furigana.py

# coding=UTF-8
#k=u"私は疲れた"
#h=u"わたしはつられた"

import re

def rubyForString(k, h):

	expression = u'([^\u3040-\u30ff]*)([\u3040-\u30ff]*)([^\u3040-\u30ff]*)([\u3040-\u30ff]*)([^\u3040-\u30ff]*)([\u3040-\u30ff]*)([^\u3040-\u30ff]*)([\u3040-\u30ff]*)'

	kresult = re.search(expression,k)
	kgroups = kresult.groups()

	krealgroups = []
	kanaexpression = ""

	for i in range(len(kgroups)):
		kgroup = kgroups[i]
		if len(kgroup) == 0:
			continue
		krealgroups.append(kgroup)
		isKanji = i % 2 == 0 # Verified
		if isKanji:
			kanaexpression += u"([\u3040-\u30ff]+)"
		else:
			kanaexpression += u"(" + kgroup + u")"



		# print "is Kanji " + str(isKanji)
		# print "group " + group
		# print ""

	#print "kanaexpression " + kanaexpression

	hresult = re.search(kanaexpression, h)
	hgroups = hresult.groups()

	if len(krealgroups) != len(hgroups):
		#print "not equal lengths"
		return "".join(hgroups)

	output = ""

	for i in range(len(hgroups)):
		kanji = krealgroups[i]
		hiragana = hgroups[i]
		if kanji == hiragana:
			output += hiragana
		else:
			output += "<ruby>" + kanji + "<rt>" + hiragana + "</rt></ruby>"

	return output