#!/usr/bin/python
# -*- coding: utf-8 -*-
#
#author: rex
#blog: http://iregex.org
#filename counter.py
#created: Mon Sep 20 21:00:52 2010
#desc: convert .py file to html with VIM.

import sys
import re
from operator import itemgetter

def readfile(f):
with file(f,"r") as pFile:
return pFile.read()

def divide(c, regex):
#the regex below is only valid for utf8 coding
return regex.findall(c)


def update_dict(di,li):
for i in li:
if di.has_key(i):
di[i]+=1
else:
di[i]=1
return di

def main():

#receive files from bash
files=sys.argv[1:]

#regex compile only once
regex=re.compile("(?x) (?: [\w-]+ | [\x80-\xff]{3} )")

dict={}

#get all words from files
for f in files:
words=divide(readfile(f), regex)
dict=update_dict(dict, words)

#sort dictionary by value
#dict is now a list.
dict=sorted(dict.items(), key=itemgetter(1), reverse=True)

#output to standard-output
for i in dict:
print i[0], i[1]


if __name__=='__main__':
main()

regex=re.compile("(?x) ( [\w-]+ | [\x80-\xff]{3} )")
words=[w for w in regex.split(line) if w]