Contacenating Rows Into 1 Column

  • Hi,

    This is driving me mad. I want to take some rows of data and put them into 1 row. There will be a max of 4 rows concatenated. Here are the tables:

    CREATE TABLE [dbo].[ResultSupp](

    [RSID] [int] IDENTITY(1,1) NOT NULL,

    [RS_FK_ResultID] [int] NOT NULL,

    [RS_FK_SuppID] [int] NOT NULL,

    [RS_Score] [bit] NOT NULL CONSTRAINT [DF_ResultSupp_RS_Score] DEFAULT ((0)),

    [RS_Active] [bit] NOT NULL CONSTRAINT [DF_ResultSupp_RS_Active] DEFAULT ((1)),

    [RS_LastModified] [datetime] NOT NULL CONSTRAINT [DF_ResultSupp_RS_LastModified] DEFAULT (getdate()),

    ) ON [PRIMARY]

    CREATE TABLE [dbo].[Results](

    [ResultID] [int] IDENTITY(1000,1) NOT NULL,

    [R_FK_LayoutID] [int] NOT NULL,

    [RScore] [int] NULL,

    [RComment] [varchar](50) NULL,

    [R_FK_HeaderID] [int] NOT NULL,

    [R_Active] [bit] NOT NULL CONSTRAINT [DF_Results_R_Active] DEFAULT ((1)),

    [R_LastModified] [datetime] NOT NULL CONSTRAINT [DF_Results_R_LastModified] DEFAULT (getdate()),

    ) ON [PRIMARY]

    CREATE TABLE [dbo].[TestLayout](

    [LayoutID] [int] IDENTITY(1,1) NOT NULL,

    [LDescription] [varchar](100) NOT NULL,

    [LMinDesc] [varchar](50) NOT NULL,

    [LMaxDesc] [varchar](50) NOT NULL,

    [L_FK_TTypeID] [int] NOT NULL,

    [LActive] [bit] NOT NULL CONSTRAINT [DF_TestLayout_LActive] DEFAULT ((1)),

    [LLastModified] [datetime] NOT NULL CONSTRAINT [DF_TestLayout_LLastModified] DEFAULT (getdate()),

    [LPageID] [int] NOT NULL,

    [SortOrder] [int] NULL,

    ) ON [PRIMARY]

    CREATE TABLE [dbo].[TestSupplement](

    [SuppID] [int] IDENTITY(1,1) NOT NULL,

    [SuppDescription] [varchar](30) NOT NULL,

    [SuppActive] [bit] NOT NULL CONSTRAINT [DF_TestSupplement_SuppActive] DEFAULT ((1)),

    [SuppLastModified] [datetime] NOT NULL CONSTRAINT [DF_TestSupplement_SuppLastModified] DEFAULT (getdate()),

    [Supp_FK_LayoutID] [int] NOT NULL,

    [SortOrder] [int] NOT NULL,

    ) ON [PRIMARY]

    Join COde:

    FROMTestSupplement INNER JOIN

    ResultSupp ON ResultSupp.RS_FK_SuppID = TestSupplement.SuppID INNER JOIN

    TestLayout ON TestSupplement.Supp_FK_LayoutID = TestLayout.LayoutID

    WHERE(ResultSupp.RS_FK_ResultID = @ResultID AND TestLayout.LayoutID = @LayoutID)

    What I want to return is Result.ResultID, TestLayout.LDescription + ': ' + TestSupplement.SuppDescription in a row for each ResultID

    I've tried this (but I can't use the output in a Crystal Report)

    SELECTTestLayout.LDescription + ': ' + TestSupplement.SuppDescription + ', ' AS 'data()'

    FROMTestSupplement INNER JOIN

    ResultSupp ON ResultSupp.RS_FK_SuppID = TestSupplement.SuppID INNER JOIN

    TestLayout ON TestSupplement.Supp_FK_LayoutID = TestLayout.LayoutID

    WHERE(ResultSupp.RS_FK_ResultID = @ResultID AND TestLayout.LayoutID = @LayoutID) for xml path('')

    Any ideas?

    Thanks.

  • There are lot of tables in the mix. I'm confused. But here is the code i use when i need to concatenate multiple rows into one single row.

