TravisDBA (1/17/2012)

---

Please, don't go out and cut the hands off of someone because they were stealing bread when they are homeless. Its the basic principles we are after, and this...

TravisDBA (1/17/2012)

---

butler-628023 (1/16/2012)

---

The pastor who performed my marriage and counseled my wife and I before the blessed event gave me this wise advice...Put family events (i.e. commitments) in your planner...

butler-628023 (1/16/2012)

---

The pastor who performed my marriage and counseled my wife and I before the blessed event gave me this wise advice...Put family events (i.e. commitments) in your planner along...

FROM PERSONAL EXPERIENCES:

As a Jr SQL dev, I've have always tried to abide by one rule: there is a median between Work and Home, and I will always stay in...

L' Eomot Inversé (1/13/2012)

---

stephen99999 (1/13/2012)

---

I think the best way to do this is to just test on various sample data. Really change it up, and see if you can "break"...

I think the best way to do this is to just test on various sample data. Really change it up, and see if you can "break" the code. If not,...

I see your point now. In all actuality, your prior code:

select count(distinct datakey) as Unique_Dataset

from (select *, row_number() over(partition by datakey order by case when Datakey!='' then 0 else 1...

Mark-101232 (1/12/2012)

---

bicky1980 (1/12/2012)

---

stephen99999 (1/12/2012)

---

in table format, bicky, please post what you are expecting your results to look like. Earlier you stated you wanted to see 9, 6, etc to be...

in table format, bicky, please post what you are expecting your results to look like. Earlier you stated you wanted to see 9, 6, etc to be returned, and now...

@bicky

It has to be one of these two options lol. For each option, I started out with a quote on your desired results should be, as well as some...

Alright, lets try this

declare @t table(indkey nvarchar(2),datakey nvarchar(4),landline nvarchar(11),mobile nvarchar(11),email nvarchar(20))

insert into @t values ('1','0001','01234567890','0712345679','1@test.co.uk')

insert into @t values('2','0001','01234567890','','1@test.co.uk')

insert into @t values('3','0002','01234567891','','2@test.co.uk')

insert into @t values('4','0002','01234567890','','2@test.co.uk')

insert into @t values('5','0002','','07123456789','')

insert into @t...

bicky1980 (1/11/2012)

---

stephen99999 (1/11/2012)

---

Try this bick, and please reply with the results... I am not on a machine with management studio, so can't test.

SELECT datakey,COUNT(datakey) OVER(Partition by datakey) AS Uniquedatakey,

...

worker bee (1/11/2012)

---

That got it, I had to do some working around on it but I got what I was looking for. Thank you for putting me in the right...

Wouldnt a date table work in this situation as well? sort of like what we find in data warehouses?

lol I am no expert either, any...