I have a large dataset where I need to do a find and replace for every group
of 3 columns. Each column as either a 1 or 0. I need to change all the "1s"
in N columns to "1", the "1s" in N+1 columns to "2", and the "1s" in N+2
columns to "3".

Then I need to merge each group of 3 columns into one column, preserving the
data. Any ideas?