    IF OBJECT_ID('TEMPDB..#CONCAT_COLUMN_VALUES') IS NOT NULL

    DROP TABLE #CONCAT_COLUMN_VALUES

    CREATE TABLE #CONCAT_COLUMN_VALUES

    (

    id INT,

    COL_VAL VARCHAR(5)

    )

    INSERT INTO #CONCAT_COLUMN_VALUES

    SELECT 1, 'A' UNION ALL

    SELECT 1, 'B' UNION ALL

    SELECT 1, 'C' UNION ALL

    SELECT 2, 'D' UNION ALL

    SELECT 2, 'E' UNION ALL

    SELECT 3, 'F' UNION ALL

    SELECT 4, 'G'

    SELECT p1.id,

    STUFF ( ( SELECT ','+COL_VAL

    FROM #CONCAT_COLUMN_VALUES p2

    WHERE p2.id = p1.id

    ORDER BY COL_VAL

    FOR XML PATH(''),TYPE

    ).value('.','VARCHAR(MAX)') , 1,1,SPACE(0)) AS Concat_Values

    FROM #CONCAT_COLUMN_VALUES p1

    GROUP BY p1.id ;

    Tweak it to make it work for you!

  • Hi,

    I've got this working but not fully happy with solution. I made a Temporary table and put the results that I wanted to add together into this first to make it a bit easier to follow. Here's what I now have (borrowed from http://www.simple-talk.com/sql/t-sql-programming/concatenating-row-values-in-transact-sql/):

    ALTER FUNCTION dbo.CollateSuppComments(@R_FK_HeaderID int

    ,@LPageID int

    )

    RETURNS varchar(1000)

    AS

    BEGIN

    DECLARE @ReturnSuppComments varchar(1000)

    DECLARE @TempComments TABLE (

    R_FK_HeaderID int

    ,CommentDesc varchar(1000)

    ,LPageID int

    )

    INSERT INTO @TempComments(

    R_FK_HeaderID

    ,CommentDesc

    ,LPageID

    )

    SELECTR_FK_HeaderID

    , TestLayout.LDescription + ': ' + TestSupplement.SuppDescription AS CommentDesc

    ,TestLayout.LPageID

    FROMTestSupplement INNER JOIN

    ResultSupp AS p1 ON p1.RS_FK_SuppID = TestSupplement.SuppID INNER JOIN

    TestLayout ON TestSupplement.Supp_FK_LayoutID = TestLayout.LayoutID INNER JOIN

    Results ON p1.RS_FK_ResultID = Results.ResultID AND TestLayout.LayoutID = Results.R_FK_LayoutID

    WHERER_FK_HeaderID = @R_FK_HeaderID AND LpageID = @LPageID AND RS_Score = 1

    SELECT @ReturnSuppComments =

    MAX( CASE seq WHEN 1 THEN CommentDesc ELSE '' END ) + ', ' +

    MAX( CASE seq WHEN 2 THEN CommentDesc ELSE '' END ) + ', ' +

    MAX( CASE seq WHEN 3 THEN CommentDesc ELSE '' END ) + ', ' +

    MAX( CASE seq WHEN 4 THEN CommentDesc ELSE '' END ) + ', ' +

    MAX( CASE seq WHEN 5 THEN CommentDesc ELSE '' END ) + ', ' +

    MAX( CASE seq WHEN 6 THEN CommentDesc ELSE '' END ) + ', ' +

    MAX( CASE seq WHEN 7 THEN CommentDesc ELSE '' END ) + ', ' +

    MAX( CASE seq WHEN 8 THEN CommentDesc ELSE '' END ) + ', ' +

    MAX( CASE seq WHEN 9 THEN CommentDesc ELSE '' END ) + ', ' +

    MAX( CASE seq WHEN 10 THEN CommentDesc ELSE '' END )

    FROM ( SELECT p1.R_FK_HeaderID, p1.CommentDesc,

    ( SELECT COUNT(*)

    FROM @TempComments p2

    WHERE p2.R_FK_HeaderID = p1.R_FK_HeaderID

    AND p2.LPageID <= p1.LPageID AND

    p2.CommentDesc <= p1.CommentDesc )

    FROM @TempComments p1 ) D ( R_FK_HeaderID, CommentDesc, seq )

    GROUP BY R_FK_HeaderID

    RETURN @ReturnSuppComments

    END

    Two questions:

    1. What does this line do:

    D ( R_FK_HeaderID, CommentDesc, seq )

    2. Is there a way to not add the extra commas if there isn't a corresponding seq i.e. if there are only three comments to add together, 7 commas are added to the end, coming from here:

    MAX( CASE seq WHEN 1 THEN CommentDesc ELSE '' END ) + ', ' +

    Thanks

  • dec_obrien (3/14/2012)


    Is there a way to not add the extra commas if there isn't a corresponding seq

    Yes. Do what ColdCoffee said in the first place.

