Related
I am writing a stored procedure with a pivot in it. The pivot field names can change depending on the data in the table.
So I have the two variables below. However this seems quite inefficient because I run two queries on the same table, this is probably due to my lack of knowledge.
declare #code nvarchar(max) = ''
select #code = #code + '[' + Code + '],' from (select Code from myTbl) as c
set #code = substring(#code , 1, len(#code ) - 1)
declare #Name nvarchar(max) = ''
select #Name = #Name + '[' + Name + '],' from (select Name from myTbl) as c
set #Name = substring(#Name , 1, len(#Name ) - 1)
Is it possible to populate both variables and only query the table once?
Yes you can, here is a simple sample
CREATE TABLE T(
Code VARCHAR(45),
Name VARCHAR(45)
);
INSERT INTO T VALUES
('Code1', 'Name1'),
('Code2', 'Name2');
DECLARE #Code VARCHAR(MAX) = '',
#Name VARCHAR(MAX) = '';
SELECT #Code = #Code + QUOTENAME(Code) + ',',
#Name = #Name + QUOTENAME(Name) + ','
FROM T;
SELECT #Code, #Name;
Returns:
+------------------+------------------+
| No column name) | (No column name) |
+------------------+------------------+
| [Code1],[Code2], | [Name1],[Name2], |
+------------------+------------------+
If you have SQL Server 2017, then no need to use substring, you can just use STRING_AGG() as
SELECT STRING_AGG(QUOTENAME(Code), ','),
STRING_AGG(QUOTENAME(Name), ',')
FROM T;
Returns:
+------------------+------------------+
| (No column name) | (No column name) |
+------------------+------------------+
| [Code1],[Code2] | [Name1],[Name2] |
+------------------+------------------+
I have stripped down the subquery (select Code from myTbl) as c since I do not think it added something in this context.
Given that. I believe it could work like this:
declare #code nvarchar(max) = ''
declare #Name nvarchar(max) = ''
select #code = #code + '[' + Code + '],', #Name = #Name + '[' + Name + '],' from myTbl
set #code = substring(#code , 1, len(#code ) - 1)
set #Name = substring(#Name , 1, len(#Name ) - 1)
Using SQL Server, how do I split a string so I can access item x?
Take a string "Hello John Smith". How can I split the string by space and access the item at index 1 which should return "John"?
I don't believe SQL Server has a built-in split function, so other than a UDF, the only other answer I know is to hijack the PARSENAME function:
SELECT PARSENAME(REPLACE('Hello John Smith', ' ', '.'), 2)
PARSENAME takes a string and splits it on the period character. It takes a number as its second argument, and that number specifies which segment of the string to return (working from back to front).
SELECT PARSENAME(REPLACE('Hello John Smith', ' ', '.'), 3) --return Hello
Obvious problem is when the string already contains a period. I still think using a UDF is the best way...any other suggestions?
You may find the solution in SQL User Defined Function to Parse a Delimited String helpful (from The Code Project).
You can use this simple logic:
Declare #products varchar(200) = '1|20|3|343|44|6|8765'
Declare #individual varchar(20) = null
WHILE LEN(#products) > 0
BEGIN
IF PATINDEX('%|%', #products) > 0
BEGIN
SET #individual = SUBSTRING(#products,
0,
PATINDEX('%|%', #products))
SELECT #individual
SET #products = SUBSTRING(#products,
LEN(#individual + '|') + 1,
LEN(#products))
END
ELSE
BEGIN
SET #individual = #products
SET #products = NULL
SELECT #individual
END
END
First, create a function (using CTE, common table expression does away with the need for a temp table)
create function dbo.SplitString
(
#str nvarchar(4000),
#separator char(1)
)
returns table
AS
return (
with tokens(p, a, b) AS (
select
1,
1,
charindex(#separator, #str)
union all
select
p + 1,
b + 1,
charindex(#separator, #str, b + 1)
from tokens
where b > 0
)
select
p-1 zeroBasedOccurance,
substring(
#str,
a,
case when b > 0 then b-a ELSE 4000 end)
AS s
from tokens
)
GO
Then, use it as any table (or modify it to fit within your existing stored proc) like this.
select s
from dbo.SplitString('Hello John Smith', ' ')
where zeroBasedOccurance=1
Update
Previous version would fail for input string longer than 4000 chars. This version takes care of the limitation:
create function dbo.SplitString
(
#str nvarchar(max),
#separator char(1)
)
returns table
AS
return (
with tokens(p, a, b) AS (
select
cast(1 as bigint),
cast(1 as bigint),
charindex(#separator, #str)
union all
select
p + 1,
b + 1,
charindex(#separator, #str, b + 1)
from tokens
where b > 0
)
select
p-1 ItemIndex,
substring(
#str,
a,
case when b > 0 then b-a ELSE LEN(#str) end)
AS s
from tokens
);
GO
Usage remains the same.
Most of the solutions here use while loops or recursive CTEs. A set-based approach will be superior, I promise, if you can use a delimiter other than a space:
CREATE FUNCTION [dbo].[SplitString]
(
#List NVARCHAR(MAX),
#Delim VARCHAR(255)
)
RETURNS TABLE
AS
RETURN ( SELECT [Value], idx = RANK() OVER (ORDER BY n) FROM
(
SELECT n = Number,
[Value] = LTRIM(RTRIM(SUBSTRING(#List, [Number],
CHARINDEX(#Delim, #List + #Delim, [Number]) - [Number])))
FROM (SELECT Number = ROW_NUMBER() OVER (ORDER BY name)
FROM sys.all_objects) AS x
WHERE Number <= LEN(#List)
AND SUBSTRING(#Delim + #List, [Number], LEN(#Delim)) = #Delim
) AS y
);
Sample usage:
SELECT Value FROM dbo.SplitString('foo,bar,blat,foo,splunge',',')
WHERE idx = 3;
Results:
----
blat
You could also add the idx you want as an argument to the function, but I'll leave that as an exercise to the reader.
You can't do this with just the native STRING_SPLIT function added in SQL Server 2016, because there is no guarantee that the output will be rendered in the order of the original list. In other words, if you pass in 3,6,1 the result will likely be in that order, but it could be 1,3,6. I have asked for the community's help in improving the built-in function here:
Please help with STRING_SPLIT improvements
With enough qualitative feedback, they may actually consider making some of these enhancements:
STRING_SPLIT is not feature complete
More on split functions, why (and proof that) while loops and recursive CTEs don't scale, and better alternatives, if splitting strings coming from the application layer:
Split strings the right way – or the next best way
Splitting Strings : A Follow-Up
Splitting Strings : Now with less T-SQL
Comparing string splitting / concatenation methods
Processing a list of integers : my approach
Splitting a list of integers : another roundup
More on splitting lists : custom delimiters, preventing duplicates, and maintaining order
Removing Duplicates from Strings in SQL Server
On SQL Server 2016 or above, though, you should look at STRING_SPLIT() and STRING_AGG():
Performance Surprises and Assumptions : STRING_SPLIT()
STRING_SPLIT() in SQL Server 2016 : Follow-Up #1
STRING_SPLIT() in SQL Server 2016 : Follow-Up #2
SQL Server v.Next : STRING_AGG() performance
Solve old problems with SQL Server’s new STRING_AGG and STRING_SPLIT functions
You can leverage a Number table to do the string parsing.
Create a physical numbers table:
create table dbo.Numbers (N int primary key);
insert into dbo.Numbers
select top 1000 row_number() over(order by number) from master..spt_values
go
Create test table with 1000000 rows
create table #yak (i int identity(1,1) primary key, array varchar(50))
insert into #yak(array)
select 'a,b,c' from dbo.Numbers n cross join dbo.Numbers nn
go
Create the function
create function [dbo].[ufn_ParseArray]
( #Input nvarchar(4000),
#Delimiter char(1) = ',',
#BaseIdent int
)
returns table as
return
( select row_number() over (order by n asc) + (#BaseIdent - 1) [i],
substring(#Input, n, charindex(#Delimiter, #Input + #Delimiter, n) - n) s
from dbo.Numbers
where n <= convert(int, len(#Input)) and
substring(#Delimiter + #Input, n, 1) = #Delimiter
)
go
Usage (outputs 3mil rows in 40s on my laptop)
select *
from #yak
cross apply dbo.ufn_ParseArray(array, ',', 1)
cleanup
drop table dbo.Numbers;
drop function [dbo].[ufn_ParseArray]
Performance here is not amazing, but calling a function over a million row table is not the best idea. If performing a string split over many rows I would avoid the function.
This question is not about a string split approach, but about how to get the nth element.
All answers here are doing some kind of string splitting using recursion, CTEs, multiple CHARINDEX, REVERSE and PATINDEX, inventing functions, call for CLR methods, number tables, CROSS APPLYs ... Most answers cover many lines of code.
But - if you really want nothing more than an approach to get the nth element - this can be done as real one-liner, no UDF, not even a sub-select... And as an extra benefit: type safe
Get part 2 delimited by a space:
DECLARE #input NVARCHAR(100)=N'part1 part2 part3';
SELECT CAST(N'<x>' + REPLACE(#input,N' ',N'</x><x>') + N'</x>' AS XML).value('/x[2]','nvarchar(max)')
Of course you can use variables for delimiter and position (use sql:column to retrieve the position directly from a query's value):
DECLARE #dlmt NVARCHAR(10)=N' ';
DECLARE #pos INT = 2;
SELECT CAST(N'<x>' + REPLACE(#input,#dlmt,N'</x><x>') + N'</x>' AS XML).value('/x[sql:variable("#pos")][1]','nvarchar(max)')
If your string might include forbidden characters (especially one among &><), you still can do it this way. Just use FOR XML PATH on your string first to replace all forbidden characters with the fitting escape sequence implicitly.
It's a very special case if - additionally - your delimiter is the semicolon. In this case I replace the delimiter first to '#DLMT#', and replace this to the XML tags finally:
SET #input=N'Some <, > and &;Other äöü#€;One more';
SET #dlmt=N';';
SELECT CAST(N'<x>' + REPLACE((SELECT REPLACE(#input,#dlmt,'#DLMT#') AS [*] FOR XML PATH('')),N'#DLMT#',N'</x><x>') + N'</x>' AS XML).value('/x[sql:variable("#pos")][1]','nvarchar(max)');
UPDATE for SQL-Server 2016+
Regretfully the developers forgot to return the part's index with STRING_SPLIT. But, using SQL-Server 2016+, there is JSON_VALUE and OPENJSON.
With JSON_VALUE we can pass in the position as the index' array.
For OPENJSON the documentation states clearly:
When OPENJSON parses a JSON array, the function returns the indexes of the elements in the JSON text as keys.
A string like 1,2,3 needs nothing more than brackets: [1,2,3].
A string of words like this is an example needs to be ["this","is","an","example"].
These are very easy string operations. Just try it out:
DECLARE #str VARCHAR(100)='Hello John Smith';
DECLARE #position INT = 2;
--We can build the json-path '$[1]' using CONCAT
SELECT JSON_VALUE('["' + REPLACE(#str,' ','","') + '"]',CONCAT('$[',#position-1,']'));
--See this for a position safe string-splitter (zero-based):
SELECT JsonArray.[key] AS [Position]
,JsonArray.[value] AS [Part]
FROM OPENJSON('["' + REPLACE(#str,' ','","') + '"]') JsonArray
In this post I tested various approaches and found, that OPENJSON is really fast. Even much faster than the famous "delimitedSplit8k()" method...
UPDATE 2 - Get the values type-safe
We can use an array within an array simply by using doubled [[]]. This allows for a typed WITH-clause:
DECLARE #SomeDelimitedString VARCHAR(100)='part1|1|20190920';
DECLARE #JsonArray NVARCHAR(MAX)=CONCAT('[["',REPLACE(#SomeDelimitedString,'|','","'),'"]]');
SELECT #SomeDelimitedString AS TheOriginal
,#JsonArray AS TransformedToJSON
,ValuesFromTheArray.*
FROM OPENJSON(#JsonArray)
WITH(TheFirstFragment VARCHAR(100) '$[0]'
,TheSecondFragment INT '$[1]'
,TheThirdFragment DATE '$[2]') ValuesFromTheArray
Here is a UDF which will do it. It will return a table of the delimited values, haven't tried all scenarios on it but your example works fine.
