HtmlSteg/steg.py

# -*- coding: utf-8 -*-#

# Copyright 2017 Weber Yann <yann.weber@member.fsf.org>
#
# This file is part of HtmlSteg.
#
#        HtmlSteg 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 3 of the License, or
#        any later version.
#
#        HtmlSteg is distributed in the hope that it will be useful,
#        but WITHOUT ANY WARRANTY; without even the implied warranty of
#        MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#        GNU General Public License for more details.
#
#        You should have received a copy of the GNU General Public License
#        along with HtmlSteg.  If not, see <http://www.gnu.org/licenses/>.
#

import warnings

R_CHARS = [' ', "\t", "\n"]

def clean_html(html):
    '''Clean html squeezing repeating R_CHARS'''
    res = ''
    prev = None
    for c in html:
        if c in R_CHARS and c == prev:
            continue
        prev = c
        res += c
    return res

def hide(html, payload):
    cur_val = payload[0]
    cur_bit = cur_val & 1
    cur_val >>= 1
    cur_shift = 1
    payload = payload[1:]

    end = False
    res = ''
    for c in html:
        if not end and c in R_CHARS:
            if cur_bit == 1:
                res += c
            if cur_shift == 8:
                cur_shift = 0
                if len(payload) == 0:
                    end = True
                else:
                    cur_val = payload[0]
                    payload = payload[1:]
            cur_bit = cur_val & 1
            cur_val >>= 1
            cur_shift += 1
        res += c
    if not end:
        if len(payload) > 0:
            warnings.warn(
                'Unable to hide all payload. %d bytes not hidden : "%s"' % (
                    len(payload), repr(payload)))
        if cur_shift != 8:
            warnings.warn(
                'last byte not complete. %d bits missing' % (8 - cur_shift))
    return res


def unhide(html):
    res = bytearray()
    cur_byte = 0
    cur_shift = 0
    prev = None
    pre_add = False
    for c in html:
        bit_added = False
        if c in R_CHARS and prev == c:
                #1
                cur_byte += 1 << cur_shift
                bit_added = True
                c = None
        elif prev in R_CHARS:
            #0
            bit_added = True
        if bit_added:
            if cur_shift == 7:
                #print("unhide byte ", bytearray([cur_byte]))
                res += bytearray([cur_byte])
                cur_byte = 0
                cur_shift = 0
            else:
                cur_shift += 1
        prev = c
    return res