when using groupBy, I often feel the need to count unique identifiers rather
than just collapse them into a comma delimited list with redundancies.
Obviously, I can then parse this list to count unique, but this is hard to with
Awk or unix utilities (at least with my knowledge) without changing the useful
groupBy output format. Basically, a groupBy uniq -c is what I'm asking for. I'm
guessing you have an easy way to do this outside of groupBy, which is why you
haven't added it yet? or am i missing some existing functionality.
Original issue reported on code.google.com by iraha...@gmail.com on 17 Sep 2011 at 8:28
Original issue reported on code.google.com by
iraha...@gmail.com
on 17 Sep 2011 at 8:28