January 20, 2014 at 5:56 am
Hi all,
I have a problem where I want to write a function to remove recurring characters from a string and replace them with a single same character.
For instance I have the string '12333345566689' and the result should be '12345689'. In Oracle I could do this with "regexp_replace('12333345566689', '(.)\1+', '\1')", but in T-SQL the only solution I could think of is something like this:
DECLARE @code NVARCHAR(255)
SET @code = '12333345566689';
SET @code = REPLACE(REPLACE(REPLACE(@Code, '1', '~1'), '1~', ''), '~1', '1');
and repeat this for 2 - 9. But I'm sure there is a more elegant version for this in SQL Server 2012. Unfortunately I haven't found any solution in other posts, so maybe someone has an idea for this?
Thanks in advance.
January 20, 2014 at 7:57 am
Probably not the best way to do this, just off the top of my head: -
DECLARE @code NVARCHAR(255);
SET @code = '12333345566689';
WITH CTE(N) AS (SELECT 1 FROM (VALUES(1),(1),(1),(1),(1),(1),(1),(1),(1),(1))a(N)),
CTE2(N) AS (SELECT 1 FROM CTE x CROSS JOIN CTE y),
CTE3(N) AS (SELECT 1 FROM CTE2 x CROSS JOIN CTE2 y),
CTE4(N) AS (SELECT 1 FROM CTE3 x CROSS JOIN CTE3 y),
CTE5(N) AS (SELECT 1 FROM CTE4 x CROSS JOIN CTE4 y),
CTE6(N) AS (SELECT TOP (LEN(@code)) 1 FROM CTE5 x CROSS JOIN CTE5 y),
TALLY(N) AS (SELECT ROW_NUMBER() OVER(ORDER BY (SELECT NULL)) FROM CTE6)
SELECT @code = REPLACE(REPLACE(REPLACE(LTRIM(RTRIM(@code)),SUBSTRING(@code,N,1)+SUBSTRING(@code,N,1),SUBSTRING(@code,N,1)+CHAR(7)),CHAR(7)+SUBSTRING(@code,N,1),''),CHAR(7),'')
FROM TALLY;
SELECT @code;
Returns "12345689"
Based on REPLACE Multiple Spaces with One[/url] and The "Numbers" or "Tally" Table: What it is and how it replaces a loop[/url].
January 20, 2014 at 8:14 am
Here's a different method that might work well. Should we start testing? 😀
DECLARE @code nvarchar(4000) = '12333345566689';
with seed1 (a)
as
(
select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all select 1 union all
select 1 union all select 1 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 (n) as
(
select top (LEN(@code)) row_number() over (order by (select null))
from seed1 s1, seed1 s2, seed1 s3
)
select a.b.value('(./text())[1]', 'nvarchar(4000)') as [text]
from (
select CASE WHEN SUBSTRING( @code, n, 1) = SUBSTRING( @code, n - 1, 1) THEN ''
ELSE SUBSTRING( @code, n, 1) END
FROM numbers
for xml path (''), type
)a(b)
January 20, 2014 at 7:37 pm
A slight variation to Luis's method and implemented as an inline table valued function
CREATE FUNCTION compressDuplicates(@code nvarchar(4000))
RETURNS TABLE WITH SCHEMABINDING AS
RETURN
WITH cteTally as (
SELECT TOP(LEN(@code) - 1) ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) N
FROM (VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) E1(N)
,(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) E2(N)
,(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) E3(N)
),
startpattern as (
SELECT SUBSTRING(@code,1,1) P
UNION ALL
SELECT SUBSTRING(@code,N+1,1) P
FROM cteTally
WHERE SUBSTRING(@code,N,1) <> SUBSTRING(@code,N+1,1) COLLATE Latin1_General_BIN -- Change to suit comparison required
)
SELECT CAST(r AS NVARCHAR(4000)) result
FROM (
SELECT P
FROM startpattern
FOR XML PATH('')
) AS A(R);
If there is non numerics in your string you will need to decide how to handle case, etc and set the collation accordingly
January 20, 2014 at 11:38 pm
Just to be different, and seeing as how we are in the SQL 2012 forum which allows me to use LAG, I will offer this:
WITH SampleData AS
(
SELECT code=CAST('12333345566689333' AS NVARCHAR(4000))
UNION ALL SELECT '12333345566689'
),
Tally (n) AS
(
SELECT TOP (4000) ROW_NUMBER() OVER (ORDER BY (SELECT NULL))
FROM sys.all_columns
),
SplitString AS
(
SELECT code, s, p, n
FROM
(
SELECT code, s, p=LAG(s, 1, '') OVER (PARTITION BY code ORDER BY n), n
FROM
(
SELECT code, n, s=SUBSTRING(code, n, 1)
FROM SampleData a
CROSS APPLY
(
SELECT TOP (LEN(code)) n
FROM Tally
) b
WHERE SUBSTRING(code, n, 1) <> ''
) a
) a
WHERE s <> p
)
SELECT code
,ReducedString=
(
SELECT s + ''
FROM SplitString b
WHERE a.code = b.code
ORDER BY n
FOR XML PATH('')
)
FROM SampleData a;
My thought question: Have you ever been told that your query runs too fast?
