the UNPIVOT function has been used to convert some columns into rows; SUBSTRING and CHARINDEX functions have been used for cleaning up the inconsistencies in the data, and the LAG function (new for SQL Server 2012) has been used in the end, as it allows referencing of previous records.
We can create a function as this
CREATE Function [dbo].[fn_CSVToTable]
(
#CSVList Varchar(max)
)
RETURNS #Table TABLE (ColumnData VARCHAR(100))
AS
BEGIN
IF RIGHT(#CSVList, 1) <> ','
SELECT #CSVList = #CSVList + ','
DECLARE #Pos BIGINT,
#OldPos BIGINT
SELECT #Pos = 1,
#OldPos = 1
WHILE #Pos < LEN(#CSVList)
BEGIN
SELECT #Pos = CHARINDEX(',', #CSVList, #OldPos)
INSERT INTO #Table
SELECT LTRIM(RTRIM(SUBSTRING(#CSVList, #OldPos, #Pos - #OldPos))) Col001
SELECT #OldPos = #Pos + 1
END
RETURN
END
We can then seperate the CSV values into our respective columns using a SELECT statement
You can use a table-valued function STRING_SPLIT, which is available only under compatibility level 130. If your database compatibility level is lower than 130, SQL Server will not be able to find and execute the STRING_SPLIT function. You can change a compatibility level of the database using the following command:
ALTER DATABASE DatabaseName SET COMPATIBILITY_LEVEL = 130
Syntax
SELECT * FROM STRING_SPLIT ( string, separator )
see documentation here
I think following function will work for you:
You have to create a function in SQL first. Like this
CREATE FUNCTION [dbo].[fn_split](
#str VARCHAR(MAX),
#delimiter CHAR(1)
)
RETURNS #returnTable TABLE (idx INT PRIMARY KEY IDENTITY, item VARCHAR(8000))
AS
BEGIN
DECLARE #pos INT
SELECT #str = #str + #delimiter
WHILE LEN(#str) > 0
BEGIN
SELECT #pos = CHARINDEX(#delimiter,#str)
IF #pos = 1
INSERT #returnTable (item)
VALUES (NULL)
ELSE
INSERT #returnTable (item)
VALUES (SUBSTRING(#str, 1, #pos-1))
SELECT #str = SUBSTRING(#str, #pos+1, LEN(#str)-#pos)
END
RETURN
END
You can call this function, like this:
select * from fn_split('1,24,5',',')
Implementation:
Declare #test TABLE (
ID VARCHAR(200),
Data VARCHAR(200)
)
insert into #test
(ID, Data)
Values
('1','Cleo,Smith')
insert into #test
(ID, Data)
Values
('2','Paul,Grim')
select ID,
(select item from fn_split(Data,',') where idx in (1)) as Name ,
(select item from fn_split(Data,',') where idx in (2)) as Surname
from #test
Result will like this:
Use Parsename() function
with cte as(
select 'Aria,Karimi' as FullName
Union
select 'Joe,Karimi' as FullName
Union
select 'Bab,Karimi' as FullName
)
SELECT PARSENAME(REPLACE(FullName,',','.'),2) as Name,
PARSENAME(REPLACE(FullName,',','.'),1) as Family
FROM cte
Result
Name Family
----- ------
Aria Karimi
Bab Karimi
Joe Karimi
Try this:
declare #csv varchar(100) ='aaa,bb,csda,daass';
set #csv = #csv+',';
with cte as
(
select SUBSTRING(#csv,1,charindex(',',#csv,1)-1) as val, SUBSTRING(#csv,charindex(',',#csv,1)+1,len(#csv)) as rem
UNION ALL
select SUBSTRING(a.rem,1,charindex(',',a.rem,1)-1)as val, SUBSTRING(a.rem,charindex(',',a.rem,1)+1,len(A.rem))
from cte a where LEN(a.rem)>=1
) select val from cte
This function is most fast:
CREATE FUNCTION dbo.F_ExtractSubString
(
#String VARCHAR(MAX),
#NroSubString INT,
#Separator VARCHAR(5)
)
RETURNS VARCHAR(MAX) AS
BEGIN
DECLARE #St INT = 0, #End INT = 0, #Ret VARCHAR(MAX)
SET #String = #String + #Separator
WHILE CHARINDEX(#Separator, #String, #End + 1) > 0 AND #NroSubString > 0
BEGIN
SET #St = #End + 1
SET #End = CHARINDEX(#Separator, #String, #End + 1)
SET #NroSubString = #NroSubString - 1
END
IF #NroSubString > 0
SET #Ret = ''
ELSE
SET #Ret = SUBSTRING(#String, #St, #End - #St)
RETURN #Ret
END
GO
Example usage:
SELECT dbo.F_ExtractSubString(COLUMN, 1, ', '),
dbo.F_ExtractSubString(COLUMN, 2, ', '),
dbo.F_ExtractSubString(COLUMN, 3, ', ')
FROM TABLE
I encountered a similar problem but a complex one and since this is the first thread i found regarding that issue i decided to post my finding. i know it is complex solution to a simple problem but i hope that i could help other people who go to this thread looking for a more complex solution. i had to split a string containing 5 numbers (column name: levelsFeed) and to show each number in a separate column.
for example: 8,1,2,2,2
should be shown as :
1 2 3 4 5
-------------
8 1 2 2 2
Solution 1: using XML functions:
this solution for the slowest solution by far
SELECT Distinct FeedbackID,
, S.a.value('(/H/r)[1]', 'INT') AS level1
, S.a.value('(/H/r)[2]', 'INT') AS level2
, S.a.value('(/H/r)[3]', 'INT') AS level3
, S.a.value('(/H/r)[4]', 'INT') AS level4
, S.a.value('(/H/r)[5]', 'INT') AS level5
FROM (
SELECT *,CAST (N'<H><r>' + REPLACE(levelsFeed, ',', '</r><r>') + '</r> </H>' AS XML) AS [vals]
FROM Feedbacks
) as d
CROSS APPLY d.[vals].nodes('/H/r') S(a)
Solution 2: using Split function and pivot. (the split function split a string to rows with the column name Data)
SELECT FeedbackID, [1],[2],[3],[4],[5]
FROM (
SELECT *, ROW_NUMBER() OVER (PARTITION BY feedbackID ORDER BY (SELECT null)) as rn
FROM (
SELECT FeedbackID, levelsFeed
FROM Feedbacks
) as a
CROSS APPLY dbo.Split(levelsFeed, ',')
) as SourceTable
PIVOT
(
MAX(data)
FOR rn IN ([1],[2],[3],[4],[5])
)as pivotTable
Solution 3: using string manipulations functions - fastest by small margin over solution 2
SELECT FeedbackID,
SUBSTRING(levelsFeed,0,CHARINDEX(',',levelsFeed)) AS level1,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),4) AS level2,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),3) AS level3,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),2) AS level4,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),1) AS level5
FROM Feedbacks
since the levelsFeed contains 5 string values i needed to use the substring function for the first string.
i hope that my solution will help other that got to this thread looking for a more complex split to columns methods
Using instring function :)
select Value,
substring(String,1,instr(String," ") -1) Fname,
substring(String,instr(String,",") +1) Sname
from tablename;
Used two functions,
1. substring(string, position, length) ==> returns string from positon to length
2. instr(string,pattern) ==> returns position of pattern.
