Yes, does my solution (bottom pg 3) work or not?

Select {col1, Col2, ...Coln} From {Table}

Where NOT (Department = 2 And Category2 = 'B')

And NOT (Department = 2 And Category3 = 'C')

And NOT (Department = 3 And Category1 =...

Hmm I know exactly how UDF's work given that I write/use them but this question must have been worded really badly (for me).

"SELECT UDF FROM Table" - to me this...

Works fine in SQL2000.. you'd get the start date column twice (once at start, once wherever it naturally sits in the order of columns on the table.. and order by...

If you can make the assumption "Smallest MID is always the 'first' row for each code" then this should work:

-- Using MID (assumption: Smallest ID = "first")

SELECT PD.Code,

PD.Remarks,

E.Status

FROMPDetails PD

INNER...

No problem at all - glad I could help 🙂

Derek Dongray (4/30/2008)

No, because this would place '192.1.1.1' before '19.2.1.12'.

Also addresses like '192.168.123.234'...

The table variable is only created for demo purposes..

Just use ORDER BY CAST(replace (IPAddr,'.','') as int)

Tell me I'm wrong.. if I am.. but isn't this the simplest way?

-===== Create a table to demo with

DECLARE @DemoTable TABLE (IPAddr VARCHAR(15))

INSERT INTO @DemoTable (IPAddr)

SELECT '192.168.215.1' UNION...

Well, since you appear to only be running it for a month at a time, why not simplify it and use DATEPART in your grouping?

SELECTUser_ID,

MIN(add_date) as 'First Log',

MAX(add_date) as 'Last...

I've made a few assumptions hence my larger query:

1, You won't always have a job within a category with a create date or filled date that falls within the desired...

If I'm understanding your requirement correctly, something like this? Sorry if I've over-complicated but I'm trying to cater for all possible scanrios here (including there being no jobs at all...