My advice:
INDEXing a poor-performing query is like putting sugar on cat food. Yeah, it probably tastes better but are you sure you want to eat it?
The path of least resistance can be a slippery slope. Take care that fixing your fixes of fixes doesn't snowball and end up costing you more than fixing the root cause would have in the first place.
Need to UNPIVOT? Why not CROSS APPLY VALUES instead?[/url]
Since random numbers are too important to be left to chance, let's generate some![/url]
Learn to understand recursive CTEs by example.[/url]
[url url=http://www.sqlservercentral.com/articles/St
January 21, 2014 at 12:01 am
OK, so the purists will argue that LAG is overkill, and for them I offer this instead:
WITH SampleData AS
(
SELECT code=CAST('12333345566689333' AS NVARCHAR(4000))
UNION ALL SELECT '12333345566689'
),
Tally (n) AS
(
SELECT TOP (4000) ROW_NUMBER() OVER (ORDER BY (SELECT NULL))
FROM sys.all_columns
),
SplitString AS
(
SELECT code, s, n
FROM
(
SELECT code, n, s=SUBSTRING(code, n, 1)
FROM SampleData a
CROSS APPLY
(
SELECT TOP (LEN(code)) n
FROM Tally
) b
WHERE SUBSTRING(code, n, 1) <> SUBSTRING(code, n-1, 1)
) a
)
SELECT code
,ReducedString=
(
SELECT s + ''
FROM SplitString b
WHERE a.code = b.code
ORDER BY n
FOR XML PATH('')
)
FROM SampleData a;
My thought question: Have you ever been told that your query runs too fast?
My advice:
INDEXing a poor-performing query is like putting sugar on cat food. Yeah, it probably tastes better but are you sure you want to eat it?
The path of least resistance can be a slippery slope. Take care that fixing your fixes of fixes doesn't snowball and end up costing you more than fixing the root cause would have in the first place.
Need to UNPIVOT? Why not CROSS APPLY VALUES instead?[/url]
Since random numbers are too important to be left to chance, let's generate some![/url]
Learn to understand recursive CTEs by example.[/url]
[url url=http://www.sqlservercentral.com/articles/St
January 21, 2014 at 2:10 am
Many thanks for your anwers.
I tried the two different approaches from dwain.c with my sample table with 100k nvarchar(50) strings. On my machine with Win 8 64, SQL Server 2012, Core i7 2.4 Ghz and 16 GB RAM the second version without LAG function took around 5 seconds. The first version with LAG executed for 12 minutes when I aborted it!! Slight difference...
With the other three approaches I have to admit that I can't get it to work with my sample table. How can I incorporate a column from my table into the code instead of the @code parameter?
Anyway, I think the 5 seconds will be hard to beat.
January 21, 2014 at 2:55 am
j.heidrich (1/21/2014)
Many thanks for your anwers.I tried the two different approaches from dwain.c with my sample table with 100k nvarchar(50) strings. On my machine with Win 8 64, SQL Server 2012, Core i7 2.4 Ghz and 16 GB RAM the second version without LAG function took around 5 seconds. The first version with LAG executed for 12 minutes when I aborted it!! Slight difference...
With the other three approaches I have to admit that I can't get it to work with my sample table. How can I incorporate a column from my table into the code instead of the @code parameter?
Anyway, I think the 5 seconds will be hard to beat.
Thanks for posting those performance results. I was just having a bit of fun with LAG, knowing it wouldn't be as good as the option without it. I'm casting my net wide to find an instance where LAG beats an earlier code pattern in performance and your test shows me this ain't one of 'em.
My thought question: Have you ever been told that your query runs too fast?
My advice:
INDEXing a poor-performing query is like putting sugar on cat food. Yeah, it probably tastes better but are you sure you want to eat it?
The path of least resistance can be a slippery slope. Take care that fixing your fixes of fixes doesn't snowball and end up costing you more than fixing the root cause would have in the first place.
Need to UNPIVOT? Why not CROSS APPLY VALUES instead?[/url]
Since random numbers are too important to be left to chance, let's generate some![/url]
Learn to understand recursive CTEs by example.[/url]
[url url=http://www.sqlservercentral.com/articles/St
January 21, 2014 at 2:58 am
MickyT's FUNCTION should work like this with your table replacing SampleData:
WITH SampleData AS
(
SELECT code=CAST('12333345566689333' AS NVARCHAR(4000))
UNION ALL SELECT '12333345566689'
)
SELECT *
FROM SampleData
CROSS APPLY compressDuplicates(code);
My thought question: Have you ever been told that your query runs too fast?