If we don’t provide length argument in substring it returns until end of string
This worked for me
CREATE FUNCTION [dbo].[SplitString](
#delimited NVARCHAR(MAX),
#delimiter NVARCHAR(100)
) RETURNS #t TABLE ( 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
mytable:
Value ColOne
--------------------
1 Cleo, Smith
The following should work if there aren't too many columns
ALTER TABLE mytable ADD ColTwo nvarchar(256);
UPDATE mytable SET ColTwo = LEFT(ColOne, Charindex(',', ColOne) - 1);
--'Cleo' = LEFT('Cleo, Smith', Charindex(',', 'Cleo, Smith') - 1)
UPDATE mytable SET ColTwo = REPLACE(ColOne, ColTwo + ',', '');
--' Smith' = REPLACE('Cleo, Smith', 'Cleo' + ',')
UPDATE mytable SET ColOne = REPLACE(ColOne, ',' + ColTwo, ''), ColTwo = LTRIM(ColTwo);
--'Cleo' = REPLACE('Cleo, Smith', ',' + ' Smith', '')
Result:
Value ColOne ColTwo
--------------------
1 Cleo Smith
DECLARE #INPUT VARCHAR (MAX)='N,A,R,E,N,D,R,A'
DECLARE #ELIMINATE_CHAR CHAR (1)=','
DECLARE #L_START INT=1
DECLARE #L_END INT=(SELECT LEN (#INPUT))
DECLARE #OUTPUT CHAR (1)
WHILE #L_START <=#L_END
BEGIN
SET #OUTPUT=(SUBSTRING (#INPUT,#L_START,1))
IF #OUTPUT!=#ELIMINATE_CHAR
BEGIN
PRINT #OUTPUT
END
SET #L_START=#L_START+1
END
You may find the solution in SQL User Defined Function to Parse a Delimited String helpful (from The Code Project).
This is the code part from this page:
CREATE FUNCTION [fn_ParseText2Table]
(#p_SourceText VARCHAR(MAX)
,#p_Delimeter VARCHAR(100)=',' --default to comma delimited.
)
RETURNS #retTable
TABLE([Position] INT IDENTITY(1,1)
,[Int_Value] INT
,[Num_Value] NUMERIC(18,3)
,[Txt_Value] VARCHAR(MAX)
,[Date_value] DATETIME
)
AS
/*
********************************************************************************
Purpose: Parse values from a delimited string
& return the result as an indexed table
Copyright 1996, 1997, 2000, 2003 Clayton Groom (Clayton_Groom#hotmail.com)
Posted to the public domain Aug, 2004
2003-06-17 Rewritten as SQL 2000 function.
Reworked to allow for delimiters > 1 character in length
and to convert Text values to numbers
2016-04-05 Added logic for date values based on "new" ISDATE() function, Updated to use XML approach, which is more efficient.
********************************************************************************
*/
BEGIN
DECLARE #w_xml xml;
SET #w_xml = N'<root><i>' + replace(#p_SourceText, #p_Delimeter,'</i><i>') + '</i></root>';
INSERT INTO #retTable
([Int_Value]
, [Num_Value]
, [Txt_Value]
, [Date_value]
)
SELECT CASE
WHEN ISNUMERIC([i].value('.', 'VARCHAR(MAX)')) = 1
THEN CAST(CAST([i].value('.', 'VARCHAR(MAX)') AS NUMERIC) AS INT)
END AS [Int_Value]
, CASE
WHEN ISNUMERIC([i].value('.', 'VARCHAR(MAX)')) = 1
THEN CAST([i].value('.', 'VARCHAR(MAX)') AS NUMERIC(18, 3))
END AS [Num_Value]
, [i].value('.', 'VARCHAR(MAX)') AS [txt_Value]
, CASE
WHEN ISDATE([i].value('.', 'VARCHAR(MAX)')) = 1
THEN CAST([i].value('.', 'VARCHAR(MAX)') AS DATETIME)
END AS [Num_Value]
FROM #w_xml.nodes('//root/i') AS [Items]([i]);
RETURN;
END;
GO
ALTER function get_occurance_index(#delimiter varchar(1),#occurence int,#String varchar(100))
returns int
AS Begin
--Declare #delimiter varchar(1)=',',#occurence int=2,#String varchar(100)='a,b,c'
Declare #result int
;with T as (
select 1 Rno,0 as row, charindex(#delimiter, #String) pos,#String st
union all
select Rno+1,pos + 1, charindex(#delimiter, #String, pos + 1), #String
from T
where pos > 0
)
select #result=pos
from T
where pos > 0 and rno = #occurence
return isnull(#result,0)
ENd
declare #data as table (data varchar(100))
insert into #data values('1,2,3')
insert into #data values('aaa,bbbbb,cccc')
select top 3 Substring (data,0,dbo.get_occurance_index( ',',1,data)) ,--First Record always starts with 0
Substring (data,dbo.get_occurance_index( ',',1,data)+1,dbo.get_occurance_index( ',',2,data)-dbo.get_occurance_index( ',',1,data)-1) ,
Substring (data,dbo.get_occurance_index( ',',2,data)+1,len(data)) , -- Last record cant be more than len of actual data
data
From #data
I found that using PARSENAME as above caused any name with a period to get nulled.
So if there was an initial or a title in the name followed by a dot they return NULL.
