Click here to monitor SSC
SQLServerCentral is supported by Red Gate Software Ltd.
 
Log in  ::  Register  ::  Not logged in
 
 


 
     
Recent Posts
   
Popular Topics
Home   
Search
    Members    Calendar    Who's On

Home » SQL Server 2008 » T-SQL (SS2K8) » Combine different rows based on sequence
Add to briefcase

Combine different rows based on sequence Expand / Collapse
Rate Topic
Display Mode
Topic Options
Author
Message
Chrissy321
Posted Monday, October 15, 2012 5:26 PM
Mr or Mrs. 500

Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500

Group: General Forum Members
Last Login: Thursday, May 16, 2013 1:01 PM
Points: 550, Visits: 2,978
Hello All,

I have some really ugly data coming in from several files. I am using SSIS to import into a single column staging table and I am incrementing a counter field so I believe I can select the data from my staging table matching the sequence of the data in the file.

Alas the sequence of the data matters.

See all my DDL below and proposed solution below.

In one scenario the 'first' two records need to be combined into one row. The self join technique works. The are other other scenarios where the first three records need to be combined. I guess I could do three joins but I would like to do something more dynamic if possible, not dynamic like dynamic sql, but rather define a parameter that defines how many rows to combine.

DECLARE @ComboFactor int --The number of rows to combine

I think I can use my parameter and maybe row_number and partition, but that is where I am stuck since my skills with these newer functions are really minimal.

Thanks if you can help.

CREATE TABLE #Staging
( RowID int,
RowData varchar(4)
)

INSERT INTO #Staging (RowID,RowData) VALUES (1,'ab')
INSERT INTO #Staging (RowID,RowData) VALUES (2,'cd')
INSERT INTO #Staging (RowID,RowData) VALUES (3,'ef')
INSERT INTO #Staging (RowID,RowData) VALUES (4,'gh')
INSERT INTO #Staging (RowID,RowData) VALUES (5,'hi')
INSERT INTO #Staging (RowID,RowData) VALUES (6,'jk')
INSERT INTO #Staging (RowID,RowData) VALUES (7,'lm')
INSERT INTO #Staging (RowID,RowData) VALUES (8,'no')
INSERT INTO #Staging (RowID,RowData) VALUES (9,'pq')

SELECT
*
FROM #Staging


SELECT
T1.RowData + T2.RowData
FROM #Staging T1
INNER JOIN #Staging T2
ON T1.RowID +1 = T2.RowID
WHERE
(t1.RowID % 2) <> 0


DROP TABLE #Staging
Post #1372993
dwain.c
Posted Monday, October 15, 2012 6:25 PM


SSCrazy

SSCrazySSCrazySSCrazySSCrazySSCrazySSCrazySSCrazySSCrazy

Group: General Forum Members
Last Login: 2 days ago @ 12:42 AM
Points: 2,338, Visits: 3,158
How about something like this?

DECLARE @ComboFactor INT = 3 --The number of rows to combine 

;WITH Groups AS (
    SELECT RowID, RowData
        ,rn=(ROW_NUMBER() OVER (ORDER BY RowID)-1) / @ComboFactor
    FROM #Staging)
SELECT (
    SELECT RowData + '' 
    FROM Groups b 
    WHERE a.rn = b.rn FOR XML PATH(''))
FROM Groups a
GROUP BY rn

No loops! No CURSORs! No RBAR! Hoo-uh!

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?

Need to UNPIVOT? Why not CROSS APPLY VALUES instead?
Since random numbers are too important to be left to chance, let's generate some!
Are you too recursively challenged?
Splitting strings based on patterns can be fast!
Post #1373001
Chrissy321
Posted Monday, October 15, 2012 7:07 PM
Mr or Mrs. 500

Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500Mr or Mrs. 500

Group: General Forum Members
Last Login: Thursday, May 16, 2013 1:01 PM
Points: 550, Visits: 2,978
Interesting. Clearly I need to study up on FOR XML PATH. Thanks, I'll continue to review.
Post #1373007
dwain.c
Posted Monday, October 15, 2012 7:25 PM


SSCrazy

SSCrazySSCrazySSCrazySSCrazySSCrazySSCrazySSCrazySSCrazy

Group: General Forum Members
Last Login: 2 days ago @ 12:42 AM
Points: 2,338, Visits: 3,158
You're welcome. I forgot something though.

WHERE a.rn = b.rn ORDER BY RowID FOR XML PATH(''))

Need to make sure the rows are concatenated in the right order.

No loops! No CURSORs! No RBAR! Hoo-uh!

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?

Need to UNPIVOT? Why not CROSS APPLY VALUES instead?
Since random numbers are too important to be left to chance, let's generate some!
Are you too recursively challenged?
Splitting strings based on patterns can be fast!
Post #1373010
russellolson
Posted Tuesday, October 16, 2012 1:17 PM
Forum Newbie

Forum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum NewbieForum Newbie

Group: General Forum Members
Last Login: Wednesday, March 27, 2013 5:58 AM
Points: 1, Visits: 38
The example code above showed this statement:

SELECT RowData + ''
FROM Groups b
WHERE a.rn = b.rn FOR XML PATH('')

Why is the + '' required? When I remove this then I get an result that looks like XML instead of a concatenated string.

If I Cast Rowdata to a varchar then the + '' is not required.

Thanks for the explanation.
Post #1373490
dwain.c
Posted Tuesday, October 16, 2012 6:18 PM


SSCrazy

SSCrazySSCrazySSCrazySSCrazySSCrazySSCrazySSCrazySSCrazy

Group: General Forum Members
Last Login: 2 days ago @ 12:42 AM
Points: 2,338, Visits: 3,158
russellolson (10/16/2012)
The example code above showed this statement:

SELECT RowData + ''
FROM Groups b
WHERE a.rn = b.rn FOR XML PATH('')

Why is the + '' required? When I remove this then I get an result that looks like XML instead of a concatenated string.

If I Cast Rowdata to a varchar then the + '' is not required.

Thanks for the explanation.


The XML tag is generated because if you remove the + '' you are now formatting a named column into the string, hence the XML tag. With the + '', it is an unnamed column so hence no XML tag.

CASTing rowdata has the same effect - producing an unnamed column.

I tend to use the former because it is fewer keystrokes. I suppose I should be more scientific and base my choice on performance, but I confess I've never tried it.

No loops! No CURSORs! No RBAR! Hoo-uh!

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?

Need to UNPIVOT? Why not CROSS APPLY VALUES instead?
Since random numbers are too important to be left to chance, let's generate some!
Are you too recursively challenged?
Splitting strings based on patterns can be fast!
Post #1373587
« Prev Topic | Next Topic »

Add to briefcase

Permissions Expand / Collapse

 
 
Copyright © 2002-2013 Simple Talk Publishing. All Rights Reserved. Privacy Policy. Terms of Use. Report Abuse.