Changed from DeDRM_WinApp to DeDRM_App

This commit is contained in:
Apprentice Alf
2012-11-01 13:05:26 +00:00
parent 4b3618246b
commit 0028027f71
40 changed files with 0 additions and 0 deletions

View File

@@ -0,0 +1,75 @@
import sys
import Tkinter
import Tkconstants
class ActivityBar(Tkinter.Frame):
def __init__(self, master, length=300, height=20, barwidth=15, interval=50, bg='white', fillcolor='orchid1',\
bd=2, relief=Tkconstants.GROOVE, *args, **kw):
Tkinter.Frame.__init__(self, master, bg=bg, width=length, height=height, *args, **kw)
self._master = master
self._interval = interval
self._maximum = length
self._startx = 0
self._barwidth = barwidth
self._bardiv = length / barwidth
if self._bardiv < 10:
self._bardiv = 10
stopx = self._startx + self._barwidth
if stopx > self._maximum:
stopx = self._maximum
# self._canv = Tkinter.Canvas(self, bg=self['bg'], width=self['width'], height=self['height'],\
# highlightthickness=0, relief='flat', bd=0)
self._canv = Tkinter.Canvas(self, bg=self['bg'], width=self['width'], height=self['height'],\
highlightthickness=0, relief=relief, bd=bd)
self._canv.pack(fill='both', expand=1)
self._rect = self._canv.create_rectangle(0, 0, self._canv.winfo_reqwidth(), self._canv.winfo_reqheight(), fill=fillcolor, width=0)
self._set()
self.bind('<Configure>', self._update_coords)
self._running = False
def _update_coords(self, event):
'''Updates the position of the rectangle inside the canvas when the size of
the widget gets changed.'''
# looks like we have to call update_idletasks() twice to make sure
# to get the results we expect
self._canv.update_idletasks()
self._maximum = self._canv.winfo_width()
self._startx = 0
self._barwidth = self._maximum / self._bardiv
if self._barwidth < 2:
self._barwidth = 2
stopx = self._startx + self._barwidth
if stopx > self._maximum:
stopx = self._maximum
self._canv.coords(self._rect, 0, 0, stopx, self._canv.winfo_height())
self._canv.update_idletasks()
def _set(self):
if self._startx < 0:
self._startx = 0
if self._startx > self._maximum:
self._startx = self._startx % self._maximum
stopx = self._startx + self._barwidth
if stopx > self._maximum:
stopx = self._maximum
self._canv.coords(self._rect, self._startx, 0, stopx, self._canv.winfo_height())
self._canv.update_idletasks()
def start(self):
self._running = True
self.after(self._interval, self._step)
def stop(self):
self._running = False
self._set()
def _step(self):
if self._running:
stepsize = self._barwidth / 4
if stepsize < 2:
stepsize = 2
self._startx += stepsize
self._set()
self.after(self._interval, self._step)

View File

@@ -0,0 +1,568 @@
#! /usr/bin/env python
"""
Routines for doing AES CBC in one file
Modified by some_updates to extract
and combine only those parts needed for AES CBC
into one simple to add python file
Original Version
Copyright (c) 2002 by Paul A. Lambert
Under:
CryptoPy Artisitic License Version 1.0
See the wonderful pure python package cryptopy-1.2.5
and read its LICENSE.txt for complete license details.
"""
class CryptoError(Exception):
""" Base class for crypto exceptions """
def __init__(self,errorMessage='Error!'):
self.message = errorMessage
def __str__(self):
return self.message
class InitCryptoError(CryptoError):
""" Crypto errors during algorithm initialization """
class BadKeySizeError(InitCryptoError):
""" Bad key size error """
class EncryptError(CryptoError):
""" Error in encryption processing """
class DecryptError(CryptoError):
""" Error in decryption processing """
class DecryptNotBlockAlignedError(DecryptError):
""" Error in decryption processing """
def xorS(a,b):
""" XOR two strings """
assert len(a)==len(b)
x = []
for i in range(len(a)):
x.append( chr(ord(a[i])^ord(b[i])))
return ''.join(x)
def xor(a,b):
""" XOR two strings """
x = []
for i in range(min(len(a),len(b))):
x.append( chr(ord(a[i])^ord(b[i])))
return ''.join(x)
"""
Base 'BlockCipher' and Pad classes for cipher instances.
BlockCipher supports automatic padding and type conversion. The BlockCipher
class was written to make the actual algorithm code more readable and
not for performance.
"""
class BlockCipher:
""" Block ciphers """
def __init__(self):
self.reset()
def reset(self):
self.resetEncrypt()
self.resetDecrypt()
def resetEncrypt(self):
self.encryptBlockCount = 0
self.bytesToEncrypt = ''
def resetDecrypt(self):
self.decryptBlockCount = 0
self.bytesToDecrypt = ''
def encrypt(self, plainText, more = None):
""" Encrypt a string and return a binary string """
self.bytesToEncrypt += plainText # append plainText to any bytes from prior encrypt
numBlocks, numExtraBytes = divmod(len(self.bytesToEncrypt), self.blockSize)
cipherText = ''
for i in range(numBlocks):
bStart = i*self.blockSize
ctBlock = self.encryptBlock(self.bytesToEncrypt[bStart:bStart+self.blockSize])
self.encryptBlockCount += 1
cipherText += ctBlock
if numExtraBytes > 0: # save any bytes that are not block aligned
self.bytesToEncrypt = self.bytesToEncrypt[-numExtraBytes:]
else:
self.bytesToEncrypt = ''
if more == None: # no more data expected from caller
finalBytes = self.padding.addPad(self.bytesToEncrypt,self.blockSize)
if len(finalBytes) > 0:
ctBlock = self.encryptBlock(finalBytes)
self.encryptBlockCount += 1
cipherText += ctBlock
self.resetEncrypt()
return cipherText
def decrypt(self, cipherText, more = None):
""" Decrypt a string and return a string """
self.bytesToDecrypt += cipherText # append to any bytes from prior decrypt
numBlocks, numExtraBytes = divmod(len(self.bytesToDecrypt), self.blockSize)
if more == None: # no more calls to decrypt, should have all the data
if numExtraBytes != 0:
raise DecryptNotBlockAlignedError, 'Data not block aligned on decrypt'
# hold back some bytes in case last decrypt has zero len
if (more != None) and (numExtraBytes == 0) and (numBlocks >0) :
numBlocks -= 1
numExtraBytes = self.blockSize
plainText = ''
for i in range(numBlocks):
bStart = i*self.blockSize
ptBlock = self.decryptBlock(self.bytesToDecrypt[bStart : bStart+self.blockSize])
self.decryptBlockCount += 1
plainText += ptBlock
if numExtraBytes > 0: # save any bytes that are not block aligned
self.bytesToEncrypt = self.bytesToEncrypt[-numExtraBytes:]
else:
self.bytesToEncrypt = ''
if more == None: # last decrypt remove padding
plainText = self.padding.removePad(plainText, self.blockSize)
self.resetDecrypt()
return plainText
class Pad:
def __init__(self):
pass # eventually could put in calculation of min and max size extension
class padWithPadLen(Pad):
""" Pad a binary string with the length of the padding """
def addPad(self, extraBytes, blockSize):
""" Add padding to a binary string to make it an even multiple
of the block size """
blocks, numExtraBytes = divmod(len(extraBytes), blockSize)
padLength = blockSize - numExtraBytes
return extraBytes + padLength*chr(padLength)
def removePad(self, paddedBinaryString, blockSize):
""" Remove padding from a binary string """
if not(0<len(paddedBinaryString)):
raise DecryptNotBlockAlignedError, 'Expected More Data'
return paddedBinaryString[:-ord(paddedBinaryString[-1])]
class noPadding(Pad):
""" No padding. Use this to get ECB behavior from encrypt/decrypt """
def addPad(self, extraBytes, blockSize):
""" Add no padding """
return extraBytes
def removePad(self, paddedBinaryString, blockSize):
""" Remove no padding """
return paddedBinaryString
"""
Rijndael encryption algorithm
This byte oriented implementation is intended to closely
match FIPS specification for readability. It is not implemented
for performance.
"""
class Rijndael(BlockCipher):
""" Rijndael encryption algorithm """
def __init__(self, key = None, padding = padWithPadLen(), keySize=16, blockSize=16 ):
self.name = 'RIJNDAEL'
self.keySize = keySize
self.strength = keySize*8
self.blockSize = blockSize # blockSize is in bytes
self.padding = padding # change default to noPadding() to get normal ECB behavior
assert( keySize%4==0 and NrTable[4].has_key(keySize/4)),'key size must be 16,20,24,29 or 32 bytes'
assert( blockSize%4==0 and NrTable.has_key(blockSize/4)), 'block size must be 16,20,24,29 or 32 bytes'
