Search Results

awk 'NF {a[$1]+=$2; b[$1]=$0} END {for (i in a) {print b[i], a[i]}}' file
This will give output as..
sam 3 pdf 6
tom 2 ltf 16
ram 5 word 12
Sorry, I could not help you much on how...

vi +<nth_occurrence>/<search_pattern> <input_file> will solve your problem

Should be comm -23 <(sort a) <(sort b)

awk 'NR==FNR {a[$1]=$2; next} $2 in a {print $0, a[$2]}' f2 f1

Please let me know if there is any more cute and compact solution.

The problem is that the loop and ssh (by default) read from the same stdin, and ssh "consumes" (eats) the input to feed it to the remote command, that would otherwise have been read by the loop....