Related
My code is:
Declare #Users table(Names nvarchar(50) not null, Flag int);
Declare #ValidUsers table(Names nvarchar(50) not null);
Declare #Office int;
Declare #NumberOfRecords int;
Declare #Count int;
Declare #IntCount int;
Declare #Binary AS nvarchar(16);
Declare #bit as nvarchar(1);
Declare #PermissionSub as nvarchar(1);
Declare #Permission as nvarchar(16);
Declare #ShouldContinue as bit;
set #ShouldContinue = 1;
set #Permission = '0001111111111111'; /* going to pass this value */
set #Count = '1';
set #IntCount = '1';
set #Office = '3'; /* going to pass this value */
Insert into #Users
Select
dbUser.usrFullName, udFeeEarnerLicence.purchaseFlag
From
[OMSBB].[dbo].[udFeeEarnerLicence]
Inner Join
[OMSBB].[dbo].[dbUser] ON udFeeEarnerLicence.feeUsrId = dbUser.usrID
Where
dbUser.brId = #Office;
select #NumberOfRecords = COUNT(Flag) from #Users;
DECLARE #Flag AS int;
select #Flag = Flag from #Users;
while(#Count <= #NumberOfRecords)
begin
WITH CTE AS
(
SELECT
Flag, ROW_NUMBER() OVER (ORDER BY Flag) AS RwNr
FROM
#Users
)
SELECT TOP(1) #Flag = Flag -- this TOP(1) is just a fail-safe
FROM CTE
WHERE RwNr = #Count;
WITH A AS
(
SELECT 0 AS ORD, #Flag AS NUMBER, CAST('' AS VARCHAR(20)) AS BITS
UNION ALL
SELECT ORD+1, NUMBER/2, CAST(BITS+CAST(NUMBER%2 AS VARCHAR(20)) AS VARCHAR(20))
FROM A
WHERE NUMBER > 0
)
SELECT #Binary = RIGHT('000000000000000'+ CASE WHEN BITS='' THEN '0' ELSE REVERSE(BITS) END,16)
FROM A
WHERE NUMBER = 0;
WHILE (#IntCount <= 16)
BEGIN
select #bit = SUBSTRING(#Binary, #IntCount, #IntCount + 1);
select #PermissionSub = SUBSTRING(#Permission, #IntCount, #IntCount + 1);
if(#PermissionSub = '1' and #bit != '1') /* if Permission selection is required and user does not have permission*/
begin
SET #ShouldContinue = 0;
break;
end
end
Set #IntCount = 0;
if(#ShouldContinue = 0)
begin
continue;
end
; WITH CTE AS
(
SELECT Names, ROW_NUMBER() OVER (ORDER BY Flag) AS RwNr
FROM #Users
)
INSERT INTO #ValidUsers
SELECT Names
FROM CTE
WHERE RwNr = #Count;
end
select * from #ValidUsers
I will be adapting this code to use it inside of an SSRS report so that's why there are comments on some parameters saying that I will be passing the parameters. This code at its basics finds all users who are from a specified office and have the specified permissions. The permission a user has are set in 5 flags in this example I'm using the purchaseFlag. This value is an int and it calculated by creating an order of permissions and set their bit values to create a string of 0's and 1's and then converting that binary number into a decimal for example '8191' which the binary value of would be '0001111111111111'. I use two while loops in this one to go through the users and the other to go through each of the 16 characters in the permissions. My issue is that this I'm almost certain that this query works but it takes so long to run that I haven't seen the result of it yet and people have recommended that I use sets instead.
SELECT REPLACE('<strong>100</strong><b>.00 GB', '%^(^-?\d*\.{0,1}\d+$)%', '');
I want to replace any markup between two parts of the number with above regex, but it does not seem to work. I'm not sure if it is regex syntax that's wrong because I tried simpler one such as '%[^0-9]%' just to test but it didn't work either. Does anyone know how can I achieve this?
You can use PATINDEX
to find the first index of the pattern (string's) occurrence. Then use STUFF to stuff another string into the pattern(string) matched.
Loop through each row. Replace each illegal characters with what you want. In your case replace non numeric with blank. The inner loop is if you have more than one illegal character in a current cell that of the loop.
DECLARE #counter int
SET #counter = 0
WHILE(#counter < (SELECT MAX(ID_COLUMN) FROM Table))
BEGIN
WHILE 1 = 1
BEGIN
DECLARE #RetVal varchar(50)
SET #RetVal = (SELECT Column = STUFF(Column, PATINDEX('%[^0-9.]%', Column),1, '')
FROM Table
WHERE ID_COLUMN = #counter)
IF(#RetVal IS NOT NULL)
UPDATE Table SET
Column = #RetVal
WHERE ID_COLUMN = #counter
ELSE
break
END
SET #counter = #counter + 1
END
Caution: This is slow though! Having a varchar column may impact. So using LTRIM RTRIM may help a bit. Regardless, it is slow.
Credit goes to this StackOverFlow answer.
EDIT
Credit also goes to #srutzky
Edit (by #Tmdean)
Instead of doing one row at a time, this answer can be adapted to a more set-based solution. It still iterates the max of the number of non-numeric characters in a single row, so it's not ideal, but I think it should be acceptable in most situations.
WHILE 1 = 1 BEGIN
WITH q AS
(SELECT ID_Column, PATINDEX('%[^0-9.]%', Column) AS n
FROM Table)
UPDATE Table
SET Column = STUFF(Column, q.n, 1, '')
FROM q
WHERE Table.ID_Column = q.ID_Column AND q.n != 0;
IF ##ROWCOUNT = 0 BREAK;
END;
You can also improve efficiency quite a lot if you maintain a bit column in the table that indicates whether the field has been scrubbed yet. (NULL represents "Unknown" in my example and should be the column default.)
DECLARE #done bit = 0;
WHILE #done = 0 BEGIN
WITH q AS
(SELECT ID_Column, PATINDEX('%[^0-9.]%', Column) AS n
FROM Table
WHERE COALESCE(Scrubbed_Column, 0) = 0)
UPDATE Table
SET Column = STUFF(Column, q.n, 1, ''),
Scrubbed_Column = 0
FROM q
WHERE Table.ID_Column = q.ID_Column AND q.n != 0;
IF ##ROWCOUNT = 0 SET #done = 1;
-- if Scrubbed_Column is still NULL, then the PATINDEX
-- must have given 0
UPDATE table
SET Scrubbed_Column = CASE
WHEN Scrubbed_Column IS NULL THEN 1
ELSE NULLIF(Scrubbed_Column, 0)
END;
END;
If you don't want to change your schema, this is easy to adapt to store intermediate results in a table valued variable which gets applied to the actual table at the end.