self.Nb = self.blockSize/4 # Nb is number of columns of 32 bit words
self.Nk = keySize/4 # Nk is the key length in 32-bit words
self.Nr = NrTable[self.Nb][self.Nk] # The number of rounds (Nr) is a function of
# the block (Nb) and key (Nk) sizes.
if key != None:
self.setKey(key)
def setKey(self, key):
""" Set a key and generate the expanded key """
assert( len(key) == (self.Nk*4) ), 'Key length must be same as keySize parameter'
self.__expandedKey = keyExpansion(self, key)
self.reset() # BlockCipher.reset()
def encryptBlock(self, plainTextBlock):
""" Encrypt a block, plainTextBlock must be a array of bytes [Nb by 4] """
self.state = self._toBlock(plainTextBlock)
AddRoundKey(self, self.__expandedKey[0:self.Nb])
for round in range(1,self.Nr): #for round = 1 step 1 to Nr
SubBytes(self)
ShiftRows(self)
MixColumns(self)
AddRoundKey(self, self.__expandedKey[round*self.Nb:(round+1)*self.Nb])
SubBytes(self)
ShiftRows(self)
AddRoundKey(self, self.__expandedKey[self.Nr*self.Nb:(self.Nr+1)*self.Nb])
return self._toBString(self.state)
def decryptBlock(self, encryptedBlock):
""" decrypt a block (array of bytes) """
self.state = self._toBlock(encryptedBlock)
AddRoundKey(self, self.__expandedKey[self.Nr*self.Nb:(self.Nr+1)*self.Nb])
for round in range(self.Nr-1,0,-1):
InvShiftRows(self)
InvSubBytes(self)
AddRoundKey(self, self.__expandedKey[round*self.Nb:(round+1)*self.Nb])
InvMixColumns(self)
InvShiftRows(self)
InvSubBytes(self)
AddRoundKey(self, self.__expandedKey[0:self.Nb])
return self._toBString(self.state)
def _toBlock(self, bs):
""" Convert binary string to array of bytes, state[col][row]"""
assert ( len(bs) == 4*self.Nb ), 'Rijndarl blocks must be of size blockSize'
return [[ord(bs[4*i]),ord(bs[4*i+1]),ord(bs[4*i+2]),ord(bs[4*i+3])] for i in range(self.Nb)]
def _toBString(self, block):
""" Convert block (array of bytes) to binary string """
l = []
for col in block:
for rowElement in col:
l.append(chr(rowElement))
return ''.join(l)
#-------------------------------------
""" Number of rounds Nr = NrTable[Nb][Nk]
Nb Nk=4 Nk=5 Nk=6 Nk=7 Nk=8
------------------------------------- """
NrTable = {4: {4:10, 5:11, 6:12, 7:13, 8:14},
5: {4:11, 5:11, 6:12, 7:13, 8:14},
6: {4:12, 5:12, 6:12, 7:13, 8:14},
7: {4:13, 5:13, 6:13, 7:13, 8:14},
8: {4:14, 5:14, 6:14, 7:14, 8:14}}
#-------------------------------------
def keyExpansion(algInstance, keyString):
""" Expand a string of size keySize into a larger array """
Nk, Nb, Nr = algInstance.Nk, algInstance.Nb, algInstance.Nr # for readability
key = [ord(byte) for byte in keyString] # convert string to list
w = [[key[4*i],key[4*i+1],key[4*i+2],key[4*i+3]] for i in range(Nk)]
for i in range(Nk,Nb*(Nr+1)):
temp = w[i-1] # a four byte column
if (i%Nk) == 0 :
temp = temp[1:]+[temp[0]] # RotWord(temp)
temp = [ Sbox[byte] for byte in temp ]
temp[0] ^= Rcon[i/Nk]
elif Nk > 6 and i%Nk == 4 :
temp = [ Sbox[byte] for byte in temp ] # SubWord(temp)
w.append( [ w[i-Nk][byte]^temp[byte] for byte in range(4) ] )
return w
Rcon = (0,0x01,0x02,0x04,0x08,0x10,0x20,0x40,0x80,0x1b,0x36, # note extra '0' !!!
0x6c,0xd8,0xab,0x4d,0x9a,0x2f,0x5e,0xbc,0x63,0xc6,
0x97,0x35,0x6a,0xd4,0xb3,0x7d,0xfa,0xef,0xc5,0x91)
#-------------------------------------
def AddRoundKey(algInstance, keyBlock):
""" XOR the algorithm state with a block of key material """
for column in range(algInstance.Nb):
for row in range(4):
algInstance.state[column][row] ^= keyBlock[column][row]
#-------------------------------------
def SubBytes(algInstance):
for column in range(algInstance.Nb):
for row in range(4):
algInstance.state[column][row] = Sbox[algInstance.state[column][row]]
def InvSubBytes(algInstance):
for column in range(algInstance.Nb):
for row in range(4):
algInstance.state[column][row] = InvSbox[algInstance.state[column][row]]
Sbox = (0x63,0x7c,0x77,0x7b,0xf2,0x6b,0x6f,0xc5,
0x30,0x01,0x67,0x2b,0xfe,0xd7,0xab,0x76,
0xca,0x82,0xc9,0x7d,0xfa,0x59,0x47,0xf0,
0xad,0xd4,0xa2,0xaf,0x9c,0xa4,0x72,0xc0,
0xb7,0xfd,0x93,0x26,0x36,0x3f,0xf7,0xcc,
0x34,0xa5,0xe5,0xf1,0x71,0xd8,0x31,0x15,
0x04,0xc7,0x23,0xc3,0x18,0x96,0x05,0x9a,
0x07,0x12,0x80,0xe2,0xeb,0x27,0xb2,0x75,
0x09,0x83,0x2c,0x1a,0x1b,0x6e,0x5a,0xa0,
0x52,0x3b,0xd6,0xb3,0x29,0xe3,0x2f,0x84,
0x53,0xd1,0x00,0xed,0x20,0xfc,0xb1,0x5b,
0x6a,0xcb,0xbe,0x39,0x4a,0x4c,0x58,0xcf,
0xd0,0xef,0xaa,0xfb,0x43,0x4d,0x33,0x85,
0x45,0xf9,0x02,0x7f,0x50,0x3c,0x9f,0xa8,
0x51,0xa3,0x40,0x8f,0x92,0x9d,0x38,0xf5,
0xbc,0xb6,0xda,0x21,0x10,0xff,0xf3,0xd2,
0xcd,0x0c,0x13,0xec,0x5f,0x97,0x44,0x17,
0xc4,0xa7,0x7e,0x3d,0x64,0x5d,0x19,0x73,
0x60,0x81,0x4f,0xdc,0x22,0x2a,0x90,0x88,
0x46,0xee,0xb8,0x14,0xde,0x5e,0x0b,0xdb,
0xe0,0x32,0x3a,0x0a,0x49,0x06,0x24,0x5c,
0xc2,0xd3,0xac,0x62,0x91,0x95,0xe4,0x79,
0xe7,0xc8,0x37,0x6d,0x8d,0xd5,0x4e,0xa9,
0x6c,0x56,0xf4,0xea,0x65,0x7a,0xae,0x08,
0xba,0x78,0x25,0x2e,0x1c,0xa6,0xb4,0xc6,
0xe8,0xdd,0x74,0x1f,0x4b,0xbd,0x8b,0x8a,
0x70,0x3e,0xb5,0x66,0x48,0x03,0xf6,0x0e,
0x61,0x35,0x57,0xb9,0x86,0xc1,0x1d,0x9e,
0xe1,0xf8,0x98,0x11,0x69,0xd9,0x8e,0x94,
0x9b,0x1e,0x87,0xe9,0xce,0x55,0x28,0xdf,
0x8c,0xa1,0x89,0x0d,0xbf,0xe6,0x42,0x68,
0x41,0x99,0x2d,0x0f,0xb0,0x54,0xbb,0x16)
InvSbox = (0x52,0x09,0x6a,0xd5,0x30,0x36,0xa5,0x38,
0xbf,0x40,0xa3,0x9e,0x81,0xf3,0xd7,0xfb,
0x7c,0xe3,0x39,0x82,0x9b,0x2f,0xff,0x87,
0x34,0x8e,0x43,0x44,0xc4,0xde,0xe9,0xcb,
0x54,0x7b,0x94,0x32,0xa6,0xc2,0x23,0x3d,
0xee,0x4c,0x95,0x0b,0x42,0xfa,0xc3,0x4e,
0x08,0x2e,0xa1,0x66,0x28,0xd9,0x24,0xb2,
0x76,0x5b,0xa2,0x49,0x6d,0x8b,0xd1,0x25,
0x72,0xf8,0xf6,0x64,0x86,0x68,0x98,0x16,
0xd4,0xa4,0x5c,0xcc,0x5d,0x65,0xb6,0x92,
0x6c,0x70,0x48,0x50,0xfd,0xed,0xb9,0xda,
0x5e,0x15,0x46,0x57,0xa7,0x8d,0x9d,0x84,
0x90,0xd8,0xab,0x00,0x8c,0xbc,0xd3,0x0a,
0xf7,0xe4,0x58,0x05,0xb8,0xb3,0x45,0x06,
0xd0,0x2c,0x1e,0x8f,0xca,0x3f,0x0f,0x02,
0xc1,0xaf,0xbd,0x03,0x01,0x13,0x8a,0x6b,
0x3a,0x91,0x11,0x41,0x4f,0x67,0xdc,0xea,
0x97,0xf2,0xcf,0xce,0xf0,0xb4,0xe6,0x73,
0x96,0xac,0x74,0x22,0xe7,0xad,0x35,0x85,
0xe2,0xf9,0x37,0xe8,0x1c,0x75,0xdf,0x6e,
0x47,0xf1,0x1a,0x71,0x1d,0x29,0xc5,0x89,
0x6f,0xb7,0x62,0x0e,0xaa,0x18,0xbe,0x1b,
0xfc,0x56,0x3e,0x4b,0xc6,0xd2,0x79,0x20,
0x9a,0xdb,0xc0,0xfe,0x78,0xcd,0x5a,0xf4,
0x1f,0xdd,0xa8,0x33,0x88,0x07,0xc7,0x31,
0xb1,0x12,0x10,0x59,0x27,0x80,0xec,0x5f,
0x60,0x51,0x7f,0xa9,0x19,0xb5,0x4a,0x0d,
0x2d,0xe5,0x7a,0x9f,0x93,0xc9,0x9c,0xef,
0xa0,0xe0,0x3b,0x4d,0xae,0x2a,0xf5,0xb0,
0xc8,0xeb,0xbb,0x3c,0x83,0x53,0x99,0x61,
0x17,0x2b,0x04,0x7e,0xba,0x77,0xd6,0x26,
0xe1,0x69,0x14,0x63,0x55,0x21,0x0c,0x7d)
#-------------------------------------
""" For each block size (Nb), the ShiftRow operation shifts row i
by the amount Ci. Note that row 0 is not shifted.
