Simple Group By

  • Hi,

    I have a table called XYZ . This table stores all this information shown below.

    AdId ClassOriginalDateAmtPaid

    2741069 1012010-06-21 125.00

    2694907 1012010-06-22 262.50

    2741898 1012010-06-22262.50

    2748110 1012010-06-22 300.00

    I want to display the above as follows

    ClassOriginalDateAmtPaid

    1012010-06-21 125.00

    1012010-06-22 825.00

    Any help appreciated .

    Thanks.

  • this almost seams to simple of a question so I feal like I am missing somehting but.

    select Class,OriginalDate,sum(AmtPaid)

    from insertyourtablename

    group by Class,OriginalDate

    Dan

    If only I could snap my figures and have all the correct indexes apear and the buffer clean and.... Start day dream here.

  • I have also used the query below

    SELECT DISTINCT A.Class,

    A.Origdate,

    SUM(A.totalpaid) As AmtPaid

    FROM dbo.ad A

    GROUP BY A.Class,A.Origdate,A.totalpaid

    HAVING A.Origdate >= '6/21/2010' AND A.Origdate < '6/25/2010'

    AND A.totalpaid <> 0.00

    But it still returns

    ClassOrigdate AmtPaid

    1012010-06-21 13:02:14.960 125.00

    1012010-06-22 14:25:37.343 262.50

    1012010-06-22 15:00:50.437 262.50

    1012010-06-22 15:52:48.647 300.00

    I want to avoid the timestamp and group based on the class and date .

  • Sorry about that I knew I was missing something. The problem is that when you display the date it has the time stamp and there for the nano secods make each vlue uniqu in the grouping so convert the the date to drop the time and poof they will all combine.

    Dan

    If only I could snap my figures and have all the correct indexes apear and the buffer clean and.... Start day dream here.

  • Sorry I meant to post the code. Man I am glad it is Friday.

    SELECT DISTINCT A.Class,

    CAST( CONVERT( CHAR(8), A.Origdate, 112) AS DATETIME),

    SUM(A.totalpaid) As AmtPaid

    FROM dbo.ad A

    GROUP BY A.Class,CAST( CONVERT( CHAR(8), A.Origdate, 112) AS DATETIME),A.totalpaid

    HAVING A.Origdate >= '6/21/2010' AND A.Origdate < '6/25/2010'

    AND A.totalpaid <> 0.00

    Dan

    If only I could snap my figures and have all the correct indexes apear and the buffer clean and.... Start day dream here.

  • I am getting the following error :

    Column 'dbo.ad.origdate' is invalid in the HAVING clause because it is not contained in either an aggregate function or the GROUP BY clause.

  • Dan.Humphries (6/25/2010)


    Sorry about that I knew I was missing something. The problem is that when you display the date it has the time stamp and there for the nano secods make each vlue uniqu in the grouping so convert the the date to drop the time and poof they will all combine.

    I think this works....without using CONVERT or CAST....

    declare @table table (id int, dy smalldatetime,amt numeric (18,2))

    Insert Into @table

    select 101,2010-06-21,125.00

    union all select 101,2010-06-22,262.50

    union all select 101,2010-06-22,262.50

    union all select 101,2010-06-22,300.00

    select * From @Table

    select id

    ,dy

    ,SUM(amt)as AmountPaid

    from @table

    GROUP BY ID,DY

    ORDER BY AmountPaid

    FO

  • If I use the query below

    SELECT DISTINCT A.Class,

    CAST( CONVERT( CHAR(8), A.Origdate, 112) AS DATETIME),

    SUM(A.totalpaid) As AmtPaid

    FROM dbo.ad A

    GROUP BY A.Class,CAST( CONVERT( CHAR(8), A.Origdate, 112) AS DATETIME),A.totalpaid

    HAVING CAST( CONVERT( CHAR(8), A.Origdate, 112)AS DATETIME) >= '6/21/2010' AND CAST( CONVERT( CHAR(8), A.Origdate, 112)AS DATETIME) < '6/25/2010'

    AND A.totalpaid <> 0.00 AND A.Class = 101

    Order By A.Class

    I get the following result :

    Class(No column name) AmtPaid

    1012010-06-21 00:00:00.000125.00

    1012010-06-22 00:00:00.000300.00

    1012010-06-22 00:00:00.000525.00

  • move the date parameter to a where clause and that should take care of it.

