A little more better...

DECLARE @MyTable TABLE

(

myVersion    VARCHAR(10),

LastDateA     DATETIME,

LastDateB     DATETIME,

LastDateC     DATETIME

)

INSERT @MyTable VALUES ('VERSION A', '01/01/2004', '02/01/2004', '01/03/2004')

INSERT @MyTable VALUES ('VERSION A', '03/01/2004', '02/04/2004', '01/09/2004')

INSERT @MyTable VALUES ('VERSION A', '04/01/2004', '04/09/2004',...

This will work for ANY No of duplicates and will keep the first row with the MIN(ID)

SET NOCOUNT ON

DECLARE @theatre TABLE

(

[ID] INT IDENTITY,

LastName VARCHAR(25),

FirstName VARCHAR(25)

)

INSERT INTO @theatre (LastName, FirstName) VALUES...

You get Null because you select NULL : )

Null As 'HireCnt'   

Null As 'ApplicantCnt'

About the numbers they might be correct. It seems that 1 row gives...

For someone to be able to help you should provide the tables and sample data with the req result.

From my undertanding (which can be wrong ) if you transform your...

SELECT AAA.OD_ID,Op_Sec_ID,Op_Val_Date,Sec_Date,Sec_Price

FROM

#tblOP AAA INNER JOIN

(SELECT OD_ID,MIN(MyNmb) MyDiff

FROM

(SELECT A.Od_Id,A.Op_Sec_Id,A.Op_Val_Date,B.Sec_ID,B.Sec_Date,B.Sec_Price,  CASE WHEN A.Op_Val_Date>B.Sec_Date THEN CAST(A.Op_Val_Date as int)-CAST(B.Sec_Date as int) ELSE CAST(B.Sec_Date as int)-CAST(A.Op_Val_Date as int) END MyNmb

FROM #tblOP A inner join #tblSecurity...

Won't work to have 4 derived tables.

The solution is based on comb of two diff sets.

so if C,N,Y,O form 2 sets the sol will work

ex: C,N can be combined...

The catch is to label the rows for each DATA and ID2 with Ordered Numbers and match the rows with diff ID2 but the same Data and RowNumber

Hmm

what if you change the design of the table

StartPeriod---DateTime

EndPeriod ---DateTime

Status ----Available/NotAvailable

and when you insert a new row upgrade EndPeriod check the status on last row and if different add...

SET NOCOUNT ON

DECLARE @T Table (ID int, ID2 char(1) , Data varchar(10))

insert into  @T (ID,ID2,Data) Values(1,'N','Data')

insert into  @T (ID,ID2,Data) Values(2,'N','Data' )

insert into  @T (ID,ID2,Data) Values(3,'C','Data' )

insert into  @T (ID,ID2,Data) Values(4,'C','Data5'...

UPS ya I was missing 20 and 30 ... didn't notice (Thks Remi : ) )

for the (10,a) pair I SAID the sol is not working... he needs to check...

At least (ID,Country) has to be key for the Solution to work.

Otherwise one more step need to be added to treat this rows depending on FirstName...

set nocount on

create table #t(IdRow...

If you have a KEY on your table you can achieve what you want otherwise it won't work (like here where IdRow is not a Key...).

set nocount on

create table...

SET QUOTED_IDENTIFIER OFF

SET NOCOUNT ON

DECLARE @SumOfMaximums FLOAT, @DataValue FLOAT, @PreviousDataValue FLOAT, @SeqNo INT,@Tid INT, @MaximumValue FLOAT

SET @Tid=11

CREATE TABLE [#TRENDDATA] ( [SEQUENCENUMBER_] [int] NOT NULL ,

 [TID_] [int] NOT NULL ,

 [DATE_STAMP_]...

SELECT CASE WHEN Step1 IS NULL THEN 0 ELSE 1 END

This will return the row as integer.

In your case it shows 1/1/1900 because it transforms that integer 0...