Nb C1 C2 C3
------------------- """
shiftOffset = { 4 : ( 0, 1, 2, 3),
5 : ( 0, 1, 2, 3),
6 : ( 0, 1, 2, 3),
7 : ( 0, 1, 2, 4),
8 : ( 0, 1, 3, 4) }
def ShiftRows(algInstance):
tmp = [0]*algInstance.Nb # list of size Nb
for r in range(1,4): # row 0 reamains unchanged and can be skipped
for c in range(algInstance.Nb):
tmp[c] = algInstance.state[(c+shiftOffset[algInstance.Nb][r]) % algInstance.Nb][r]
for c in range(algInstance.Nb):
algInstance.state[c][r] = tmp[c]
def InvShiftRows(algInstance):
tmp = [0]*algInstance.Nb # list of size Nb
for r in range(1,4): # row 0 reamains unchanged and can be skipped
for c in range(algInstance.Nb):
tmp[c] = algInstance.state[(c+algInstance.Nb-shiftOffset[algInstance.Nb][r]) % algInstance.Nb][r]
for c in range(algInstance.Nb):
algInstance.state[c][r] = tmp[c]
#-------------------------------------
def MixColumns(a):
Sprime = [0,0,0,0]
for j in range(a.Nb): # for each column
Sprime[0] = mul(2,a.state[j][0])^mul(3,a.state[j][1])^mul(1,a.state[j][2])^mul(1,a.state[j][3])
Sprime[1] = mul(1,a.state[j][0])^mul(2,a.state[j][1])^mul(3,a.state[j][2])^mul(1,a.state[j][3])
Sprime[2] = mul(1,a.state[j][0])^mul(1,a.state[j][1])^mul(2,a.state[j][2])^mul(3,a.state[j][3])
Sprime[3] = mul(3,a.state[j][0])^mul(1,a.state[j][1])^mul(1,a.state[j][2])^mul(2,a.state[j][3])
for i in range(4):
a.state[j][i] = Sprime[i]
def InvMixColumns(a):
""" Mix the four bytes of every column in a linear way
This is the opposite operation of Mixcolumn """
Sprime = [0,0,0,0]
for j in range(a.Nb): # for each column
Sprime[0] = mul(0x0E,a.state[j][0])^mul(0x0B,a.state[j][1])^mul(0x0D,a.state[j][2])^mul(0x09,a.state[j][3])
Sprime[1] = mul(0x09,a.state[j][0])^mul(0x0E,a.state[j][1])^mul(0x0B,a.state[j][2])^mul(0x0D,a.state[j][3])
Sprime[2] = mul(0x0D,a.state[j][0])^mul(0x09,a.state[j][1])^mul(0x0E,a.state[j][2])^mul(0x0B,a.state[j][3])
Sprime[3] = mul(0x0B,a.state[j][0])^mul(0x0D,a.state[j][1])^mul(0x09,a.state[j][2])^mul(0x0E,a.state[j][3])
for i in range(4):
a.state[j][i] = Sprime[i]
#-------------------------------------
def mul(a, b):
""" Multiply two elements of GF(2^m)
needed for MixColumn and InvMixColumn """
if (a !=0 and b!=0):
return Alogtable[(Logtable[a] + Logtable[b])%255]
else:
return 0
Logtable = ( 0, 0, 25, 1, 50, 2, 26, 198, 75, 199, 27, 104, 51, 238, 223, 3,
100, 4, 224, 14, 52, 141, 129, 239, 76, 113, 8, 200, 248, 105, 28, 193,
125, 194, 29, 181, 249, 185, 39, 106, 77, 228, 166, 114, 154, 201, 9, 120,
101, 47, 138, 5, 33, 15, 225, 36, 18, 240, 130, 69, 53, 147, 218, 142,
150, 143, 219, 189, 54, 208, 206, 148, 19, 92, 210, 241, 64, 70, 131, 56,
102, 221, 253, 48, 191, 6, 139, 98, 179, 37, 226, 152, 34, 136, 145, 16,
126, 110, 72, 195, 163, 182, 30, 66, 58, 107, 40, 84, 250, 133, 61, 186,
43, 121, 10, 21, 155, 159, 94, 202, 78, 212, 172, 229, 243, 115, 167, 87,
175, 88, 168, 80, 244, 234, 214, 116, 79, 174, 233, 213, 231, 230, 173, 232,
44, 215, 117, 122, 235, 22, 11, 245, 89, 203, 95, 176, 156, 169, 81, 160,
127, 12, 246, 111, 23, 196, 73, 236, 216, 67, 31, 45, 164, 118, 123, 183,
204, 187, 62, 90, 251, 96, 177, 134, 59, 82, 161, 108, 170, 85, 41, 157,
151, 178, 135, 144, 97, 190, 220, 252, 188, 149, 207, 205, 55, 63, 91, 209,
83, 57, 132, 60, 65, 162, 109, 71, 20, 42, 158, 93, 86, 242, 211, 171,
68, 17, 146, 217, 35, 32, 46, 137, 180, 124, 184, 38, 119, 153, 227, 165,
103, 74, 237, 222, 197, 49, 254, 24, 13, 99, 140, 128, 192, 247, 112, 7)
Alogtable= ( 1, 3, 5, 15, 17, 51, 85, 255, 26, 46, 114, 150, 161, 248, 19, 53,
95, 225, 56, 72, 216, 115, 149, 164, 247, 2, 6, 10, 30, 34, 102, 170,
229, 52, 92, 228, 55, 89, 235, 38, 106, 190, 217, 112, 144, 171, 230, 49,
83, 245, 4, 12, 20, 60, 68, 204, 79, 209, 104, 184, 211, 110, 178, 205,
76, 212, 103, 169, 224, 59, 77, 215, 98, 166, 241, 8, 24, 40, 120, 136,
131, 158, 185, 208, 107, 189, 220, 127, 129, 152, 179, 206, 73, 219, 118, 154,
181, 196, 87, 249, 16, 48, 80, 240, 11, 29, 39, 105, 187, 214, 97, 163,
254, 25, 43, 125, 135, 146, 173, 236, 47, 113, 147, 174, 233, 32, 96, 160,
251, 22, 58, 78, 210, 109, 183, 194, 93, 231, 50, 86, 250, 21, 63, 65,
195, 94, 226, 61, 71, 201, 64, 192, 91, 237, 44, 116, 156, 191, 218, 117,
159, 186, 213, 100, 172, 239, 42, 126, 130, 157, 188, 223, 122, 142, 137, 128,
155, 182, 193, 88, 232, 35, 101, 175, 234, 37, 111, 177, 200, 67, 197, 84,
252, 31, 33, 99, 165, 244, 7, 9, 27, 45, 119, 153, 176, 203, 70, 202,
69, 207, 74, 222, 121, 139, 134, 145, 168, 227, 62, 66, 198, 81, 243, 14,
18, 54, 90, 238, 41, 123, 141, 140, 143, 138, 133, 148, 167, 242, 13, 23,
57, 75, 221, 124, 132, 151, 162, 253, 28, 36, 108, 180, 199, 82, 246, 1)
"""
AES Encryption Algorithm
The AES algorithm is just Rijndael algorithm restricted to the default
blockSize of 128 bits.
"""
class AES(Rijndael):
""" The AES algorithm is the Rijndael block cipher restricted to block
sizes of 128 bits and key sizes of 128, 192 or 256 bits
"""
def __init__(self, key = None, padding = padWithPadLen(), keySize=16):
""" Initialize AES, keySize is in bytes """
if not (keySize == 16 or keySize == 24 or keySize == 32) :
raise BadKeySizeError, 'Illegal AES key size, must be 16, 24, or 32 bytes'
Rijndael.__init__( self, key, padding=padding, keySize=keySize, blockSize=16 )
self.name = 'AES'
"""
CBC mode of encryption for block ciphers.
This algorithm mode wraps any BlockCipher to make a
Cipher Block Chaining mode.
"""
from random import Random # should change to crypto.random!!!
class CBC(BlockCipher):
""" The CBC class wraps block ciphers to make cipher block chaining (CBC) mode
algorithms. The initialization (IV) is automatic if set to None. Padding
is also automatic based on the Pad class used to initialize the algorithm
"""
def __init__(self, blockCipherInstance, padding = padWithPadLen()):
""" CBC algorithms are created by initializing with a BlockCipher instance """
self.baseCipher = blockCipherInstance
self.name = self.baseCipher.name + '_CBC'
self.blockSize = self.baseCipher.blockSize
self.keySize = self.baseCipher.keySize
self.padding = padding
self.baseCipher.padding = noPadding() # baseCipher should NOT pad!!
self.r = Random() # for IV generation, currently uses
# mediocre standard distro version <----------------
import time
newSeed = time.ctime()+str(self.r) # seed with instance location
self.r.seed(newSeed) # to make unique
self.reset()
def setKey(self, key):
self.baseCipher.setKey(key)
# Overload to reset both CBC state and the wrapped baseCipher
def resetEncrypt(self):
BlockCipher.resetEncrypt(self) # reset CBC encrypt state (super class)
self.baseCipher.resetEncrypt() # reset base cipher encrypt state
def resetDecrypt(self):
BlockCipher.resetDecrypt(self) # reset CBC state (super class)
self.baseCipher.resetDecrypt() # reset base cipher decrypt state
def encrypt(self, plainText, iv=None, more=None):
""" CBC encryption - overloads baseCipher to allow optional explicit IV
when iv=None, iv is auto generated!
"""
if self.encryptBlockCount == 0:
self.iv = iv
else:
assert(iv==None), 'IV used only on first call to encrypt'
return BlockCipher.encrypt(self,plainText, more=more)
def decrypt(self, cipherText, iv=None, more=None):
""" CBC decryption - overloads baseCipher to allow optional explicit IV
when iv=None, iv is auto generated!
"""
if self.decryptBlockCount == 0:
self.iv = iv
else:
assert(iv==None), 'IV used only on first call to decrypt'
return BlockCipher.decrypt(self, cipherText, more=more)
def encryptBlock(self, plainTextBlock):
""" CBC block encryption, IV is set with 'encrypt' """
auto_IV = ''
if self.encryptBlockCount == 0:
if self.iv == None:
# generate IV and use
self.iv = ''.join([chr(self.r.randrange(256)) for i in range(self.blockSize)])
self.prior_encr_CT_block = self.iv
auto_IV = self.prior_encr_CT_block # prepend IV if it's automatic
else: # application provided IV
assert(len(self.iv) == self.blockSize ),'IV must be same length as block'
self.prior_encr_CT_block = self.iv
""" encrypt the prior CT XORed with the PT """
ct = self.baseCipher.encryptBlock( xor(self.prior_encr_CT_block, plainTextBlock) )
self.prior_encr_CT_block = ct
return auto_IV+ct
def decryptBlock(self, encryptedBlock):
""" Decrypt a single block """
if self.decryptBlockCount == 0: # first call, process IV
if self.iv == None: # auto decrypt IV?
