#!/usr/bin/awk -f
#
# count the number each word is used

BEGIN {
	FS = "[^A-Za-z]+"
}

{
	for (i = 1; i <= NF; i++) {
		words[tolower($i)]++
		nwords++
		if (length($i) > maxlen) {
			maxlen = length($i)
		}
	}
}

END {
	delete words[""]
	for (i in words) {
		printf "%4d %-"maxlen"s (%2.2f%%)\n", words[i], i, (100.0*words[i]/nwords)
	}
}