Instead of stripping out the found character by its sole position, using Replace(Column, BadFoundCharacter, '') could be substantially faster. Additionally, instead of just replacing the one bad character found next in each column, this replaces all those found.
WHILE 1 = 1 BEGIN
UPDATE dbo.YourTable
SET Column = Replace(Column, Substring(Column, PatIndex('%[^0-9.-]%', Column), 1), '')
WHERE Column LIKE '%[^0-9.-]%'
If ##RowCount = 0 BREAK;
END;
I am convinced this will work better than the accepted answer, if only because it does fewer operations. There are other ways that might also be faster, but I don't have time to explore those right now.
In a general sense, SQL Server does not support regular expressions and you cannot use them in the native T-SQL code.
You could write a CLR function to do that. See here, for example.
For those looking for a performant and easy solution and are willing to enable CLR:
CREATE database TestSQLFunctions
go
use TestSQLFunctions
go
ALTER database TestSQLFunctions set trustworthy on
EXEC sp_configure 'clr enabled', 1
RECONFIGURE WITH OVERRIDE
go
CREATE ASSEMBLY [SQLFunctions]
AUTHORIZATION [dbo]
FROM 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
WITH PERMISSION_SET = SAFE
go
CREATE FUNCTION RegexReplace(
#input nvarchar(max),
#pattern nvarchar(max),
#replacement nvarchar(max)
) RETURNS nvarchar (max)
AS EXTERNAL NAME SQLFunctions.[SQLFunctions.Regex].Replace;
go
-- outputs This is a test
SELECT dbo.RegexReplace('This is a test 12345','[0-9]','')
Content of the DLL:
I stumbled across this post looking for something else but thought I'd mention a solution I use which is far more efficient - and really should be the default implementation of any function when used with a set based query - which is to use a cross applied table function. Seems the topic is still active so hopefully this is useful to someone.
Example runtime on a few of the answers so far based on running recursive set based queries or scalar function, based on 1m rows test set removing the chars from a random newid, ranges from 34s to 2m05s for the WHILE loop examples and from 1m3s to {forever} for the function examples.
Using a table function with cross apply achieves the same goal in 10s. You may need to adjust it to suit your needs such as the max length it handles.
Function:
CREATE FUNCTION [dbo].[RemoveChars](#InputUnit VARCHAR(40))
RETURNS TABLE
AS
RETURN
(
WITH Numbers_prep(Number) 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
)
,Numbers(Number) AS
(
SELECT TOP (ISNULL(LEN(#InputUnit),0))
row_number() OVER (ORDER BY (SELECT NULL))
FROM Numbers_prep a
CROSS JOIN Numbers_prep b
)
SELECT
OutputUnit
FROM
(
SELECT
substring(#InputUnit,Number,1)
FROM Numbers
WHERE substring(#InputUnit,Number,1) like '%[0-9]%'
ORDER BY Number
FOR XML PATH('')
) Sub(OutputUnit)
)
Usage:
UPDATE t
SET column = o.OutputUnit
FROM ##t t
CROSS APPLY [dbo].[RemoveChars](t.column) o
Here is a function I wrote to accomplish this based off of the previous answers.
CREATE FUNCTION dbo.RepetitiveReplace
(
#P_String VARCHAR(MAX),
#P_Pattern VARCHAR(MAX),
#P_ReplaceString VARCHAR(MAX),
#P_ReplaceLength INT = 1
)
RETURNS VARCHAR(MAX)
BEGIN
DECLARE #Index INT;
-- Get starting point of pattern
SET #Index = PATINDEX(#P_Pattern, #P_String);
while #Index > 0
begin
--replace matching charactger at index
SET #P_String = STUFF(#P_String, PATINDEX(#P_Pattern, #P_String), #P_ReplaceLength, #P_ReplaceString);
SET #Index = PATINDEX(#P_Pattern, #P_String);
end
RETURN #P_String;
END;
[Gist][1]
[1]: https://gist.github.com/jkdba/ca13fe8f2a9855c4bdbfd0a5d3dfcda2
Edit:
Originally I had a recursive function here which does not play well with sql server as it has a 32 nesting level limit which would result in an error like the below any time you attempt to make 32+ replacements with the function. Instead of trying to make a server level change to allow more nesting (which could be dangerous like allow never ending loops) switching to a while loop makes a lot more sense.
Maximum stored procedure, function, trigger, or view nesting level exceeded (limit 32).
Wrapping the solution inside a SQL function could be useful if you want to reuse it.
I'm even doing it at the cell level, that's why I'm putting this as a different answer:
CREATE FUNCTION [dbo].[fnReplaceInvalidChars] (#string VARCHAR(300))
RETURNS VARCHAR(300)
BEGIN
DECLARE #str VARCHAR(300) = #string;
DECLARE #Pattern VARCHAR (20) = '%[^a-zA-Z0-9]%';
DECLARE #Len INT;
SELECT #Len = LEN(#String);
WHILE #Len > 0
BEGIN
SET #Len = #Len - 1;
IF (PATINDEX(#Pattern,#str) > 0)
BEGIN
SELECT #str = STUFF(#str, PATINDEX(#Pattern,#str),1,'');
END
ELSE
BEGIN
BREAK;
END
END
RETURN #str
END
A more speedy approach for large strings would look something like this:
CREATE FUNCTION [dbo].[fnReplaceInvalidChars] (#string VARCHAR(MAX))
RETURNS VARCHAR(MAX)
BEGIN
DECLARE #str VARCHAR(MAX) = #string;
DECLARE #Pattern VARCHAR (MAX) = '%[^a-zA-Z0-9]%';
WHILE PATINDEX(#Pattern,#str) > 0
BEGIN
SELECT #str = STUFF(#str, PATINDEX(#Pattern,#str),1,'');
END
RETURN #str
END
I've created this function to clean up a string that contained non numeric characters in a time field. The time contained question marks when they did not added the minutes, something like this 20:??. Function loops through each character and replaces the ? with a 0 :
CREATE FUNCTION [dbo].[CleanTime]
(
-- Add the parameters for the function here
#intime nvarchar(10)
)
RETURNS nvarchar(5)
AS
BEGIN
-- Declare the return variable here
DECLARE #ResultVar nvarchar(5)
DECLARE #char char(1)
-- Add the T-SQL statements to compute the return value here
DECLARE #i int = 1
WHILE #i <= LEN(#intime)
BEGIN
SELECT #char = CASE WHEN substring(#intime,#i,1) like '%[0-9:]%' THEN substring(#intime,#i,1) ELSE '0' END
SELECT #ResultVar = concat(#ResultVar,#char)
set #i = #i + 1
END;
-- Return the result of the function
RETURN #ResultVar
END
I think this solution is faster and simple. I use always CTE/recursive because WHILE is so slow on SQL Server.