CREATE FUNCTION SplitString
(
-- Add the parameters for the function here
#myString varchar(500),
#deliminator varchar(10)
)
RETURNS
#ReturnTable TABLE
(
-- Add the column definitions for the TABLE variable here
[id] [int] IDENTITY(1,1) NOT NULL,
[part] [varchar](50) NULL
)
AS
BEGIN
Declare #iSpaces int
Declare #part varchar(50)
--initialize spaces
Select #iSpaces = charindex(#deliminator,#myString,0)
While #iSpaces > 0
Begin
Select #part = substring(#myString,0,charindex(#deliminator,#myString,0))
Insert Into #ReturnTable(part)
Select #part
Select #myString = substring(#mystring,charindex(#deliminator,#myString,0)+ len(#deliminator),len(#myString) - charindex(' ',#myString,0))
Select #iSpaces = charindex(#deliminator,#myString,0)
end
If len(#myString) > 0
Insert Into #ReturnTable
Select #myString
RETURN
END
GO
You would call it like this:
Select * From SplitString('Hello John Smith',' ')
Edit: Updated solution to handle delimters with a len>1 as in :
select * From SplitString('Hello**John**Smith','**')
Here I post a simple way of solution
CREATE FUNCTION [dbo].[split](
#delimited NVARCHAR(MAX),
#delimiter NVARCHAR(100)
) RETURNS #t TABLE (id INT IDENTITY(1,1), val NVARCHAR(MAX))
AS
BEGIN
DECLARE #xml XML
SET #xml = N'<t>' + REPLACE(#delimited,#delimiter,'</t><t>') + '</t>'
INSERT INTO #t(val)
SELECT r.value('.','varchar(MAX)') as item
FROM #xml.nodes('/t') as records(r)
RETURN
END
Execute the function like this
select * from dbo.split('Hello John Smith',' ')
In my opinion you guys are making it way too complicated. Just create a CLR UDF and be done with it.
using System;
using System.Data;
using System.Data.SqlClient;
using System.Data.SqlTypes;
using Microsoft.SqlServer.Server;
using System.Collections.Generic;
public partial class UserDefinedFunctions {
[SqlFunction]
public static SqlString SearchString(string Search) {
List<string> SearchWords = new List<string>();
foreach (string s in Search.Split(new char[] { ' ' })) {
if (!s.ToLower().Equals("or") && !s.ToLower().Equals("and")) {
SearchWords.Add(s);
}
}
return new SqlString(string.Join(" OR ", SearchWords.ToArray()));
}
};
What about using string and values() statement?
DECLARE #str varchar(max)
SET #str = 'Hello John Smith'
DECLARE #separator varchar(max)
SET #separator = ' '
DECLARE #Splited TABLE(id int IDENTITY(1,1), item varchar(max))
SET #str = REPLACE(#str, #separator, '''),(''')
SET #str = 'SELECT * FROM (VALUES(''' + #str + ''')) AS V(A)'
INSERT INTO #Splited
EXEC(#str)
SELECT * FROM #Splited
Result-set achieved.
id item
1 Hello
2 John
3 Smith
I use the answer of frederic but this did not work in SQL Server 2005
I modified it and I'm using select with union all and it works
DECLARE #str varchar(max)
SET #str = 'Hello John Smith how are you'
DECLARE #separator varchar(max)
SET #separator = ' '
DECLARE #Splited table(id int IDENTITY(1,1), item varchar(max))
SET #str = REPLACE(#str, #separator, ''' UNION ALL SELECT ''')
SET #str = ' SELECT ''' + #str + ''' '
INSERT INTO #Splited
EXEC(#str)
SELECT * FROM #Splited
And the result-set is:
id item
1 Hello
2 John
3 Smith
4 how
5 are
6 you
This pattern works fine and you can generalize
Convert(xml,'<n>'+Replace(FIELD,'.','</n><n>')+'</n>').value('(/n[INDEX])','TYPE')
^^^^^ ^^^^^ ^^^^
note FIELD, INDEX and TYPE.
Let some table with identifiers like
sys.message.1234.warning.A45
sys.message.1235.error.O98
....
Then, you can write
SELECT Source = q.value('(/n[1])', 'varchar(10)'),
RecordType = q.value('(/n[2])', 'varchar(20)'),
RecordNumber = q.value('(/n[3])', 'int'),
Status = q.value('(/n[4])', 'varchar(5)')
FROM (
SELECT q = Convert(xml,'<n>'+Replace(fieldName,'.','</n><n>')+'</n>')
FROM some_TABLE
) Q
splitting and casting all parts.
Yet another get n'th part of string by delimeter function:
create function GetStringPartByDelimeter (
#value as nvarchar(max),
#delimeter as nvarchar(max),
#position as int
) returns NVARCHAR(MAX)
AS BEGIN
declare #startPos as int
declare #endPos as int
set #endPos = -1
while (#position > 0 and #endPos != 0) begin
set #startPos = #endPos + 1
set #endPos = charindex(#delimeter, #value, #startPos)
if(#position = 1) begin
if(#endPos = 0)
set #endPos = len(#value) + 1
return substring(#value, #startPos, #endPos - #startPos)
end
set #position = #position - 1
end
return null
end
and the usage:
select dbo.GetStringPartByDelimeter ('a;b;c;d;e', ';', 3)
which returns:
c
If your database has compatibility level of 130 or higher then you can use the STRING_SPLIT function along with OFFSET FETCH clauses to get the specific item by index.
To get the item at index N (zero based), you can use the following code
SELECT value
FROM STRING_SPLIT('Hello John Smith',' ')
ORDER BY (SELECT NULL)
OFFSET N ROWS
FETCH NEXT 1 ROWS ONLY
To check the compatibility level of your database, execute this code:
SELECT compatibility_level
FROM sys.databases WHERE name = 'YourDBName';
Try this:
CREATE function [SplitWordList]
(
#list varchar(8000)
)
returns #t table
(
Word varchar(50) not null,
Position int identity(1,1) not null
)
as begin
declare
#pos int,
#lpos int,
#item varchar(100),
#ignore varchar(100),
#dl int,
#a1 int,
#a2 int,
#z1 int,
#z2 int,
#n1 int,
#n2 int,
#c varchar(1),
#a smallint
select
#a1 = ascii('a'),
#a2 = ascii('A'),
#z1 = ascii('z'),
#z2 = ascii('Z'),
#n1 = ascii('0'),
#n2 = ascii('9')
set #ignore = '''"'
set #pos = 1
set #dl = datalength(#list)
set #lpos = 1
set #item = ''
while (#pos <= #dl) begin
set #c = substring(#list, #pos, 1)
if (#ignore not like '%' + #c + '%') begin
set #a = ascii(#c)
if ((#a >= #a1) and (#a <= #z1))
or ((#a >= #a2) and (#a <= #z2))
or ((#a >= #n1) and (#a <= #n2))
begin
set #item = #item + #c
end else if (#item > '') begin
insert into #t values (#item)
set #item = ''
end
end
set #pos = #pos + 1
end
if (#item > '') begin
insert into #t values (#item)
end
return
end
Test it like this:
select * from SplitWordList('Hello John Smith')
I was looking for the solution on net and the below works for me.
Ref.
And you call the function like this :
SELECT * FROM dbo.split('ram shyam hari gopal',' ')
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE FUNCTION [dbo].[Split](#String VARCHAR(8000), #Delimiter CHAR(1))
RETURNS #temptable TABLE (items VARCHAR(8000))
AS
BEGIN
DECLARE #idx INT
DECLARE #slice VARCHAR(8000)
SELECT #idx = 1
IF len(#String)<1 OR #String IS NULL RETURN
WHILE #idx!= 0
BEGIN
SET #idx = charindex(#Delimiter,#String)
IF #idx!=0
SET #slice = LEFT(#String,#idx - 1)
ELSE
SET #slice = #String
IF(len(#slice)>0)
INSERT INTO #temptable(Items) VALUES(#slice)
SET #String = RIGHT(#String,len(#String) - #idx)
IF len(#String) = 0 break
END
RETURN
END
The following example uses a recursive CTE
Update 18.09.2013
CREATE FUNCTION dbo.SplitStrings_CTE(#List nvarchar(max), #Delimiter nvarchar(1))
RETURNS #returns TABLE (val nvarchar(max), [level] int, PRIMARY KEY CLUSTERED([level]))
AS
BEGIN
;WITH cte AS
(
SELECT SUBSTRING(#List, 0, CHARINDEX(#Delimiter, #List + #Delimiter)) AS val,
CAST(STUFF(#List + #Delimiter, 1, CHARINDEX(#Delimiter, #List + #Delimiter), '') AS nvarchar(max)) AS stval,
1 AS [level]
UNION ALL
SELECT SUBSTRING(stval, 0, CHARINDEX(#Delimiter, stval)),
CAST(STUFF(stval, 1, CHARINDEX(#Delimiter, stval), '') AS nvarchar(max)),
[level] + 1
FROM cte
WHERE stval != ''
)
INSERT #returns
SELECT REPLACE(val, ' ','' ) AS val, [level]
FROM cte
WHERE val > ''
RETURN
END
Demo on SQLFiddle
Alter Function dbo.fn_Split
(
#Expression nvarchar(max),
#Delimiter nvarchar(20) = ',',
#Qualifier char(1) = Null
)
RETURNS #Results TABLE (id int IDENTITY(1,1), value nvarchar(max))
AS
BEGIN
/* USAGE
Select * From dbo.fn_Split('apple pear grape banana orange honeydew cantalope 3 2 1 4', ' ', Null)
Select * From dbo.fn_Split('1,abc,"Doe, John",4', ',', '"')
Select * From dbo.fn_Split('Hello 0,"&""&&&&', ',', '"')
*/
-- Declare Variables
DECLARE
#X xml,
#Temp nvarchar(max),
#Temp2 nvarchar(max),
#Start int,
#End int
-- HTML Encode #Expression
Select #Expression = (Select #Expression For XML Path(''))
-- Find all occurences of #Delimiter within #Qualifier and replace with |||***|||
While PATINDEX('%' + #Qualifier + '%', #Expression) > 0 AND Len(IsNull(#Qualifier, '')) > 0
BEGIN
Select
-- Starting character position of #Qualifier
#Start = PATINDEX('%' + #Qualifier + '%', #Expression),
-- #Expression starting at the #Start position
#Temp = SubString(#Expression, #Start + 1, LEN(#Expression)-#Start+1),
-- Next position of #Qualifier within #Expression
#End = PATINDEX('%' + #Qualifier + '%', #Temp) - 1,
-- The part of Expression found between the #Qualifiers
#Temp2 = Case When #End < 0 Then #Temp Else Left(#Temp, #End) End,
-- New #Expression
#Expression = REPLACE(#Expression,
#Qualifier + #Temp2 + Case When #End < 0 Then '' Else #Qualifier End,
Replace(#Temp2, #Delimiter, '|||***|||')
)
END
-- Replace all occurences of #Delimiter within #Expression with '</fn_Split><fn_Split>'
-- And convert it to XML so we can select from it
SET
#X = Cast('<fn_Split>' +
Replace(#Expression, #Delimiter, '</fn_Split><fn_Split>') +
'</fn_Split>' as xml)
-- Insert into our returnable table replacing '|||***|||' back to #Delimiter
INSERT #Results
SELECT
"Value" = LTRIM(RTrim(Replace(C.value('.', 'nvarchar(max)'), '|||***|||', #Delimiter)))
FROM
#X.nodes('fn_Split') as X(C)
-- Return our temp table
RETURN
END
You can split a string in SQL without needing a function:
DECLARE #bla varchar(MAX)
SET #bla = 'BED40DFC-F468-46DD-8017-00EF2FA3E4A4,64B59FC5-3F4D-4B0E-9A48-01F3D4F220B0,A611A108-97CA-42F3-A2E1-057165339719,E72D95EA-578F-45FC-88E5-075F66FD726C'
-- http://stackoverflow.com/questions/14712864/how-to-query-values-from-xml-nodes
SELECT
x.XmlCol.value('.', 'varchar(36)') AS val
FROM
(
SELECT
CAST('<e>' + REPLACE(#bla, ',', '</e><e>') + '</e>' AS xml) AS RawXml
) AS b
CROSS APPLY b.RawXml.nodes('e') x(XmlCol);
If you need to support arbitrary strings (with xml special characters)
DECLARE #bla NVARCHAR(MAX)
SET #bla = '<html>unsafe & safe Utf8CharsDon''tGetEncoded ÄöÜ - "Conex"<html>,Barnes & Noble,abc,def,ghi'
-- http://stackoverflow.com/questions/14712864/how-to-query-values-from-xml-nodes
SELECT
x.XmlCol.value('.', 'nvarchar(MAX)') AS val
FROM
(
SELECT
CAST('<e>' + REPLACE((SELECT #bla FOR XML PATH('')), ',', '</e><e>') + '</e>' AS xml) AS RawXml
) AS b
CROSS APPLY b.RawXml.nodes('e') x(XmlCol);
In Azure SQL Database (based on Microsoft SQL Server but not exactly the same thing) the signature of STRING_SPLIT function looks like:
STRING_SPLIT ( string , separator [ , enable_ordinal ] )
When enable_ordinal flag is set to 1 the result will include a column named ordinal that consists of the 1‑based position of the substring within the input string:
SELECT *
FROM STRING_SPLIT('hello john smith', ' ', 1)
| value | ordinal |
|-------|---------|
| hello | 1 |
| john | 2 |
| smith | 3 |
This allows us to do this:
SELECT value
FROM STRING_SPLIT('hello john smith', ' ', 1)
WHERE ordinal = 2
| value |
|-------|
| john |
If enable_ordinal is not available then there is a trick which assumes that the substrings within the input string are unique. In this scenario, CHAR_INDEX could be used to find the position of the substring within the input string:
SELECT value, ROW_NUMBER() OVER (ORDER BY CHARINDEX(value, input_str)) AS ord_pos
FROM (VALUES
('hello john smith')
) AS x(input_str)
CROSS APPLY STRING_SPLIT(input_str, ' ')
| value | ord_pos |
|-------+---------|
| hello | 1 |
| john | 2 |
| smith | 3 |
I know it's an old Question, but i think some one can benefit from my solution.
