As you see here you need to be very precise in the desciption of your problem as you might get different results.
The main difference in the two solutions from Dieter and me is that Dieter is counting the number of days per weekdays with data in a date range and my solution is counting the days per weekdays in a time range - without reference of the presence to data.
In case for each day at least one records exists in your table the result is the same but the result will be different if at least one day is not present for all group by values in your base table.
↧