I use it in projects I work with and large databases.
/*
Function: dbo.kSql_ReplaceRegExp
Create Date: 20.02.2021
Author: Karcan Ozbal
Description: The given string value will be replaced according to the given regexp/pattern.
Parameter(s): #Value : Value/Text to REPLACE.
#RegExp : The regexp/pattern to be used for REPLACE operation.
Usage: select dbo.kSql_ReplaceRegExp('2T3EST5','%[0-9]%')
Output: 'TEST'
*/
ALTER FUNCTION [dbo].[kSql_ReplaceRegExp](
#Value nvarchar(max),
#RegExp nvarchar(50)
)
RETURNS nvarchar(max)
AS
BEGIN
DECLARE #Result nvarchar(max)
;WITH CTE AS (
SELECT NUM = 1, VALUE = #Value, IDX = PATINDEX(#RegExp, #Value)
UNION ALL
SELECT NUM + 1, VALUE = REPLACE(VALUE, SUBSTRING(VALUE,IDX,1),''), IDX = PATINDEX(#RegExp, REPLACE(VALUE, SUBSTRING(VALUE,IDX,1),''))
FROM CTE
WHERE IDX > 0
)
SELECT TOP(1) #Result = VALUE
FROM CTE
ORDER BY NUM DESC
OPTION (maxrecursion 0)
RETURN #Result
END
If you are doing this just for a parameter coming into a Stored Procedure, you can use the following:
declare #badIndex int
set #badIndex = PatIndex('%[^0-9]%', #Param)
while #badIndex > 0
set #Param = Replace(#Param, Substring(#Param, #badIndex, 1), '')
set #badIndex = PatIndex('%[^0-9]%', #Param)
I thought this was clearer:
ALTER FUNCTION [dbo].[func_ReplaceChars](
#Value nvarchar(max),
#Chars nvarchar(50)
)
RETURNS nvarchar(max)
AS
BEGIN
DECLARE #cLen int = len(#Chars);
DECLARE #curChar int = 0;
WHILE #curChar<#cLen
BEGIN
set #Value = replace(#Value,substring(#Chars,#curChar,1),'');
set #curChar = #curChar + 1;
END;
RETURN #Value
END
I'm using this code similar to several codes above:
DROP FUNCTION [dbo].[fnCleanString]
GO
CREATE FUNCTION [dbo].[fnCleanString] (#input VARCHAR(max), #Pattern
VARCHAR (20))
RETURNS VARCHAR(max)
BEGIN
DECLARE #str VARCHAR(max) = #input;
DECLARE #Len INT;
DECLARE #INDEX INT;
SELECT #Len = LEN(#input);
WHILE #Len > 0
BEGIN
SET #INDEX = PATINDEX(#Pattern,#str);
IF (#INDEX > 0)
BEGIN
SET #str=REPLACE(#str,SUBSTRING(#str,#INDEX, 1), '');
END
ELSE
BEGIN
BREAK;
END
END
RETURN #str
END
You can use it like this:
SELECT CleanName = dbo.[fnCleanString](Name, '%[0-9]%') from YourTable
I think a simpler and faster approach is iterate by each character of the alphabet:
DECLARE #i int
SET #i = 0
WHILE(#i < 256)
BEGIN
IF char(#i) NOT IN ('0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '.')
UPDATE Table SET Column = replace(Column, char(#i), '')
SET #i = #i + 1
END
SELECT REPLACE('<strong>100</strong><b>.00 GB', '%^(^-?\d*\.{0,1}\d+$)%', '');
I want to replace any markup between two parts of the number with above regex, but it does not seem to work. I'm not sure if it is regex syntax that's wrong because I tried simpler one such as '%[^0-9]%' just to test but it didn't work either. Does anyone know how can I achieve this?
You can use PATINDEX
to find the first index of the pattern (string's) occurrence. Then use STUFF to stuff another string into the pattern(string) matched.
Loop through each row. Replace each illegal characters with what you want. In your case replace non numeric with blank. The inner loop is if you have more than one illegal character in a current cell that of the loop.
DECLARE #counter int
SET #counter = 0
WHILE(#counter < (SELECT MAX(ID_COLUMN) FROM Table))
BEGIN
WHILE 1 = 1
BEGIN
DECLARE #RetVal varchar(50)
SET #RetVal = (SELECT Column = STUFF(Column, PATINDEX('%[^0-9.]%', Column),1, '')
FROM Table
WHERE ID_COLUMN = #counter)
IF(#RetVal IS NOT NULL)
UPDATE Table SET
Column = #RetVal
WHERE ID_COLUMN = #counter
ELSE
break
END
SET #counter = #counter + 1
END
Caution: This is slow though! Having a varchar column may impact. So using LTRIM RTRIM may help a bit. Regardless, it is slow.
Credit goes to this StackOverFlow answer.
EDIT
Credit also goes to #srutzky
Edit (by #Tmdean)
Instead of doing one row at a time, this answer can be adapted to a more set-based solution. It still iterates the max of the number of non-numeric characters in a single row, so it's not ideal, but I think it should be acceptable in most situations.
WHILE 1 = 1 BEGIN
WITH q AS
(SELECT ID_Column, PATINDEX('%[^0-9.]%', Column) AS n
FROM Table)
UPDATE Table
SET Column = STUFF(Column, q.n, 1, '')
FROM q
WHERE Table.ID_Column = q.ID_Column AND q.n != 0;
IF ##ROWCOUNT = 0 BREAK;
END;
You can also improve efficiency quite a lot if you maintain a bit column in the table that indicates whether the field has been scrubbed yet. (NULL represents "Unknown" in my example and should be the column default.)