I found this worked for me:
SELECT
REPLACE(SUBSTRING(FullName, 1,CHARINDEX(',', FullName)), ',','') as Name,
REPLACE(SUBSTRING(FullName, CHARINDEX(',', FullName), LEN(FullName)), ',', '') as Surname
FROM Table1
it is so easy, you can take it by below query:
DECLARE #str NVARCHAR(MAX)='ControlID_05436b78-04ba-9667-fa01-9ff8c1b7c235,3'
SELECT LEFT(#str, CHARINDEX(',',#str)-1),RIGHT(#str,LEN(#str)-(CHARINDEX(',',#str)))
select distinct modelFileId,F4.*
from contract
cross apply (select XmlList=convert(xml, '<x>'+replace(modelFileId,';','</x><x>')+'</x>').query('.')) F2
cross apply (select mfid1=XmlNode.value('/x[1]','varchar(512)')
,mfid2=XmlNode.value('/x[2]','varchar(512)')
,mfid3=XmlNode.value('/x[3]','varchar(512)')
,mfid4=XmlNode.value('/x[4]','varchar(512)') from XmlList.nodes('x') F3(XmlNode)) F4
where modelFileId like '%;%'
order by modelFileId
Select distinct PROJ_UID,PROJ_NAME,RES_UID from E2E_ProjectWiseTimesheetActuals
where CHARINDEX(','+cast(PROJ_UID as varchar(8000))+',', #params) > 0 and CHARINDEX(','+cast(RES_UID as varchar(8000))+',', #res) > 0
I re-wrote an answer above and made it better:
CREATE FUNCTION [dbo].[CSVParser]
(
#s VARCHAR(255),
#idx NUMERIC
)
RETURNS VARCHAR(12)
BEGIN
DECLARE #comma int
SET #comma = CHARINDEX(',', #s)
WHILE 1=1
BEGIN
IF #comma=0
IF #idx=1
RETURN #s
ELSE
RETURN ''
IF #idx=1
BEGIN
DECLARE #word VARCHAR(12)
SET #word=LEFT(#s, #comma - 1)
RETURN #word
END
SET #s = RIGHT(#s,LEN(#s)-#comma)
SET #comma = CHARINDEX(',', #s)
SET #idx = #idx - 1
END
RETURN 'not used'
END
Example usage:
SELECT dbo.CSVParser(COLUMN, 1),
dbo.CSVParser(COLUMN, 2),
dbo.CSVParser(COLUMN, 3)
FROM TABLE
question is simple, but problem is hot :)
So I create some wrapper for string_split() which pivot result in more generic way. It's table function which returns values (nn, value1, value2, ... , value50) - enough for most CSV lines. If there are more values, they will wrap to next line - nn indicate line number. Set third parameter #columnCnt = [yourNumber] to wrap at specific position:
alter FUNCTION fn_Split50
(
#str varchar(max),
#delim char(1),
#columnCnt int = 50
)
RETURNS TABLE
AS
RETURN
(
SELECT *
FROM (SELECT
nn = (nn - 1) / #columnCnt + 1,
nnn = 'value' + cast(((nn - 1) % #columnCnt) + 1 as varchar(10)),
value
FROM (SELECT
nn = ROW_NUMBER() over (order by (select null)),
value
FROM string_split(#str, #delim) aa
) aa
where nn > 0
) bb
PIVOT
(
max(value)
FOR nnn IN (
value1, value2, value3, value4, value5, value6, value7, value8, value9, value10,
value11, value12, value13, value14, value15, value16, value17, value18, value19, value20,
value21, value22, value23, value24, value25, value26, value27, value28, value29, value30,
value31, value32, value33, value34, value35, value36, value37, value38, value39, value40,
value41, value42, value43, value44, value45, value46, value47, value48, value49, value50
)
) AS PivotTable
)
Example of using:
select * from dbo.fn_split50('zz1,aa2,ss3,dd4,ff5', ',', DEFAULT)
select * from dbo.fn_split50('zz1,aa2,ss3,dd4,ff5,gg6,hh7,jj8,ww9,qq10', ',', 3)
select * from dbo.fn_split50('zz1,11,aa2,22,ss3,33,dd4,44,ff5,55,gg6,66,hh7,77,jj8,88,ww9,99,qq10,1010', ',',2)
Hope, it will helps :)

SQL split a colon and Space from the Description Field/Column [duplicate]

I have a table like this
Value String
-------------------
1 Cleo, Smith
I want to separate the comma delimited string into two columns
Value Name Surname
-------------------
1 Cleo Smith
I need only two fixed extra columns
Your purpose can be solved using following query -
Select Value , Substring(FullName, 1,Charindex(',', FullName)-1) as Name,
Substring(FullName, Charindex(',', FullName)+1, LEN(FullName)) as Surname
from Table1
There is no readymade Split function in sql server, so we need to create user defined function.
CREATE FUNCTION Split (
#InputString VARCHAR(8000),
#Delimiter VARCHAR(50)
)
RETURNS #Items TABLE (
Item VARCHAR(8000)
)
AS
BEGIN
IF #Delimiter = ' '
BEGIN
SET #Delimiter = ','
SET #InputString = REPLACE(#InputString, ' ', #Delimiter)
END
IF (#Delimiter IS NULL OR #Delimiter = '')
SET #Delimiter = ','
--INSERT INTO #Items VALUES (#Delimiter) -- Diagnostic
--INSERT INTO #Items VALUES (#InputString) -- Diagnostic
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 = #ItemList minus one less item
SET #ItemList = SUBSTRING(#ItemList, #DelimIndex+1, LEN(#ItemList)-#DelimIndex)
SET #DelimIndex = CHARINDEX(#Delimiter, #ItemList, 0)
END -- End WHILE
IF #Item IS NOT NULL -- At least one delimiter was encountered in #InputString
BEGIN
SET #Item = #ItemList
INSERT INTO #Items VALUES (#Item)
END
-- No delimiters were encountered in #InputString, so just return #InputString
ELSE INSERT INTO #Items VALUES (#InputString)
RETURN
END -- End Function
GO
---- Set Permissions
--GRANT SELECT ON Split TO UserRole1
--GRANT SELECT ON Split TO UserRole2
--GO
;WITH Split_Names (Value,Name, xmlname)
AS
(
SELECT Value,
Name,
CONVERT(XML,'<Names><name>'
+ REPLACE(Name,',', '</name><name>') + '</name></Names>') AS xmlname
FROM tblnames
)
SELECT Value,
xmlname.value('/Names[1]/name[1]','varchar(100)') AS Name,
xmlname.value('/Names[1]/name[2]','varchar(100)') AS Surname
FROM Split_Names
and also check the link below for reference
http://jahaines.blogspot.in/2009/06/converting-delimited-string-of-values.html
xml based answer is simple and clean
refer this
DECLARE #S varchar(max),
#Split char(1),
#X xml
SELECT #S = 'ab,cd,ef,gh,ij',
#Split = ','
SELECT #X = CONVERT(xml,' <root> <myvalue>' +
REPLACE(#S,#Split,'</myvalue> <myvalue>') + '</myvalue> </root> ')
SELECT T.c.value('.','varchar(20)'), --retrieve ALL values at once
T.c.value('(/root/myvalue)[1]','VARCHAR(20)') , --retrieve index 1 only, which is the 'ab'
T.c.value('(/root/myvalue)[2]','VARCHAR(20)')
FROM #X.nodes('/root/myvalue') T(c)
I think this is cool
SELECT value,
PARSENAME(REPLACE(String,',','.'),2) 'Name' ,
PARSENAME(REPLACE(String,',','.'),1) 'Surname'
FROM table WITH (NOLOCK)
With CROSS APPLY
select ParsedData.*
from MyTable mt
cross apply ( select str = mt.String + ',,' ) f1
cross apply ( select p1 = charindex( ',', str ) ) ap1
cross apply ( select p2 = charindex( ',', str, p1 + 1 ) ) ap2
cross apply ( select Nmame = substring( str, 1, p1-1 )
, Surname = substring( str, p1+1, p2-p1-1 )
) ParsedData
There are multiple ways to solve this and many different ways have been proposed already. Simplest would be to use LEFT / SUBSTRING and other string functions to achieve the desired result.
