http://blog.henix.info/blog/linux-cmd-freq.html

reducewhat.py

#!/usr/bin/python2
import sys
import re

records = []
for line in sys.stdin:
	records.append(re.match(r'[ ]*(\d)+ ([^ ]+)(.*)', line.rstrip()).groups())

for record in sorted(records, key=lambda d:int(d[0])*(len(d[1])+len(d[2])), reverse=True):
	print "%s %s%s" % record