DECLARE #done bit = 0;
WHILE #done = 0 BEGIN
WITH q AS
(SELECT ID_Column, PATINDEX('%[^0-9.]%', Column) AS n
FROM Table
WHERE COALESCE(Scrubbed_Column, 0) = 0)
UPDATE Table
SET Column = STUFF(Column, q.n, 1, ''),
Scrubbed_Column = 0
FROM q
WHERE Table.ID_Column = q.ID_Column AND q.n != 0;
IF ##ROWCOUNT = 0 SET #done = 1;
-- if Scrubbed_Column is still NULL, then the PATINDEX
-- must have given 0
UPDATE table
SET Scrubbed_Column = CASE
WHEN Scrubbed_Column IS NULL THEN 1
ELSE NULLIF(Scrubbed_Column, 0)
END;
END;
If you don't want to change your schema, this is easy to adapt to store intermediate results in a table valued variable which gets applied to the actual table at the end.
Instead of stripping out the found character by its sole position, using Replace(Column, BadFoundCharacter, '') could be substantially faster. Additionally, instead of just replacing the one bad character found next in each column, this replaces all those found.
WHILE 1 = 1 BEGIN
UPDATE dbo.YourTable
SET Column = Replace(Column, Substring(Column, PatIndex('%[^0-9.-]%', Column), 1), '')
WHERE Column LIKE '%[^0-9.-]%'
If ##RowCount = 0 BREAK;
END;
I am convinced this will work better than the accepted answer, if only because it does fewer operations. There are other ways that might also be faster, but I don't have time to explore those right now.
In a general sense, SQL Server does not support regular expressions and you cannot use them in the native T-SQL code.
You could write a CLR function to do that. See here, for example.
For those looking for a performant and easy solution and are willing to enable CLR:
CREATE database TestSQLFunctions
go
use TestSQLFunctions
go
ALTER database TestSQLFunctions set trustworthy on
EXEC sp_configure 'clr enabled', 1
RECONFIGURE WITH OVERRIDE
go
CREATE ASSEMBLY [SQLFunctions]
AUTHORIZATION [dbo]
FROM 0x4D5A90000300000004000000FFFF0000B800000000000000400000000000000000000000000000000000000000000000000000000000000000000000800000000E1FBA0E00B409CD21B8014CCD21546869732070726F6772616D2063616E6E6F742062652072756E20696E20444F53206D6F64652E0D0D0A2400000000000000504500004C0103004BE8B85F0000000000000000E00022200B013000000800000006000000000000C2270000002000000040000000000010002000000002000004000000000000000600000000000000008000000002000000000000030060850000100000100000000010000010000000000000100000000000000000000000702700004F000000004000009803000000000000000000000000000000000000006000000C00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000200000080000000000000000000000082000004800000000000000000000002E74657874000000C8070000002000000008000000020000000000000000000000000000200000602E72737263000000980300000040000000040000000A0000000000000000000000000000400000402E72656C6F6300000C0000000060000000020000000E00000000000000000000000000004000004200000000000000000000000000000000A4270000000000004800000002000500682000000807000001000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000003A022D02022A020304281000000A2A1E02281100000A2A0042534A4201000100000000000C00000076342E302E33303331390000000005006C00000018020000237E000084020000CC02000023537472696E6773000000005005000004000000235553005405000010000000234755494400000064050000A401000023426C6F620000000000000002000001471500000900000000FA0133001600000100000013000000020000000200000003000000110000000F00000001000000030000000000CA01010000000000060025014F02060092014F02060044001D020F006F02000006006C00E20106000801E2010600D400E20106007901E20106004501E20106005E01E20106008300E2010600580030020600360030020600B700E20106009E00B0010600AA02DB010A00F300FC010A001F00FC010E00C3027E02000000000100000000000100010001001000C3029D0241000100010050200000000096002E001A0001005F2000000000861817020600040000000100BD0200000200F40100000300B102090017020100110017020600190017020A0029001702100031001702100039001702100041001702100049001702100051001702100059001702100061001702150069001702100071001702100079001702100089001702060099002E001A0081001702060020007B0010012E000B002A002E00130033002E001B0052002E0023005B002E002B006D002E0033006D002E003B006D002E0043005B002E004B0073002E0053006D002E005B006D002E0063008B002E006B00B5002E007300C2000480000001000000000000000000000000009D020000040000000000000000000000210016000000000004000000000000000000000021000A00000000000400000000000000000000002100DB010000000000000000003C4D6F64756C653E0053797374656D2E44617461006D73636F726C696200446174614163636573734B696E64005265706C61636500477569644174747269627574650044656275676761626C6541747472696275746500436F6D56697369626C6541747472696275746500417373656D626C795469746C6541747472696275746500417373656D626C7954726164656D61726B417474726962757465005461726765744672616D65776F726B41747472696275746500417373656D626C7946696C6556657273696F6E41747472696275746500417373656D626C79436F6E66696775726174696F6E4174747269627574650053716C46756E6374696F6E41747472696275746500417373656D626C794465736372697074696F6E41747472696275746500436F6D70696C6174696F6E52656C61786174696F6E7341747472696275746500417373656D626C7950726F6475637441747472696275746500417373656D626C79436F7079726967687441747472696275746500417373656D626C79436F6D70616E794174747269627574650052756E74696D65436F6D7061746962696C6974794174747269627574650053797374656D2E52756E74696D652E56657273696F6E696E670053514C46756E6374696F6E732E646C6C0053797374656D0053797374656D2E5265666C656374696F6E007061747465726E004D6963726F736F66742E53716C5365727665722E536572766572002E63746F720053797374656D2E446961676E6F73746963730053797374656D2E52756E74696D652E496E7465726F7053657276696365730053797374656D2E52756E74696D652E436F6D70696C6572536572766963657300446562756767696E674D6F6465730053797374656D2E546578742E526567756C617245787072657373696F6E730053514C46756E6374696F6E73004F626A656374007265706C6163656D656E7400696E70757400526567657800000000000000003A1617E607071B47B964858BCD87458B00042001010803200001052001011111042001010E04200101020600030E0E0E0E08B77A5C561934E0890801000800000000001E01000100540216577261