unicode字符计数

count.py

#!/usr/bin/env python
#-*- coding:utf8 -*-
import re
import urllib

def counter(url, start, end):
	p = urllib.urlopen(url).read().decode('utf8')
	art = r'(?s)(?=%s).*?(?<=%s)' % (start, end)
	article = re.findall(art, p)[0]
	# print article
	cn = re.compile(u'(?:<.*?>)|([\u4e00-\u9fa5]+)|([a-zA-Z0-9_-]+)')
	num_zh = num_en = 0
	for chr in cn.finditer(article):
		zh, en = chr.groups()
		if zh:
			num_zh += len(zh.encode('utf8'))/3
			print zh.encode('utf8'),
		if en:
			num_en += 1
			print en,
	return num_zh, num_en

if __name__ == '__main__':
	url = r'http://qd.58.com/zufang/14336862785801x.shtml'
	start, end = '<article class="description_con " >', '</article>'
	num_zh, num_en = counter(url, start, end)

	print 
	print '中文字数: %d' % num_zh
	print '英文字数: %d' % num_en
	print '总字数: %d' % (num_zh + num_en)