select
SUBSTRING(column_name,1,CHARINDEX(' ',column_name,1)-1)
,SUBSTRING(SUBSTRING(column_name,CHARINDEX(' ',column_name,1)+1,LEN(column_name))
,1
,CHARINDEX(' ',SUBSTRING(column_name,CHARINDEX(' ',column_name,1)+1,LEN(column_name)),1)-1)
,SUBSTRING(SUBSTRING(column_name,CHARINDEX(' ',column_name,1)+1,LEN(column_name))
,CHARINDEX(' ',SUBSTRING(column_name,CHARINDEX(' ',column_name,1)+1,LEN(column_name)),1)+1
,LEN(column_name))
from table_name
SQL FIDDLE
Advantages:
It separates all the 3 sub-strings deliminator by ' '.
One must not use while loop, as it decreases the performance.
No need to Pivot as all the resultant sub-string will be displayed in
one Row
Limitations:
One must know the total no. of spaces (sub-string).
Note: the solution can give sub-string up to to N.
To overcame the limitation we can use the following ref.
But again the above solution can't be use in a table (Actaully i wasn't able to use it).
Again i hope this solution can help some-one.
Update: In case of Records > 50000 it is not advisable to use LOOPS as it will degrade the Performance
Pure set-based solution using TVF with recursive CTE. You can JOIN and APPLY this function to any dataset.
create function [dbo].[SplitStringToResultSet] (#value varchar(max), #separator char(1))
returns table
as return
with r as (
select value, cast(null as varchar(max)) [x], -1 [no] from (select rtrim(cast(#value as varchar(max))) [value]) as j
union all
select right(value, len(value)-case charindex(#separator, value) when 0 then len(value) else charindex(#separator, value) end) [value]
, left(r.[value], case charindex(#separator, r.value) when 0 then len(r.value) else abs(charindex(#separator, r.[value])-1) end ) [x]
, [no] + 1 [no]
from r where value > '')
select ltrim(x) [value], [no] [index] from r where x is not null;
go
Usage:
select *
from [dbo].[SplitStringToResultSet]('Hello John Smith', ' ')
where [index] = 1;
Result:
value index
-------------
John 1
Almost all the other answers are replacing the string being split which wastes CPU cycles and performs unnecessary memory allocations.
I cover a much better way to do a string split here: http://www.digitalruby.com/split-string-sql-server/
Here is the code:
SET NOCOUNT ON
-- You will want to change nvarchar(MAX) to nvarchar(50), varchar(50) or whatever matches exactly with the string column you will be searching against
DECLARE #SplitStringTable TABLE (Value nvarchar(MAX) NOT NULL)
DECLARE #StringToSplit nvarchar(MAX) = 'your|string|to|split|here'
DECLARE #SplitEndPos int
DECLARE #SplitValue nvarchar(MAX)
DECLARE #SplitDelim nvarchar(1) = '|'
DECLARE #SplitStartPos int = 1
SET #SplitEndPos = CHARINDEX(#SplitDelim, #StringToSplit, #SplitStartPos)
WHILE #SplitEndPos > 0
BEGIN
SET #SplitValue = SUBSTRING(#StringToSplit, #SplitStartPos, (#SplitEndPos - #SplitStartPos))
INSERT #SplitStringTable (Value) VALUES (#SplitValue)
SET #SplitStartPos = #SplitEndPos + 1
SET #SplitEndPos = CHARINDEX(#SplitDelim, #StringToSplit, #SplitStartPos)
END
SET #SplitValue = SUBSTRING(#StringToSplit, #SplitStartPos, 2147483647)
INSERT #SplitStringTable (Value) VALUES(#SplitValue)
SET NOCOUNT OFF
-- You can select or join with the values in #SplitStringTable at this point.
Recursive CTE solution with server pain, test it
MS SQL Server 2008 Schema Setup:
create table Course( Courses varchar(100) );
insert into Course values ('Hello John Smith');
Query 1:
with cte as
( select
left( Courses, charindex( ' ' , Courses) ) as a_l,
cast( substring( Courses,
charindex( ' ' , Courses) + 1 ,
len(Courses ) ) + ' '
as varchar(100) ) as a_r,
Courses as a,
0 as n
from Course t
union all
select
left(a_r, charindex( ' ' , a_r) ) as a_l,
substring( a_r, charindex( ' ' , a_r) + 1 , len(a_R ) ) as a_r,
cte.a,
cte.n + 1 as n
from Course t inner join cte
on t.Courses = cte.a and len( a_r ) > 0
)
select a_l, n from cte
--where N = 1
Results:
| A_L | N |
|--------|---|
| Hello | 0 |
| John | 1 |
| Smith | 2 |
while similar to the xml based answer by josejuan, i found that processing the xml path only once, then pivoting was moderately more efficient:
select ID,
[3] as PathProvidingID,
[4] as PathProvider,
[5] as ComponentProvidingID,
[6] as ComponentProviding,
[7] as InputRecievingID,
[8] as InputRecieving,
[9] as RowsPassed,
[10] as InputRecieving2
from
(
select id,message,d.* from sysssislog cross apply (
SELECT Item = y.i.value('(./text())[1]', 'varchar(200)'),
row_number() over(order by y.i) as rn
FROM
(
SELECT x = CONVERT(XML, '<i>' + REPLACE(Message, ':', '</i><i>') + '</i>').query('.')
) AS a CROSS APPLY x.nodes('i') AS y(i)
) d
WHERE event
=
'OnPipelineRowsSent'
) as tokens
pivot
( max(item) for [rn] in ([3],[4],[5],[6],[7],[8],[9],[10])
) as data
ran in 8:30
select id,
tokens.value('(/n[3])', 'varchar(100)')as PathProvidingID,
tokens.value('(/n[4])', 'varchar(100)') as PathProvider,
tokens.value('(/n[5])', 'varchar(100)') as ComponentProvidingID,
tokens.value('(/n[6])', 'varchar(100)') as ComponentProviding,
tokens.value('(/n[7])', 'varchar(100)') as InputRecievingID,
tokens.value('(/n[8])', 'varchar(100)') as InputRecieving,
tokens.value('(/n[9])', 'varchar(100)') as RowsPassed
from
(
select id, Convert(xml,'<n>'+Replace(message,'.','</n><n>')+'</n>') tokens
from sysssislog
WHERE event
=
'OnPipelineRowsSent'
) as data
ran in 9:20
CREATE FUNCTION [dbo].[fnSplitString]
(
#string NVARCHAR(MAX),
#delimiter CHAR(1)
)
RETURNS #output TABLE(splitdata NVARCHAR(MAX)
)
BEGIN
DECLARE #start INT, #end INT
SELECT #start = 1, #end = CHARINDEX(#delimiter, #string)
WHILE #start < LEN(#string) + 1 BEGIN
IF #end = 0
SET #end = LEN(#string) + 1
INSERT INTO #output (splitdata)
VALUES(SUBSTRING(#string, #start, #end - #start))
SET #start = #end + 1
SET #end = CHARINDEX(#delimiter, #string, #start)
END
RETURN
END
AND USE IT
select *from dbo.fnSplitString('Querying SQL Server','')
if anyone wants to get only one part of the seperatured text can use this
select * from fromSplitStringSep('Word1 wordr2 word3',' ')
CREATE function [dbo].[SplitStringSep]
(
#str nvarchar(4000),
#separator char(1)
)
returns table
AS
return (
with tokens(p, a, b) AS (
select
1,
1,
charindex(#separator, #str)
union all
select
p + 1,
b + 1,
charindex(#separator, #str, b + 1)
from tokens
where b > 0
)
select
p-1 zeroBasedOccurance,
substring(
#str,
a,
case when b > 0 then b-a ELSE 4000 end)
AS s
from tokens
)
I devoloped this,
declare #x nvarchar(Max) = 'ali.veli.deli.';
declare #item nvarchar(Max);
declare #splitter char='.';
while CHARINDEX(#splitter,#x) != 0
begin
set #item = LEFT(#x,CHARINDEX(#splitter,#x))
set #x = RIGHT(#x,len(#x)-len(#item) )
select #item as item, #x as x;
end
the only attention you should is dot '.' that end of the #x is always should be there.
building on #NothingsImpossible solution, or, rather, comment on the most voted answer (just below the accepted one), i found the following quick-and-dirty solution fulfill my own needs - it has a benefit of being solely within SQL domain.
given a string "first;second;third;fourth;fifth", say, I want to get the third token. this works only if we know how many tokens the string is going to have - in this case it's 5. so my way of action is to chop the last two tokens away (inner query), and then to chop the first two tokens away (outer query)
i know that this is ugly and covers the specific conditions i was in, but am posting it just in case somebody finds it useful. cheers
select
REVERSE(
SUBSTRING(
reverse_substring,
0,
CHARINDEX(';', reverse_substring)
)
)
from
(
select
msg,
SUBSTRING(
REVERSE(msg),
CHARINDEX(
';',
REVERSE(msg),
CHARINDEX(
';',
REVERSE(msg)
)+1
)+1,
1000
) reverse_substring
from
(
select 'first;second;third;fourth;fifth' msg
) a
) b
declare #strng varchar(max)='hello john smith'
select (
substring(
#strng,
charindex(' ', #strng) + 1,
(
(charindex(' ', #strng, charindex(' ', #strng) + 1))
- charindex(' ',#strng)
)
))
Using SQL Server, how do I split a string so I can access item x?
Take a string "Hello John Smith". How can I split the string by space and access the item at index 1 which should return "John"?
I don't believe SQL Server has a built-in split function, so other than a UDF, the only other answer I know is to hijack the PARSENAME function:
SELECT PARSENAME(REPLACE('Hello John Smith', ' ', '.'), 2)
PARSENAME takes a string and splits it on the period character. It takes a number as its second argument, and that number specifies which segment of the string to return (working from back to front).
SELECT PARSENAME(REPLACE('Hello John Smith', ' ', '.'), 3) --return Hello
Obvious problem is when the string already contains a period. I still think using a UDF is the best way...any other suggestions?
You may find the solution in SQL User Defined Function to Parse a Delimited String helpful (from The Code Project).