Sample Data
DECLARE #tbl1 TABLE (Value INT,String VARCHAR(MAX))
INSERT INTO #tbl1 VALUES(1,'Cleo, Smith');
INSERT INTO #tbl1 VALUES(2,'John, Mathew');
Using String Functions like LEFT
SELECT
Value,
LEFT(String,CHARINDEX(',',String)-1) as Fname,
LTRIM(RIGHT(String,LEN(String) - CHARINDEX(',',String) )) AS Lname
FROM #tbl1
This approach fails if there are more 2 items in a String.
In such a scenario, we can use a splitter and then use PIVOT or convert the string into an XML and use .nodes to get string items. XML based solution have been detailed out by aads and bvr in their solution.
The answers for this question which use splitter, all use WHILE which is inefficient for splitting. Check this performance comparison. One of the best splitters around is DelimitedSplit8K, created by Jeff Moden. You can read more about it here
Splitter with PIVOT
DECLARE #tbl1 TABLE (Value INT,String VARCHAR(MAX))
INSERT INTO #tbl1 VALUES(1,'Cleo, Smith');
INSERT INTO #tbl1 VALUES(2,'John, Mathew');
SELECT t3.Value,[1] as Fname,[2] as Lname
FROM #tbl1 as t1
CROSS APPLY [dbo].[DelimitedSplit8K](String,',') as t2
PIVOT(MAX(Item) FOR ItemNumber IN ([1],[2])) as t3
Output
Value Fname Lname
1 Cleo Smith
2 John Mathew
DelimitedSplit8K by Jeff Moden
CREATE FUNCTION [dbo].[DelimitedSplit8K]
/**********************************************************************************************************************
Purpose:
Split a given string at a given delimiter and return a list of the split elements (items).
Notes:
1. Leading a trailing delimiters are treated as if an empty string element were present.
2. Consecutive delimiters are treated as if an empty string element were present between them.
3. Except when spaces are used as a delimiter, all spaces present in each element are preserved.
Returns:
iTVF containing the following:
ItemNumber = Element position of Item as a BIGINT (not converted to INT to eliminate a CAST)
Item = Element value as a VARCHAR(8000)
Statistics on this function may be found at the following URL:
http://www.sqlservercentral.com/Forums/Topic1101315-203-4.aspx
CROSS APPLY Usage Examples and Tests:
--=====================================================================================================================
-- TEST 1:
-- This tests for various possible conditions in a string using a comma as the delimiter. The expected results are
-- laid out in the comments
--=====================================================================================================================
--===== Conditionally drop the test tables to make reruns easier for testing.
-- (this is NOT a part of the solution)
IF OBJECT_ID('tempdb..#JBMTest') IS NOT NULL DROP TABLE #JBMTest
;
--===== Create and populate a test table on the fly (this is NOT a part of the solution).
-- In the following comments, "b" is a blank and "E" is an element in the left to right order.
-- Double Quotes are used to encapsulate the output of "Item" so that you can see that all blanks
-- are preserved no matter where they may appear.
SELECT *
INTO #JBMTest
FROM ( --# & type of Return Row(s)
SELECT 0, NULL UNION ALL --1 NULL
SELECT 1, SPACE(0) UNION ALL --1 b (Empty String)
SELECT 2, SPACE(1) UNION ALL --1 b (1 space)
SELECT 3, SPACE(5) UNION ALL --1 b (5 spaces)
SELECT 4, ',' UNION ALL --2 b b (both are empty strings)
SELECT 5, '55555' UNION ALL --1 E
SELECT 6, ',55555' UNION ALL --2 b E
SELECT 7, ',55555,' UNION ALL --3 b E b
SELECT 8, '55555,' UNION ALL --2 b B
SELECT 9, '55555,1' UNION ALL --2 E E
SELECT 10, '1,55555' UNION ALL --2 E E
SELECT 11, '55555,4444,333,22,1' UNION ALL --5 E E E E E
SELECT 12, '55555,4444,,333,22,1' UNION ALL --6 E E b E E E
SELECT 13, ',55555,4444,,333,22,1,' UNION ALL --8 b E E b E E E b
SELECT 14, ',55555,4444,,,333,22,1,' UNION ALL --9 b E E b b E E E b
SELECT 15, ' 4444,55555 ' UNION ALL --2 E (w/Leading Space) E (w/Trailing Space)
SELECT 16, 'This,is,a,test.' --E E E E
) d (SomeID, SomeValue)
;
--===== Split the CSV column for the whole table using CROSS APPLY (this is the solution)
SELECT test.SomeID, test.SomeValue, split.ItemNumber, Item = QUOTENAME(split.Item,'"')
FROM #JBMTest test
CROSS APPLY dbo.DelimitedSplit8K(test.SomeValue,',') split
;
--=====================================================================================================================
-- TEST 2:
-- This tests for various "alpha" splits and COLLATION using all ASCII characters from 0 to 255 as a delimiter against
-- a given string. Note that not all of the delimiters will be visible and some will show up as tiny squares because
-- they are "control" characters. More specifically, this test will show you what happens to various non-accented
-- letters for your given collation depending on the delimiter you chose.
--=====================================================================================================================
WITH
cteBuildAllCharacters (String,Delimiter) AS
(
SELECT TOP 256
'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789',
CHAR(ROW_NUMBER() OVER (ORDER BY (SELECT NULL))-1)
FROM master.sys.all_columns
)
SELECT ASCII_Value = ASCII(c.Delimiter), c.Delimiter, split.ItemNumber, Item = QUOTENAME(split.Item,'"')
FROM cteBuildAllCharacters c
CROSS APPLY dbo.DelimitedSplit8K(c.String,c.Delimiter) split
ORDER BY ASCII_Value, split.ItemNumber
;
-----------------------------------------------------------------------------------------------------------------------
Other Notes:
1. Optimized for VARCHAR(8000) or less. No testing or error reporting for truncation at 8000 characters is done.
2. Optimized for single character delimiter. Multi-character delimiters should be resolvedexternally from this
function.
3. Optimized for use with CROSS APPLY.
4. Does not "trim" elements just in case leading or trailing blanks are intended.
5. If you don't know how a Tally table can be used to replace loops, please see the following...
http://www.sqlservercentral.com/articles/T-SQL/62867/
6. Changing this function to use NVARCHAR(MAX) will cause it to run twice as slow. It's just the nature of
VARCHAR(MAX) whether it fits in-row or not.
7. Multi-machine testing for the method of using UNPIVOT instead of 10 SELECT/UNION ALLs shows that the UNPIVOT method
is quite machine dependent and can slow things down quite a bit.
