Related
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 :)
I ran this query :
select * From dbo.fn_split('Age,15,14,193,188 ',',')
It's returning the values but cutting off one character in front of each value
I tried adding space after every comma like
select * From dbo.fn_split('Age, 15, 14, 193, 188 ',',')
And it worked. But I want to know why is not working with commas
select * From dbo.fn_split ('Age,15,14,193,188 ',',')
You're not posting the code of fn_split which is where the problem resides. But if your strings are less than 8000 chars long, this function will help you to split them in an optimal way. This function is a modified version of Jeff Moden's splitter made by Eirikur Eirikson.
CREATE FUNCTION [dbo].[DelimitedSplit8K_LEAD]
--===== 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 VARCHAR(8000)
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 "zero base" and limits the number of rows right up front
-- for both a performance gain and prevention of accidental "overruns"
SELECT 0 UNION ALL
SELECT TOP (DATALENGTH(ISNULL(#pString,1))) 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 t.N+1
FROM cteTally t
WHERE (SUBSTRING(#pString,t.N,1) = #pDelimiter OR t.N = 0)
)
--===== 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 s.N1),
Item = SUBSTRING(#pString,s.N1,ISNULL(NULLIF((LEAD(s.N1,1,1) OVER (ORDER BY s.N1) - 1),0)-s.N1,8000))
FROM cteStart s
;
GO
ALTER FUNCTION [dbo].[fn_Split](#text varchar(8000), #delimiter varchar(20) = ',')
RETURNS #Strings TABLE
(
position int IDENTITY PRIMARY KEY,
value varchar(8000)
)
AS
BEGIN
DECLARE #index int
SET #index = -1
WHILE (LEN(#text) > 0)
BEGIN
SET #index = CHARINDEX(#delimiter , #text)
IF (#index = 0) AND (LEN(#text) > 0)
BEGIN
INSERT INTO #Strings VALUES (#text)
BREAK
END
IF (#index > 1)
BEGIN
INSERT INTO #Strings VALUES (LEFT(#text, #index - 1))
SET #text = RIGHT(#text, (LEN(#text) - #index))
END
ELSE
SET #text = RIGHT(#text, (LEN(#text) - #index))
END
RETURN
END
I figured out what was I doing wrong. Basically I was adding an extra space at the end of string that I want to split. If you change
Select * From dbo.fn_split(dbo.fn_split('Age,15,14,193,188 ',',')) to Select * From dbo.fn_split('Age,15,14,193,188',','). In other words, get rid of the extra space after number 188 you will get the desired result.
Solution
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 :)
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 :)
I want to return integers from rather complex strings which combined unicode characters such as - and . with characters and integers.
I've come a long way in achieving this, but I still have troubles with some strings of a more complex structure. For instance:
DECLARE #Tabl as table
(
dats nvarchar(15)
)
INSERT INTO #Tabl VALUES
('103-P705hh'),
('115-xxx-44'),
('103-705.13'),
('525-hheef4')
select LEFT(SUBSTRING(REPLACE(REPLACE(dats, '.',''),'-',''), PATINDEX('%[0-9.-]%', REPLACE(REPLACE(dats, '.',''),'-','')), 8000),
PATINDEX('%[^0-9.-]%', SUBSTRING(REPLACE(REPLACE(dats, '.',''),'-',''), PATINDEX('%[0-9.-]%', REPLACE(REPLACE(dats, '.',''),'-','')), 8000) + 'X')-1)
from #tabl
Gives
Raw Input Actual return: Desired return:
103-P705hh 103 103705
115-xxx-44 115 11544
103-705.13 10370513 10370513
525-hheef4 525 5254
I had a topic regarding this yesterday to cover the case when multiple - or . are present, but as seen in the return this is actually taken care of now. However, expanding the databases I work with I encountered much more complex string such as those I presented here.
Does anyone have any idea what to do when characters and integers are "mixed up" in the string?
Regards,
Cenderze
I have seen loads of solutions that use a scalar udf with a loop, but I don't like either of these things, so throwing my hat into the ring with a different approach.
With the help of a numbers table you can deconstruct each value into its individual characters, remove non-numeric characters, then reconstruct it using FOR XML to concatenate rows, e.g.
