Shell Programming and Scripting

Note that you could also put the sort and tail pipeline in the shell instead of doing it inside awk by changing the last line of the gawk script from:

Code:

END {for(ip in C) printf "%7d %s\n", C[ip], ip}' $FILES

to:

Code:

END {for(ip in C) printf "%7d %s\n", C[ip], ip}' $FILES | sort -k1,1rn | tail -5

Doing it this way should be slightly faster since gawk won't have to spawn an additional shell to run the pipeline.

XXXXXXXXXXXXXXXXXXXX

Please ignore the above suggestion; I lost track of some of the changes that had been made since I last touched this problem and confused this thread with a problem I was working on for another thread.

I agree doing it outside of gawk should be much faster and easier to implement. However, I am getting less output when I change the code to what you suggested. This would be OK except I used the same range I had before so I know there were several ip's that matched the criteria before. Whereas, it is only showing 1 that matches.

So, I will have to do some checking on the syntax to see what is going on.

Nevertheless, I appreciate you getting me pointed in the right direction.

I apologize. I was working on an awk problem in another thread and forgot how the code in this thread worked. Please ignore my suggestion in post #36 in this thread.

I sure appreciate you letting me know to not continue down that path for my answer.