One approach could be to use a WHILE loop to increment grp values based on row counts and the DATEDIFF interval

Starting with Jeff's code:

--=================================================================================================
--   ...

The fellow with the "Accepted" answer made the same mistake and worse because he used DENSE_RANK on the days.  Unfortunately for him, there were enough days in the original...

Before the pivot it needs to enumerate the rows partitioned within portfoliocode(s).  Once you have the 1's and 2's (if any) you could pivot using conditional aggregation, or the built-in...

I want to defer this process as long as possible. To me that means always aiming for the latest and greatest version. SQL Server releases roughly every 2-3 years,...

Not sure about SSIS specifically but in T-SQL the following seems to be equivalent afaik and unless it can be proven otherwise 🙂

select case when '1900-01-01'='1900-01-01' then...

• This reply was modified 3 years, 9 months ago by Steve Collins.

Imo this is a guessing puzzle unless there's more of an explanation.  What are the rules to apply here?  Why not remove all of the dashes '-' and then concatenate...

What's the new question?

By default SQL Server performs integer division on integers which means decimal remainders are truncated.  The DB Engine switches  to floating point math if the denominator of a fraction is...

Sorry  not able to replicate the issue.  The OP code relies on implicit conversion between INT and NVARCHAR which generally works but is an error prone (and awful for the...

TRIGGER_NESTLEVEL?  I usually try to persuade novices out of it.  Maybe the non-INSTEAD OF triggers could ask themselves: "Have I been executed at an inappropriate nest level?"

The textbooks say writing your own loops in relational database code is something to be avoided.  Are you sure it's necessary to write loops?

SGAB, does any one have a decent example of an Instead Of Update trigger that will handle more than one bloody row at a time and also ONLY update...

To generate sample data you could use a Tally table and calculate a datetime column.  This generates 400 rows and calculates a column 'sample_dt' which starts with the current...

• This reply was modified 3 years, 9 months ago by Steve Collins.
• This reply was modified 3 years, 9 months ago by Steve Collins.

Interesting and much appreciated.  Happy Thanksgiving!

If you load up one of the strings with a bunch of commas it causes an error afaik.  JSON doesn't care how many commas in a row.  A tiny point...