Hi all

I'm hoping someone out there can suggest a way forward with a spreadsheet
application Ive recently developed.

The spreadsheet contains data from a third party application which in effect
produces a "csv" file with the planned shifts for a group of employees. My
application is designed to analyse the hours being worked in various
categories from the data contained in the spreadsheet.

The user has the option from the third party application to select more than
one employee and more than one week. If that option is selected the output is
sorted by employee then by date. So employee 1 has a number of lines under
week 1 followed by a number of lines under week 2 followed by employee 2 with
week 1 and employee 2 with week 2 etc.

Part of my analysis of the data requires that I know the number of weeks in
the report. Presently that is done by way of a text box on the user form
asking the user to select 1 or 2 or 3 etc.

I'd prefer if the number of weeks could be identified programatically so my
question is :

Is there a way we can count the number of unique values from all of the
values in a column .....

Emp1, 22/05/06
Emp1, 29/05/06
Emp2, 22/05/06
Emp2, 29/05/06

would give the result 2

Note: the dates are not in consecutive rows. There is data in rows between
the firstd ate for Emp1 and second date for Emp2 etc., however the only date
format which exists in the column is that against the employee name. None of
the other rows between employee names contains a date format

I hope I've explained my problem and add that I'm happy with a suggestion
for the principle of how to do it as I'm confident in my own coding to be
able to move forward if I can just have an idea as to how I could appraoch it.

Any help or suggestions would be greatly appreciated.

Regards

Michael
in employee order