self.prior_CT_block = encryptedBlock
return ''
else:
assert(len(self.iv)==self.blockSize),"Bad IV size on CBC decryption"
self.prior_CT_block = self.iv
dct = self.baseCipher.decryptBlock(encryptedBlock)
""" XOR the prior decrypted CT with the prior CT """
dct_XOR_priorCT = xor( self.prior_CT_block, dct )
self.prior_CT_block = encryptedBlock
return dct_XOR_priorCT
"""
AES_CBC Encryption Algorithm
"""
class AES_CBC(CBC):
""" AES encryption in CBC feedback mode """
def __init__(self, key=None, padding=padWithPadLen(), keySize=16):
CBC.__init__( self, AES(key, noPadding(), keySize), padding)
self.name = 'AES_CBC'

View File

@@ -0,0 +1,290 @@
#! /usr/bin/env python
import sys, os
import hmac
from struct import pack
import hashlib
# interface to needed routines libalfcrypto
def _load_libalfcrypto():
import ctypes
from ctypes import CDLL, byref, POINTER, c_void_p, c_char_p, c_int, c_long, \
Structure, c_ulong, create_string_buffer, addressof, string_at, cast, sizeof
pointer_size = ctypes.sizeof(ctypes.c_voidp)
name_of_lib = None
if sys.platform.startswith('darwin'):
name_of_lib = 'libalfcrypto.dylib'
elif sys.platform.startswith('win'):
if pointer_size == 4:
name_of_lib = 'alfcrypto.dll'
else:
name_of_lib = 'alfcrypto64.dll'
else:
if pointer_size == 4:
name_of_lib = 'libalfcrypto32.so'
else:
name_of_lib = 'libalfcrypto64.so'
libalfcrypto = sys.path[0] + os.sep + name_of_lib
if not os.path.isfile(libalfcrypto):
raise Exception('libalfcrypto not found')
libalfcrypto = CDLL(libalfcrypto)
c_char_pp = POINTER(c_char_p)
c_int_p = POINTER(c_int)
def F(restype, name, argtypes):
func = getattr(libalfcrypto, name)
func.restype = restype
func.argtypes = argtypes
return func
# aes cbc decryption
#
# struct aes_key_st {
# unsigned long rd_key[4 *(AES_MAXNR + 1)];
# int rounds;
# };
#
# typedef struct aes_key_st AES_KEY;
#
# int AES_set_decrypt_key(const unsigned char *userKey, const int bits, AES_KEY *key);
#
#
# void AES_cbc_encrypt(const unsigned char *in, unsigned char *out,
# const unsigned long length, const AES_KEY *key,
# unsigned char *ivec, const int enc);
AES_MAXNR = 14
class AES_KEY(Structure):
_fields_ = [('rd_key', c_long * (4 * (AES_MAXNR + 1))), ('rounds', c_int)]
AES_KEY_p = POINTER(AES_KEY)
AES_cbc_encrypt = F(None, 'AES_cbc_encrypt',[c_char_p, c_char_p, c_ulong, AES_KEY_p, c_char_p, c_int])
AES_set_decrypt_key = F(c_int, 'AES_set_decrypt_key',[c_char_p, c_int, AES_KEY_p])
# Pukall 1 Cipher
# unsigned char *PC1(const unsigned char *key, unsigned int klen, const unsigned char *src,
# unsigned char *dest, unsigned int len, int decryption);
PC1 = F(c_char_p, 'PC1', [c_char_p, c_ulong, c_char_p, c_char_p, c_ulong, c_ulong])
# Topaz Encryption
# typedef struct _TpzCtx {
# unsigned int v[2];
# } TpzCtx;
#
# void topazCryptoInit(TpzCtx *ctx, const unsigned char *key, int klen);
# void topazCryptoDecrypt(const TpzCtx *ctx, const unsigned char *in, unsigned char *out, int len);
class TPZ_CTX(Structure):
_fields_ = [('v', c_long * 2)]
TPZ_CTX_p = POINTER(TPZ_CTX)
topazCryptoInit = F(None, 'topazCryptoInit', [TPZ_CTX_p, c_char_p, c_ulong])
topazCryptoDecrypt = F(None, 'topazCryptoDecrypt', [TPZ_CTX_p, c_char_p, c_char_p, c_ulong])
class AES_CBC(object):
def __init__(self):
self._blocksize = 0
self._keyctx = None
self._iv = 0
def set_decrypt_key(self, userkey, iv):
self._blocksize = len(userkey)
if (self._blocksize != 16) and (self._blocksize != 24) and (self._blocksize != 32) :
raise Exception('AES CBC improper key used')
return
keyctx = self._keyctx = AES_KEY()
self._iv = iv
rv = AES_set_decrypt_key(userkey, len(userkey) * 8, keyctx)
if rv < 0:
raise Exception('Failed to initialize AES CBC key')
def decrypt(self, data):
out = create_string_buffer(len(data))
mutable_iv = create_string_buffer(self._iv, len(self._iv))
rv = AES_cbc_encrypt(data, out, len(data), self._keyctx, mutable_iv, 0)
if rv == 0:
raise Exception('AES CBC decryption failed')
return out.raw
class Pukall_Cipher(object):
def __init__(self):
self.key = None
def PC1(self, key, src, decryption=True):
self.key = key
out = create_string_buffer(len(src))
de = 0
if decryption:
de = 1
rv = PC1(key, len(key), src, out, len(src), de)
return out.raw
class Topaz_Cipher(object):
def __init__(self):
self._ctx = None
def ctx_init(self, key):
tpz_ctx = self._ctx = TPZ_CTX()
topazCryptoInit(tpz_ctx, key, len(key))
return tpz_ctx
def decrypt(self, data, ctx=None):
if ctx == None:
ctx = self._ctx
out = create_string_buffer(len(data))
topazCryptoDecrypt(ctx, data, out, len(data))
return out.raw
print "Using Library AlfCrypto DLL/DYLIB/SO"
return (AES_CBC, Pukall_Cipher, Topaz_Cipher)
def _load_python_alfcrypto():
import aescbc
class Pukall_Cipher(object):
def __init__(self):
self.key = None
def PC1(self, key, src, decryption=True):
sum1 = 0;
sum2 = 0;
keyXorVal = 0;
if len(key)!=16:
print "Bad key length!"
return None
wkey = []
for i in xrange(8):
wkey.append(ord(key[i*2])<<8 | ord(key[i*2+1]))
dst = ""
for i in xrange(len(src)):
temp1 = 0;
byteXorVal = 0;
for j in xrange(8):
temp1 ^= wkey[j]
sum2 = (sum2+j)*20021 + sum1
sum1 = (temp1*346)&0xFFFF
sum2 = (sum2+sum1)&0xFFFF
temp1 = (temp1*20021+1)&0xFFFF
byteXorVal ^= temp1 ^ sum2
curByte = ord(src[i])
if not decryption:
keyXorVal = curByte * 257;
curByte = ((curByte ^ (byteXorVal >> 8)) ^ byteXorVal) & 0xFF
if decryption:
keyXorVal = curByte * 257;
for j in xrange(8):
wkey[j] ^= keyXorVal;
dst+=chr(curByte)
return dst
class Topaz_Cipher(object):
def __init__(self):
self._ctx = None
def ctx_init(self, key):
ctx1 = 0x0CAFFE19E
for keyChar in key:
keyByte = ord(keyChar)
ctx2 = ctx1
ctx1 = ((((ctx1 >>2) * (ctx1 >>7))&0xFFFFFFFF) ^ (keyByte * keyByte * 0x0F902007)& 0xFFFFFFFF )
self._ctx = [ctx1, ctx2]
return [ctx1,ctx2]
def decrypt(self, data, ctx=None):
if ctx == None:
ctx = self._ctx
ctx1 = ctx[0]
ctx2 = ctx[1]
plainText = ""
for dataChar in data:
dataByte = ord(dataChar)
m = (dataByte ^ ((ctx1 >> 3) &0xFF) ^ ((ctx2<<3) & 0xFF)) &0xFF
ctx2 = ctx1
ctx1 = (((ctx1 >> 2) * (ctx1 >> 7)) &0xFFFFFFFF) ^((m * m * 0x0F902007) &0xFFFFFFFF)
plainText += chr(m)
return plainText
class AES_CBC(object):
def __init__(self):
self._key = None
self._iv = None
self.aes = None
def set_decrypt_key(self, userkey, iv):
self._key = userkey
self._iv = iv
self.aes = aescbc.AES_CBC(userkey, aescbc.noPadding(), len(userkey))
def decrypt(self, data):
iv = self._iv
cleartext = self.aes.decrypt(iv + data)
return cleartext
return (AES_CBC, Pukall_Cipher, Topaz_Cipher)
def _load_crypto():
AES_CBC = Pukall_Cipher = Topaz_Cipher = None
cryptolist = (_load_libalfcrypto, _load_python_alfcrypto)
for loader in cryptolist:
try:
AES_CBC, Pukall_Cipher, Topaz_Cipher = loader()
break
except (ImportError, Exception):
pass
return AES_CBC, Pukall_Cipher, Topaz_Cipher
AES_CBC, Pukall_Cipher, Topaz_Cipher = _load_crypto()
class KeyIVGen(object):
# this only exists in openssl so we will use pure python implementation instead
# PKCS5_PBKDF2_HMAC_SHA1 = F(c_int, 'PKCS5_PBKDF2_HMAC_SHA1',
# [c_char_p, c_ulong, c_char_p, c_ulong, c_ulong, c_ulong, c_char_p])
def pbkdf2(self, passwd, salt, iter, keylen):
def xorstr( a, b ):
if len(a) != len(b):
raise Exception("xorstr(): lengths differ")
return ''.join((chr(ord(x)^ord(y)) for x, y in zip(a, b)))
def prf( h, data ):
hm = h.copy()
hm.update( data )
return hm.digest()
def pbkdf2_F( h, salt, itercount, blocknum ):
U = prf( h, salt + pack('>i',blocknum ) )
T = U
for i in range(2, itercount+1):
U = prf( h, U )
T = xorstr( T, U )
return T
sha = hashlib.sha1
digest_size = sha().digest_size
# l - number of output blocks to produce
l = keylen / digest_size
if keylen % digest_size != 0:
l += 1
h = hmac.new( passwd, None, sha )
T = ""
for i in range(1, l+1):
T += pbkdf2_F( h, salt, iter, i )
return T[0: keylen]

View File

@@ -0,0 +1,899 @@
#! /usr/bin/python
"""
Comprehensive Mazama Book DRM with Topaz Cryptography V2.2
-----BEGIN PUBLIC KEY-----
MIGfMA0GCSqGSIb3DQEBAQUAA4GNADCBiQKBgQDdBHJ4CNc6DNFCw4MRCw4SWAK6
M8hYfnNEI0yQmn5Ti+W8biT7EatpauE/5jgQMPBmdNrDr1hbHyHBSP7xeC2qlRWC
