#!/usr/bin/python

# Copyright 2007 Frederik Gladhorn


#/***************************************************************************
 #*                                                                         *
 #*   This program is free software; you can redistribute it and/or modify  *
 #*   it under the terms of the GNU General Public License as published by  *
 #*   the Free Software Foundation; either version 2 of the License, or     *
 #*   (at your option) any later version.                                   *
 #*                                                                         *
 #***************************************************************************/



#convert wordtypes of files that were created during the beta phase
#paste the output at the end of the kvtml file, before the </kvtml> tag


#import xml.dom.minidom

#xml - sax parsing
from xml.sax import make_parser
from xml.sax.handler import ContentHandler

#xml - sax - writing the meta files
from xml.sax.saxutils import XMLGenerator
from xml.sax import saxutils

#from textnormalize import text_normalize_filter

#directory stuff
from dircache import listdir
from os.path import isdir

#for the time - this still has to be improved or done in a different way - file time?
import time

#unicode
import codecs


wordtypeFile = codecs.open("wordtypes_xml", "w", "utf-8")



class KVTML_2_Handler(ContentHandler):
	#important tags
	CAPTURE_ENTRY = 1
	CAPTURE_WORDTYPE = 2
	CAPTURE_SUBTYPE = 3

	def __init__(self):
		self.wordtype=""
		self.subtype=""
		
		self._state = None
		return


	def startDocument(self):
		pass

	def startElement(self, name, attrs):
		if (name == u"entry"):
			self.entry_id = attrs["id"]
				
		if (name == u"translation"):
			self.trans_id = attrs["id"]
		
		if (name == u"typename"):
			self._state = self.CAPTURE_WORDTYPE
			
		if (name == u"subtypename"):
			self._state = self.CAPTURE_SUBTYPE


	def endElement(self, name):
		if (name == u"translation"):
			if (len(self.wordtype) > 0):
				print self.entry_id, " - ", self.trans_id, " ", self.wordtype, " :: ", self.subtype
				
				wordtypeFile.write(u"<container><name>")
				wordtypeFile.write(self.wordtype)
				wordtypeFile.write(u"</name>")
				
				
				
				if (len(self.subtype) > 0):
					wordtypeFile.write(u"<container><name>")
					wordtypeFile.write(self.subtype)
					wordtypeFile.write(u"</name>")
					
				wordtypeFile.write(u"<entry id=\"" + str(self.entry_id) + "\">")
				wordtypeFile.write(u"<translation id=\"" + str(self.trans_id) + "\"/>")
				wordtypeFile.write(u"</entry>")
				
				if (len(self.subtype) > 0):
					wordtypeFile.write(u"</container>\n")

				wordtypeFile.write(u"</container>\n")
			self.wordtype=""
			self.subtype=""
				
		

	def characters(self, text):
		if self._state == self.CAPTURE_WORDTYPE:
			self.wordtype = text
		if self._state == self.CAPTURE_SUBTYPE:
			self.subtype = text

		self._state = None
		return


	def endDocument(self):
		pass




def processFile(voc_file):
	#kvtml = parse(open(voc_file))
	#root = kvtml.getElementsByTagName("kvtml")[0]
	#readLanguages(root)

	language = KVTML_2_Handler()
	saxparser = make_parser()
	saxparser.setContentHandler(language)
	
	datasource = open(voc_file,"r")
	
	wordtypeFile.write(u"<wordtypes>\n")
	
	saxparser.parse(datasource)
	
	wordtypeFile.write(u"</wordtypes>")
	wordtypeFile.close()



def readFile(path):
	print "Reading " + path
	processFile(path)


def readDirectory(path):
	for entry in listdir(path):
		if isdir(path + "/" + entry):
			#print path + "/" + entry
			readDirectory(path + "/" + entry)
		else:
			#print "a file: " + path + "/" + entry
			if entry[-5:] == "kvtml":
				readFile(path + "/" + entry)


def main():
	print "paste the output into the kvtml file"
	rootPath = "."	
	readDirectory(rootPath)
	
main()