    Untested, but it should be something like this: -

    ALTER FUNCTION dbo.CollateSuppComments (@R_FK_HeaderID INT, @LPageID INT)

    RETURNS VARCHAR(1000)

    AS

    BEGIN

    SELECT @ReturnSuppComments =

    STUFF((SELECT ', ' + CommentDesc

    FROM (SELECT R_FK_HeaderID, TestLayout.LDescription + ': ' + TestSupplement.SuppDescription AS CommentDesc

    FROM TestSupplement

    INNER JOIN ResultSupp AS p1 ON p1.RS_FK_SuppID = TestSupplement.SuppID

    INNER JOIN TestLayout ON TestSupplement.Supp_FK_LayoutID = TestLayout.LayoutID

    INNER JOIN Results ON p1.RS_FK_ResultID = Results.ResultID AND TestLayout.LayoutID = Results.R_FK_LayoutID

    WHERE R_FK_HeaderID = @R_FK_HeaderID AND LpageID = @LPageID AND RS_Score = 1) a

    FOR XML PATH(''), TYPE).value('.', 'VARCHAR(MAX)'), 1, 2, '');

    RETURN @ReturnSuppComments

    END


    Forever trying to learn
    My blog - http://www.cadavre.co.uk/
    For better, quicker answers on T-SQL questions, click on the following...http://www.sqlservercentral.com/articles/Best+Practices/61537/
    For better, quicker answers on SQL Server performance related questions, click on the following...http://www.sqlservercentral.com/articles/SQLServerCentral/66909/

  • Hi,

    Yes it works (although I've no idea why!).

    THanks

  • dec_obrien (3/14/2012)


    Hi,

    Yes it works (although I've no idea why!).

    THanks

    I'll try an example that you can follow in your SQL window

    First, set your results to text and execute this: -

    SELECT TOP 5 t1.[name]

    FROM sys.columns t1

    INNER JOIN sys.columns t2 on t1.[column_id] = t2.[column_id]

    WHERE t1.[column_id] = t2.[column_id]

    FOR XML PATH

    You'll get back something like this: -

    <row><name>addr</name></row><row><name>addr</name></row><row><name>addr</name></row><row><name>addr</name></row><row><name>addr</name></row>

    Next, we'll elimate the row tags, which is done simply by adding ('')

    SELECT TOP 5 t1.[name]

    FROM sys.columns t1

    INNER JOIN sys.columns t2 on t1.[column_id] = t2.[column_id]

    WHERE t1.[column_id] = t2.[column_id]

    FOR XML PATH('')

    Now we have this: -

    <name>addr</name><name>addr</name><name>addr</name><name>addr</name><name>addr</name>

    Now, we'll add a comma that comes before the result. This makes the result into a string.

    SELECT TOP 5 ', ' + t1.[name]

    FROM sys.columns t1

    INNER JOIN sys.columns t2 on t1.[column_id] = t2.[column_id]

    WHERE t1.[column_id] = t2.[column_id]

    FOR XML PATH('')

    , addr, addr, addr, addr, addr

    STUFF allows us to get rid of the comma at the beginning, giving us a list.

    SELECT STUFF((SELECT TOP 5 ', ' + t1.[name]

    FROM sys.columns t1

    INNER JOIN sys.columns t2

    ON t1.[column_id] = t2.[column_id]

    WHERE t1.[column_id] = t2.[column_id]

    FOR XML PATH('')), 1, 2, '')

    Which gives us this: -

    addr, addr, addr, addr, addr

    Does that help?


    Forever trying to learn
    My blog - http://www.cadavre.co.uk/
    For better, quicker answers on T-SQL questions, click on the following...http://www.sqlservercentral.com/articles/Best+Practices/61537/
    For better, quicker answers on SQL Server performance related questions, click on the following...http://www.sqlservercentral.com/articles/SQLServerCentral/66909/

  • Helps a lot. Thank you for going to the trouble of explaining it. And if you're up for one last question, teh last line of my previous solution containd the following:

    FROM @TempComments p1 ) D ( R_FK_HeaderID, CommentDesc, seq )

    What does that do?

  • FROM @TempComments p1 ) D ( R_FK_HeaderID, CommentDesc, seq )

    D is called table-alias.. This is dynamically assinging a name to a sub-query so that i can be referenced elswhere in the query. And the values in the brackets are the column names that the table-alias is going to contain.

    To know more about tables aliases, read here Using Table Aliases MSDN

    Thanks to cadavre for explaining the XML so lucidly, i couldnt have done it better!

  • Spot on guys.

    Great help and thanks again!

Viewing 9 posts - 1 through 8 (of 8 total)

You must be logged in to reply to this topic. Login to reply