-----------------------------------------------------------------------------------------------------------------------
Credits:
This code is the product of many people's efforts including but not limited to the following:
cteTally concept originally by Iztek Ben Gan and "decimalized" by Lynn Pettis (and others) for a bit of extra speed
and finally redacted by Jeff Moden for a different slant on readability and compactness. Hat's off to Paul White for
his simple explanations of CROSS APPLY and for his detailed testing efforts. Last but not least, thanks to
Ron "BitBucket" McCullough and Wayne Sheffield for their extreme performance testing across multiple machines and
versions of SQL Server. The latest improvement brought an additional 15-20% improvement over Rev 05. Special thanks
to "Nadrek" and "peter-757102" (aka Peter de Heer) for bringing such improvements to light. Nadrek's original
improvement brought about a 10% performance gain and Peter followed that up with the content of Rev 07.
I also thank whoever wrote the first article I ever saw on "numbers tables" which is located at the following URL
and to Adam Machanic for leading me to it many years ago.
http://sqlserver2000.databases.aspfaq.com/why-should-i-consider-using-an-auxiliary-numbers-table.html
-----------------------------------------------------------------------------------------------------------------------
Revision History:
Rev 00 - 20 Jan 2010 - Concept for inline cteTally: Lynn Pettis and others.
Redaction/Implementation: Jeff Moden
- Base 10 redaction and reduction for CTE. (Total rewrite)
Rev 01 - 13 Mar 2010 - Jeff Moden
- Removed one additional concatenation and one subtraction from the SUBSTRING in the SELECT List for that tiny
bit of extra speed.
Rev 02 - 14 Apr 2010 - Jeff Moden
- No code changes. Added CROSS APPLY usage example to the header, some additional credits, and extra
documentation.
Rev 03 - 18 Apr 2010 - Jeff Moden
- No code changes. Added notes 7, 8, and 9 about certain "optimizations" that don't actually work for this
type of function.
Rev 04 - 29 Jun 2010 - Jeff Moden
- Added WITH SCHEMABINDING thanks to a note by Paul White. This prevents an unnecessary "Table Spool" when the
function is used in an UPDATE statement even though the function makes no external references.
Rev 05 - 02 Apr 2011 - Jeff Moden
- Rewritten for extreme performance improvement especially for larger strings approaching the 8K boundary and
for strings that have wider elements. The redaction of this code involved removing ALL concatenation of
delimiters, optimization of the maximum "N" value by using TOP instead of including it in the WHERE clause,
and the reduction of all previous calculations (thanks to the switch to a "zero based" cteTally) to just one
instance of one add and one instance of a subtract. The length calculation for the final element (not
followed by a delimiter) in the string to be split has been greatly simplified by using the ISNULL/NULLIF
combination to determine when the CHARINDEX returned a 0 which indicates there are no more delimiters to be
had or to start with. Depending on the width of the elements, this code is between 4 and 8 times faster on a
single CPU box than the original code especially near the 8K boundary.
- Modified comments to include more sanity checks on the usage example, etc.
- Removed "other" notes 8 and 9 as they were no longer applicable.
Rev 06 - 12 Apr 2011 - Jeff Moden
- Based on a suggestion by Ron "Bitbucket" McCullough, additional test rows were added to the sample code and
the code was changed to encapsulate the output in pipes so that spaces and empty strings could be perceived
in the output. The first "Notes" section was added. Finally, an extra test was added to the comments above.
Rev 07 - 06 May 2011 - Peter de Heer, a further 15-20% performance enhancement has been discovered and incorporated
into this code which also eliminated the need for a "zero" position in the cteTally table.
**********************************************************************************************************************/
--===== Define I/O parameters
(#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
;
GO
With SQL Server 2016 we can use string_split to accomplish this:
create table commasep (
id int identity(1,1)
,string nvarchar(100) )
insert into commasep (string) values ('John, Adam'), ('test1,test2,test3')
select id, [value] as String from commasep
cross apply string_split(string,',')
CREATE FUNCTION [dbo].[fn_split_string_to_column] (
#string NVARCHAR(MAX),
#delimiter CHAR(1)
)
RETURNS #out_put TABLE (
[column_id] INT IDENTITY(1, 1) NOT NULL,
[value] NVARCHAR(MAX)
)
AS
BEGIN
DECLARE #value NVARCHAR(MAX),
#pos INT = 0,
#len INT = 0
SET #string = CASE
WHEN RIGHT(#string, 1) != #delimiter
THEN #string + #delimiter
ELSE #string
END
WHILE CHARINDEX(#delimiter, #string, #pos + 1) > 0
BEGIN
SET #len = CHARINDEX(#delimiter, #string, #pos + 1) - #pos
SET #value = SUBSTRING(#string, #pos, #len)
INSERT INTO #out_put ([value])
SELECT LTRIM(RTRIM(#value)) AS [column]
SET #pos = CHARINDEX(#delimiter, #string, #pos + #len) + 1
END
RETURN
END
SELECT id,
Substring(NAME, 0, Charindex(',', NAME)) AS firstname,
Substring(NAME, Charindex(',', NAME), Len(NAME) + 1) AS lastname
FROM spilt
Try this (change instances of ' ' to ',' or whatever delimiter you want to use)
CREATE FUNCTION dbo.Wordparser
(
#multiwordstring VARCHAR(255),
#wordnumber NUMERIC
)
returns VARCHAR(255)
AS
BEGIN
DECLARE #remainingstring VARCHAR(255)
SET #remainingstring=#multiwordstring
DECLARE #numberofwords NUMERIC
SET #numberofwords=(LEN(#remainingstring) - LEN(REPLACE(#remainingstring, ' ', '')) + 1)
DECLARE #word VARCHAR(50)
DECLARE #parsedwords TABLE
(
line NUMERIC IDENTITY(1, 1),
word VARCHAR(255)
)
WHILE #numberofwords > 1
BEGIN
SET #word=LEFT(#remainingstring, CHARINDEX(' ', #remainingstring) - 1)
INSERT INTO #parsedwords(word)
SELECT #word
SET #remainingstring= REPLACE(#remainingstring, Concat(#word, ' '), '')
SET #numberofwords=(LEN(#remainingstring) - LEN(REPLACE(#remainingstring, ' ', '')) + 1)
IF #numberofwords = 1
BREAK
ELSE
CONTINUE
END
IF #numberofwords = 1
SELECT #word = #remainingstring
INSERT INTO #parsedwords(word)
SELECT #word
RETURN
(SELECT word
FROM #parsedwords
WHERE line = #wordnumber)
END
Example usage:
SELECT dbo.Wordparser(COLUMN, 1),
dbo.Wordparser(COLUMN, 2),
dbo.Wordparser(COLUMN, 3)
FROM TABLE
I think PARSENAME is the neat function to use for this example, as described in this article: http://www.sqlshack.com/parsing-and-rotating-delimited-data-in-sql-server-2012/
The PARSENAME function is logically designed to parse four-part object names. The nice thing about PARSENAME is that it’s not limited to parsing just SQL Server four-part object names – it will parse any function or string data that is delimited by dots.
The first parameter is the object to parse, and the second is the integer value of the object piece to return. The article is discussing parsing and rotating delimited data - company phone numbers, but it can be used to parse name/surname data also.
