auto number via a query

Hello,
I am wondering if somone could provide me with a sample SELECT for an
auto number query. For example I have a table called People with two columns
first_name, and last_name. There isn't a unique id to correspond with the
table but would like to dynamically make one during the return of the query.
So if there was 5 rows in the table it would return
1 John Alpha
2 John Beta
3 John Cat
4 John Delta
5 John Echo
Where the query was an order by last_name. Thanks in advance.
Jake"Jake Smythe" <someone@.microsoft.com> wrote in message
news:O5zxrnZmGHA.4100@.TK2MSFTNGP05.phx.gbl...
> Hello,
> I am wondering if somone could provide me with a sample SELECT for an
> auto number query. For example I have a table called People with two
> columns first_name, and last_name. There isn't a unique id to correspond
> with the table but would like to dynamically make one during the return of
> the query. So if there was 5 rows in the table it would return
> 1 John Alpha
> 2 John Beta
> 3 John Cat
> 4 John Delta
> 5 John Echo
> Where the query was an order by last_name. Thanks in advance.
Something like this will work, you'll need to add the first name to the
comparison also.
CREATE TABLE People (FirstName VARCHAR(100), LastName VARCHAR(100))
INSERT INTO People VALUES('John','Alpha')
INSERT INTO People VALUES('John','Beta')
INSERT INTO People VALUES('John','Cat')
INSERT INTO People VALUES('John','Delta')
INSERT INTO People VALUES('John','Echo')
SELECT *, (SELECT COUNT(*) FROM People AS P1 WHERE P1.LastName <=
People.LastName) FROM People
ORDER BY LastName, FirstName
DROP TABLE People

> Jake
>|||While this kludge may work for the immediate need, you must be warned that t
here is no certainly that the order will be static. Each time the query exec
utes, the order may be different. And if will have problems with perfectly d
uplicate names. Try adding duplicate names and watch what happens...
SELECT
( SELECT sum(1)
FROM People p
WHERE ( p.LastName + p.FirstName ) <= ( p.LastName + p.FirstName )
) AS rownum
, p2.LastName
, p2.FirstName
FROM People p2
--
Arnie Rowland, YACE*
"To be successful, your heart must accompany your knowledge."
*Yet Another certification Exam
"Michael C" <nospam@.nospam.com> wrote in message news:%23j%23u8yZmGHA.4052@.TK2MSFTNGP05.phx
.gbl...
> "Jake Smythe" <someone@.microsoft.com> wrote in message
> news:O5zxrnZmGHA.4100@.TK2MSFTNGP05.phx.gbl...
>
> Something like this will work, you'll need to add the first name to the
> comparison also.
>
> CREATE TABLE People (FirstName VARCHAR(100), LastName VARCHAR(100))
> INSERT INTO People VALUES('John','Alpha')
> INSERT INTO People VALUES('John','Beta')
> INSERT INTO People VALUES('John','Cat')
> INSERT INTO People VALUES('John','Delta')
> INSERT INTO People VALUES('John','Echo')
> SELECT *, (SELECT COUNT(*) FROM People AS P1 WHERE P1.LastName <=
> People.LastName) FROM People
> ORDER BY LastName, FirstName
> DROP TABLE People
>
>
>
>|||"Arnie Rowland" <arnie@.1568.com> wrote in message
news:eQ0TcDamGHA.4076@.TK2MSFTNGP05.phx.gbl...
While this kludge may work for the immediate need, you must be warned that
there is no certainly that the order will be static. Each time the query
executes, the order may be different. And if will have problems with
perfectly duplicate names. Try adding duplicate names and watch what
happens...
I was going to add a warning that this method wasn't perfect but I had to
race off so just hit send. Of course the value will change if the order
changes :-) As for duplicates you'd just need to use the fields that make up
the primary key, if rows are duplicated then maybe they should have the same
values anyway. Maybe performance of this method might be a problem?
Michael|||Guys thanks for the responses. It's fine if the order changes each time I am
just looking for a identifier at run time.
"Michael C" <nospam@.nospam.com> wrote in message
news:e0x5fZbmGHA.4064@.TK2MSFTNGP02.phx.gbl...
> "Arnie Rowland" <arnie@.1568.com> wrote in message
> news:eQ0TcDamGHA.4076@.TK2MSFTNGP05.phx.gbl...
> While this kludge may work for the immediate need, you must be warned that
> there is no certainly that the order will be static. Each time the query
> executes, the order may be different. And if will have problems with
> perfectly duplicate names. Try adding duplicate names and watch what
> happens...
> I was going to add a warning that this method wasn't perfect but I had to
> race off so just hit send. Of course the value will change if the order
> changes :-) As for duplicates you'd just need to use the fields that make
> up the primary key, if rows are duplicated then maybe they should have the
> same values anyway. Maybe performance of this method might be a problem?
> Michael
>