    SELECT DISTINCT A.Class,

    CAST( CONVERT( CHAR(8), A.Origdate, 112) AS DATETIME),

    SUM(A.totalpaid) As AmtPaid

    FROM dbo.ad A

    where CAST( CONVERT( CHAR(8),A.Origdate, 112) AS DATETIME), between A.Origdate '6/21/2010' AND '6/25/2010'

    GROUP BY A.Class,CAST( CONVERT( CHAR(8), A.Origdate, 112) AS DATETIME),A.totalpaid

    HAVING AND A.totalpaid <> 0.00

    sorry I was posting at the same time. I am not sure why it is doig that but I will try a few things. your query would work fine.

    Dan

    If only I could snap my figures and have all the correct indexes apear and the buffer clean and.... Start day dream here.

  • The following query

    SELECT DISTINCT A.Class,

    CAST( CONVERT( CHAR(8), A.Origdate, 112) AS DATETIME),

    SUM(A.totalpaid) As AmtPaid

    FROM dbo.ad A

    where CAST( CONVERT( CHAR(8),A.Origdate, 112) AS DATETIME)>= '6/21/2010' AND CAST( CONVERT( CHAR(8),A.Origdate, 112) AS DATETIME) < '6/25/2010'

    AND A.totalpaid <> 0.00

    GROUP BY A.Class,CAST( CONVERT( CHAR(8), A.Origdate, 112) AS DATETIME),A.totalpaid

    Order By A.Class

    results in

    Class(No column name)AmtPaid

    1012010-06-21 00:00:00.000125.00

    1012010-06-22 00:00:00.000525.00

    1012010-06-22 00:00:00.000300.00

  • declare @table table (id int, dy smalldatetime,amt numeric (18,2))

    Insert Into @table

    select 101,2010-06-21,125.00

    union all select 101,2010-06-22,262.50

    union all select 101,2010-06-22,262.50

    union all select 101,2010-06-22,300.00

    select * From @Table

    select id

    ,dy

    ,SUM(amt)as AmountPaid

    from @table

    GROUP BY ID,DY

    ORDER BY AmountPaid

    FO

  • you had one two many criteria in your group by also I would move the having to a where clause.

    declare @table table (AdId int, class int, Origdate datetime,totalpaid numeric (10,2))

    insert into @table (AdId,class,Origdate,totalpaid) values(2741069,101,'06/21/2010 13:02:14.960',125.00)

    insert into @table (AdId,class,Origdate,totalpaid) values(2694907,101,'06/22/2010 14:25:37.343',262.50)

    insert into @table (AdId,class,Origdate,totalpaid) values(2741898,101,'06/22/2010 15:00:50.437',262.50)

    insert into @table (AdId,class,Origdate,totalpaid) values(2748110,101,'06/22/2010 15:52:48.647',300.00)

    SELECT DISTINCT A.Class,

    CAST( CONVERT( CHAR(8), A.Origdate, 112) AS DATETIME),

    SUM(A.totalpaid) As AmtPaid

    FROM @table A

    WHERE CAST( CONVERT( CHAR(8), A.Origdate, 112)AS DATETIME) >= '6/21/2010' AND CAST( CONVERT( CHAR(8), A.Origdate, 112)AS DATETIME) < '6/25/2010'

    AND A.totalpaid <> 0.00 AND A.Class = 101

    GROUP BY A.Class,CAST( CONVERT( CHAR(8), A.Origdate, 112) AS DATETIME)

    Order By A.Class

    Dan

    If only I could snap my figures and have all the correct indexes apear and the buffer clean and.... Start day dream here.

  • falgunoza,

    If the data was stored as smalldatetime I doubt he would be having this problem to start with. It is fairly typical to have to do a convert in my expierence smalldatetime is seldom used.

    Dan

    If only I could snap my figures and have all the correct indexes apear and the buffer clean and.... Start day dream here.

  • Thanks that worked .

  • Dan.Humphries (6/25/2010)


    falgunoza,

    If the data was stored as smalldatetime I doubt he would be having this problem to start with. It is fairly typical to have to do a convert in my expierence smalldatetime is seldom used.

    But I changed table variable's date column data type to datetime and it still worked...Am I missing something I wonder....:unsure:

    FO

Viewing 15 posts - 1 through 15 (of 15 total)

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