Example:
USE COMPANY;
SELECT PARSENAME('Whatever.you.want.parsed',3) AS 'ReturnValue';
The article also describes using a Common Table Expression (CTE) called ‘replaceChars’, to run PARSENAME against the delimiter-replaced values. A CTE is useful for returning a temporary view or result set.
After that, the UNPIVOT function has been used to convert some columns into rows; SUBSTRING and CHARINDEX functions have been used for cleaning up the inconsistencies in the data, and the LAG function (new for SQL Server 2012) has been used in the end, as it allows referencing of previous records.
We can create a function as this
CREATE Function [dbo].[fn_CSVToTable]
(
#CSVList Varchar(max)
)
RETURNS #Table TABLE (ColumnData VARCHAR(100))
AS
BEGIN
IF RIGHT(#CSVList, 1) <> ','
SELECT #CSVList = #CSVList + ','
DECLARE #Pos BIGINT,
#OldPos BIGINT
SELECT #Pos = 1,
#OldPos = 1
WHILE #Pos < LEN(#CSVList)
BEGIN
SELECT #Pos = CHARINDEX(',', #CSVList, #OldPos)
INSERT INTO #Table
SELECT LTRIM(RTRIM(SUBSTRING(#CSVList, #OldPos, #Pos - #OldPos))) Col001
SELECT #OldPos = #Pos + 1
END
RETURN
END
We can then seperate the CSV values into our respective columns using a SELECT statement
You can use a table-valued function STRING_SPLIT, which is available only under compatibility level 130. If your database compatibility level is lower than 130, SQL Server will not be able to find and execute the STRING_SPLIT function. You can change a compatibility level of the database using the following command:
ALTER DATABASE DatabaseName SET COMPATIBILITY_LEVEL = 130
Syntax
SELECT * FROM STRING_SPLIT ( string, separator )
see documentation here
I think following function will work for you:
You have to create a function in SQL first. Like this
CREATE FUNCTION [dbo].[fn_split](
#str VARCHAR(MAX),
#delimiter CHAR(1)
)
RETURNS #returnTable TABLE (idx INT PRIMARY KEY IDENTITY, item VARCHAR(8000))
AS
BEGIN
DECLARE #pos INT
SELECT #str = #str + #delimiter
WHILE LEN(#str) > 0
BEGIN
SELECT #pos = CHARINDEX(#delimiter,#str)
IF #pos = 1
INSERT #returnTable (item)
VALUES (NULL)
ELSE
INSERT #returnTable (item)
VALUES (SUBSTRING(#str, 1, #pos-1))
SELECT #str = SUBSTRING(#str, #pos+1, LEN(#str)-#pos)
END
RETURN
END
You can call this function, like this:
select * from fn_split('1,24,5',',')
Implementation:
Declare #test TABLE (
ID VARCHAR(200),
Data VARCHAR(200)
)
insert into #test
(ID, Data)
Values
('1','Cleo,Smith')
insert into #test
(ID, Data)
Values
('2','Paul,Grim')
select ID,
(select item from fn_split(Data,',') where idx in (1)) as Name ,
(select item from fn_split(Data,',') where idx in (2)) as Surname
from #test
Result will like this:
Use Parsename() function
with cte as(
select 'Aria,Karimi' as FullName
Union
select 'Joe,Karimi' as FullName
Union
select 'Bab,Karimi' as FullName
)
SELECT PARSENAME(REPLACE(FullName,',','.'),2) as Name,
PARSENAME(REPLACE(FullName,',','.'),1) as Family
FROM cte
Result
Name Family
----- ------
Aria Karimi
Bab Karimi
Joe Karimi
Try this:
declare #csv varchar(100) ='aaa,bb,csda,daass';
set #csv = #csv+',';
with cte as
(
select SUBSTRING(#csv,1,charindex(',',#csv,1)-1) as val, SUBSTRING(#csv,charindex(',',#csv,1)+1,len(#csv)) as rem
UNION ALL
select SUBSTRING(a.rem,1,charindex(',',a.rem,1)-1)as val, SUBSTRING(a.rem,charindex(',',a.rem,1)+1,len(A.rem))
from cte a where LEN(a.rem)>=1
) select val from cte
This function is most fast:
CREATE FUNCTION dbo.F_ExtractSubString
(
#String VARCHAR(MAX),
#NroSubString INT,
#Separator VARCHAR(5)
)
RETURNS VARCHAR(MAX) AS
BEGIN
DECLARE #St INT = 0, #End INT = 0, #Ret VARCHAR(MAX)
SET #String = #String + #Separator
WHILE CHARINDEX(#Separator, #String, #End + 1) > 0 AND #NroSubString > 0
BEGIN
SET #St = #End + 1
SET #End = CHARINDEX(#Separator, #String, #End + 1)
SET #NroSubString = #NroSubString - 1
END
IF #NroSubString > 0
SET #Ret = ''
ELSE
SET #Ret = SUBSTRING(#String, #St, #End - #St)
RETURN #Ret
END
GO
Example usage:
SELECT dbo.F_ExtractSubString(COLUMN, 1, ', '),
dbo.F_ExtractSubString(COLUMN, 2, ', '),
dbo.F_ExtractSubString(COLUMN, 3, ', ')
FROM TABLE
I encountered a similar problem but a complex one and since this is the first thread i found regarding that issue i decided to post my finding. i know it is complex solution to a simple problem but i hope that i could help other people who go to this thread looking for a more complex solution. i had to split a string containing 5 numbers (column name: levelsFeed) and to show each number in a separate column.
for example: 8,1,2,2,2
should be shown as :
1 2 3 4 5
-------------
8 1 2 2 2
Solution 1: using XML functions:
this solution for the slowest solution by far
SELECT Distinct FeedbackID,
, S.a.value('(/H/r)[1]', 'INT') AS level1
, S.a.value('(/H/r)[2]', 'INT') AS level2
, S.a.value('(/H/r)[3]', 'INT') AS level3
, S.a.value('(/H/r)[4]', 'INT') AS level4
, S.a.value('(/H/r)[5]', 'INT') AS level5
FROM (
SELECT *,CAST (N'<H><r>' + REPLACE(levelsFeed, ',', '</r><r>') + '</r> </H>' AS XML) AS [vals]
FROM Feedbacks
) as d
CROSS APPLY d.[vals].nodes('/H/r') S(a)
Solution 2: using Split function and pivot. (the split function split a string to rows with the column name Data)
SELECT FeedbackID, [1],[2],[3],[4],[5]
FROM (
SELECT *, ROW_NUMBER() OVER (PARTITION BY feedbackID ORDER BY (SELECT null)) as rn
FROM (
SELECT FeedbackID, levelsFeed
FROM Feedbacks
) as a
CROSS APPLY dbo.Split(levelsFeed, ',')
) as SourceTable
PIVOT
(
MAX(data)
FOR rn IN ([1],[2],[3],[4],[5])
)as pivotTable
Solution 3: using string manipulations functions - fastest by small margin over solution 2
SELECT FeedbackID,
SUBSTRING(levelsFeed,0,CHARINDEX(',',levelsFeed)) AS level1,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),4) AS level2,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),3) AS level3,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),2) AS level4,
PARSENAME(REPLACE(SUBSTRING(levelsFeed,CHARINDEX(',',levelsFeed)+1,LEN(levelsFeed)),',','.'),1) AS level5
FROM Feedbacks
since the levelsFeed contains 5 string values i needed to use the substring function for the first string.
i hope that my solution will help other that got to this thread looking for a more complex split to columns methods
Using instring function :)
select Value,
substring(String,1,instr(String," ") -1) Fname,
substring(String,instr(String,",") +1) Sname
from tablename;
Used two functions,
1. substring(string, position, length) ==> returns string from positon to length
2. instr(string,pattern) ==> returns position of pattern.
