#!/bin/sh URL="intel-ussur.ru" MAXWORD=100 w3m -dump -T text/html ${URL} | \ sed s/" "/"\n"/g | \ sort | \ sed 's/[[:upper:]]/\l&/g' | \ sed s/[^a-zа-я-]//g | \ uniq -c -d | \ sort -n | \ awk 'length($2) > 4' | \ tail -n ${MAXWORD} | \ sort -n -r