You can use this simple logic:
Declare #products varchar(200) = '1|20|3|343|44|6|8765'
Declare #individual varchar(20) = null
WHILE LEN(#products) > 0
BEGIN
IF PATINDEX('%|%', #products) > 0
BEGIN
SET #individual = SUBSTRING(#products,
0,
PATINDEX('%|%', #products))
SELECT #individual
SET #products = SUBSTRING(#products,
LEN(#individual + '|') + 1,
LEN(#products))
END
ELSE
BEGIN
SET #individual = #products
SET #products = NULL
SELECT #individual
END
END
First, create a function (using CTE, common table expression does away with the need for a temp table)
create function dbo.SplitString
(
#str nvarchar(4000),
#separator char(1)
)
returns table
AS
return (
with tokens(p, a, b) AS (
select
1,
1,
charindex(#separator, #str)
union all
select
p + 1,
b + 1,
charindex(#separator, #str, b + 1)
from tokens
where b > 0
)
select
p-1 zeroBasedOccurance,
substring(
#str,
a,
case when b > 0 then b-a ELSE 4000 end)
AS s
from tokens
)
GO
Then, use it as any table (or modify it to fit within your existing stored proc) like this.
select s
from dbo.SplitString('Hello John Smith', ' ')
where zeroBasedOccurance=1
Update
Previous version would fail for input string longer than 4000 chars. This version takes care of the limitation:
create function dbo.SplitString
(
#str nvarchar(max),
#separator char(1)
)
returns table
AS
return (
with tokens(p, a, b) AS (
select
cast(1 as bigint),
cast(1 as bigint),
charindex(#separator, #str)
union all
select
p + 1,
b + 1,
charindex(#separator, #str, b + 1)
from tokens
where b > 0
)
select
p-1 ItemIndex,
substring(
#str,
a,
case when b > 0 then b-a ELSE LEN(#str) end)
AS s
from tokens
);
GO
Usage remains the same.
Most of the solutions here use while loops or recursive CTEs. A set-based approach will be superior, I promise, if you can use a delimiter other than a space:
CREATE FUNCTION [dbo].[SplitString]
(
#List NVARCHAR(MAX),
#Delim VARCHAR(255)
)
RETURNS TABLE
AS
RETURN ( SELECT [Value], idx = RANK() OVER (ORDER BY n) FROM
(
SELECT n = Number,
[Value] = LTRIM(RTRIM(SUBSTRING(#List, [Number],
CHARINDEX(#Delim, #List + #Delim, [Number]) - [Number])))
FROM (SELECT Number = ROW_NUMBER() OVER (ORDER BY name)
FROM sys.all_objects) AS x
WHERE Number <= LEN(#List)
AND SUBSTRING(#Delim + #List, [Number], LEN(#Delim)) = #Delim
) AS y
);
Sample usage:
SELECT Value FROM dbo.SplitString('foo,bar,blat,foo,splunge',',')
WHERE idx = 3;
Results:
----
blat
You could also add the idx you want as an argument to the function, but I'll leave that as an exercise to the reader.
You can't do this with just the native STRING_SPLIT function added in SQL Server 2016, because there is no guarantee that the output will be rendered in the order of the original list. In other words, if you pass in 3,6,1 the result will likely be in that order, but it could be 1,3,6. I have asked for the community's help in improving the built-in function here:
Please help with STRING_SPLIT improvements
With enough qualitative feedback, they may actually consider making some of these enhancements:
STRING_SPLIT is not feature complete
More on split functions, why (and proof that) while loops and recursive CTEs don't scale, and better alternatives, if splitting strings coming from the application layer:
Split strings the right way – or the next best way
Splitting Strings : A Follow-Up
Splitting Strings : Now with less T-SQL
Comparing string splitting / concatenation methods
Processing a list of integers : my approach
Splitting a list of integers : another roundup
More on splitting lists : custom delimiters, preventing duplicates, and maintaining order
Removing Duplicates from Strings in SQL Server
On SQL Server 2016 or above, though, you should look at STRING_SPLIT() and STRING_AGG():
Performance Surprises and Assumptions : STRING_SPLIT()
STRING_SPLIT() in SQL Server 2016 : Follow-Up #1
STRING_SPLIT() in SQL Server 2016 : Follow-Up #2
SQL Server v.Next : STRING_AGG() performance
Solve old problems with SQL Server’s new STRING_AGG and STRING_SPLIT functions
You can leverage a Number table to do the string parsing.
Create a physical numbers table:
create table dbo.Numbers (N int primary key);
insert into dbo.Numbers
select top 1000 row_number() over(order by number) from master..spt_values
go
Create test table with 1000000 rows
create table #yak (i int identity(1,1) primary key, array varchar(50))
insert into #yak(array)
select 'a,b,c' from dbo.Numbers n cross join dbo.Numbers nn
go
Create the function
create function [dbo].[ufn_ParseArray]
( #Input nvarchar(4000),
#Delimiter char(1) = ',',
#BaseIdent int
)
returns table as
return
( select row_number() over (order by n asc) + (#BaseIdent - 1) [i],
substring(#Input, n, charindex(#Delimiter, #Input + #Delimiter, n) - n) s
from dbo.Numbers
where n <= convert(int, len(#Input)) and
substring(#Delimiter + #Input, n, 1) = #Delimiter
)
go
Usage (outputs 3mil rows in 40s on my laptop)
select *
from #yak
cross apply dbo.ufn_ParseArray(array, ',', 1)
cleanup
drop table dbo.Numbers;
drop function [dbo].[ufn_ParseArray]
Performance here is not amazing, but calling a function over a million row table is not the best idea. If performing a string split over many rows I would avoid the function.
This question is not about a string split approach, but about how to get the nth element.
All answers here are doing some kind of string splitting using recursion, CTEs, multiple CHARINDEX, REVERSE and PATINDEX, inventing functions, call for CLR methods, number tables, CROSS APPLYs ... Most answers cover many lines of code.
But - if you really want nothing more than an approach to get the nth element - this can be done as real one-liner, no UDF, not even a sub-select... And as an extra benefit: type safe
Get part 2 delimited by a space:
DECLARE #input NVARCHAR(100)=N'part1 part2 part3';
SELECT CAST(N'<x>' + REPLACE(#input,N' ',N'</x><x>') + N'</x>' AS XML).value('/x[2]','nvarchar(max)')
Of course you can use variables for delimiter and position (use sql:column to retrieve the position directly from a query's value):
DECLARE #dlmt NVARCHAR(10)=N' ';
DECLARE #pos INT = 2;
SELECT CAST(N'<x>' + REPLACE(#input,#dlmt,N'</x><x>') + N'</x>' AS XML).value('/x[sql:variable("#pos")][1]','nvarchar(max)')
If your string might include forbidden characters (especially one among &><), you still can do it this way. Just use FOR XML PATH on your string first to replace all forbidden characters with the fitting escape sequence implicitly.
It's a very special case if - additionally - your delimiter is the semicolon. In this case I replace the delimiter first to '#DLMT#', and replace this to the XML tags finally:
SET #input=N'Some <, > and &;Other äöü#€;One more';
SET #dlmt=N';';
SELECT CAST(N'<x>' + REPLACE((SELECT REPLACE(#input,#dlmt,'#DLMT#') AS [*] FOR XML PATH('')),N'#DLMT#',N'</x><x>') + N'</x>' AS XML).value('/x[sql:variable("#pos")][1]','nvarchar(max)');
UPDATE for SQL-Server 2016+
Regretfully the developers forgot to return the part's index with STRING_SPLIT. But, using SQL-Server 2016+, there is JSON_VALUE and OPENJSON.
With JSON_VALUE we can pass in the position as the index' array.
For OPENJSON the documentation states clearly:
When OPENJSON parses a JSON array, the function returns the indexes of the elements in the JSON text as keys.
A string like 1,2,3 needs nothing more than brackets: [1,2,3].
A string of words like this is an example needs to be ["this","is","an","example"].
These are very easy string operations. Just try it out:
DECLARE #str VARCHAR(100)='Hello John Smith';
DECLARE #position INT = 2;
--We can build the json-path '$[1]' using CONCAT
SELECT JSON_VALUE('["' + REPLACE(#str,' ','","') + '"]',CONCAT('$[',#position-1,']'));
--See this for a position safe string-splitter (zero-based):
SELECT JsonArray.[key] AS [Position]
,JsonArray.[value] AS [Part]
FROM OPENJSON('["' + REPLACE(#str,' ','","') + '"]') JsonArray
In this post I tested various approaches and found, that OPENJSON is really fast. Even much faster than the famous "delimitedSplit8k()" method...
UPDATE 2 - Get the values type-safe
We can use an array within an array simply by using doubled [[]]. This allows for a typed WITH-clause:
DECLARE #SomeDelimitedString VARCHAR(100)='part1|1|20190920';
DECLARE #JsonArray NVARCHAR(MAX)=CONCAT('[["',REPLACE(#SomeDelimitedString,'|','","'),'"]]');
SELECT #SomeDelimitedString AS TheOriginal
,#JsonArray AS TransformedToJSON
,ValuesFromTheArray.*
FROM OPENJSON(#JsonArray)
WITH(TheFirstFragment VARCHAR(100) '$[0]'
,TheSecondFragment INT '$[1]'
,TheThirdFragment DATE '$[2]') ValuesFromTheArray
Here is a UDF which will do it. It will return a table of the delimited values, haven't tried all scenarios on it but your example works fine.
CREATE FUNCTION SplitString
(
-- Add the parameters for the function here
#myString varchar(500),
#deliminator varchar(10)
)
RETURNS
#ReturnTable TABLE
(
-- Add the column definitions for the TABLE variable here
[id] [int] IDENTITY(1,1) NOT NULL,
[part] [varchar](50) NULL
)
AS
BEGIN
Declare #iSpaces int
Declare #part varchar(50)
--initialize spaces
Select #iSpaces = charindex(#deliminator,#myString,0)
While #iSpaces > 0
Begin
Select #part = substring(#myString,0,charindex(#deliminator,#myString,0))
Insert Into #ReturnTable(part)
Select #part
Select #myString = substring(#mystring,charindex(#deliminator,#myString,0)+ len(#deliminator),len(#myString) - charindex(' ',#myString,0))
Select #iSpaces = charindex(#deliminator,#myString,0)
end
If len(#myString) > 0
Insert Into #ReturnTable
Select #myString
RETURN
END
GO
You would call it like this:
Select * From SplitString('Hello John Smith',' ')
Edit: Updated solution to handle delimters with a len>1 as in :
select * From SplitString('Hello**John**Smith','**')
Here I post a simple way of solution
CREATE FUNCTION [dbo].[split](
#delimited NVARCHAR(MAX),
#delimiter NVARCHAR(100)
) RETURNS #t TABLE (id INT IDENTITY(1,1), val NVARCHAR(MAX))
AS
BEGIN
DECLARE #xml XML
SET #xml = N'<t>' + REPLACE(#delimited,#delimiter,'</t><t>') + '</t>'
INSERT INTO #t(val)
SELECT r.value('.','varchar(MAX)') as item
FROM #xml.nodes('/t') as records(r)
RETURN
END
Execute the function like this
select * from dbo.split('Hello John Smith',' ')
In my opinion you guys are making it way too complicated. Just create a CLR UDF and be done with it.
using System;
using System.Data;
using System.Data.SqlClient;
using System.Data.SqlTypes;
using Microsoft.SqlServer.Server;
using System.Collections.Generic;
public partial class UserDefinedFunctions {
[SqlFunction]
public static SqlString SearchString(string Search) {
List<string> SearchWords = new List<string>();
foreach (string s in Search.Split(new char[] { ' ' })) {
if (!s.ToLower().Equals("or") && !s.ToLower().Equals("and")) {
SearchWords.Add(s);
}
}
return new SqlString(string.Join(" OR ", SearchWords.ToArray()));
}
};
What about using string and values() statement?
DECLARE #str varchar(max)
SET #str = 'Hello John Smith'
DECLARE #separator varchar(max)
SET #separator = ' '
DECLARE #Splited TABLE(id int IDENTITY(1,1), item varchar(max))
SET #str = REPLACE(#str, #separator, '''),(''')
SET #str = 'SELECT * FROM (VALUES(''' + #str + ''')) AS V(A)'
INSERT INTO #Splited
EXEC(#str)
SELECT * FROM #Splited
Result-set achieved.
id item
1 Hello
2 John
3 Smith
I use the answer of frederic but this did not work in SQL Server 2005
I modified it and I'm using select with union all and it works
DECLARE #str varchar(max)
SET #str = 'Hello John Smith how are you'
DECLARE #separator varchar(max)
SET #separator = ' '
DECLARE #Splited table(id int IDENTITY(1,1), item varchar(max))
SET #str = REPLACE(#str, #separator, ''' UNION ALL SELECT ''')
SET #str = ' SELECT ''' + #str + ''' '
INSERT INTO #Splited
EXEC(#str)
SELECT * FROM #Splited
And the result-set is:
id item
1 Hello
2 John
3 Smith
4 how
5 are
6 you
This pattern works fine and you can generalize
Convert(xml,'<n>'+Replace(FIELD,'.','</n><n>')+'</n>').value('(/n[INDEX])','TYPE')
^^^^^ ^^^^^ ^^^^
note FIELD, INDEX and TYPE.