WITH Numbers (Number) AS
( SELECT ROW_NUMBER() OVER(ORDER BY N1.N)
FROM (VALUES (1),(1),(1),(1),(1),(1),(1),(1),(1),(1)) AS N1 (N) -- 100
CROSS JOIN (VALUES (1),(1),(1),(1),(1),(1),(1),(1),(1),(1)) AS N2 (N) -- 100
CROSS JOIN (VALUES (1),(1),(1),(1),(1),(1),(1),(1),(1),(1)) AS N3 (N) -- 1,000
--CROSS JOIN (VALUES (1),(1),(1),(1),(1),(1),(1),(1),(1),(1)) AS N4 (N) -- 10,000
--CROSS JOIN (VALUES (1),(1),(1),(1),(1),(1),(1),(1),(1),(1)) AS N5 (N) -- 100,000
--COMMENT OR UNCOMMENT ROWS AS NECESSARY DEPENDING ON YOU MAX STRING LENGTH
)
SELECT t.dats,
Stripped = x.data.value('.', 'INT')
FROM #tabl AS t
CROSS APPLY
( SELECT SUBSTRING(t.dats, n.Number, 1)
FROM Numbers n
WHERE n.Number <= LEN(t.dats)
AND SUBSTRING(t.dats, n.Number, 1) LIKE '[0-9]'
ORDER BY n.Number
FOR XML PATH(''), TYPE
) x (data);
Gives:
dats Stripped
----------------------
103-P705hh 103705
115-xxx-44 11544
103-705.13 10370513
525-hheef4 5254
I haven't done any testing so it could be that the added overhead of expanding each string into individual characters and reconstructing it is actually a lot more overhead than than a UDF with a loop.
I decided to bench mark this
1. Set up functions
CREATE FUNCTION dbo.ExtractNumeric_TVF (#Input VARCHAR(8000))
RETURNS TABLE
AS
RETURN
( WITH Numbers (Number) AS
( SELECT TOP (LEN(#Input)) ROW_NUMBER() OVER(ORDER BY N1.N)
FROM (VALUES (1),(1),(1),(1),(1),(1),(1),(1),(1),(1)) AS N1 (N) -- 100
CROSS JOIN (VALUES (1),(1),(1),(1),(1),(1),(1),(1),(1),(1)) AS N2 (N) -- 100
CROSS JOIN (VALUES (1),(1),(1),(1),(1),(1),(1),(1),(1),(1)) AS N3 (N) -- 1,000
CROSS JOIN (VALUES (1),(1),(1),(1),(1),(1),(1),(1),(1),(1)) AS N4 (N) -- 10,000
)
SELECT Stripped = x.data.value('.', 'VARCHAR(MAX)')
FROM ( SELECT SUBSTRING(#Input, n.Number, 1)
FROM Numbers n
WHERE n.Number <= LEN(#Input)
AND SUBSTRING(#Input, n.Number, 1) LIKE '[0-9]'
ORDER BY n.Number
FOR XML PATH(''), TYPE
) x (data)
);
GO
create function dbo.ExtractNumeric_UDF(#s varchar(8000))
returns varchar(8000)
as
begin
declare #out varchar(max) = ''
declare #c char(1)
while len(#s) > 0 begin
set #c = left(#s,1)
if #c like '[0123456789]' set #out += #c
set #s = substring(#s, 2, len(#s) -1)
end
return #out
end
GO
2. Create first set of sample data and log table
CREATE TABLE dbo.T (Value VARCHAR(8000) NOT NULL);
INSERT dbo.T (Value)
SELECT TOP 1000 LEFT(NEWID(), CEILING(RAND(CHECKSUM(NEWID())) * 36))
FROM sys.all_objects a
CROSS JOIN sys.all_objects b;
CREATE TABLE dbo.TestLog (Fx VARCHAR(255), NumberOfRows INT, TimeStart DATETIME2(7), TimeEnd DATETIME2(7))
3. Run Tests
GO
DECLARE #T TABLE (Val VARCHAR(8000));
INSERT dbo.TestLog (fx, NumberOfRows, TimeStart)
VALUES ('dbo.ExtractNumeric_UDF', 1000, SYSDATETIME());
INSERT #T (Val)
SELECT dbo.ExtractNumeric_UDF(Value)
FROM dbo.T;
UPDATE dbo.TestLog
SET TimeEnd = SYSDATETIME()
WHERE TimeEnd IS NULL;
GO 100
DECLARE #T TABLE (Val VARCHAR(8000));
INSERT dbo.TestLog (fx, NumberOfRows, TimeStart)
VALUES ('dbo.ExtractNumeric_TVF', 1000, SYSDATETIME());
INSERT #T (Val)
SELECT f.Stripped
FROM dbo.T
CROSS APPLY dbo.ExtractNumeric_TVF(Value) f;
UPDATE dbo.TestLog
SET TimeEnd = SYSDATETIME()
WHERE TimeEnd IS NULL;
GO 100
4. Get Results
SELECT Fx,
NumberOfRows,
RunTime = AVG(DATEDIFF(MILLISECOND, TimeStart, TimeEnd))
FROM dbo.TestLog
GROUP BY fx, NumberOfRows;
I did the following (using just NEWID() so only a maximum of 36 characters) over 1,000 and 10,000 rows, the results were:
Fx NumberOfRows RunTime
--------------------------------------------------------
dbo.ExtractNumeric_TVF 1000 31
dbo.ExtractNumeric_UDF 1000 56
dbo.ExtractNumeric_TVF 10000 280
dbo.ExtractNumeric_UDF 10000 510
So the TVF coming in at just under half the time of the UDF.
