Hi James,
I've had a quick play around with the existing Utility - Collection Manipulation and have modified the Remove Null Rows action to do what I think you are after. Here's the code:
'Set local variables
dim iRow as integer
dim count as integer
dim rows as integer
Try
'Initialise row counts and prepopulate output table
rows = Collection_In.Rows.Count - 1
Collection_Out = Collection_In
iRow = 0
'Loop through each row in the input collection
Do While iRow < rows
'Reset counter
count = 0
'Select all rows in the output collection which match the specified value in the input collection
For Each row2 As DataRow in Collection_Out.Select( String.Format("[{0}] ='{1}' or trim([{0}]) = '{1}'", Key_Field,Collection_In.Rows(iRow).Item(Key_Field)))
'If this row is the first found, move on to the next one, otherwise delete it
If count > 0 then
'Delete duplicated row
row2.Delete()
'Reduce the total number of rows in the output collection
rows = rows - 1
End If
'Increase the count of duplicates
count = count + 1
Next
'Increase the count of rows being checked
iRow = iRow + 1
Loop
Success = True
Catch ex as Exception
Error_Message = ex.ToString()
Success = False
End Try
The basic premise is to copy the input collection to the output collection, then loop through the input collection and search the output for duplicates. Then, delete any duplicates from the output and adjust the count values to suit. I've tested this against a small dataset and it seems to work as expected. I've also attached the full object (including my test dataset in the action) to set you on your way.
Hopefully this helps, but feel free to ask for clarification if anything isn't clear or quite right!
Kind regards,
------------------------------
Sam Stone
CoE Build Lead
Blue Prism
------------------------------
Sam Stone
CoE Build Lead
[CompanyName]