My advice:
INDEXing a poor-performing query is like putting sugar on cat food. Yeah, it probably tastes better but are you sure you want to eat it?
The path of least resistance can be a slippery slope. Take care that fixing your fixes of fixes doesn't snowball and end up costing you more than fixing the root cause would have in the first place.
Need to UNPIVOT? Why not CROSS APPLY VALUES instead?[/url]
Since random numbers are too important to be left to chance, let's generate some![/url]
Learn to understand recursive CTEs by example.[/url]
[url url=http://www.sqlservercentral.com/articles/St
January 21, 2014 at 6:46 am
Couple of small changes to MickeyT's otherwise excellent function:
DECLARE @code nvarchar(4000);
SET @code = '99944777777775588888888888812' ;
WITH cteTally as (
SELECT TOP(LEN(@code) - 1) ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) N
FROM (VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) E1(N)
,(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) E2(N)
,(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) E3(N)
),
startpattern as (
SELECT SUBSTRING(@code,1,1) P
UNION ALL
SELECT SUBSTRING(@code,N+1,1) P
FROM cteTally
WHERE SUBSTRING(@code,N,1) <> SUBSTRING(@code,N+1,1) COLLATE Latin1_General_BIN -- Change to suit comparison required
)
SELECT CAST(r AS NVARCHAR(4000)) result
FROM (
SELECT P
FROM startpattern
FOR XML PATH('')
) AS A(R);
-- result: "<P>9</P><P>4</P><P>7</P><P>5</P><P>8</P><P>1</P><P>2</P>"
--------------------------------------------------------------------------------------
WITH cteTally as (
SELECT TOP(LEN(@code) - 1) ROW_NUMBER() OVER (ORDER BY (SELECT NULL)) N
FROM (VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) E1(N)
,(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) E2(N)
,(VALUES (0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) E3(N)
),
startpattern as (
SELECT FirstNumOfSet = SUBSTRING(@code,n,1)
FROM cteTally
WHERE SUBSTRING(@code,N,1) <> SUBSTRING(@code,N-1,1) COLLATE Latin1_General_BIN -- Change to suit comparison required
)
SELECT result = CAST(XMLresult AS NVARCHAR(4000))
FROM (
SELECT FirstNumOfSet AS [text()]
FROM startpattern
FOR XML PATH('')
) d (XMLresult);
-- result: "947581"
For fast, accurate and documented assistance in answering your questions, please read this article.
Understanding and using APPLY, (I) and (II) Paul White
Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden
January 21, 2014 at 8:54 am
To use the solutions provided with a table column instead of a variable. You just need to change the variable declaration to a function definition and call it with cross apply.
e.g. Change this:
DECLARE @code nvarchar(4000);
SET @code = '99944777777775588888888888812' ;
To this:
CREATE FUNCTION FunctionName(@code nvarchar(4000))
RETURNS TABLE WITH SCHEMABINDING AS
RETURN
January 21, 2014 at 11:03 am
ChrisM@Work (1/21/2014)
Couple of small changes to MickeyT's otherwise excellent function:
Thanks Chris, I missed that I had messed up the result. I made some changes for performance to what I was originally going to post and didn't recheck the results, just the speed.
Insert a quote from Homer Simpson here:-D
January 22, 2014 at 12:39 am
Thanks guys,
mickyT's solution is even faster for large data sets (> 500k) 🙂
January 22, 2014 at 1:36 am
mickyT (1/21/2014)
ChrisM@Work (1/21/2014)
Couple of small changes to MickeyT's otherwise excellent function:Thanks Chris, I missed that I had messed up the result. I made some changes for performance to what I was originally going to post and didn't recheck the results, just the speed.
Insert a quote from Homer Simpson here:-D
No worries Mickey - been there many times;-)
For fast, accurate and documented assistance in answering your questions, please read this article.
Understanding and using APPLY, (I) and (II) Paul White
Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden
January 22, 2014 at 4:28 am
This problem was heavily discussed back in 2009. It only delt with multiple blanks and at the time I had proposed the best SQL solution. Naturally there were CLR solutions that were better.
Have a look at
http://www.sqlservercentral.com/Forums/Topic819042-203-1.aspx
and
http://www.sqlservercentral.com/Forums/Topic819042-203-12.aspx
Agreed, "multiple any character" is quite a different problem.
Viewing 15 posts - 1 through 14 (of 14 total)
You must be logged in to reply to this topic. Login to reply