I wanted to test edge cases so put 1,000 rows of longer strings (5,400 characters)
TRUNCATE TABLE dbo.T;
INSERT dbo.T (Value)
SELECT TOP 1000
REPLICATE(CONCAT(NEWID(), NEWID(), NEWID(), NEWID(), NEWID()), 30)
FROM sys.all_objects a
CROSS JOIN sys.all_objects b;
And this is where the TVF came into its own, running over 5x faster:
Fx NumberOfRows RunTime
------------------------------------------------
dbo.ExtractNumeric_TVF 1000 2485
dbo.ExtractNumeric_UDF 1000 12955
I also really don't like the looping solutions so I decided to try my hand at one. This is using a predefined tally table but is quite similar to others posted here already.
This is my tally table. I keep this as a view on my system.
create View [dbo].[cteTally] as
WITH
E1(N) AS (select 1 from (values (1),(1),(1),(1),(1),(1),(1),(1),(1),(1))dt(n)),
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
(
SELECT ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) FROM E4
)
select N from cteTally
GO
Because I don't like looping I decided to use the table valued function approach which let me reuse this functionality in other queries with little to no effort. Here is one way to write such a function.
create function GetOnlyNumbers
(
#SearchVal varchar(8000)
) returns table as return
with MyValues as
(
select substring(#SearchVal, N, 1) as number
, t.N
from cteTally t
where N <= len(#SearchVal)
and substring(#SearchVal, N, 1) like '[0-9]'
)
select distinct NumValue = STUFF((select number + ''
from MyValues mv2
order by mv2.N
for xml path('')), 1, 0, '')
from MyValues mv
That looks good but the proof is in the pudding. Let's take this out with our sample data and kick the tires a few times.
DECLARE #Tabl as table
(
dats nvarchar(15)
)
INSERT INTO #Tabl VALUES
('103-P705hh'),
('115-xxx-44'),
('103-705.13'),
('525-hheef4')
select *
from #Tabl t
cross apply dbo.GetOnlyNumbers(t.dats) x
Sure looks nice and tidy. I tested against several of the other solutions posted here and without going into deep testing this appears to be significantly faster than the other approaches posted at this time.
DECLARE #Tabl as table
(
ID INT,
dats nvarchar(15)
)
INSERT INTO #Tabl VALUES
(1, '103-P705hh'),
(2, '115-xxx-44'),
(3, '103-705.13'),
(4, '525-hheef4')
SELECT T.ID, t.dats
,(
SELECT SUBSTRING(tt.dats,V.number,1)
FROM #Tabl tt
JOIN master.dbo.spt_values V ON V.type='P' AND V.number BETWEEN 1 AND LEN(tt.dats)
WHERE tt.ID=T.ID AND SUBSTRING(TT.dats,V.number,1) LIKE '[0-9]'
ORDER BY V.number
FOR XML PATH('')
) S
FROM #Tabl t
ORDER BY T.ID;
Can you use a udf ? If so, try this
create alter function numerals(#s varchar(max))
returns varchar(max)
as
begin
declare #out varchar(max) = ''
declare #c char(1)
while len(#s) > 0 begin
set #c = left(#s,1)
if #c like '[0123456789]' set #out += #c
set #s = substring(#s, 2, len(#s) -1)
end
return #out
end
to use it on your temp table...
select dbo.numerals(dats) from #Tabl
another solution, that does not use a UDF, but will work only if your table has a primary key, uses a recursive CTE. It is:
DECLARE #Tabl as table
(pk int identity not null, -- <=== added a primary key
dats nvarchar(max) )
INSERT INTO #Tabl VALUES
('103-P705hh'),
('115-xxx-44'),
('103-705.13'),
('525-hheef4');
with newVals(pk, pos, newD) as
(select pk, 1,
case when left(Dats,1) like '[0123456789]'
then left(Dats,1) else '' end
from #tabl
Union All
Select t.pk, pos + 1, n.newD +
case when substring(dats, pos+1, 1) like '[0123456789]'
then substring(dats, pos+1, 1) else '' end
from #tabl t join newVals n on n.pk = t.pk
where pos+1 <= len(dats) )
Select newD from newVals x
where pos = (Select Max(pos)
from newVals
where pk = x.pk)