Question

using linux commands,
I have a quoted csv file which I sorted by first column and then second column, now I want to remove duplicates where they match in the first and second column, how can this be done? uniq doesn’t seem to be enough, or is it?

Answer 1

You could reverse (rev) the file, then uniq ignoring the first N-2 fields (everything but the first two columns), then rev again.

rev | uniq -f N-2 -u | rev