704E6F6E457863657074696F6E5468726F7773010801000200000000001101000C53514C46756E6374696F6E73000005010000000017010012436F7079726967687420C2A920203230323000002901002434346436386231632D393735312D343938612D396665352D32316666333934303738303900000C010007312E302E302E3000004D01001C2E4E45544672616D65776F726B2C56657273696F6E3D76342E352E320100540E144672616D65776F726B446973706C61794E616D65142E4E4554204672616D65776F726B20342E352E32808F010001005455794D6963726F736F66742E53716C5365727665722E5365727665722E446174614163636573734B696E642C2053797374656D2E446174612C2056657273696F6E3D342E302E302E302C2043756C747572653D6E65757472616C2C205075626C69634B6579546F6B656E3D623737613563353631393334653038390A4461746141636365737301000000000000982700000000000000000000B2270000002000000000000000000000000000000000000000000000A4270000000000000000000000005F436F72446C6C4D61696E006D73636F7265652E646C6C0000000000FF25002000100000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000001001000000018000080000000000000000000000000000001000100000030000080000000000000000000000000000001000000000048000000584000003C03000000000000000000003C0334000000560053005F00560045005200530049004F004E005F0049004E0046004F0000000000BD04EFFE00000100000001000000000000000100000000003F000000000000000400000002000000000000000000000000000000440000000100560061007200460069006C00650049006E0066006F00000000002400040000005400720061006E0073006C006100740069006F006E00000000000000B0049C020000010053007400720069006E006700460069006C00650049006E0066006F0000007802000001003000300030003000300034006200300000001A000100010043006F006D006D0065006E007400730000000000000022000100010043006F006D00700061006E0079004E0061006D006500000000000000000042000D000100460069006C0065004400650073006300720069007000740069006F006E0000000000530051004C00460075006E006300740069006F006E00730000000000300008000100460069006C006500560065007200730069006F006E000000000031002E0030002E0030002E003000000042001100010049006E007400650072006E0061006C004E0061006D0065000000530051004C00460075006E006300740069006F006E0073002E0064006C006C00000000004800120001004C006500670061006C0043006F007000790072006900670068007400000043006F0070007900720069006700680074002000A90020002000320030003200300000002A00010001004C006500670061006C00540072006100640065006D00610072006B00730000000000000000004A00110001004F0072006900670069006E0061006C00460069006C0065006E0061006D0065000000530051004C00460075006E006300740069006F006E0073002E0064006C006C00000000003A000D000100500072006F0064007500630074004E0061006D00650000000000530051004C00460075006E006300740069006F006E00730000000000340008000100500072006F006400750063007400560065007200730069006F006E00000031002E0030002E0030002E003000000038000800010041007300730065006D0062006C0079002000560065007200730069006F006E00000031002E0030002E0030002E0030000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000002000000C000000C43700000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
WITH PERMISSION_SET = SAFE
go
CREATE FUNCTION RegexReplace(
#input nvarchar(max),
#pattern nvarchar(max),
#replacement nvarchar(max)
) RETURNS nvarchar (max)
AS EXTERNAL NAME SQLFunctions.[SQLFunctions.Regex].Replace;
go
-- outputs This is a test
SELECT dbo.RegexReplace('This is a test 12345','[0-9]','')
Content of the DLL:
I stumbled across this post looking for something else but thought I'd mention a solution I use which is far more efficient - and really should be the default implementation of any function when used with a set based query - which is to use a cross applied table function. Seems the topic is still active so hopefully this is useful to someone.
Example runtime on a few of the answers so far based on running recursive set based queries or scalar function, based on 1m rows test set removing the chars from a random newid, ranges from 34s to 2m05s for the WHILE loop examples and from 1m3s to {forever} for the function examples.
Using a table function with cross apply achieves the same goal in 10s. You may need to adjust it to suit your needs such as the max length it handles.
Function:
CREATE FUNCTION [dbo].[RemoveChars](#InputUnit VARCHAR(40))
RETURNS TABLE
AS
RETURN
(
WITH Numbers_prep(Number) 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
)
,Numbers(Number) AS
(
SELECT TOP (ISNULL(LEN(#InputUnit),0))
row_number() OVER (ORDER BY (SELECT NULL))
FROM Numbers_prep a
CROSS JOIN Numbers_prep b
)
SELECT
OutputUnit
FROM
(
SELECT
substring(#InputUnit,Number,1)
FROM Numbers
WHERE substring(#InputUnit,Number,1) like '%[0-9]%'
ORDER BY Number
FOR XML PATH('')
) Sub(OutputUnit)
)
Usage:
UPDATE t
SET column = o.OutputUnit
FROM ##t t
CROSS APPLY [dbo].[RemoveChars](t.column) o
Here is a function I wrote to accomplish this based off of the previous answers.
CREATE FUNCTION dbo.RepetitiveReplace
(
#P_String VARCHAR(MAX),
#P_Pattern VARCHAR(MAX),
#P_ReplaceString VARCHAR(MAX),
#P_ReplaceLength INT = 1
)
RETURNS VARCHAR(MAX)
BEGIN
DECLARE #Index INT;
-- Get starting point of pattern
SET #Index = PATINDEX(#P_Pattern, #P_String);
while #Index > 0
begin
--replace matching charactger at index
SET #P_String = STUFF(#P_String, PATINDEX(#P_Pattern, #P_String), #P_ReplaceLength, #P_ReplaceString);
SET #Index = PATINDEX(#P_Pattern, #P_String);
end
RETURN #P_String;
END;
[Gist][1]
[1]: https://gist.github.com/jkdba/ca13fe8f2a9855c4bdbfd0a5d3dfcda2
Edit:
Originally I had a recursive function here which does not play well with sql server as it has a 32 nesting level limit which would result in an error like the below any time you attempt to make 32+ replacements with the function. Instead of trying to make a server level change to allow more nesting (which could be dangerous like allow never ending loops) switching to a while loop makes a lot more sense.
Maximum stored procedure, function, trigger, or view nesting level exceeded (limit 32).
Wrapping the solution inside a SQL function could be useful if you want to reuse it.
