#!/usr/bin/python2
# coding: utf-8

import codecs
from lxml import etree
from copy import deepcopy
from random import choice

filename = "alutor.lift"
TOTAL_SIZE = float(raw_input("Input amount of MB: ")) * 1000000

letters = u'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ-'
letters += u'абсгдеёжзийклмнопрстуфхцчшщъыьЭЮЯАБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯ'
letters += u'jəɣŋ'

def morph_word(word):
    l = []
    for i in xrange(len(word)):
        if word[i] in letters:
            l.append(i)
    if 0 in l:
        l.remove(0)

    if l:
        for k in xrange(3):
            i = choice(l)
            j = choice(l)
            if i > j: i, j = j, i

            if i != j:
                word = word[:i] + word[j] + word[i+1:j] + word[i] + word[j+1:]
    return word


def morph_sentence(text):
    new_text = ''
    for word in text.split(' '):
        new_text += morph_word(word) + ' '
    return new_text.strip()


f = codecs.open(filename, 'r', 'utf-8')
tree = etree.parse(f)
f.close()

old_lift = tree.getroot()

for i in xrange(int(TOTAL_SIZE // len(etree.tostring(old_lift, encoding='unicode')))):
    lift = deepcopy( old_lift )

    header = lift.find('header')
    fields = header.find('fields')
    for field in fields.findall('field'):
        form = field.find('form')
        text = form.find('text')
        if text.text:
            text.text = morph_sentence(text.text)

    for entry in lift.findall('entry'):
        for el in entry.getchildren():
            if el.tag == 'sense':
                for el1 in el.getchildren():
                    text = el1.find('text')
                    if text is not None: # если имеется тег <text>
                        if text.text: # если этот тег непустой
                            text.text = morph_sentence(text.text)

                    for form in el1.findall('form'):
                        text1 = form.find('text')
                        if text1 is not None: # если имеется тег <text>
                            if text1.text: # если этот тег непустой
                                text1.text = morph_sentence(text1.text)
            else:
                form = el.find('form')
                if form is not None: # если имеется тег <form>
                    text = form.find('text')
                    if text is not None: # если имеется тег <text>
                        if text.text: # если этот тег непустой
                            text.text = morph_sentence(text.text)
        
    f = codecs.open('new_%03i_%s'%(i+1, filename), 'w', 'utf-8')
    f.write(etree.tostring(lift, encoding='unicode'))
    f.close()