If we don’t provide length argument in substring it returns until end of string
This worked for me
CREATE FUNCTION [dbo].[SplitString](
#delimited NVARCHAR(MAX),
#delimiter NVARCHAR(100)
) RETURNS #t TABLE ( 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
mytable:
Value ColOne
--------------------
1 Cleo, Smith
The following should work if there aren't too many columns
ALTER TABLE mytable ADD ColTwo nvarchar(256);
UPDATE mytable SET ColTwo = LEFT(ColOne, Charindex(',', ColOne) - 1);
--'Cleo' = LEFT('Cleo, Smith', Charindex(',', 'Cleo, Smith') - 1)
UPDATE mytable SET ColTwo = REPLACE(ColOne, ColTwo + ',', '');
--' Smith' = REPLACE('Cleo, Smith', 'Cleo' + ',')
UPDATE mytable SET ColOne = REPLACE(ColOne, ',' + ColTwo, ''), ColTwo = LTRIM(ColTwo);
--'Cleo' = REPLACE('Cleo, Smith', ',' + ' Smith', '')
Result:
Value ColOne ColTwo
--------------------
1 Cleo Smith
DECLARE #INPUT VARCHAR (MAX)='N,A,R,E,N,D,R,A'
DECLARE #ELIMINATE_CHAR CHAR (1)=','
DECLARE #L_START INT=1
DECLARE #L_END INT=(SELECT LEN (#INPUT))
DECLARE #OUTPUT CHAR (1)
WHILE #L_START <=#L_END
BEGIN
SET #OUTPUT=(SUBSTRING (#INPUT,#L_START,1))
IF #OUTPUT!=#ELIMINATE_CHAR
BEGIN
PRINT #OUTPUT
END
SET #L_START=#L_START+1
END
You may find the solution in SQL User Defined Function to Parse a Delimited String helpful (from The Code Project).
This is the code part from this page:
CREATE FUNCTION [fn_ParseText2Table]
(#p_SourceText VARCHAR(MAX)
,#p_Delimeter VARCHAR(100)=',' --default to comma delimited.
)
RETURNS #retTable
TABLE([Position] INT IDENTITY(1,1)
,[Int_Value] INT
,[Num_Value] NUMERIC(18,3)
,[Txt_Value] VARCHAR(MAX)
,[Date_value] DATETIME
)
AS
/*
********************************************************************************
Purpose: Parse values from a delimited string
& return the result as an indexed table
Copyright 1996, 1997, 2000, 2003 Clayton Groom (Clayton_Groom#hotmail.com)
Posted to the public domain Aug, 2004
2003-06-17 Rewritten as SQL 2000 function.
Reworked to allow for delimiters > 1 character in length
and to convert Text values to numbers
2016-04-05 Added logic for date values based on "new" ISDATE() function, Updated to use XML approach, which is more efficient.
********************************************************************************
*/
BEGIN
DECLARE #w_xml xml;
SET #w_xml = N'<root><i>' + replace(#p_SourceText, #p_Delimeter,'</i><i>') + '</i></root>';
INSERT INTO #retTable
([Int_Value]
, [Num_Value]
, [Txt_Value]
, [Date_value]
)
SELECT CASE
WHEN ISNUMERIC([i].value('.', 'VARCHAR(MAX)')) = 1
THEN CAST(CAST([i].value('.', 'VARCHAR(MAX)') AS NUMERIC) AS INT)
END AS [Int_Value]
, CASE
WHEN ISNUMERIC([i].value('.', 'VARCHAR(MAX)')) = 1
THEN CAST([i].value('.', 'VARCHAR(MAX)') AS NUMERIC(18, 3))
END AS [Num_Value]
, [i].value('.', 'VARCHAR(MAX)') AS [txt_Value]
, CASE
WHEN ISDATE([i].value('.', 'VARCHAR(MAX)')) = 1
THEN CAST([i].value('.', 'VARCHAR(MAX)') AS DATETIME)
END AS [Num_Value]
FROM #w_xml.nodes('//root/i') AS [Items]([i]);
RETURN;
END;
GO
ALTER function get_occurance_index(#delimiter varchar(1),#occurence int,#String varchar(100))
returns int
AS Begin
--Declare #delimiter varchar(1)=',',#occurence int=2,#String varchar(100)='a,b,c'
Declare #result int
;with T as (
select 1 Rno,0 as row, charindex(#delimiter, #String) pos,#String st
union all
select Rno+1,pos + 1, charindex(#delimiter, #String, pos + 1), #String
from T
where pos > 0
)
select #result=pos
from T
where pos > 0 and rno = #occurence
return isnull(#result,0)
ENd
declare #data as table (data varchar(100))
insert into #data values('1,2,3')
insert into #data values('aaa,bbbbb,cccc')
select top 3 Substring (data,0,dbo.get_occurance_index( ',',1,data)) ,--First Record always starts with 0
Substring (data,dbo.get_occurance_index( ',',1,data)+1,dbo.get_occurance_index( ',',2,data)-dbo.get_occurance_index( ',',1,data)-1) ,
Substring (data,dbo.get_occurance_index( ',',2,data)+1,len(data)) , -- Last record cant be more than len of actual data
data
From #data
I found that using PARSENAME as above caused any name with a period to get nulled.
So if there was an initial or a title in the name followed by a dot they return NULL.