B62UCxeu/fpfnvNHDN/wPWWH4jynZ2M6cdcnE5LQ+FfeKqZn7gnG2No1U9h7oOHx
y2/pHuYme7U1TsgSjwIDAQAB
-----END PUBLIC KEY-----
"""
from __future__ import with_statement
import csv
import sys
import os
import getopt
import zlib
from struct import pack
from struct import unpack
from ctypes import windll, c_char_p, c_wchar_p, c_uint, POINTER, byref, \
create_unicode_buffer, create_string_buffer, CFUNCTYPE, addressof, \
string_at, Structure, c_void_p, cast
import _winreg as winreg
import Tkinter
import Tkconstants
import tkMessageBox
import traceback
import hashlib
MAX_PATH = 255
kernel32 = windll.kernel32
advapi32 = windll.advapi32
crypt32 = windll.crypt32
global kindleDatabase
global bookFile
global bookPayloadOffset
global bookHeaderRecords
global bookMetadata
global bookKey
global command
#
# Various character maps used to decrypt books. Probably supposed to act as obfuscation
#
charMap1 = "n5Pr6St7Uv8Wx9YzAb0Cd1Ef2Gh3Jk4M"
charMap2 = "AaZzB0bYyCc1XxDdW2wEeVv3FfUuG4g-TtHh5SsIiR6rJjQq7KkPpL8lOoMm9Nn_"
charMap3 = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"
charMap4 = "ABCDEFGHIJKLMNPQRSTUVWXYZ123456789"
#
# Exceptions for all the problems that might happen during the script
#
class CMBDTCError(Exception):
pass
class CMBDTCFatal(Exception):
pass
#
# Stolen stuff
#
class DataBlob(Structure):
_fields_ = [('cbData', c_uint),
('pbData', c_void_p)]
DataBlob_p = POINTER(DataBlob)
def GetSystemDirectory():
GetSystemDirectoryW = kernel32.GetSystemDirectoryW
GetSystemDirectoryW.argtypes = [c_wchar_p, c_uint]
GetSystemDirectoryW.restype = c_uint
def GetSystemDirectory():
buffer = create_unicode_buffer(MAX_PATH + 1)
GetSystemDirectoryW(buffer, len(buffer))
return buffer.value
return GetSystemDirectory
GetSystemDirectory = GetSystemDirectory()
def GetVolumeSerialNumber():
GetVolumeInformationW = kernel32.GetVolumeInformationW
GetVolumeInformationW.argtypes = [c_wchar_p, c_wchar_p, c_uint,
POINTER(c_uint), POINTER(c_uint),
POINTER(c_uint), c_wchar_p, c_uint]
GetVolumeInformationW.restype = c_uint
def GetVolumeSerialNumber(path):
vsn = c_uint(0)
GetVolumeInformationW(path, None, 0, byref(vsn), None, None, None, 0)
return vsn.value
return GetVolumeSerialNumber
GetVolumeSerialNumber = GetVolumeSerialNumber()
def GetUserName():
GetUserNameW = advapi32.GetUserNameW
GetUserNameW.argtypes = [c_wchar_p, POINTER(c_uint)]
GetUserNameW.restype = c_uint
def GetUserName():
buffer = create_unicode_buffer(32)
size = c_uint(len(buffer))
while not GetUserNameW(buffer, byref(size)):
buffer = create_unicode_buffer(len(buffer) * 2)
size.value = len(buffer)
return buffer.value.encode('utf-16-le')[::2]
return GetUserName
GetUserName = GetUserName()
def CryptUnprotectData():
_CryptUnprotectData = crypt32.CryptUnprotectData
_CryptUnprotectData.argtypes = [DataBlob_p, c_wchar_p, DataBlob_p,
c_void_p, c_void_p, c_uint, DataBlob_p]
_CryptUnprotectData.restype = c_uint
def CryptUnprotectData(indata, entropy):
indatab = create_string_buffer(indata)
indata = DataBlob(len(indata), cast(indatab, c_void_p))
entropyb = create_string_buffer(entropy)
entropy = DataBlob(len(entropy), cast(entropyb, c_void_p))
outdata = DataBlob()
if not _CryptUnprotectData(byref(indata), None, byref(entropy),
None, None, 0, byref(outdata)):
raise CMBDTCFatal("Failed to Unprotect Data")
return string_at(outdata.pbData, outdata.cbData)
return CryptUnprotectData
CryptUnprotectData = CryptUnprotectData()
#
# Returns the MD5 digest of "message"
#
def MD5(message):
ctx = hashlib.md5()
ctx.update(message)
return ctx.digest()
#
# Returns the MD5 digest of "message"
#
def SHA1(message):
ctx = hashlib.sha1()
ctx.update(message)
return ctx.digest()
#
# Open the book file at path
#
def openBook(path):
try:
return open(path,'rb')
except:
raise CMBDTCFatal("Could not open book file: " + path)
#
# Encode the bytes in data with the characters in map
#
def encode(data, map):
result = ""
for char in data:
value = ord(char)
Q = (value ^ 0x80) // len(map)
R = value % len(map)
result += map[Q]
result += map[R]
return result
#
# Hash the bytes in data and then encode the digest with the characters in map
#
def encodeHash(data,map):
return encode(MD5(data),map)
#
# Decode the string in data with the characters in map. Returns the decoded bytes
#
def decode(data,map):
result = ""
for i in range (0,len(data),2):
high = map.find(data[i])
low = map.find(data[i+1])
value = (((high * 0x40) ^ 0x80) & 0xFF) + low
result += pack("B",value)
return result
#
# Locate and open the Kindle.info file (Hopefully in the way it is done in the Kindle application)
#
def openKindleInfo():
regkey = winreg.OpenKey(winreg.HKEY_CURRENT_USER, "Software\\Microsoft\\Windows\\CurrentVersion\\Explorer\\Shell Folders\\")
path = winreg.QueryValueEx(regkey, 'Local AppData')[0]
return open(path+'\\Amazon\\Kindle For PC\\{AMAwzsaPaaZAzmZzZQzgZCAkZ3AjA_AY}\\kindle.info','r')
#
# Parse the Kindle.info file and return the records as a list of key-values
#
def parseKindleInfo():
DB = {}
infoReader = openKindleInfo()
infoReader.read(1)
data = infoReader.read()
items = data.split('{')
for item in items:
splito = item.split(':')
DB[splito[0]] =splito[1]
return DB
#
# Find if the original string for a hashed/encoded string is known. If so return the original string othwise return an empty string. (Totally not optimal)
#
def findNameForHash(hash):
names = ["kindle.account.tokens","kindle.cookie.item","eulaVersionAccepted","login_date","kindle.token.item","login","kindle.key.item","kindle.name.info","kindle.device.info", "MazamaRandomNumber"]
result = ""
for name in names:
if hash == encodeHash(name, charMap2):
result = name
break
return name
#
# Print all the records from the kindle.info file (option -i)
#
def printKindleInfo():
for record in kindleDatabase:
name = findNameForHash(record)
if name != "" :
print (name)
print ("--------------------------\n")
else :
print ("Unknown Record")
print getKindleInfoValueForHash(record)
print "\n"
#
# Get a record from the Kindle.info file for the key "hashedKey" (already hashed and encoded). Return the decoded and decrypted record
#
def getKindleInfoValueForHash(hashedKey):
global kindleDatabase
encryptedValue = decode(kindleDatabase[hashedKey],charMap2)
return CryptUnprotectData(encryptedValue,"")
#
# Get a record from the Kindle.info file for the string in "key" (plaintext). Return the decoded and decrypted record
#
def getKindleInfoValueForKey(key):
return getKindleInfoValueForHash(encodeHash(key,charMap2))
#
# Get a 7 bit encoded number from the book file
#
def bookReadEncodedNumber():
flag = False
data = ord(bookFile.read(1))
if data == 0xFF:
flag = True
data = ord(bookFile.read(1))
if data >= 0x80:
datax = (data & 0x7F)
while data >= 0x80 :
data = ord(bookFile.read(1))
datax = (datax <<7) + (data & 0x7F)
data = datax
if flag:
data = -data
return data
#
# Encode a number in 7 bit format
#
def encodeNumber(number):
result = ""
negative = False
flag = 0
if number < 0 :
number = -number + 1
negative = True
while True:
byte = number & 0x7F
number = number >> 7
byte += flag
result += chr(byte)
flag = 0x80
if number == 0 :
if (byte == 0xFF and negative == False) :
result += chr(0x80)
break
if negative:
result += chr(0xFF)
return result[::-1]
#
# Get a length prefixed string from the file
#
def bookReadString():
stringLength = bookReadEncodedNumber()
return unpack(str(stringLength)+"s",bookFile.read(stringLength))[0]
#
# Returns a length prefixed string
#
def lengthPrefixString(data):
return encodeNumber(len(data))+data
#
# Read and return the data of one header record at the current book file position [[offset,compressedLength,decompressedLength],...]
#
def bookReadHeaderRecordData():
nbValues = bookReadEncodedNumber()
values = []
for i in range (0,nbValues):
values.append([bookReadEncodedNumber(),bookReadEncodedNumber(),bookReadEncodedNumber()])
return values
#
# Read and parse one header record at the current book file position and return the associated data [[offset,compressedLength,decompressedLength],...]
