I like the idea of source control.  Makes sense.   I prefer an on-prem solution vs github etc.  Can I do this in SS?  Or do I need an external program? ...

Actually...I ended up moving them manually.  No need for code.

i used Ken's solution.  works great!

i just tried the clustered index approach.  it worked.  thank you for this!!

• This reply was modified 3 months, 2 weeks ago by water490.

thank you both.  the code is working great now.

Thank you both.  The line numbers got added.  I ran the BULK INSERT code and the entire text file got added to one row vs separate rows.  I don't think...

• This reply was modified 3 months, 2 weeks ago by water490.

it won't let me post the formatfile so here is a picture of it:

Not sure what is going on here.  I posted a reply twice but it never got posted.  I will try again.  Sorry for any spam.

I did some research and it...

I tried the batch approach yesterday.  There are over 5000 units and one unit took about 6 seconds to run.  If I did this in batches of one unit at...

this works!  thank you so much!!!

I tried the query on a much larger dataset and after 4.5 hours it is still running.  The dataset has about 10million records.  Is there...

• This reply was modified 3 months, 3 weeks ago by water490. Reason: asked a follow up question
• This reply was modified 3 months, 3 weeks ago by water490.

I forgot to include the expected results.  Sorry about that.  Here you go:

I think this is what you are looking for.

-- For each test_date, determine the last date to count in a 10 date range
; WITH DateAnalysis AS
(SELECT...

My SQL skills are not too sharp.  I am...

np.  here you go:

drop table if...

Can you not insert a rolling 10 days window? Something along the lines of

PERCENT_RANK() OVER (PARTITION BY year ORDER BY date DESC ROWS BETWEEN 10 PRECEDING...