I found this worked for me:
SELECT
REPLACE(SUBSTRING(FullName, 1,CHARINDEX(',', FullName)), ',','') as Name,
REPLACE(SUBSTRING(FullName, CHARINDEX(',', FullName), LEN(FullName)), ',', '') as Surname
FROM Table1
it is so easy, you can take it by below query:
DECLARE #str NVARCHAR(MAX)='ControlID_05436b78-04ba-9667-fa01-9ff8c1b7c235,3'
SELECT LEFT(#str, CHARINDEX(',',#str)-1),RIGHT(#str,LEN(#str)-(CHARINDEX(',',#str)))
select distinct modelFileId,F4.*
from contract
cross apply (select XmlList=convert(xml, '<x>'+replace(modelFileId,';','</x><x>')+'</x>').query('.')) F2
cross apply (select mfid1=XmlNode.value('/x[1]','varchar(512)')
,mfid2=XmlNode.value('/x[2]','varchar(512)')
,mfid3=XmlNode.value('/x[3]','varchar(512)')
,mfid4=XmlNode.value('/x[4]','varchar(512)') from XmlList.nodes('x') F3(XmlNode)) F4
where modelFileId like '%;%'
order by modelFileId
Select distinct PROJ_UID,PROJ_NAME,RES_UID from E2E_ProjectWiseTimesheetActuals
where CHARINDEX(','+cast(PROJ_UID as varchar(8000))+',', #params) > 0 and CHARINDEX(','+cast(RES_UID as varchar(8000))+',', #res) > 0
I re-wrote an answer above and made it better:
CREATE FUNCTION [dbo].[CSVParser]
(
#s VARCHAR(255),
#idx NUMERIC
)
RETURNS VARCHAR(12)
BEGIN
DECLARE #comma int
SET #comma = CHARINDEX(',', #s)
WHILE 1=1
BEGIN
IF #comma=0
IF #idx=1
RETURN #s
ELSE
RETURN ''
IF #idx=1
BEGIN
DECLARE #word VARCHAR(12)
SET #word=LEFT(#s, #comma - 1)
RETURN #word
END
SET #s = RIGHT(#s,LEN(#s)-#comma)
SET #comma = CHARINDEX(',', #s)
SET #idx = #idx - 1
END
RETURN 'not used'
END
Example usage:
SELECT dbo.CSVParser(COLUMN, 1),
dbo.CSVParser(COLUMN, 2),
dbo.CSVParser(COLUMN, 3)
FROM TABLE
question is simple, but problem is hot :)
So I create some wrapper for string_split() which pivot result in more generic way. It's table function which returns values (nn, value1, value2, ... , value50) - enough for most CSV lines. If there are more values, they will wrap to next line - nn indicate line number. Set third parameter #columnCnt = [yourNumber] to wrap at specific position:
alter FUNCTION fn_Split50
(
#str varchar(max),
#delim char(1),
#columnCnt int = 50
)
RETURNS TABLE
AS
RETURN
(
SELECT *
FROM (SELECT
nn = (nn - 1) / #columnCnt + 1,
nnn = 'value' + cast(((nn - 1) % #columnCnt) + 1 as varchar(10)),
value
FROM (SELECT
nn = ROW_NUMBER() over (order by (select null)),
value
FROM string_split(#str, #delim) aa
) aa
where nn > 0
) bb
PIVOT
(
max(value)
FOR nnn IN (
value1, value2, value3, value4, value5, value6, value7, value8, value9, value10,
value11, value12, value13, value14, value15, value16, value17, value18, value19, value20,
value21, value22, value23, value24, value25, value26, value27, value28, value29, value30,
value31, value32, value33, value34, value35, value36, value37, value38, value39, value40,
value41, value42, value43, value44, value45, value46, value47, value48, value49, value50
)
) AS PivotTable
)
Example of using:
select * from dbo.fn_split50('zz1,aa2,ss3,dd4,ff5', ',', DEFAULT)
select * from dbo.fn_split50('zz1,aa2,ss3,dd4,ff5,gg6,hh7,jj8,ww9,qq10', ',', 3)
select * from dbo.fn_split50('zz1,11,aa2,22,ss3,33,dd4,44,ff5,55,gg6,66,hh7,77,jj8,88,ww9,99,qq10,1010', ',',2)
Hope, it will helps :)

subquery with table value function not working

DECLARE #temp AS TABLE (id INT, NAME VARCHAR(20) )
DECLARE #str VARCHAR(20) = '1,2'
INSERT INTO #temp (id, NAME)
VALUES (1, ''), (2, ''), (2, '')
SELECT *
FROM #temp a
WHERE id IN ((SELECT String FROM dbo.FN_SplitStrings(#str,',')))
I'm getting the following error while running this
Conversion failed when converting the varchar value '1,2' to data type
int.
Code:
CREATE function [dbo].[FN_SplitStrings]
(
#StringToSplit varchar(8000),
#Separator varchar(128)
)
RETURN TABLE
AS
RETURN
with indices as
(
select
0 S, 1 E
union all
select
E, charindex(#Separator, #StringToSplit, E) + len(#Separator)
from
indices
where E > S
)
select
substring(#StringToSplit,S, case when E > len(#Separator)
then e-s-len(#Separator) else len(#StringToSplit) - s + 1 end) String ,
S StartIndex
from
indices
where
S > 0
Try this. This splitting can be used without a function
DECLARE #temp AS TABLE
(
id INT,
NAME VARCHAR(20)
)
DECLARE #str VARCHAR(20)='1,2'
INSERT INTO #temp
( id, NAME )
VALUES ( 1, '' ),
( 2, ''),
( 2, '')
SELECT * FROM #temp a
WHERE id IN
(
SELECT LTRIM(RTRIM(Split.a.value('.', 'VARCHAR(100)'))) 'KeyWords'
FROM
(
-- To change ',' to any other delimeter, just change ',' before '</M><M>' to your desired one
SELECT CAST ('<M>' + REPLACE(#str, ',', '</M><M>') + '</M>' AS XML) AS Data
) AS A
CROSS APPLY Data.nodes ('/M') AS Split(a)
)
Click here to view the result
EDIT:
You had some problems in function for splitting.
Function
ALTER FUNCTION dbo.FN_SplitStrings(#StringToSplit varchar(8000),#Separator char(1))
RETURNS table
AS
RETURN (
WITH splitter_cte AS (
SELECT CHARINDEX(#Separator, #StringToSplit) as pos, 0 as lastPos
UNION ALL
SELECT CHARINDEX(#Separator, #StringToSplit, pos + 1), pos
FROM splitter_cte
WHERE pos > 0
)
SELECT SUBSTRING(#StringToSplit, lastPos + 1,
case when pos = 0 then 80000
else pos - lastPos -1 end) as String
FROM splitter_cte
)
Query
DECLARE #temp AS TABLE (id INT, NAME VARCHAR(20) )
DECLARE #str VARCHAR(20) = '1,2'
INSERT INTO #temp (id, NAME)
VALUES (1, ''), (2, ''), (2, '')
SELECT *
FROM #temp a
WHERE id IN ((SELECT String FROM dbo.FN_SplitStrings(#str,',')))
Click here to view result
The reason is a mix in data types and the fact that your function is an inline table valued functions which means that it is embedded into the query before query optimization takes place.
If you remove where S > 0 from your function and execute it with 1,2, the result of the function is:
String
------
1,2
1
2
Notice the first row where the value is 1,2.
When the optimizer does its job with your query the comparison against the column id is done before the where clause of the function is evaluated. In that comparison you have an implicit cast to int and 1,2 can not be casted to an int.
To fix this you can make sure that the String column of your split function is always an int (and perhaps changing the name of the column in the process).
select
cast(substring(#StringToSplit,S, case when E > len(#Separator)
then e-s-len(#Separator)
else len(#StringToSplit) - s + 1
end) as int) String ,

Resources