Let some table with identifiers like
sys.message.1234.warning.A45
sys.message.1235.error.O98
....
Then, you can write
SELECT Source = q.value('(/n[1])', 'varchar(10)'),
RecordType = q.value('(/n[2])', 'varchar(20)'),
RecordNumber = q.value('(/n[3])', 'int'),
Status = q.value('(/n[4])', 'varchar(5)')
FROM (
SELECT q = Convert(xml,'<n>'+Replace(fieldName,'.','</n><n>')+'</n>')
FROM some_TABLE
) Q
splitting and casting all parts.
Yet another get n'th part of string by delimeter function:
create function GetStringPartByDelimeter (
#value as nvarchar(max),
#delimeter as nvarchar(max),
#position as int
) returns NVARCHAR(MAX)
AS BEGIN
declare #startPos as int
declare #endPos as int
set #endPos = -1
while (#position > 0 and #endPos != 0) begin
set #startPos = #endPos + 1
set #endPos = charindex(#delimeter, #value, #startPos)
if(#position = 1) begin
if(#endPos = 0)
set #endPos = len(#value) + 1
return substring(#value, #startPos, #endPos - #startPos)
end
set #position = #position - 1
end
return null
end
and the usage:
select dbo.GetStringPartByDelimeter ('a;b;c;d;e', ';', 3)
which returns:
c
If your database has compatibility level of 130 or higher then you can use the STRING_SPLIT function along with OFFSET FETCH clauses to get the specific item by index.
To get the item at index N (zero based), you can use the following code
SELECT value
FROM STRING_SPLIT('Hello John Smith',' ')
ORDER BY (SELECT NULL)
OFFSET N ROWS
FETCH NEXT 1 ROWS ONLY
To check the compatibility level of your database, execute this code:
SELECT compatibility_level
FROM sys.databases WHERE name = 'YourDBName';
Try this:
CREATE function [SplitWordList]
(
#list varchar(8000)
)
returns #t table
(
Word varchar(50) not null,
Position int identity(1,1) not null
)
as begin
declare
#pos int,
#lpos int,
#item varchar(100),
#ignore varchar(100),
#dl int,
#a1 int,
#a2 int,
#z1 int,
#z2 int,
#n1 int,
#n2 int,
#c varchar(1),
#a smallint
select
#a1 = ascii('a'),
#a2 = ascii('A'),
#z1 = ascii('z'),
#z2 = ascii('Z'),
#n1 = ascii('0'),
#n2 = ascii('9')
set #ignore = '''"'
set #pos = 1
set #dl = datalength(#list)
set #lpos = 1
set #item = ''
while (#pos <= #dl) begin
set #c = substring(#list, #pos, 1)
if (#ignore not like '%' + #c + '%') begin
set #a = ascii(#c)
if ((#a >= #a1) and (#a <= #z1))
or ((#a >= #a2) and (#a <= #z2))
or ((#a >= #n1) and (#a <= #n2))
begin
set #item = #item + #c
end else if (#item > '') begin
insert into #t values (#item)
set #item = ''
end
end
set #pos = #pos + 1
end
if (#item > '') begin
insert into #t values (#item)
end
return
end
Test it like this:
select * from SplitWordList('Hello John Smith')
I was looking for the solution on net and the below works for me.
Ref.
And you call the function like this :
SELECT * FROM dbo.split('ram shyam hari gopal',' ')
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE FUNCTION [dbo].[Split](#String VARCHAR(8000), #Delimiter CHAR(1))
RETURNS #temptable TABLE (items VARCHAR(8000))
AS
BEGIN
DECLARE #idx INT
DECLARE #slice VARCHAR(8000)
SELECT #idx = 1
IF len(#String)<1 OR #String IS NULL RETURN
WHILE #idx!= 0
BEGIN
SET #idx = charindex(#Delimiter,#String)
IF #idx!=0
SET #slice = LEFT(#String,#idx - 1)
ELSE
SET #slice = #String
IF(len(#slice)>0)
INSERT INTO #temptable(Items) VALUES(#slice)
SET #String = RIGHT(#String,len(#String) - #idx)
IF len(#String) = 0 break
END
RETURN
END
The following example uses a recursive CTE
Update 18.09.2013
CREATE FUNCTION dbo.SplitStrings_CTE(#List nvarchar(max), #Delimiter nvarchar(1))
RETURNS #returns TABLE (val nvarchar(max), [level] int, PRIMARY KEY CLUSTERED([level]))
AS
BEGIN
;WITH cte AS
(
SELECT SUBSTRING(#List, 0, CHARINDEX(#Delimiter, #List + #Delimiter)) AS val,
CAST(STUFF(#List + #Delimiter, 1, CHARINDEX(#Delimiter, #List + #Delimiter), '') AS nvarchar(max)) AS stval,
1 AS [level]
UNION ALL
SELECT SUBSTRING(stval, 0, CHARINDEX(#Delimiter, stval)),
CAST(STUFF(stval, 1, CHARINDEX(#Delimiter, stval), '') AS nvarchar(max)),
[level] + 1
FROM cte
WHERE stval != ''
)
INSERT #returns
SELECT REPLACE(val, ' ','' ) AS val, [level]
FROM cte
WHERE val > ''
RETURN
END
Demo on SQLFiddle
Alter Function dbo.fn_Split
(
#Expression nvarchar(max),
#Delimiter nvarchar(20) = ',',
#Qualifier char(1) = Null
)
RETURNS #Results TABLE (id int IDENTITY(1,1), value nvarchar(max))
AS
BEGIN
/* USAGE
Select * From dbo.fn_Split('apple pear grape banana orange honeydew cantalope 3 2 1 4', ' ', Null)
Select * From dbo.fn_Split('1,abc,"Doe, John",4', ',', '"')
Select * From dbo.fn_Split('Hello 0,"&""&&&&', ',', '"')
*/
-- Declare Variables
DECLARE
#X xml,
#Temp nvarchar(max),
#Temp2 nvarchar(max),
#Start int,
#End int
-- HTML Encode #Expression
Select #Expression = (Select #Expression For XML Path(''))
-- Find all occurences of #Delimiter within #Qualifier and replace with |||***|||
While PATINDEX('%' + #Qualifier + '%', #Expression) > 0 AND Len(IsNull(#Qualifier, '')) > 0
BEGIN
Select
-- Starting character position of #Qualifier
#Start = PATINDEX('%' + #Qualifier + '%', #Expression),
-- #Expression starting at the #Start position
#Temp = SubString(#Expression, #Start + 1, LEN(#Expression)-#Start+1),
-- Next position of #Qualifier within #Expression
#End = PATINDEX('%' + #Qualifier + '%', #Temp) - 1,
-- The part of Expression found between the #Qualifiers
#Temp2 = Case When #End < 0 Then #Temp Else Left(#Temp, #End) End,
-- New #Expression
#Expression = REPLACE(#Expression,
#Qualifier + #Temp2 + Case When #End < 0 Then '' Else #Qualifier End,
Replace(#Temp2, #Delimiter, '|||***|||')
)
END
-- Replace all occurences of #Delimiter within #Expression with '</fn_Split><fn_Split>'
-- And convert it to XML so we can select from it
SET
#X = Cast('<fn_Split>' +
Replace(#Expression, #Delimiter, '</fn_Split><fn_Split>') +
'</fn_Split>' as xml)
-- Insert into our returnable table replacing '|||***|||' back to #Delimiter
INSERT #Results
SELECT
"Value" = LTRIM(RTrim(Replace(C.value('.', 'nvarchar(max)'), '|||***|||', #Delimiter)))
FROM
#X.nodes('fn_Split') as X(C)
-- Return our temp table
RETURN
END
You can split a string in SQL without needing a function:
DECLARE #bla varchar(MAX)
SET #bla = 'BED40DFC-F468-46DD-8017-00EF2FA3E4A4,64B59FC5-3F4D-4B0E-9A48-01F3D4F220B0,A611A108-97CA-42F3-A2E1-057165339719,E72D95EA-578F-45FC-88E5-075F66FD726C'
-- http://stackoverflow.com/questions/14712864/how-to-query-values-from-xml-nodes
SELECT
x.XmlCol.value('.', 'varchar(36)') AS val
FROM
(
SELECT
CAST('<e>' + REPLACE(#bla, ',', '</e><e>') + '</e>' AS xml) AS RawXml
) AS b
CROSS APPLY b.RawXml.nodes('e') x(XmlCol);
If you need to support arbitrary strings (with xml special characters)
DECLARE #bla NVARCHAR(MAX)
SET #bla = '<html>unsafe & safe Utf8CharsDon''tGetEncoded ÄöÜ - "Conex"<html>,Barnes & Noble,abc,def,ghi'
-- http://stackoverflow.com/questions/14712864/how-to-query-values-from-xml-nodes
SELECT
x.XmlCol.value('.', 'nvarchar(MAX)') AS val
FROM
(
SELECT
CAST('<e>' + REPLACE((SELECT #bla FOR XML PATH('')), ',', '</e><e>') + '</e>' AS xml) AS RawXml
) AS b
CROSS APPLY b.RawXml.nodes('e') x(XmlCol);
In Azure SQL Database (based on Microsoft SQL Server but not exactly the same thing) the signature of STRING_SPLIT function looks like:
STRING_SPLIT ( string , separator [ , enable_ordinal ] )
When enable_ordinal flag is set to 1 the result will include a column named ordinal that consists of the 1‑based position of the substring within the input string:
SELECT *
FROM STRING_SPLIT('hello john smith', ' ', 1)
| value | ordinal |
|-------|---------|
| hello | 1 |
| john | 2 |
| smith | 3 |
This allows us to do this:
SELECT value
FROM STRING_SPLIT('hello john smith', ' ', 1)
WHERE ordinal = 2
| value |
|-------|
| john |
If enable_ordinal is not available then there is a trick which assumes that the substrings within the input string are unique. In this scenario, CHAR_INDEX could be used to find the position of the substring within the input string:
SELECT value, ROW_NUMBER() OVER (ORDER BY CHARINDEX(value, input_str)) AS ord_pos
FROM (VALUES
('hello john smith')
) AS x(input_str)
CROSS APPLY STRING_SPLIT(input_str, ' ')
| value | ord_pos |
|-------+---------|
| hello | 1 |
| john | 2 |
| smith | 3 |
I know it's an old Question, but i think some one can benefit from my solution.
select
SUBSTRING(column_name,1,CHARINDEX(' ',column_name,1)-1)
,SUBSTRING(SUBSTRING(column_name,CHARINDEX(' ',column_name,1)+1,LEN(column_name))
,1
,CHARINDEX(' ',SUBSTRING(column_name,CHARINDEX(' ',column_name,1)+1,LEN(column_name)),1)-1)
,SUBSTRING(SUBSTRING(column_name,CHARINDEX(' ',column_name,1)+1,LEN(column_name))
,CHARINDEX(' ',SUBSTRING(column_name,CHARINDEX(' ',column_name,1)+1,LEN(column_name)),1)+1
,LEN(column_name))
from table_name
SQL FIDDLE
Advantages:
It separates all the 3 sub-strings deliminator by ' '.
One must not use while loop, as it decreases the performance.
No need to Pivot as all the resultant sub-string will be displayed in
one Row
Limitations:
One must know the total no. of spaces (sub-string).
Note: the solution can give sub-string up to to N.
To overcame the limitation we can use the following ref.
But again the above solution can't be use in a table (Actaully i wasn't able to use it).
Again i hope this solution can help some-one.
Update: In case of Records > 50000 it is not advisable to use LOOPS as it will degrade the Performance
Pure set-based solution using TVF with recursive CTE. You can JOIN and APPLY this function to any dataset.
create function [dbo].[SplitStringToResultSet] (#value varchar(max), #separator char(1))
returns table
as return
with r as (
select value, cast(null as varchar(max)) [x], -1 [no] from (select rtrim(cast(#value as varchar(max))) [value]) as j
union all
select right(value, len(value)-case charindex(#separator, value) when 0 then len(value) else charindex(#separator, value) end) [value]
, left(r.[value], case charindex(#separator, r.value) when 0 then len(r.value) else abs(charindex(#separator, r.[value])-1) end ) [x]
, [no] + 1 [no]
from r where value > '')
select ltrim(x) [value], [no] [index] from r where x is not null;
go
Usage:
select *
from [dbo].[SplitStringToResultSet]('Hello John Smith', ' ')
where [index] = 1;
Result:
value index
-------------
John 1
Almost all the other answers are replacing the string being split which wastes CPU cycles and performs unnecessary memory allocations.
