rsample - reservoir sampling script in awk

rsample

#!/usr/bin/env awk -f
# Slight variation of this script: https://stackoverflow.com/a/48869665

BEGIN {
    if (k !~ /^[1-9][0-9]*$/) {
      help()
      exit(0)
    }
    srand();
}

NR <= k {
    reservoir[NR-1] = $0;
    next;
}

{ i = int(NR * rand()) }

i < k { reservoir[i] = $0 }

END {
    for (i in reservoir) {
        print reservoir[i];
    }
}

function help() {
    print "rsample - reservoir sampling"
    printf "\n"
    print "Usage: rsample -v k=[INTEGER] [FILE]"
    printf "\n"
    print "Description:"
    print "Use reservoir sampling to collect k lines from a pipe, stdin or"
    print "a file in O(N) time with N being the total number of lines."
}