#
def parseTopazHeaderRecord():
if ord(bookFile.read(1)) != 0x63:
raise CMBDTCFatal("Parse Error : Invalid Header")
tag = bookReadString()
record = bookReadHeaderRecordData()
return [tag,record]
#
# Parse the header of a Topaz file, get all the header records and the offset for the payload
#
def parseTopazHeader():
global bookHeaderRecords
global bookPayloadOffset
magic = unpack("4s",bookFile.read(4))[0]
if magic != 'TPZ0':
raise CMBDTCFatal("Parse Error : Invalid Header, not a Topaz file")
nbRecords = bookReadEncodedNumber()
bookHeaderRecords = {}
for i in range (0,nbRecords):
result = parseTopazHeaderRecord()
bookHeaderRecords[result[0]] = result[1]
if ord(bookFile.read(1)) != 0x64 :
raise CMBDTCFatal("Parse Error : Invalid Header")
bookPayloadOffset = bookFile.tell()
#
# Get a record in the book payload, given its name and index. If necessary the record is decrypted. The record is not decompressed
#
def getBookPayloadRecord(name, index):
encrypted = False
try:
recordOffset = bookHeaderRecords[name][index][0]
except:
raise CMBDTCFatal("Parse Error : Invalid Record, record not found")
bookFile.seek(bookPayloadOffset + recordOffset)
tag = bookReadString()
if tag != name :
raise CMBDTCFatal("Parse Error : Invalid Record, record name doesn't match")
recordIndex = bookReadEncodedNumber()
if recordIndex < 0 :
encrypted = True
recordIndex = -recordIndex -1
if recordIndex != index :
raise CMBDTCFatal("Parse Error : Invalid Record, index doesn't match")
if bookHeaderRecords[name][index][2] != 0 :
record = bookFile.read(bookHeaderRecords[name][index][2])
else:
record = bookFile.read(bookHeaderRecords[name][index][1])
if encrypted:
ctx = topazCryptoInit(bookKey)
record = topazCryptoDecrypt(record,ctx)
return record
#
# Extract, decrypt and decompress a book record indicated by name and index and print it or save it in "filename"
#
def extractBookPayloadRecord(name, index, filename):
compressed = False
try:
compressed = bookHeaderRecords[name][index][2] != 0
record = getBookPayloadRecord(name,index)
except:
print("Could not find record")
if compressed:
try:
record = zlib.decompress(record)
except:
raise CMBDTCFatal("Could not decompress record")
if filename != "":
try:
file = open(filename,"wb")
file.write(record)
file.close()
except:
raise CMBDTCFatal("Could not write to destination file")
else:
print(record)
#
# return next record [key,value] from the book metadata from the current book position
#
def readMetadataRecord():
return [bookReadString(),bookReadString()]
#
# Parse the metadata record from the book payload and return a list of [key,values]
#
def parseMetadata():
global bookHeaderRecords
global bookPayloadAddress
global bookMetadata
bookMetadata = {}
bookFile.seek(bookPayloadOffset + bookHeaderRecords["metadata"][0][0])
tag = bookReadString()
if tag != "metadata" :
raise CMBDTCFatal("Parse Error : Record Names Don't Match")
flags = ord(bookFile.read(1))
nbRecords = ord(bookFile.read(1))
for i in range (0,nbRecords) :
record =readMetadataRecord()
bookMetadata[record[0]] = record[1]
#
# Returns two bit at offset from a bit field
#
def getTwoBitsFromBitField(bitField,offset):
byteNumber = offset // 4
bitPosition = 6 - 2*(offset % 4)
return ord(bitField[byteNumber]) >> bitPosition & 3
#
# Returns the six bits at offset from a bit field
#
def getSixBitsFromBitField(bitField,offset):
offset *= 3
value = (getTwoBitsFromBitField(bitField,offset) <<4) + (getTwoBitsFromBitField(bitField,offset+1) << 2) +getTwoBitsFromBitField(bitField,offset+2)
return value
#
# 8 bits to six bits encoding from hash to generate PID string
#
def encodePID(hash):
global charMap3
PID = ""
for position in range (0,8):
PID += charMap3[getSixBitsFromBitField(hash,position)]
return PID
#
# Context initialisation for the Topaz Crypto
#
def topazCryptoInit(key):
ctx1 = 0x0CAFFE19E
for keyChar in key:
keyByte = ord(keyChar)
ctx2 = ctx1
ctx1 = ((((ctx1 >>2) * (ctx1 >>7))&0xFFFFFFFF) ^ (keyByte * keyByte * 0x0F902007)& 0xFFFFFFFF )
return [ctx1,ctx2]
#
# decrypt data with the context prepared by topazCryptoInit()
#
def topazCryptoDecrypt(data, ctx):
ctx1 = ctx[0]
ctx2 = ctx[1]
plainText = ""
for dataChar in data:
dataByte = ord(dataChar)
m = (dataByte ^ ((ctx1 >> 3) &0xFF) ^ ((ctx2<<3) & 0xFF)) &0xFF
ctx2 = ctx1
ctx1 = (((ctx1 >> 2) * (ctx1 >> 7)) &0xFFFFFFFF) ^((m * m * 0x0F902007) &0xFFFFFFFF)
plainText += chr(m)
return plainText
#
# Decrypt a payload record with the PID
#
def decryptRecord(data,PID):
ctx = topazCryptoInit(PID)
return topazCryptoDecrypt(data, ctx)
#
# Try to decrypt a dkey record (contains the book PID)
#
def decryptDkeyRecord(data,PID):
record = decryptRecord(data,PID)
fields = unpack("3sB8sB8s3s",record)
if fields[0] != "PID" or fields[5] != "pid" :
raise CMBDTCError("Didn't find PID magic numbers in record")
elif fields[1] != 8 or fields[3] != 8 :
raise CMBDTCError("Record didn't contain correct length fields")
elif fields[2] != PID :
raise CMBDTCError("Record didn't contain PID")
return fields[4]
#
# Decrypt all the book's dkey records (contain the book PID)
#
def decryptDkeyRecords(data,PID):
nbKeyRecords = ord(data[0])
records = []
data = data[1:]
for i in range (0,nbKeyRecords):
length = ord(data[0])
try:
key = decryptDkeyRecord(data[1:length+1],PID)
records.append(key)
except CMBDTCError:
pass
data = data[1+length:]
return records
#
# Encryption table used to generate the device PID
#
def generatePidEncryptionTable() :
table = []
for counter1 in range (0,0x100):
value = counter1
for counter2 in range (0,8):
if (value & 1 == 0) :
value = value >> 1
else :
value = value >> 1
value = value ^ 0xEDB88320
table.append(value)
return table
#
# Seed value used to generate the device PID
#
def generatePidSeed(table,dsn) :
value = 0
for counter in range (0,4) :
index = (ord(dsn[counter]) ^ value) &0xFF
value = (value >> 8) ^ table[index]
return value
#
# Generate the device PID
#
def generateDevicePID(table,dsn,nbRoll):
seed = generatePidSeed(table,dsn)
pidAscii = ""
pid = [(seed >>24) &0xFF,(seed >> 16) &0xff,(seed >> 8) &0xFF,(seed) & 0xFF,(seed>>24) & 0xFF,(seed >> 16) &0xff,(seed >> 8) &0xFF,(seed) & 0xFF]
index = 0
for counter in range (0,nbRoll):
pid[index] = pid[index] ^ ord(dsn[counter])
index = (index+1) %8
for counter in range (0,8):
index = ((((pid[counter] >>5) & 3) ^ pid[counter]) & 0x1f) + (pid[counter] >> 7)
pidAscii += charMap4[index]
return pidAscii
#
# Create decrypted book payload
#
def createDecryptedPayload(payload):
# store data to be able to create the header later
headerData= []
currentOffset = 0
# Add social DRM to decrypted files
try:
data = getKindleInfoValueForKey("kindle.name.info")+":"+ getKindleInfoValueForKey("login")
if payload!= None:
payload.write(lengthPrefixString("sdrm"))
payload.write(encodeNumber(0))
payload.write(data)
else:
currentOffset += len(lengthPrefixString("sdrm"))
currentOffset += len(encodeNumber(0))
currentOffset += len(data)
except:
pass
for headerRecord in bookHeaderRecords:
name = headerRecord
newRecord = []
if name != "dkey" :
for index in range (0,len(bookHeaderRecords[name])) :
offset = currentOffset
if payload != None:
# write tag
payload.write(lengthPrefixString(name))
# write data
payload.write(encodeNumber(index))
payload.write(getBookPayloadRecord(name, index))
else :
currentOffset += len(lengthPrefixString(name))
currentOffset += len(encodeNumber(index))
currentOffset += len(getBookPayloadRecord(name, index))
newRecord.append([offset,bookHeaderRecords[name][index][1],bookHeaderRecords[name][index][2]])
headerData.append([name,newRecord])
return headerData
#
# Create decrypted book
#
def createDecryptedBook(outputFile):
outputFile = open(outputFile,"wb")
# Write the payload in a temporary file
headerData = createDecryptedPayload(None)
outputFile.write("TPZ0")
outputFile.write(encodeNumber(len(headerData)))
for header in headerData :
outputFile.write(chr(0x63))
outputFile.write(lengthPrefixString(header[0]))
outputFile.write(encodeNumber(len(header[1])))
for numbers in header[1] :
outputFile.write(encodeNumber(numbers[0]))
outputFile.write(encodeNumber(numbers[1]))
outputFile.write(encodeNumber(numbers[2]))
outputFile.write(chr(0x64))
createDecryptedPayload(outputFile)
outputFile.close()
#
# Set the command to execute by the programm according to cmdLine parameters
#
def setCommand(name) :
global command
if command != "" :
raise CMBDTCFatal("Invalid command line parameters")
else :
command = name
#
# Program usage
#
def usage():
print("\nUsage:")
print("\nCMBDTC.py [options] bookFileName\n")
print("-p Adds a PID to the list of PIDs that are tried to decrypt the book key (can be used several times)")
print("-d Saves a decrypted copy of the book")
print("-r Prints or writes to disk a record indicated in the form name:index (e.g \"img:0\")")
print("-o Output file name to write records and decrypted books")
print("-v Verbose (can be used several times)")
print("-i Prints kindle.info database")
#
# Main
#
def main(argv=sys.argv):
global kindleDatabase
global bookMetadata
global bookKey
global bookFile
global command
progname = os.path.basename(argv[0])
verbose = 0
recordName = ""
recordIndex = 0
outputFile = ""
PIDs = []
kindleDatabase = None
command = ""
try:
opts, args = getopt.getopt(sys.argv[1:], "vdir:o:p:")
except getopt.GetoptError, err:
# print help information and exit:
print str(err) # will print something like "option -a not recognized"
usage()
sys.exit(2)
if len(opts) == 0 and len(args) == 0 :
usage()
sys.exit(2)
for o, a in opts:
if o == "-v":
verbose+=1
if o == "-i":
setCommand("printInfo")
if o =="-o":
if a == None :
raise CMBDTCFatal("Invalid parameter for -o")
outputFile = a
if o =="-r":
setCommand("printRecord")
try:
recordName,recordIndex = a.split(':')
except:
raise CMBDTCFatal("Invalid parameter for -r")
if o =="-p":
PIDs.append(a)
if o =="-d":
setCommand("doit")
if command == "" :
raise CMBDTCFatal("No action supplied on command line")
#
# Read the encrypted database
#
try:
kindleDatabase = parseKindleInfo()
except Exception, message:
if verbose>0:
print(message)
if kindleDatabase != None :
if command == "printInfo" :
printKindleInfo()
#
# Compute the DSN
#
# Get the Mazama Random number
MazamaRandomNumber = getKindleInfoValueForKey("MazamaRandomNumber")
# Get the HDD serial
encodedSystemVolumeSerialNumber = encodeHash(str(GetVolumeSerialNumber(GetSystemDirectory().split('\\')[0] + '\\')),charMap1)
# Get the current user name
encodedUsername = encodeHash(GetUserName(),charMap1)
# concat, hash and encode
DSN = encode(SHA1(MazamaRandomNumber+encodedSystemVolumeSerialNumber+encodedUsername),charMap1)
if verbose >1:
print("DSN: " + DSN)
#
# Compute the device PID
#
table = generatePidEncryptionTable()
devicePID = generateDevicePID(table,DSN,4)
PIDs.append(devicePID)
if verbose > 0:
print("Device PID: " + devicePID)
#
# Open book and parse metadata
#
if len(args) == 1:
bookFile = openBook(args[0])
parseTopazHeader()
parseMetadata()
#
# Compute book PID
#
# Get the account token
if kindleDatabase != None:
kindleAccountToken = getKindleInfoValueForKey("kindle.account.tokens")
if verbose >1:
print("Account Token: " + kindleAccountToken)
keysRecord = bookMetadata["keys"]
keysRecordRecord = bookMetadata[keysRecord]
pidHash = SHA1(DSN+kindleAccountToken+keysRecord+keysRecordRecord)
bookPID = encodePID(pidHash)
PIDs.append(bookPID)
if verbose > 0:
print ("Book PID: " + bookPID )
#
# Decrypt book key
#
dkey = getBookPayloadRecord('dkey', 0)
bookKeys = []
for PID in PIDs :
bookKeys+=decryptDkeyRecords(dkey,PID)
if len(bookKeys) == 0 :
if verbose > 0 :
print ("Book key could not be found. Maybe this book is not registered with this device.")