I cover a much better way to do a string split here: http://www.digitalruby.com/split-string-sql-server/
Here is the code:
SET NOCOUNT ON
-- You will want to change nvarchar(MAX) to nvarchar(50), varchar(50) or whatever matches exactly with the string column you will be searching against
DECLARE #SplitStringTable TABLE (Value nvarchar(MAX) NOT NULL)
DECLARE #StringToSplit nvarchar(MAX) = 'your|string|to|split|here'
DECLARE #SplitEndPos int
DECLARE #SplitValue nvarchar(MAX)
DECLARE #SplitDelim nvarchar(1) = '|'
DECLARE #SplitStartPos int = 1
SET #SplitEndPos = CHARINDEX(#SplitDelim, #StringToSplit, #SplitStartPos)
WHILE #SplitEndPos > 0
BEGIN
SET #SplitValue = SUBSTRING(#StringToSplit, #SplitStartPos, (#SplitEndPos - #SplitStartPos))
INSERT #SplitStringTable (Value) VALUES (#SplitValue)
SET #SplitStartPos = #SplitEndPos + 1
SET #SplitEndPos = CHARINDEX(#SplitDelim, #StringToSplit, #SplitStartPos)
END
SET #SplitValue = SUBSTRING(#StringToSplit, #SplitStartPos, 2147483647)
INSERT #SplitStringTable (Value) VALUES(#SplitValue)
SET NOCOUNT OFF
-- You can select or join with the values in #SplitStringTable at this point.
Recursive CTE solution with server pain, test it
MS SQL Server 2008 Schema Setup:
create table Course( Courses varchar(100) );
insert into Course values ('Hello John Smith');
Query 1:
with cte as
( select
left( Courses, charindex( ' ' , Courses) ) as a_l,
cast( substring( Courses,
charindex( ' ' , Courses) + 1 ,
len(Courses ) ) + ' '
as varchar(100) ) as a_r,
Courses as a,
0 as n
from Course t
union all
select
left(a_r, charindex( ' ' , a_r) ) as a_l,
substring( a_r, charindex( ' ' , a_r) + 1 , len(a_R ) ) as a_r,
cte.a,
cte.n + 1 as n
from Course t inner join cte
on t.Courses = cte.a and len( a_r ) > 0
)
select a_l, n from cte
--where N = 1
Results:
| A_L | N |
|--------|---|
| Hello | 0 |
| John | 1 |
| Smith | 2 |
while similar to the xml based answer by josejuan, i found that processing the xml path only once, then pivoting was moderately more efficient:
select ID,
[3] as PathProvidingID,
[4] as PathProvider,
[5] as ComponentProvidingID,
[6] as ComponentProviding,
[7] as InputRecievingID,
[8] as InputRecieving,
[9] as RowsPassed,
[10] as InputRecieving2
from
(
select id,message,d.* from sysssislog cross apply (
SELECT Item = y.i.value('(./text())[1]', 'varchar(200)'),
row_number() over(order by y.i) as rn
FROM
(
SELECT x = CONVERT(XML, '<i>' + REPLACE(Message, ':', '</i><i>') + '</i>').query('.')
) AS a CROSS APPLY x.nodes('i') AS y(i)
) d
WHERE event
=
'OnPipelineRowsSent'
) as tokens
pivot
( max(item) for [rn] in ([3],[4],[5],[6],[7],[8],[9],[10])
) as data
ran in 8:30
select id,
tokens.value('(/n[3])', 'varchar(100)')as PathProvidingID,
tokens.value('(/n[4])', 'varchar(100)') as PathProvider,
tokens.value('(/n[5])', 'varchar(100)') as ComponentProvidingID,
tokens.value('(/n[6])', 'varchar(100)') as ComponentProviding,
tokens.value('(/n[7])', 'varchar(100)') as InputRecievingID,
tokens.value('(/n[8])', 'varchar(100)') as InputRecieving,
tokens.value('(/n[9])', 'varchar(100)') as RowsPassed
from
(
select id, Convert(xml,'<n>'+Replace(message,'.','</n><n>')+'</n>') tokens
from sysssislog
WHERE event
=
'OnPipelineRowsSent'
) as data
ran in 9:20
CREATE FUNCTION [dbo].[fnSplitString]
(
#string NVARCHAR(MAX),
#delimiter CHAR(1)
)
RETURNS #output TABLE(splitdata NVARCHAR(MAX)
)
BEGIN
DECLARE #start INT, #end INT
SELECT #start = 1, #end = CHARINDEX(#delimiter, #string)
WHILE #start < LEN(#string) + 1 BEGIN
IF #end = 0
SET #end = LEN(#string) + 1
INSERT INTO #output (splitdata)
VALUES(SUBSTRING(#string, #start, #end - #start))
SET #start = #end + 1
SET #end = CHARINDEX(#delimiter, #string, #start)
END
RETURN
END
AND USE IT
select *from dbo.fnSplitString('Querying SQL Server','')
if anyone wants to get only one part of the seperatured text can use this
select * from fromSplitStringSep('Word1 wordr2 word3',' ')
CREATE function [dbo].[SplitStringSep]
(
#str nvarchar(4000),
#separator char(1)
)
returns table
AS
return (
with tokens(p, a, b) AS (
select
1,
1,
charindex(#separator, #str)
union all
select
p + 1,
b + 1,
charindex(#separator, #str, b + 1)
from tokens
where b > 0
)
select
p-1 zeroBasedOccurance,
substring(
#str,
a,
case when b > 0 then b-a ELSE 4000 end)
AS s
from tokens
)
I devoloped this,
declare #x nvarchar(Max) = 'ali.veli.deli.';
declare #item nvarchar(Max);
declare #splitter char='.';
while CHARINDEX(#splitter,#x) != 0
begin
set #item = LEFT(#x,CHARINDEX(#splitter,#x))
set #x = RIGHT(#x,len(#x)-len(#item) )
select #item as item, #x as x;
end
the only attention you should is dot '.' that end of the #x is always should be there.
building on #NothingsImpossible solution, or, rather, comment on the most voted answer (just below the accepted one), i found the following quick-and-dirty solution fulfill my own needs - it has a benefit of being solely within SQL domain.
given a string "first;second;third;fourth;fifth", say, I want to get the third token. this works only if we know how many tokens the string is going to have - in this case it's 5. so my way of action is to chop the last two tokens away (inner query), and then to chop the first two tokens away (outer query)
i know that this is ugly and covers the specific conditions i was in, but am posting it just in case somebody finds it useful. cheers
select
REVERSE(
SUBSTRING(
reverse_substring,
0,
CHARINDEX(';', reverse_substring)
)
)
from
(
select
msg,
SUBSTRING(
REVERSE(msg),
CHARINDEX(
';',
REVERSE(msg),
CHARINDEX(
';',
REVERSE(msg)
)+1
)+1,
1000
) reverse_substring
from
(
select 'first;second;third;fourth;fifth' msg
) a
) b
declare #strng varchar(max)='hello john smith'
select (
substring(
#strng,
charindex(' ', #strng) + 1,
(
(charindex(' ', #strng, charindex(' ', #strng) + 1))
- charindex(' ',#strng)
)
))
I have an string
declare #s varchar(100),
set #s='aaaa,bbbb,cccc'
declare #first varchar(100),
declare #second varchar(100),
declare #third varchar(100)
Now I need to split these strings into three variable holding there values like this
#first=aaaa
#second=bbbb
#third=cccc
If I am using the split function then I get the output like this
aaaa
bbbb
cccc
Is there any better way we can achieve this result? Any help would be appreciated.
I want this result into a variable because I need it for further processing
Thanks
Here's a quick and dirty hack, assuming that your input strings always follow that format.
DECLARE #s VARCHAR(100)
SET #s = 'aaaa,bbbb,cccc'
DECLARE #first VARCHAR(100)
DECLARE #second VARCHAR(100)
DECLARE #third VARCHAR(100)
SET #s = '<row>' + REPLACE(#s, ',', '</row><row>') + '</row>'
SELECT #first = CONVERT(XML, #s).value('(/row)[1]', 'varchar(100)')
, #second = CONVERT(XML, #s).value('(/row)[2]', 'varchar(100)')
, #third = CONVERT(XML, #s).value('(/row)[3]', 'varchar(100)')
SELECT #first
, #second
, #third
If the split function you're using returns the three values in a table, one option might be to use that output to insert into a table variable with an integer identity, and then pick which variable pairs with which identity value:
DECLARE #first VARCHAR(100)
DECLARE #second VARCHAR(100)
DECLARE #third VARCHAR(100)
declare #split_output table (block varchar(100) )
declare #split_identity table (in_order int identity(1,1), block varchar(100))
/* recreate output of split fn */
insert into #split_output
select 'aaaa'
union
select 'bbbb'
union
select 'cccc'
select * from #split_output
/* push split output into table with identity column */
insert into #split_identity (block)
select block from #split_output
select * from #split_identity
/* Use identity value to pick which row of the table goes with which variable */
set #first = (select block from #split_identity where in_order = 1)
set #second = (select block from #split_identity where in_order = 2)
set #third = (select block from #split_identity where in_order = 3)
select #first, #second, #third
declare #s varchar(100)
set #s='aaaa,bbbb,cccc'
declare #first varchar(100)
declare #second varchar(100)
declare #third varchar(100)
select #first = left(#s, T.C1-1),
#second = substring(#s, T.C1+1, T.C2-T.C1),
#third = stuff(#s, 1, T.C2+1, '')
from (select charindex(',', #s),
len(#s)-charindex(',', reverse(#s))) as T(C1, C2)
String Functions (Transact-SQL)
How could you remove all characters that are not alphabetic from a string?
What about non-alphanumeric?
Does this have to be a custom function or are there also more generalizable solutions?
Try this function:
Create Function [dbo].[RemoveNonAlphaCharacters](#Temp VarChar(1000))
Returns VarChar(1000)
AS
Begin
Declare #KeepValues as varchar(50)
Set #KeepValues = '%[^a-z]%'
While PatIndex(#KeepValues, #Temp) > 0
Set #Temp = Stuff(#Temp, PatIndex(#KeepValues, #Temp), 1, '')
Return #Temp
End
Call it like this:
Select dbo.RemoveNonAlphaCharacters('abc1234def5678ghi90jkl')
Once you understand the code, you should see that it is relatively simple to change it to remove other characters, too. You could even make this dynamic enough to pass in your search pattern.
Parameterized version of George Mastros' awesome answer:
CREATE FUNCTION [dbo].[fn_StripCharacters]
(
#String NVARCHAR(MAX),
#MatchExpression VARCHAR(255)
)
RETURNS NVARCHAR(MAX)
AS
BEGIN
SET #MatchExpression = '%['+#MatchExpression+']%'
WHILE PatIndex(#MatchExpression, #String) > 0
SET #String = Stuff(#String, PatIndex(#MatchExpression, #String), 1, '')
RETURN #String
END
Alphabetic only:
SELECT dbo.fn_StripCharacters('a1!s2#d3#f4$', '^a-z')
Numeric only:
SELECT dbo.fn_StripCharacters('a1!s2#d3#f4$', '^0-9')
Alphanumeric only:
SELECT dbo.fn_StripCharacters('a1!s2#d3#f4$', '^a-z0-9')
Non-alphanumeric:
SELECT dbo.fn_StripCharacters('a1!s2#d3#f4$', 'a-z0-9')
Believe it or not, in my system this ugly function performs better than G Mastros elegant one.
CREATE FUNCTION dbo.RemoveSpecialChar (#s VARCHAR(256))
RETURNS VARCHAR(256)
WITH SCHEMABINDING
BEGIN
IF #s IS NULL
RETURN NULL
DECLARE #s2 VARCHAR(256) = '',
#l INT = LEN(#s),
#p INT = 1
WHILE #p <= #l
BEGIN
DECLARE #c INT
SET #c = ASCII(SUBSTRING(#s, #p, 1))
IF #c BETWEEN 48 AND 57
OR #c BETWEEN 65 AND 90
OR #c BETWEEN 97 AND 122
SET #s2 = #s2 + CHAR(#c)
SET #p = #p + 1
END
IF LEN(#s2) = 0
RETURN NULL
RETURN #s2
I knew that SQL was bad at string manipulation, but I didn't think it would be this difficult. Here's a simple function to strip out all the numbers from a string. There would be better ways to do this, but this is a start.
