Subquery with case statement

  • Hi:

    I have two tables T1 and T2, both have the same structure.

    Tables contain Quarterly data. I usually produce charts from T1 table.

    Table structure is :

    Event_Qtr Total ....

    Q1-2009 21644865 ....

    Q2-2009 152899 .....

    Both table have from 2007 to 2009 quarterly data.

    For some management requirement I have to select total from table T2 if quarter is else total will be selected from T1 table(for Q4-2009 and up).

    Is it possible to select data by subquery (using case statement) ?

    I need help badly pls. Any coding help is highly appreciated

    Thanking in advance.

    Maksuda

  • Some sample data along with an example of the result set you would like to see based off of your sample data would help you get a better quality answer. See this thread for putting together a post that will get you a quick answer.

    http://www.sqlservercentral.com/articles/Best+Practices/61537/"> http://www.sqlservercentral.com/articles/Best+Practices/61537/

    John Rowan

    ======================================================
    ======================================================
    Forum Etiquette: How to post data/code on a forum to get the best help[/url] - by Jeff Moden

  • Thanks John. You are right. I should provide enough information to get the quick response.

    Here is my table:Qtr_Event_TotalInfo(T1)

    CREATE TABLE [dbo].[Qtr_TotalInfo](

    Qtr_ID INT IDENTITY(1,1) PRIMARY KEY CLUSTERED,

    Event_Qtr varchar(7) ,

    Event_Total int NULL

    )

    Table T2 also have the same structure

    data in table(T1)

    insert into Qtr_Event_TotalInfo values('Q1-2009',21644865)

    insert into Qtr_Event_TotalInfo values('Q2-2009', 38533950)

    insert into Qtr_Event_TotalInfo values('Q3-2009',23143204)

    insert into Qtr_Event_TotalInfo values('Q4-2009',6355608)

    insert into Qtr_Event_TotalInfo values('Q1-2008',21376661)

    insert into Qtr_Event_TotalInfo values('Q2-2008',16412433)

    insert into Qtr_Event_TotalInfo values('Q3-2008',18035155

    insert into Qtr_Event_TotalInfo values('Q4-2008',17574502)

    Data in table T2:

    insert into T2 values('Q1-2009', 68200 )

    insert into T2 values('Q2-2009', 3161591)

    insert into T2 values('Q3-2009',5950028)

    insert into T2 values('Q1-2008',168212)

    insert into T2 values('Q2-2008', 152899)

    insert into T2 values('Q3-2008',2265710)

    insert into T2 values('Q4-2008',168212)

    Infact table T2 has data from Q1-2007 to Q3-2009.

    Table Qtr_TotalInfo (T1) has data from Q1-2007 to Q4-2009.

    Now in my report user selects year from a list

    if selected year is 2008 or 2007 data will show up from T2 table,

    if selected year is 2009 then Q1-2009,Q2-2009 and Q3-2009 data will show up from table T2 and Q4-2009 data will show up from Qtr_TotalInfo (T1) table.

    Is it possible to select data for this condition by a single query (using subquery) ?

    Please need help. code sample is highly appreciated.

    Thanks,

    Maksuda

  • I think this will give you what you are looking for.

    SELECT IsNull(t2.Event_Total, t1.Event_Total) AS Event_Total

    FROM qtr_totalinfo AS t1

    LEFT OUTER JOIN t2

    ON t1.Event_Qtr = t2.Event_Qtr

    Drew

    J. Drew Allen
    Business Intelligence Analyst
    Philadelphia, PA

  • Thanks Drew for the reply.

    I tried and it giving me Cartesian Product.

    Thanking you,

    Maksuda

  • h_maksuda (8/19/2009)


    Thanks Drew for the reply.

    I tried and it giving me Cartesian Product.

    Is your data representative of the actual data? Your data only has one value for each date in each table. If that is the case you will not get a Cartesian Product.

    If either or both of your tables has multiple rows for a single date, I suggest that you update your sample data to be more representative of your actual data. As they say, "garbage in, garbage out". If you want a good solution, your going to need to provide good data.

    Drew

    J. Drew Allen
    Business Intelligence Analyst
    Philadelphia, PA

  • If you're aggregating quarters that happen to exist in both tables, I think you want UNION ALL rather than a join. Here's one way - not super pretty, but it does what I think you want to do with your sample data (Q1-2008 from Qtr_Event_TotalInfo plus Q1-2008 from T2 = 21,544,873. If that's not it, we may need a little more clarification.

    SELECT

    A.Event_Qtr,

    SUM(A.Event_Total) AS Event_Total

    FROM (

    SELECT * FROM Qtr_Event_TotalInfo Q

    UNION ALL

    SELECT * FROM T2 Q

    ) A

    GROUP BY

    A.Event_Qtr

    ORDER BY RIGHT(A.Event_Qtr, 4) + LEFT(A.Event_Qtr,2)

    Regards,

    John


    Regards,

    John Hopkins

Viewing 7 posts - 1 through 6 (of 6 total)

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