I have broken up the solution for explanation. I broke it into 3 logical parts.

1) Use 'cross apply' with string_split to list each word for each CompanyName

2) Use left to...

You can use "CHARINDEX" to find out the position of that character. Then you would use "SUBSTRING" to extract each of the two strings you want. That is the basics.

• This reply was modified 1 year, 7 months ago by  aitorzur@hotmail.com.

Modified to do the group by using the "Timestamp_datetime" field in case someone has the scenario where the time stamp is not already available as a seconds value. I just...

I would  use ' '  (space) between the date and time value. I would then cast the result as a 'datetime' data type.

Also, some of the test data has invalid...

You may want to use "ALTER COLUMN"  syntax rather than "ADD/DROP COLUMN" to preserve any security configuration the that could exist on the column.

Based on the stated assumptions I would suggest the following.

/*--- Initial setup
  --- Cretae synonyms to reference the remote table\s
  I use a synonym to...

lalitprg2017, for giving you a solution I would have to make some assumptions. Please tell me if you see then as reasonable for you situation. The assumptions are:

1. The tables...

lalitprg2017, please confirm whether my understanding is correct.

When things are running both DB-s hold the same data. What you want is that after an outage happens in one DB, lets...

• This reply was modified 1 year, 9 months ago by  aitorzur@hotmail.com.

Can we see how "scratch.dbo.DelimitedSplit8K"  does the spliting of the string?

Here is a function of my own that does the spliting:

create    FUNCTION [dbo].[fn_tbl_StringParseByDelimiter](@InputString ...

The " ORDER BY id DESC " portion on your solution is unnecessary. I don't think it help in any way, not even performance. Am I wrong?

If you want the...

• This reply was modified 1 year, 9 months ago by  aitorzur@hotmail.com.

I agree with Steve Jones. Based on what you say I think using linked tables in Microsoft Access linked to the tables in SQL server will do the job for...

• This reply was modified 1 year, 9 months ago by  aitorzur@hotmail.com.

I modified "pietlinden" solution to use a left join rather than EXCEPT. I find it more clear.

select xj.UserId, xj.ProductId
from (
select u.UserId,  ProductId = p.id
from ( select...

I modified "pietlinden" solution to use a left join rather than EXCEPT. I find it more clear.

select xj.UserId, xj.ProductId
from (
select u.UserId,  ProductId = p.id
from ( select...

Look at the  OPENJSON (Transact-SQL) function.

https://www.sqlservercentral.com/forums/topic/converting-nested-json-document-to-sql-data

OPENJSON is a table-valued function that parses JSON text and returns objects and properties from the JSON input as rows and columns.

Assuming your fact table is something like this:

create table ##Temp_CustomerAppointments

(

Customer varchar(10)

, AppointmentType varchar(20)

, Other varchar(30)

)

This query will get you the result: (The list of customers that have appointment of all...