else :
bookKey = bookKeys[0]
if verbose > 0:
print("Book key: " + bookKey.encode('hex'))
if command == "printRecord" :
extractBookPayloadRecord(recordName,int(recordIndex),outputFile)
if outputFile != "" and verbose>0 :
print("Wrote record to file: "+outputFile)
elif command == "doit" :
if outputFile!="" :
createDecryptedBook(outputFile)
if verbose >0 :
print ("Decrypted book saved. Don't pirate!")
elif verbose > 0:
print("Output file name was not supplied.")
return 0
if __name__ == '__main__':
sys.exit(main())

View File

@@ -0,0 +1,59 @@
from PyQt4.Qt import QWidget, QVBoxLayout, QLabel, QLineEdit
from calibre.utils.config import JSONConfig
# This is where all preferences for this plugin will be stored
# You should always prefix your config file name with plugins/,
# so as to ensure you dont accidentally clobber a calibre config file
prefs = JSONConfig('plugins/K4MobiDeDRM')
# Set defaults
prefs.defaults['pids'] = ""
prefs.defaults['serials'] = ""
prefs.defaults['WINEPREFIX'] = None
class ConfigWidget(QWidget):
def __init__(self):
QWidget.__init__(self)
self.l = QVBoxLayout()
self.setLayout(self.l)
self.serialLabel = QLabel('Kindle Serial numbers (separate with commas, no spaces)')
self.l.addWidget(self.serialLabel)
self.serials = QLineEdit(self)
self.serials.setText(prefs['serials'])
self.l.addWidget(self.serials)
self.serialLabel.setBuddy(self.serials)
self.pidLabel = QLabel('Mobipocket PIDs (separate with commas, no spaces)')
self.l.addWidget(self.pidLabel)
self.pids = QLineEdit(self)
self.pids.setText(prefs['pids'])
self.l.addWidget(self.pids)
self.pidLabel.setBuddy(self.serials)
self.wpLabel = QLabel('For Linux only: WINEPREFIX (enter absolute path)')
self.l.addWidget(self.wpLabel)
self.wineprefix = QLineEdit(self)
wineprefix = prefs['WINEPREFIX']
if wineprefix is not None:
self.wineprefix.setText(wineprefix)
else:
self.wineprefix.setText('')
self.l.addWidget(self.wineprefix)
self.wpLabel.setBuddy(self.wineprefix)
def save_settings(self):
prefs['pids'] = str(self.pids.text())
prefs['serials'] = str(self.serials.text())
winepref=str(self.wineprefix.text())
if winepref.strip() != '':
prefs['WINEPREFIX'] = winepref
else:
prefs['WINEPREFIX'] = None

View File

@@ -0,0 +1,846 @@
#! /usr/bin/python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
# For use with Topaz Scripts Version 2.6
class Unbuffered:
def __init__(self, stream):
self.stream = stream
def write(self, data):
self.stream.write(data)
self.stream.flush()
def __getattr__(self, attr):
return getattr(self.stream, attr)
import sys
sys.stdout=Unbuffered(sys.stdout)
import csv
import os
import getopt
from struct import pack
from struct import unpack
class TpzDRMError(Exception):
pass
# Get a 7 bit encoded number from string. The most
# significant byte comes first and has the high bit (8th) set
def readEncodedNumber(file):
flag = False
c = file.read(1)
if (len(c) == 0):
return None
data = ord(c)
if data == 0xFF:
flag = True
c = file.read(1)
if (len(c) == 0):
return None
data = ord(c)
if data >= 0x80:
datax = (data & 0x7F)
while data >= 0x80 :
c = file.read(1)
if (len(c) == 0):
return None
data = ord(c)
datax = (datax <<7) + (data & 0x7F)
data = datax
if flag:
data = -data
return data
# returns a binary string that encodes a number into 7 bits
# most significant byte first which has the high bit set
def encodeNumber(number):
result = ""
negative = False
flag = 0
if number < 0 :
number = -number + 1
negative = True
while True:
byte = number & 0x7F
number = number >> 7
byte += flag
result += chr(byte)
flag = 0x80
if number == 0 :
if (byte == 0xFF and negative == False) :
result += chr(0x80)
break
if negative:
result += chr(0xFF)
return result[::-1]
# create / read a length prefixed string from the file
def lengthPrefixString(data):
return encodeNumber(len(data))+data
def readString(file):
stringLength = readEncodedNumber(file)
if (stringLength == None):
return ""
sv = file.read(stringLength)
if (len(sv) != stringLength):
return ""
return unpack(str(stringLength)+"s",sv)[0]
# convert a binary string generated by encodeNumber (7 bit encoded number)
# to the value you would find inside the page*.dat files to be processed
def convert(i):
result = ''
val = encodeNumber(i)
for j in xrange(len(val)):
c = ord(val[j:j+1])
result += '%02x' % c
return result
# the complete string table used to store all book text content
# as well as the xml tokens and values that make sense out of it
class Dictionary(object):
def __init__(self, dictFile):
self.filename = dictFile
self.size = 0
self.fo = file(dictFile,'rb')
self.stable = []
self.size = readEncodedNumber(self.fo)
for i in xrange(self.size):
self.stable.append(self.escapestr(readString(self.fo)))
self.pos = 0
def escapestr(self, str):
str = str.replace('&','&amp;')
str = str.replace('<','&lt;')
str = str.replace('>','&gt;')
str = str.replace('=','&#61;')
return str
def lookup(self,val):
if ((val >= 0) and (val < self.size)) :
self.pos = val
return self.stable[self.pos]
else:
print "Error - %d outside of string table limits" % val
raise TpzDRMError('outside of string table limits')
# sys.exit(-1)
def getSize(self):
return self.size
def getPos(self):
return self.pos
def dumpDict(self):
for i in xrange(self.size):
print "%d %s %s" % (i, convert(i), self.stable[i])
return
# parses the xml snippets that are represented by each page*.dat file.
# also parses the other0.dat file - the main stylesheet
# and information used to inject the xml snippets into page*.dat files
class PageParser(object):
def __init__(self, filename, dict, debug, flat_xml):
self.fo = file(filename,'rb')
self.id = os.path.basename(filename).replace('.dat','')
self.dict = dict
self.debug = debug
self.flat_xml = flat_xml
self.tagpath = []
self.doc = []
self.snippetList = []
# hash table used to enable the decoding process
# This has all been developed by trial and error so it may still have omissions or
# contain errors
# Format:
# tag : (number of arguments, argument type, subtags present, special case of subtags presents when escaped)
token_tags = {
'x' : (1, 'scalar_number', 0, 0),
'y' : (1, 'scalar_number', 0, 0),
'h' : (1, 'scalar_number', 0, 0),
'w' : (1, 'scalar_number', 0, 0),
'firstWord' : (1, 'scalar_number', 0, 0),
'lastWord' : (1, 'scalar_number', 0, 0),
'rootID' : (1, 'scalar_number', 0, 0),
'stemID' : (1, 'scalar_number', 0, 0),
'type' : (1, 'scalar_text', 0, 0),
'info' : (0, 'number', 1, 0),
'info.word' : (0, 'number', 1, 1),
'info.word.ocrText' : (1, 'text', 0, 0),
'info.word.firstGlyph' : (1, 'raw', 0, 0),
'info.word.lastGlyph' : (1, 'raw', 0, 0),
'info.word.bl' : (1, 'raw', 0, 0),
'info.word.link_id' : (1, 'number', 0, 0),
'glyph' : (0, 'number', 1, 1),
'glyph.x' : (1, 'number', 0, 0),
'glyph.y' : (1, 'number', 0, 0),
'glyph.glyphID' : (1, 'number', 0, 0),
'dehyphen' : (0, 'number', 1, 1),
'dehyphen.rootID' : (1, 'number', 0, 0),
'dehyphen.stemID' : (1, 'number', 0, 0),
'dehyphen.stemPage' : (1, 'number', 0, 0),
'dehyphen.sh' : (1, 'number', 0, 0),
'links' : (0, 'number', 1, 1),
'links.page' : (1, 'number', 0, 0),
'links.rel' : (1, 'number', 0, 0),
'links.row' : (1, 'number', 0, 0),
'links.title' : (1, 'text', 0, 0),
'links.href' : (1, 'text', 0, 0),
'links.type' : (1, 'text', 0, 0),
'links.id' : (1, 'number', 0, 0),
'paraCont' : (0, 'number', 1, 1),
'paraCont.rootID' : (1, 'number', 0, 0),
'paraCont.stemID' : (1, 'number', 0, 0),
'paraCont.stemPage' : (1, 'number', 0, 0),
'paraStems' : (0, 'number', 1, 1),
'paraStems.stemID' : (1, 'number', 0, 0),
'wordStems' : (0, 'number', 1, 1),
'wordStems.stemID' : (1, 'number', 0, 0),
'empty' : (1, 'snippets', 1, 0),
'page' : (1, 'snippets', 1, 0),
'page.pageid' : (1, 'scalar_text', 0, 0),
'page.pagelabel' : (1, 'scalar_text', 0, 0),
'page.type' : (1, 'scalar_text', 0, 0),
