#! /bin/sh

total=`cat $1 | wc -l`

cat $1 | \
  sed -e 's/via proxy gateway  CERN[^ ][^ ]*//g' -e 's/libwww\/[^ ][^ ]*//g' \
   -e 's/([^)][^)]*)//g' -e 's/modified//g' -e 's/for the X Window System\///'\
   -e 's/for Windows\/Version//' -e 's/libwww-perl\///' -e 's/for Windows\///'\
   -e 's/NCSA//' -e 's/^\(Mozilla\)..*$/\1/' \
   -e 's/via Gateway  CERN-HTTPD[^ ][^ ]*//g' \
   -e 's/  */ /g' -e 's/^ //' |\
  sort | grep -v '^' | uniq -c | sort -rn |\
  gawk -v "linecount=$total"  \
          '{ clientname=sprintf ("%s %s %s %s", $2, $3, $4, $5) ; 
            printf "%-7s %-40s %3.2f\%\n", $1, clientname, ($1 / linecount) * 100.0 }'