CREATE FUNCTION dbo.AlphaOnly (
#String varchar(100)
)
RETURNS varchar(100)
AS BEGIN
RETURN (
REPLACE(
REPLACE(
REPLACE(
REPLACE(
REPLACE(
REPLACE(
REPLACE(
REPLACE(
REPLACE(
REPLACE(
#String,
'9', ''),
'8', ''),
'7', ''),
'6', ''),
'5', ''),
'4', ''),
'3', ''),
'2', ''),
'1', ''),
'0', '')
)
END
GO
-- ==================
DECLARE #t TABLE (
ColID int,
ColString varchar(50)
)
INSERT INTO #t VALUES (1, 'abc1234567890')
SELECT ColID, ColString, dbo.AlphaOnly(ColString)
FROM #t
Output
ColID ColString
----- ------------- ---
1 abc1234567890 abc
Round 2 - Data-Driven Blacklist
-- ============================================
-- Create a table of blacklist characters
-- ============================================
IF EXISTS (SELECT * FROM sys.tables WHERE [object_id] = OBJECT_ID('dbo.CharacterBlacklist'))
DROP TABLE dbo.CharacterBlacklist
GO
CREATE TABLE dbo.CharacterBlacklist (
CharID int IDENTITY,
DisallowedCharacter nchar(1) NOT NULL
)
GO
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'0')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'1')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'2')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'3')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'4')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'5')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'6')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'7')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'8')
INSERT INTO dbo.CharacterBlacklist (DisallowedCharacter) VALUES (N'9')
GO
-- ====================================
IF EXISTS (SELECT * FROM sys.objects WHERE [object_id] = OBJECT_ID('dbo.StripBlacklistCharacters'))
DROP FUNCTION dbo.StripBlacklistCharacters
GO
CREATE FUNCTION dbo.StripBlacklistCharacters (
#String nvarchar(100)
)
RETURNS varchar(100)
AS BEGIN
DECLARE #blacklistCt int
DECLARE #ct int
DECLARE #c nchar(1)
SELECT #blacklistCt = COUNT(*) FROM dbo.CharacterBlacklist
SET #ct = 0
WHILE #ct < #blacklistCt BEGIN
SET #ct = #ct + 1
SELECT #String = REPLACE(#String, DisallowedCharacter, N'')
FROM dbo.CharacterBlacklist
WHERE CharID = #ct
END
RETURN (#String)
END
GO
-- ====================================
DECLARE #s nvarchar(24)
SET #s = N'abc1234def5678ghi90jkl'
SELECT
#s AS OriginalString,
dbo.StripBlacklistCharacters(#s) AS ResultString
Output
OriginalString ResultString
------------------------ ------------
abc1234def5678ghi90jkl abcdefghijkl
My challenge to readers: Can you make this more efficient? What about using recursion?
Here's a solution that doesn't require creating a function or listing all instances of characters to replace. It uses a recursive WITH statement in combination with a PATINDEX to find unwanted chars. It will replace all unwanted chars in a column - up to 100 unique bad characters contained in any given string. (E.G. "ABC123DEF234" would contain 4 bad characters 1, 2, 3 and 4) The 100 limit is the maximum number of recursions allowed in a WITH statement, but this doesn't impose a limit on the number of rows to process, which is only limited by the memory available.
If you don't want DISTINCT results, you can remove the two options from the code.
-- Create some test data:
SELECT * INTO #testData
FROM (VALUES ('ABC DEF,K.l(p)'),('123H,J,234'),('ABCD EFG')) as t(TXT)
-- Actual query:
-- Remove non-alpha chars: '%[^A-Z]%'
-- Remove non-alphanumeric chars: '%[^A-Z0-9]%'
DECLARE #BadCharacterPattern VARCHAR(250) = '%[^A-Z]%';
WITH recurMain as (
SELECT DISTINCT CAST(TXT AS VARCHAR(250)) AS TXT, PATINDEX(#BadCharacterPattern, TXT) AS BadCharIndex
FROM #testData
UNION ALL
SELECT CAST(TXT AS VARCHAR(250)) AS TXT, PATINDEX(#BadCharacterPattern, TXT) AS BadCharIndex
FROM (
SELECT
CASE WHEN BadCharIndex > 0
THEN REPLACE(TXT, SUBSTRING(TXT, BadCharIndex, 1), '')
ELSE TXT
END AS TXT
FROM recurMain
WHERE BadCharIndex > 0
) badCharFinder
)
SELECT DISTINCT TXT
FROM recurMain
WHERE BadCharIndex = 0;
Having looked at all the given solutions I thought that there has to be a pure SQL method that does not require a function or a CTE / XML query, and doesn't involve difficult to maintain nested REPLACE statements. Here is my solution:
SELECT
x
,CASE WHEN a NOT LIKE '%' + SUBSTRING(x, 1, 1) + '%' THEN '' ELSE SUBSTRING(x, 1, 1) END
+ CASE WHEN a NOT LIKE '%' + SUBSTRING(x, 2, 1) + '%' THEN '' ELSE SUBSTRING(x, 2, 1) END
+ CASE WHEN a NOT LIKE '%' + SUBSTRING(x, 3, 1) + '%' THEN '' ELSE SUBSTRING(x, 3, 1) END
+ CASE WHEN a NOT LIKE '%' + SUBSTRING(x, 4, 1) + '%' THEN '' ELSE SUBSTRING(x, 4, 1) END
+ CASE WHEN a NOT LIKE '%' + SUBSTRING(x, 5, 1) + '%' THEN '' ELSE SUBSTRING(x, 5, 1) END
+ CASE WHEN a NOT LIKE '%' + SUBSTRING(x, 6, 1) + '%' THEN '' ELSE SUBSTRING(x, 6, 1) END
-- Keep adding rows until you reach the column size
AS stripped_column
FROM (SELECT
column_to_strip AS x
,'ABCDEFGHIJKLMNOPQRSTUVWXYZ' AS a
FROM my_table) a
The advantage of doing it this way is that the valid characters are contained in the one string in the sub query making easy to reconfigure for a different set of characters.
The downside is that you have to add a row of SQL for each character up to the size of your column. To make that task easier I just used the Powershell script below, this example if for a VARCHAR(64):
1..64 | % {
" + CASE WHEN a NOT LIKE '%' + SUBSTRING(x, {0}, 1) + '%' THEN '' ELSE SUBSTRING(x, {0}, 1) END" -f $_
} | clip.exe
If you are like me and don't have access to just add functions to your production data but still want to perform this kind of filtering, here's a pure SQL solution using a PIVOT table to put the filtered pieces back together again.
N.B. I hardcoded the table up to 40 characters, you'll have to add more if you have longer strings to filter.
SET CONCAT_NULL_YIELDS_NULL OFF;
with
ToBeScrubbed
as (
select 1 as id, '*SOME 222# !#* #* BOGUS !#*&! DATA' as ColumnToScrub
),
Scrubbed as (
select
P.Number as ValueOrder,
isnull ( substring ( t.ColumnToScrub , number , 1 ) , '' ) as ScrubbedValue,
t.id
from
ToBeScrubbed t
left join master..spt_values P
on P.number between 1 and len(t.ColumnToScrub)
and type ='P'
where
PatIndex('%[^a-z]%', substring(t.ColumnToScrub,P.number,1) ) = 0
)
SELECT
id,
[1]+ [2]+ [3]+ [4]+ [5]+ [6]+ [7]+ [8] +[9] +[10]
+ [11]+ [12]+ [13]+ [14]+ [15]+ [16]+ [17]+ [18] +[19] +[20]
+ [21]+ [22]+ [23]+ [24]+ [25]+ [26]+ [27]+ [28] +[29] +[30]
+ [31]+ [32]+ [33]+ [34]+ [35]+ [36]+ [37]+ [38] +[39] +[40] as ScrubbedData
FROM (
select
*
from
Scrubbed
)
src
PIVOT (
MAX(ScrubbedValue) FOR ValueOrder IN (
[1], [2], [3], [4], [5], [6], [7], [8], [9], [10],
[11], [12], [13], [14], [15], [16], [17], [18], [19], [20],
[21], [22], [23], [24], [25], [26], [27], [28], [29], [30],
[31], [32], [33], [34], [35], [36], [37], [38], [39], [40]
)
) pvt
Here is another way to remove non-alphabetic characters using an iTVF. First, you need a pattern-based string splitter. Here is one taken from Dwain Camp's article:
-- PatternSplitCM will split a string based on a pattern of the form
-- supported by LIKE and PATINDEX
--
-- Created by: Chris Morris 12-Oct-2012
CREATE FUNCTION [dbo].[PatternSplitCM]
(
#List VARCHAR(8000) = NULL
,#Pattern VARCHAR(50)
) RETURNS TABLE WITH SCHEMABINDING
AS
RETURN
WITH numbers AS (
SELECT TOP(ISNULL(DATALENGTH(#List), 0))
n = ROW_NUMBER() OVER(ORDER BY (SELECT NULL))
FROM
(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) d (n),
(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) e (n),
(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) f (n),
(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) g (n)
)
SELECT
ItemNumber = ROW_NUMBER() OVER(ORDER BY MIN(n)),
Item = SUBSTRING(#List,MIN(n),1+MAX(n)-MIN(n)),
[Matched]
FROM (
SELECT n, y.[Matched], Grouper = n - ROW_NUMBER() OVER(ORDER BY y.[Matched],n)
FROM numbers
CROSS APPLY (
SELECT [Matched] = CASE WHEN SUBSTRING(#List,n,1) LIKE #Pattern THEN 1 ELSE 0 END
) y
) d
GROUP BY [Matched], Grouper
Now that you have a pattern-based splitter, you need to split the strings that match the pattern:
[a-z]
and then concatenate them back to get the desired result:
SELECT *
FROM tbl t
CROSS APPLY(
SELECT Item + ''
FROM dbo.PatternSplitCM(t.str, '[a-z]')
WHERE Matched = 1
ORDER BY ItemNumber
FOR XML PATH('')
) x (a)
SAMPLE
Result:
| Id | str | a |
|----|------------------|----------------|
| 1 | test“te d'abc | testtedabc |
| 2 | anr¤a | anra |
| 3 | gs-re-C“te d'ab | gsreCtedab |
| 4 | M‚fe, DF | MfeDF |
| 5 | R™temd | Rtemd |
| 6 | ™jad”ji | jadji |
| 7 | Cje y ret¢n | Cjeyretn |
| 8 | J™kl™balu | Jklbalu |
| 9 | le“ne-iokd | leneiokd |
| 10 | liode-Pyr‚n‚ie | liodePyrnie |
| 11 | V„s G”ta | VsGta |
| 12 | Sƒo Paulo | SoPaulo |
| 13 | vAstra gAtaland | vAstragAtaland |
| 14 | ¥uble / Bio-Bio | ubleBioBio |
| 15 | U“pl™n/ds VAsb-y | UplndsVAsby |
This solution, inspired by Mr. Allen's solution, requires a Numbers table of integers (which you should have on hand if you want to do serious query operations with good performance). It does not require a CTE. You can change the NOT IN (...) expression to exclude specific characters, or change it to an IN (...) OR LIKE expression to retain only certain characters.
SELECT (
SELECT SUBSTRING([YourString], N, 1)
FROM dbo.Numbers
WHERE N > 0 AND N <= CONVERT(INT, LEN([YourString]))
AND SUBSTRING([YourString], N, 1) NOT IN ('(',')',',','.')
FOR XML PATH('')
) AS [YourStringTransformed]
FROM ...
From performance perspective I'd use Inline Function:
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE FUNCTION [dbo].[udf_RemoveNumericCharsFromString]
(
#List NVARCHAR(4000)
)
RETURNS TABLE
AS RETURN
WITH GetNums AS (
SELECT TOP(ISNULL(DATALENGTH(#List), 0))
n = ROW_NUMBER() OVER(ORDER BY (SELECT NULL))
FROM
(VALUES (0),(0),(0),(0)) d (n),
(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) e (n),
(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) f (n),
(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) g (n)
)
SELECT StrOut = ''+
(SELECT Chr
FROM GetNums
CROSS APPLY (SELECT SUBSTRING(#List , n,1)) X(Chr)
WHERE Chr LIKE '%[^0-9]%'
ORDER BY N
FOR XML PATH (''),TYPE).value('.','NVARCHAR(MAX)')
/*How to Use
SELECT StrOut FROM dbo.udf_RemoveNumericCharsFromString ('vv45--9gut')
Result: vv--gut
*/
Here's another recursive CTE solution, based on #Gerhard Weiss's answer here. You should be able to copy and paste the whole code block into SSMS and play with it there. The results include a few extra columns to help us understand what's going on. It took me a while until I understood all that's going on with both PATINDEX (RegEx) and the recursive CTE.