I'm even doing it at the cell level, that's why I'm putting this as a different answer:
CREATE FUNCTION [dbo].[fnReplaceInvalidChars] (#string VARCHAR(300))
RETURNS VARCHAR(300)
BEGIN
DECLARE #str VARCHAR(300) = #string;
DECLARE #Pattern VARCHAR (20) = '%[^a-zA-Z0-9]%';
DECLARE #Len INT;
SELECT #Len = LEN(#String);
WHILE #Len > 0
BEGIN
SET #Len = #Len - 1;
IF (PATINDEX(#Pattern,#str) > 0)
BEGIN
SELECT #str = STUFF(#str, PATINDEX(#Pattern,#str),1,'');
END
ELSE
BEGIN
BREAK;
END
END
RETURN #str
END
A more speedy approach for large strings would look something like this:
CREATE FUNCTION [dbo].[fnReplaceInvalidChars] (#string VARCHAR(MAX))
RETURNS VARCHAR(MAX)
BEGIN
DECLARE #str VARCHAR(MAX) = #string;
DECLARE #Pattern VARCHAR (MAX) = '%[^a-zA-Z0-9]%';
WHILE PATINDEX(#Pattern,#str) > 0
BEGIN
SELECT #str = STUFF(#str, PATINDEX(#Pattern,#str),1,'');
END
RETURN #str
END
I've created this function to clean up a string that contained non numeric characters in a time field. The time contained question marks when they did not added the minutes, something like this 20:??. Function loops through each character and replaces the ? with a 0 :
CREATE FUNCTION [dbo].[CleanTime]
(
-- Add the parameters for the function here
#intime nvarchar(10)
)
RETURNS nvarchar(5)
AS
BEGIN
-- Declare the return variable here
DECLARE #ResultVar nvarchar(5)
DECLARE #char char(1)
-- Add the T-SQL statements to compute the return value here
DECLARE #i int = 1
WHILE #i <= LEN(#intime)
BEGIN
SELECT #char = CASE WHEN substring(#intime,#i,1) like '%[0-9:]%' THEN substring(#intime,#i,1) ELSE '0' END
SELECT #ResultVar = concat(#ResultVar,#char)
set #i = #i + 1
END;
-- Return the result of the function
RETURN #ResultVar
END
I think this solution is faster and simple. I use always CTE/recursive because WHILE is so slow on SQL Server.
I use it in projects I work with and large databases.
/*
Function: dbo.kSql_ReplaceRegExp
Create Date: 20.02.2021
Author: Karcan Ozbal
Description: The given string value will be replaced according to the given regexp/pattern.
Parameter(s): #Value : Value/Text to REPLACE.
#RegExp : The regexp/pattern to be used for REPLACE operation.
Usage: select dbo.kSql_ReplaceRegExp('2T3EST5','%[0-9]%')
Output: 'TEST'
*/
ALTER FUNCTION [dbo].[kSql_ReplaceRegExp](
#Value nvarchar(max),
#RegExp nvarchar(50)
)
RETURNS nvarchar(max)
AS
BEGIN
DECLARE #Result nvarchar(max)
;WITH CTE AS (
SELECT NUM = 1, VALUE = #Value, IDX = PATINDEX(#RegExp, #Value)
UNION ALL
SELECT NUM + 1, VALUE = REPLACE(VALUE, SUBSTRING(VALUE,IDX,1),''), IDX = PATINDEX(#RegExp, REPLACE(VALUE, SUBSTRING(VALUE,IDX,1),''))
FROM CTE
WHERE IDX > 0
)
SELECT TOP(1) #Result = VALUE
FROM CTE
ORDER BY NUM DESC
OPTION (maxrecursion 0)
RETURN #Result
END
If you are doing this just for a parameter coming into a Stored Procedure, you can use the following:
declare #badIndex int
set #badIndex = PatIndex('%[^0-9]%', #Param)
while #badIndex > 0
set #Param = Replace(#Param, Substring(#Param, #badIndex, 1), '')
set #badIndex = PatIndex('%[^0-9]%', #Param)
I thought this was clearer:
ALTER FUNCTION [dbo].[func_ReplaceChars](
#Value nvarchar(max),
#Chars nvarchar(50)
)
RETURNS nvarchar(max)
AS
BEGIN
DECLARE #cLen int = len(#Chars);
DECLARE #curChar int = 0;
WHILE #curChar<#cLen
BEGIN
set #Value = replace(#Value,substring(#Chars,#curChar,1),'');
set #curChar = #curChar + 1;
END;
RETURN #Value
END
I'm using this code similar to several codes above:
DROP FUNCTION [dbo].[fnCleanString]
GO
CREATE FUNCTION [dbo].[fnCleanString] (#input VARCHAR(max), #Pattern
VARCHAR (20))
RETURNS VARCHAR(max)
BEGIN
DECLARE #str VARCHAR(max) = #input;
DECLARE #Len INT;
DECLARE #INDEX INT;
SELECT #Len = LEN(#input);
WHILE #Len > 0
BEGIN
SET #INDEX = PATINDEX(#Pattern,#str);
IF (#INDEX > 0)
BEGIN
SET #str=REPLACE(#str,SUBSTRING(#str,#INDEX, 1), '');
END
ELSE
BEGIN
BREAK;
END
END
RETURN #str
END
You can use it like this:
SELECT CleanName = dbo.[fnCleanString](Name, '%[0-9]%') from YourTable
I think a simpler and faster approach is iterate by each character of the alphabet:
DECLARE #i int
SET #i = 0
WHILE(#i < 256)
BEGIN
IF char(#i) NOT IN ('0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '.')
UPDATE Table SET Column = replace(Column, char(#i), '')
SET #i = #i + 1
END
How can I make this possible..really need advice? I want to get the id where my condition is met, then used it in my queries.
IF EXISTS (Select sn_id as snid FROM device.sn WHERE dname_id = 62 and sn_value = '123415')
BEGIN
SELECT MAX(id) AS maxid FROM device.list
INSERT INTO parts (sn_id,device_id) VALUES (snid, maxid)
END
ELSE
BEGIN
PRINT 'id does not exist'
return
END
You can use variables to store the results from the two queries and then use those values in your INSERT statement.
If you're using Microsoft SQL Server then the following may work (but there may be superficial syntax errors as it hasn't been tested). Note that I've assumed the type of your columns is int.
