you* could write one in half a day or an even better one in a full day

* from your post you sound experienced so the T-SQL should be a breeze

ok, the Jet OLEDB connection is breaking the text file up in to object compenents which are then returned to your query, the order of the columns is never important...

this works for integer values up to ~700

isnull(ceiling(log(nullif(floor(exp( @value )), 0))), 0)

as I said previously though, I would avoid like the plague

Hi Adam,

I've seen and made similar work arounds but they look so hacky I can't bring myself to use them in real life applications.

For some complex queries, like this one...

test it out the second query a little and see what happens 😉

No problem 🙂

remember I said how if i found a more elegant solution I'd post it.

Well with my mathematical head on this morning a slightly neater approach dawned on me:

take...

Change NCHAR to VARCHAR

works for me 🙂

Edit: Beaten to it!!

It was correct, its your specifications that were wrong 😛

select top 1

date, session

from #rr

where convert(varchar, date, 103) + '-'+ cast(session...

kgunnarsson,

That doesn't return anything for me!

select top 1

date, session

from #rr

where session not in (select session from #rr where message = 'error')

order by date desc

here's a good article on how to post:

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

as you're new to SQL, have a look at my first reply if any confuses you then you need to...

Nice bit of your code in your post, certainly speeds things up! 🙂

select top 1

date, session

from #rr

where message <> 'error'

order...

Well this has been fun and brightened up a quiet morning 😀

Heres a shift calculating function you can use:

alter function dbo.calc_shift (@Time_start as datetime, @Time_end as datetime, @shift_start as int,...

use a distinct on the select from table A:

Select Distinct col1, col2, col3

from tableA

add a clause if you want to exclude everytihng from tableB

Select Distinct a.col1, a.col2, a.col3

...

Here's a more complete solution:

select

-- zero column, between 3pm and 11pm

-- first 2 cases, work...