Thom, this is great. Thank you for taking the time to write it out so clearly.

but I'm always surprised more people don't use it

Almost all of my development is done...

This can be done, if you think laterally: make " your delimiter and your formulas become:

TOKEN( SourceRecord,"\"",1)
TOKEN( SourceRecord,"\"",3)
TOKEN( SourceRecord,"\"",5)
TOKEN( SourceRecord,"\"",7)
TOKEN( SourceRecord,"\"",9)

Having said that, Frederico's suggestion is better,...

Years later - if someone stumbles over this: WITH(NOLOCK) can get locks for non-SELECT statements, for example, it will get locks if it is part of an UPDATE statement....

In this i want to add condition if @frdate Month = 3 and Date = 31 ...

Try adding this to your WHERE clause:

      AND NOT (
            ...

No need to apologise ... probably a false positive, but I put it there just in case.

Wish I could help with your actual problem, but I'm just a lowly developer....

No idea how to ban someone, but I...

I was half expecting/hoping that there would at least...

Note to others about the jasonbrimhall link – MBAM blocked it for me:

If you would like help creating a solution in code, please provide your data in consumable format (ie, one which can be cut & pasted into SSMS).

Start date is not relevant ... at least based on the sample data provided.

SELECT *
FROM #TEMP t
WHERE t.EDate >= '20190101';

That's effectively the same answer, from three posters I trust, so I'll just carry on doing it this way! Thanks to you all for taking the time to respond.

Have you tried using COUNT(DISTINCT col)?

Your query will never get total record counts for the tables because you are filtering out rows in the WHERE clause before the COUNT is applied..

Well, all csv files belongign to a specific table are structured the same. BUt based on some characters in the file names, they are differently structured nad go into...

• This reply was modified 6 years, 4 months ago by Phil Parkin.