DECLARE #snid int
SET #snid = NULL
Select #snid = sn_id FROM device.sn WHERE dname_id = 62 and sn_value = '123415'
IF #snid IS NULL
BEGIN
PRINT 'id does not exist'
END
ELSE
BEGIN
DECLARE #maxid int
SELECT #maxid = MAX(id) AS maxid FROM device.list
INSERT INTO parts (sn_id,device_id) VALUES (#snid, #maxid)
END
In SQLServer. This script at first insert records and after checks count of the inserted rows
INSERT INTO parts (sn_id, device_id)
SELECT sn_id, (SELECT MAX(id) FROM device.list)
FROM device.sn
WHERE dname_id = 62 and sn_value = '123415'
IF ##ROWCOUNT = 0 PRINT 'id does not exist'
Declare #snid int=null
Declare #maxid int=1 -- if no value exists in device.list table
set #snid = (select sn_id from device.sn WHERE dname_id = 62 and sn_value = '123415')
set #maxid = (select MAX(id) AS maxid FROM device.list)
if #snid is not null
Begin
insert into parts(sn_id,device_id)
values(#snid,#maxid)
End
else
Begin
print 'ID does not exist'
End
Using only SQL Server 2008 R2 (this is going to be in a stored proc), how can I determine if two variables of type XML are equivalent?
Here is what I want to do:
DECLARE #XmlA XML
DECLARE #XmlB XML
SET #XmlA = '[Really long Xml value]'
SET #XmlB = '[Really long Xml value]'
IF #XmlA = #XmlB
SELECT 'Matching Xml!'
But as you probably know, it returns:
Msg 305, Level 16, State 1, Line 7 The XML data type cannot be
compared or sorted, except when using the IS NULL operator.
I can convert to VarChar(MAX) and compare, but that only compares the first 2MB. Is there another way?
Check this SQL function:
CREATE FUNCTION [dbo].[CompareXml]
(
#xml1 XML,
#xml2 XML
)
RETURNS INT
AS
BEGIN
DECLARE #ret INT
SELECT #ret = 0
-- -------------------------------------------------------------
-- If one of the arguments is NULL then we assume that they are
-- not equal.
-- -------------------------------------------------------------
IF #xml1 IS NULL OR #xml2 IS NULL
BEGIN
RETURN 1
END
-- -------------------------------------------------------------
-- Match the name of the elements
-- -------------------------------------------------------------
IF (SELECT #xml1.value('(local-name((/*)[1]))','VARCHAR(MAX)'))
<>
(SELECT #xml2.value('(local-name((/*)[1]))','VARCHAR(MAX)'))
BEGIN
RETURN 1
END
---------------------------------------------------------------
--Match the value of the elements
---------------------------------------------------------------
IF((#xml1.query('count(/*)').value('.','INT') = 1) AND (#xml2.query('count(/*)').value('.','INT') = 1))
BEGIN
DECLARE #elValue1 VARCHAR(MAX), #elValue2 VARCHAR(MAX)
SELECT
#elValue1 = #xml1.value('((/*)[1])','VARCHAR(MAX)'),
#elValue2 = #xml2.value('((/*)[1])','VARCHAR(MAX)')
IF #elValue1 <> #elValue2
BEGIN
RETURN 1
END
END
-- -------------------------------------------------------------
-- Match the number of attributes
-- -------------------------------------------------------------
DECLARE #attCnt1 INT, #attCnt2 INT
SELECT
#attCnt1 = #xml1.query('count(/*/#*)').value('.','INT'),
#attCnt2 = #xml2.query('count(/*/#*)').value('.','INT')
IF #attCnt1 <> #attCnt2 BEGIN
RETURN 1
END
-- -------------------------------------------------------------
-- Match the attributes of attributes
-- Here we need to run a loop over each attribute in the
-- first XML element and see if the same attribut exists
-- in the second element. If the attribute exists, we
-- need to check if the value is the same.
-- -------------------------------------------------------------
DECLARE #cnt INT, #cnt2 INT
DECLARE #attName VARCHAR(MAX)
DECLARE #attValue VARCHAR(MAX)
SELECT #cnt = 1
WHILE #cnt <= #attCnt1
BEGIN
SELECT #attName = NULL, #attValue = NULL
SELECT
#attName = #xml1.value(
'local-name((/*/#*[sql:variable("#cnt")])[1])',
'varchar(MAX)'),
#attValue = #xml1.value(
'(/*/#*[sql:variable("#cnt")])[1]',
'varchar(MAX)')
-- check if the attribute exists in the other XML document
IF #xml2.exist(
'(/*/#*[local-name()=sql:variable("#attName")])[1]'
) = 0
BEGIN
RETURN 1
END
IF #xml2.value(
'(/*/#*[local-name()=sql:variable("#attName")])[1]',
'varchar(MAX)')
<>
#attValue
BEGIN
RETURN 1
END
SELECT #cnt = #cnt + 1
END
-- -------------------------------------------------------------
-- Match the number of child elements
-- -------------------------------------------------------------
DECLARE #elCnt1 INT, #elCnt2 INT
SELECT
#elCnt1 = #xml1.query('count(/*/*)').value('.','INT'),
#elCnt2 = #xml2.query('count(/*/*)').value('.','INT')
IF #elCnt1 <> #elCnt2
BEGIN
RETURN 1
END
-- -------------------------------------------------------------
-- Start recursion for each child element
-- -------------------------------------------------------------
SELECT #cnt = 1
SELECT #cnt2 = 1
DECLARE #x1 XML, #x2 XML
DECLARE #noMatch INT
WHILE #cnt <= #elCnt1
BEGIN
SELECT #x1 = #xml1.query('/*/*[sql:variable("#cnt")]')
--RETURN CONVERT(VARCHAR(MAX),#x1)
WHILE #cnt2 <= #elCnt2
BEGIN
SELECT #x2 = #xml2.query('/*/*[sql:variable("#cnt2")]')
SELECT #noMatch = dbo.CompareXml( #x1, #x2 )
IF #noMatch = 0 BREAK
SELECT #cnt2 = #cnt2 + 1
END
SELECT #cnt2 = 1
IF #noMatch = 1
BEGIN
RETURN 1
END
SELECT #cnt = #cnt + 1
END
RETURN #ret
END
Here is the Source
The function fails to compare XML fragments e.g. when there is not a single root element, like:
SELECT dbo.CompareXml('<data/>', '<data/><data234/>')
In order to fix this, you must wrap your XMLs in root elements, when they are passed to the function or edit the function to do this. For, example:
SELECT dbo.CompareXml('<r><data/></r>', '<r><data/><data234/></r>')
There are many different ways of comparing two XML documents, and a lot depends on what kind of differences you want to tolerate: you definitely need to tolerate differences in encoding, attribute order, insignificant whitespace, numeric character references, and use of attribute delimiters, and you should probably also tolerate differences in use of comments, namespace prefixes, and CDATA. So comparing two XML documents as strings is definitely not a good idea - unless you invoke XML canonicalization first.