DECLARE #DefineBadCharPattern varchar(30)
SET #DefineBadCharPattern = '%[^A-z]%' --Means anything NOT between A and z characters (according to ascii char value) is "bad"
SET #DefineBadCharPattern = '%[^a-z0-9]%' --Means anything NOT between a and z characters or numbers 0 through 9 (according to ascii char value) are "bad"
SET #DefineBadCharPattern = '%[^ -~]%' --Means anything NOT between space and ~ characters (all non-printable characters) is "bad"
--Change #ReplaceBadCharWith to '' to strip "bad" characters from string
--Change to some character if you want to 'see' what's being replaced. NOTE: It must be allowed accoring to #DefineBadCharPattern above
DECLARE #ReplaceBadCharWith varchar(1) = '#' --Change this to whatever you want to replace non-printable chars with
IF patindex(#DefineBadCharPattern COLLATE Latin1_General_BIN, #ReplaceBadCharWith) > 0
BEGIN
RAISERROR('#ReplaceBadCharWith value (%s) must be a character allowed by PATINDEX pattern of %s',16,1,#ReplaceBadCharWith, #DefineBadCharPattern)
RETURN
END
--A table of values to play with:
DECLARE #temp TABLE (OriginalString varchar(100))
INSERT #temp SELECT ' 1hello' + char(13) + char(10) + 'there' + char(30) + char(9) + char(13) + char(10)
INSERT #temp SELECT '2hello' + char(30) + 'there' + char(30)
INSERT #temp SELECT ' 3hello there'
INSERT #temp SELECT ' tab' + char(9) + ' character'
INSERT #temp SELECT 'good bye'
--Let the magic begin:
;WITH recurse AS (
select
OriginalString,
OriginalString as CleanString,
patindex(#DefineBadCharPattern COLLATE Latin1_General_BIN, OriginalString) as [Position],
substring(OriginalString,patindex(#DefineBadCharPattern COLLATE Latin1_General_BIN, OriginalString),1) as [InvalidCharacter],
ascii(substring(OriginalString,patindex(#DefineBadCharPattern COLLATE Latin1_General_BIN, OriginalString),1)) as [ASCIICode]
from #temp
UNION ALL
select
OriginalString,
CONVERT(varchar(100),REPLACE(CleanString,InvalidCharacter,#ReplaceBadCharWith)),
patindex(#DefineBadCharPattern COLLATE Latin1_General_BIN,CleanString) as [Position],
substring(CleanString,patindex(#DefineBadCharPattern COLLATE Latin1_General_BIN,CleanString),1),
ascii(substring(CleanString,patindex(#DefineBadCharPattern COLLATE Latin1_General_BIN,CleanString),1))
from recurse
where patindex(#DefineBadCharPattern COLLATE Latin1_General_BIN,CleanString) > 0
)
SELECT * FROM recurse
--optionally comment out this last WHERE clause to see more of what the recursion is doing:
WHERE patindex(#DefineBadCharPattern COLLATE Latin1_General_BIN,CleanString) = 0
Another possibe option for SQL Server 2017+, without loops and/or recursion, is a string-based approach using TRANSLATE() and REPLACE().
T-SQL statement:
DECLARE #pattern varchar(52) = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
SELECT
v.[Text],
REPLACE(
TRANSLATE(
v.[Text],
REPLACE(TRANSLATE(v.[Text], #pattern, REPLICATE('a', LEN(#pattern))), 'a', ''),
REPLICATE('0', LEN(REPLACE(TRANSLATE(v.[Text], #pattern, REPLICATE('a', LEN(#pattern))), 'a', '')))
),
'0',
''
) AS AlphabeticCharacters
FROM (VALUES
('abc1234def5678ghi90jkl##$&'),
('1234567890'),
('JAHDBESBN%*#*#*($E*sd55bn')
) v ([Text])
or as a function:
CREATE FUNCTION dbo.RemoveNonAlphabeticCharacters (#Text varchar(1000))
RETURNS varchar(1000)
AS BEGIN
DECLARE #pattern varchar(52) = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ'
SET #text = REPLACE(
TRANSLATE(
#Text,
REPLACE(TRANSLATE(#Text, #pattern, REPLICATE('a', LEN(#pattern))), 'a', ''),
REPLICATE('0', LEN(REPLACE(TRANSLATE(#Text, #pattern, REPLICATE('a', LEN(#pattern))), 'a', '')))
),
'0',
''
)
RETURN #Text
END
I put this in both places where PatIndex is called.
PatIndex('%[^A-Za-z0-9]%', #Temp)
for the custom function above RemoveNonAlphaCharacters and renamed it RemoveNonAlphaNumericCharacters
--First create one function
CREATE FUNCTION [dbo].[GetNumericonly]
(#strAlphaNumeric VARCHAR(256))
RETURNS VARCHAR(256)
AS
BEGIN
DECLARE #intAlpha INT
SET #intAlpha = PATINDEX('%[^0-9]%', #strAlphaNumeric)
BEGIN
WHILE #intAlpha > 0
BEGIN
SET #strAlphaNumeric = STUFF(#strAlphaNumeric, #intAlpha, 1, '' )
SET #intAlpha = PATINDEX('%[^0-9]%', #strAlphaNumeric )
END
END
RETURN ISNULL(#strAlphaNumeric,0)
END
Now call this function like
select [dbo].[GetNumericonly]('Abhi12shek23jaiswal')
Its result like
1223
For SQL Server >= 2017...
declare #text varchar(max)
-- create some sample text
select
#text=
'
Lorem #ipsum *&dolor-= sit?! amet, {consectetur } adipiscing\ elit. Vivamus commodo justo metus, sed facilisis ante
congue eget. Proin ac bibendum sem/.
'
-- the characters to be removed
declare #unwanted varchar(max)='''.,!?/<>"[]{}|`~##$%^&*()-+=/\:;'+char(13)+char(10)
-- interim replaced with
declare #replace_with char(1)=' '
-- call the translate function that will change unwanted characters to spaces
-- in this sample
declare #translated varchar(max)
select #translated=TRANSLATE(#text,#unwanted,REPLICATE(#replace_with,len(#unwanted)))
-- In this case, I want to preserve one space
select string_agg(trim(value),' ')
from STRING_SPLIT(#translated,' ')
where trim(value)<>''
-- Result
'Lorem ipsum dolor sit amet consectetur adipiscing elit Vivamus commodo justo metus sed facilisis ante congue eget Proin ac bibendum sem'
Using a CTE generated numbers table to examine each character, then FOR XML to concat to a string of kept values you can...
CREATE FUNCTION [dbo].[PatRemove](
#pattern varchar(50),
#expression varchar(8000)
)
RETURNS varchar(8000)
AS
BEGIN
WITH
d(d) AS (SELECT d FROM (VALUES (0),(1),(2),(3),(4),(5),(6),(7),(8),(9)) digits(d)),
nums(n) AS (SELECT ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) FROM d d1, d d2, d d3, d d4),
chars(c) AS (SELECT SUBSTRING(#expression, n, 1) FROM nums WHERE n <= LEN(#expression))
SELECT
#expression = (SELECT c AS [text()] FROM chars WHERE c NOT LIKE #pattern FOR XML PATH(''));
RETURN #expression;
END
DECLARE #vchVAlue NVARCHAR(255) = 'SWP, Lettering Position 1: 4 Ω, 2: 8 Ω, 3: 16 Ω, 4: , 5: , 6: , Voltage Selector, Solder, 6, Step switch, : w/o fuseholder '
WHILE PATINDEX('%?%' , CAST(#vchVAlue AS VARCHAR(255))) > 0
BEGIN
SELECT #vchVAlue = STUFF(#vchVAlue,PATINDEX('%?%' , CAST(#vchVAlue AS VARCHAR(255))),1,' ')
END
SELECT #vchVAlue
this way didn't work for me as i was trying to keep the Arabic letters i tried to replace the regular expression but also it didn't work. i wrote another method to work on ASCII level as it was my only choice and it worked.
Create function [dbo].[RemoveNonAlphaCharacters] (#s varchar(4000)) returns varchar(4000)
with schemabinding
begin
if #s is null
return null
declare #s2 varchar(4000)
set #s2 = ''
declare #l int
set #l = len(#s)
declare #p int
set #p = 1
while #p <= #l begin
declare #c int
set #c = ascii(substring(#s, #p, 1))
if #c between 48 and 57 or #c between 65 and 90 or #c between 97 and 122 or #c between 165 and 253 or #c between 32 and 33
set #s2 = #s2 + char(#c)
set #p = #p + 1
end
if len(#s2) = 0
return null
return #s2
end
GO
CREATE FUNCTION remove_spc_char(#str VARCHAR(MAX))
RETURNS VARCHAR(MAX)
AS
BEGIN
DECLARE #resp VARCHAR(MAX) = '';
DECLARE #str_val VARCHAR(MAX) = UPPER(#str);
DECLARE #i INTEGER= 1;
DECLARE #v_asc INTEGER;
WHILE #i <= (LEN(#str_val))
BEGIN
SET #v_asc = (ASCII(SUBSTRING(#str_val, #i, 1)))
BEGIN
IF #v_asc in (192,193,194,195,196,65)
begin
SET #v_asc = 65;
SET #resp = concat(#resp, CHAR(#v_asc));
end;
IF #v_asc in (200,201,202,203,233,69)
begin
SET #v_asc = 69;
SET #resp = concat(#resp, CHAR(#v_asc));
end;
IF #v_asc in (204,205,206,207,296,73)
begin
SET #v_asc = 73;
SET #resp = concat(#resp, CHAR(#v_asc));
end;
IF #v_asc in (210,211,212,213,214,79)
begin
SET #v_asc = 79;
SET #resp = concat(#resp, CHAR(#v_asc));
end;
IF #v_asc in (217,218,219,220,85)
begin
SET #v_asc = 85;
SET #resp = concat(#resp, CHAR(#v_asc));
end;
IF #v_asc in (199,231,67)
begin
SET #v_asc = 67;
SET #resp = concat(#resp, CHAR(#v_asc));
end;
IF #v_asc in (209,78)
begin
SET #v_asc = 78;
SET #resp = concat(#resp, CHAR(#v_asc));
end;
IF #v_asc in (924,181,358,216,222,330,272,208,198,42,37,38,34,36,35,
64,33,39,41,40,43,61,95,45,62,60,63,47,176,183,124,166,174,359,248,254,
180,170,186,126,312,331,273,172,178,179,163,162,123,91,93,125,92,167,240,
223,230,171,187,169,185,168)
begin
SET #resp = concat(#resp, '');
end;
ELSE
begin
if #v_asc not in (65,67,69,73,78,79,85)
begin
SET #resp = concat(#resp, CHAR(#v_asc));
end;
end;
END;
SET #i = #i + 1
END;
RETURN #resp;
END;
Though post is a bit old, I would like to say the following.
Issue I had with above solution is that it does not filter out characters like ç, ë, ï, etc. I adapted a function as follows (I only used an 80 varchar string to save memory):
create FUNCTION dbo.udf_Cleanchars (#InputString varchar(80))
RETURNS varchar(80)
AS
BEGIN
declare #return varchar(80) , #length int , #counter int , #cur_char char(1)
SET #return = ''
SET #length = 0
SET #counter = 1
SET #length = LEN(#InputString)
IF #length > 0
BEGIN WHILE #counter <= #length
BEGIN SET #cur_char = SUBSTRING(#InputString, #counter, 1) IF ((ascii(#cur_char) in (32,44,46)) or (ascii(#cur_char) between 48 and 57) or (ascii(#cur_char) between 65 and 90) or (ascii(#cur_char) between 97 and 122))
BEGIN SET #return = #return + #cur_char END
SET #counter = #counter + 1
END END
RETURN #return END
I just found this built into Oracle 10g if that is what you're using. I had to strip all the special characters out for a phone number compare.
regexp_replace(c.phone, '[^0-9]', '')