For many purposes the XQuery deep-equals() function does the right thing (and is more-or-less equivalent to comparing the canonical forms of the two XML documents). I don't know enough about Microsoft's SQL Server implementation of XQuery to tell you how to invoke this from the SQL level.
You may cast fields to varbinary(max), hash them and compare hashes. But you definitely miss if XMLs are equivalent but not identical
To calculate hash you may use either CLR function:
using System;
using System.Data.SqlTypes;
using System.IO;
namespace ClrHelpers
{
public partial class UserDefinedFunctions {
[Microsoft.SqlServer.Server.SqlFunction]
public static Guid HashMD5(SqlBytes data) {
System.Security.Cryptography.MD5CryptoServiceProvider md5 = new System.Security.Cryptography.MD5CryptoServiceProvider();
md5.Initialize();
int len = 0;
byte[] b = new byte[8192];
Stream s = data.Stream;
do {
len = s.Read(b, 0, 8192);
md5.TransformBlock(b, 0, len, b, 0);
} while(len > 0);
md5.TransformFinalBlock(b, 0, 0);
Guid g = new Guid(md5.Hash);
return g;
}
};
}
Or sql function:
CREATE FUNCTION dbo.GetMyLongHash(#data VARBINARY(MAX))
RETURNS VARBINARY(MAX)
WITH RETURNS NULL ON NULL INPUT
AS
BEGIN
DECLARE #res VARBINARY(MAX) = 0x
DECLARE #position INT = 1, #len INT = DATALENGTH(#data)
WHILE 1 = 1
BEGIN
SET #res = #res + HASHBYTES('MD5', SUBSTRING(#data, #position, 8000))
SET #position = #position+8000
IF #Position > #len
BREAK
END
WHILE DATALENGTH(#res) > 16 SET #res= dbo.GetMyLongHash(#res)
RETURN #res
END
If you can use SQL CLR, I suggest to write a function using XNode.DeepEquals Method:
var xmlTree1 = new XElement("Root",
new XAttribute("Att1", 1),
new XAttribute("Att2", 2),
new XElement("Child1", 1),
new XElement("Child2", "some content")
);
var xmlTree2 = new XElement("Root",
new XAttribute("Att1", 1),
new XAttribute("Att2", 2),
new XElement("Child1", 1),
new XElement("Child2", "some content")
);
Console.WriteLine(XNode.DeepEquals(xmlTree1, xmlTree2));
If you cannot, you can write your own function (see SQL FIDDLE EXAMPLE):
CREATE function [dbo].[udf_XML_Is_Equal]
(
#Data1 xml,
#Data2 xml
)
returns bit
as
begin
declare
#i bigint, #cnt1 bigint, #cnt2 bigint,
#Sub_Data1 xml, #Sub_Data2 xml,
#Name varchar(max), #Value1 nvarchar(max), #Value2 nvarchar(max)
if #Data1 is null or #Data2 is null
return 1
--=========================================================================================================
-- If more than one root - recurse for each element
--=========================================================================================================
select
#cnt1 = #Data1.query('count(/*)').value('.','int'),
#cnt2 = #Data1.query('count(/*)').value('.','int')
if #cnt1 <> #cnt2
return 0
if #cnt1 > 1
begin
select #i = 1
while #i <= #cnt1
begin
select
#Sub_Data1 = #Data1.query('/*[sql:variable("#i")]'),
#Sub_Data2 = #Data2.query('/*[sql:variable("#i")]')
if dbo.udf_XML_Is_Equal_New(#Sub_Data1, #Sub_Data2) = 0
return 0
select #i = #i + 1
end
return 1
end
--=========================================================================================================
-- Comparing root data
--=========================================================================================================
if #Data1.value('local-name(/*[1])','nvarchar(max)') <> #Data2.value('local-name(/*[1])','nvarchar(max)')
return 0
if #Data1.value('/*[1]', 'nvarchar(max)') <> #Data2.value('/*[1]', 'nvarchar(max)')
return 0
--=========================================================================================================
-- Comparing attributes
--=========================================================================================================
select
#cnt1 = #Data1.query('count(/*[1]/#*)').value('.','int'),
#cnt2 = #Data1.query('count(/*[1]/#*)').value('.','int')
if #cnt1 <> #cnt2
return 0
if exists (
select *
from
(
select
T.C.value('local-name(.)', 'nvarchar(max)') as Name,
T.C.value('.', 'nvarchar(max)') as Value
from #Data1.nodes('/*[1]/#*') as T(C)
) as D1
full outer join
(
select
T.C.value('local-name(.)', 'nvarchar(max)') as Name,
T.C.value('.', 'nvarchar(max)') as Value
from #Data2.nodes('/*[1]/#*') as T(C)
) as D2
on D1.Name = D2.Name
where
not
(
D1.Value is null and D2.Value is null or
D1.Value is not null and D2.Value is not null and D1.Value = D2.Value
)
)
return 0
--=========================================================================================================
-- Recursively running for each child
--=========================================================================================================
select
#cnt1 = #Data1.query('count(/*[1]/*)').value('.','int'),
#cnt2 = #Data2.query('count(/*[1]/*)').value('.','int')
if #cnt1 <> #cnt2
return 0
select #i = 1
while #i <= #cnt1
begin
select
#Sub_Data1 = #Data1.query('/*/*[sql:variable("#i")]'),
#Sub_Data2 = #Data2.query('/*/*[sql:variable("#i")]')
if dbo.udf_XML_Is_Equal(#Sub_Data1, #Sub_Data2) = 0
return 0
select #i = #i + 1
end
return 1
END
I stumbled upon this fairly comprehensive article which goes into more detail of actually comparing the CONTENT of 2 XML entries to determine whether they are the same. It makes sense, as the ordering of attributes in nodes CAN differ, even though their values are exactly the same. I'd recommend you read through it and even implement the function to see if it works for you... I tried it out quickly and it seemed to work for me?