'page.h' : (1, 'scalar_number', 0, 0),
'page.w' : (1, 'scalar_number', 0, 0),
'page.startID' : (1, 'scalar_number', 0, 0),
'group' : (1, 'snippets', 1, 0),
'group.type' : (1, 'scalar_text', 0, 0),
'group._tag' : (1, 'scalar_text', 0, 0),
'group.orientation': (1, 'scalar_text', 0, 0),
'region' : (1, 'snippets', 1, 0),
'region.type' : (1, 'scalar_text', 0, 0),
'region.x' : (1, 'scalar_number', 0, 0),
'region.y' : (1, 'scalar_number', 0, 0),
'region.h' : (1, 'scalar_number', 0, 0),
'region.w' : (1, 'scalar_number', 0, 0),
'region.orientation' : (1, 'scalar_text', 0, 0),
'empty_text_region' : (1, 'snippets', 1, 0),
'img' : (1, 'snippets', 1, 0),
'img.x' : (1, 'scalar_number', 0, 0),
'img.y' : (1, 'scalar_number', 0, 0),
'img.h' : (1, 'scalar_number', 0, 0),
'img.w' : (1, 'scalar_number', 0, 0),
'img.src' : (1, 'scalar_number', 0, 0),
'img.color_src' : (1, 'scalar_number', 0, 0),
'paragraph' : (1, 'snippets', 1, 0),
'paragraph.class' : (1, 'scalar_text', 0, 0),
'paragraph.firstWord' : (1, 'scalar_number', 0, 0),
'paragraph.lastWord' : (1, 'scalar_number', 0, 0),
'paragraph.lastWord' : (1, 'scalar_number', 0, 0),
'paragraph.gridSize' : (1, 'scalar_number', 0, 0),
'paragraph.gridBottomCenter' : (1, 'scalar_number', 0, 0),
'paragraph.gridTopCenter' : (1, 'scalar_number', 0, 0),
'paragraph.gridBeginCenter' : (1, 'scalar_number', 0, 0),
'paragraph.gridEndCenter' : (1, 'scalar_number', 0, 0),
'word_semantic' : (1, 'snippets', 1, 1),
'word_semantic.type' : (1, 'scalar_text', 0, 0),
'word_semantic.firstWord' : (1, 'scalar_number', 0, 0),
'word_semantic.lastWord' : (1, 'scalar_number', 0, 0),
'word' : (1, 'snippets', 1, 0),
'word.type' : (1, 'scalar_text', 0, 0),
'word.class' : (1, 'scalar_text', 0, 0),
'word.firstGlyph' : (1, 'scalar_number', 0, 0),
'word.lastGlyph' : (1, 'scalar_number', 0, 0),
'_span' : (1, 'snippets', 1, 0),
'_span.firstWord' : (1, 'scalar_number', 0, 0),
'_span.lastWord' : (1, 'scalar_number', 0, 0),
'_span.gridSize' : (1, 'scalar_number', 0, 0),
'_span.gridBottomCenter' : (1, 'scalar_number', 0, 0),
'_span.gridTopCenter' : (1, 'scalar_number', 0, 0),
'_span.gridBeginCenter' : (1, 'scalar_number', 0, 0),
'_span.gridEndCenter' : (1, 'scalar_number', 0, 0),
'span' : (1, 'snippets', 1, 0),
'span.firstWord' : (1, 'scalar_number', 0, 0),
'span.lastWord' : (1, 'scalar_number', 0, 0),
'span.gridSize' : (1, 'scalar_number', 0, 0),
'span.gridBottomCenter' : (1, 'scalar_number', 0, 0),
'span.gridTopCenter' : (1, 'scalar_number', 0, 0),
'span.gridBeginCenter' : (1, 'scalar_number', 0, 0),
'span.gridEndCenter' : (1, 'scalar_number', 0, 0),
'extratokens' : (1, 'snippets', 1, 0),
'extratokens.type' : (1, 'scalar_text', 0, 0),
'extratokens.firstGlyph' : (1, 'scalar_number', 0, 0),
'extratokens.lastGlyph' : (1, 'scalar_number', 0, 0),
'glyph.h' : (1, 'number', 0, 0),
'glyph.w' : (1, 'number', 0, 0),
'glyph.use' : (1, 'number', 0, 0),
'glyph.vtx' : (1, 'number', 0, 1),
'glyph.len' : (1, 'number', 0, 1),
'glyph.dpi' : (1, 'number', 0, 0),
'vtx' : (0, 'number', 1, 1),
'vtx.x' : (1, 'number', 0, 0),
'vtx.y' : (1, 'number', 0, 0),
'len' : (0, 'number', 1, 1),
'len.n' : (1, 'number', 0, 0),
'book' : (1, 'snippets', 1, 0),
'version' : (1, 'snippets', 1, 0),
'version.FlowEdit_1_id' : (1, 'scalar_text', 0, 0),
'version.FlowEdit_1_version' : (1, 'scalar_text', 0, 0),
'version.Schema_id' : (1, 'scalar_text', 0, 0),
'version.Schema_version' : (1, 'scalar_text', 0, 0),
'version.Topaz_version' : (1, 'scalar_text', 0, 0),
'version.WordDetailEdit_1_id' : (1, 'scalar_text', 0, 0),
'version.WordDetailEdit_1_version' : (1, 'scalar_text', 0, 0),
'version.ZoneEdit_1_id' : (1, 'scalar_text', 0, 0),
'version.ZoneEdit_1_version' : (1, 'scalar_text', 0, 0),
'version.chapterheaders' : (1, 'scalar_text', 0, 0),
'version.creation_date' : (1, 'scalar_text', 0, 0),
'version.header_footer' : (1, 'scalar_text', 0, 0),
'version.init_from_ocr' : (1, 'scalar_text', 0, 0),
'version.letter_insertion' : (1, 'scalar_text', 0, 0),
'version.xmlinj_convert' : (1, 'scalar_text', 0, 0),
'version.xmlinj_reflow' : (1, 'scalar_text', 0, 0),
'version.xmlinj_transform' : (1, 'scalar_text', 0, 0),
'version.findlists' : (1, 'scalar_text', 0, 0),
'version.page_num' : (1, 'scalar_text', 0, 0),
'version.page_type' : (1, 'scalar_text', 0, 0),
'version.bad_text' : (1, 'scalar_text', 0, 0),
'version.glyph_mismatch' : (1, 'scalar_text', 0, 0),
'version.margins' : (1, 'scalar_text', 0, 0),
'version.staggered_lines' : (1, 'scalar_text', 0, 0),
'version.paragraph_continuation' : (1, 'scalar_text', 0, 0),
'version.toc' : (1, 'scalar_text', 0, 0),
'stylesheet' : (1, 'snippets', 1, 0),
'style' : (1, 'snippets', 1, 0),
'style._tag' : (1, 'scalar_text', 0, 0),
'style.type' : (1, 'scalar_text', 0, 0),
'style._parent_type' : (1, 'scalar_text', 0, 0),
'style.class' : (1, 'scalar_text', 0, 0),
'style._after_class' : (1, 'scalar_text', 0, 0),
'rule' : (1, 'snippets', 1, 0),
'rule.attr' : (1, 'scalar_text', 0, 0),
'rule.value' : (1, 'scalar_text', 0, 0),
'original' : (0, 'number', 1, 1),
'original.pnum' : (1, 'number', 0, 0),
'original.pid' : (1, 'text', 0, 0),
'pages' : (0, 'number', 1, 1),
'pages.ref' : (1, 'number', 0, 0),
'pages.id' : (1, 'number', 0, 0),
'startID' : (0, 'number', 1, 1),
'startID.page' : (1, 'number', 0, 0),
'startID.id' : (1, 'number', 0, 0),
}
# full tag path record keeping routines
def tag_push(self, token):
self.tagpath.append(token)
def tag_pop(self):
if len(self.tagpath) > 0 :
self.tagpath.pop()
def tagpath_len(self):
return len(self.tagpath)
def get_tagpath(self, i):
cnt = len(self.tagpath)
if i < cnt : result = self.tagpath[i]
for j in xrange(i+1, cnt) :
result += '.' + self.tagpath[j]
return result
# list of absolute command byte values values that indicate
# various types of loop meachanisms typically used to generate vectors
cmd_list = (0x76, 0x76)
# peek at and return 1 byte that is ahead by i bytes
def peek(self, aheadi):
c = self.fo.read(aheadi)
if (len(c) == 0):
return None
self.fo.seek(-aheadi,1)
c = c[-1:]
return ord(c)
# get the next value from the file being processed
def getNext(self):
nbyte = self.peek(1);
if (nbyte == None):
return None
val = readEncodedNumber(self.fo)
return val
# format an arg by argtype
def formatArg(self, arg, argtype):
if (argtype == 'text') or (argtype == 'scalar_text') :
result = self.dict.lookup(arg)
elif (argtype == 'raw') or (argtype == 'number') or (argtype == 'scalar_number') :
result = arg
elif (argtype == 'snippets') :
result = arg
else :
print "Error Unknown argtype %s" % argtype
sys.exit(-2)
return result
# process the next tag token, recursively handling subtags,
# arguments, and commands
def procToken(self, token):
known_token = False
self.tag_push(token)
if self.debug : print 'Processing: ', self.get_tagpath(0)
cnt = self.tagpath_len()
for j in xrange(cnt):
tkn = self.get_tagpath(j)
if tkn in self.token_tags :
num_args = self.token_tags[tkn][0]
argtype = self.token_tags[tkn][1]
subtags = self.token_tags[tkn][2]
splcase = self.token_tags[tkn][3]
ntags = -1
known_token = True
break
if known_token :
# handle subtags if present
subtagres = []
if (splcase == 1):
# this type of tag uses of escape marker 0x74 indicate subtag count
if self.peek(1) == 0x74:
skip = readEncodedNumber(self.fo)
subtags = 1
num_args = 0
if (subtags == 1):
ntags = readEncodedNumber(self.fo)
if self.debug : print 'subtags: ' + token + ' has ' + str(ntags)
for j in xrange(ntags):
val = readEncodedNumber(self.fo)
subtagres.append(self.procToken(self.dict.lookup(val)))
# arguments can be scalars or vectors of text or numbers
argres = []
if num_args > 0 :
firstarg = self.peek(1)
if (firstarg in self.cmd_list) and (argtype != 'scalar_number') and (argtype != 'scalar_text'):
# single argument is a variable length vector of data
arg = readEncodedNumber(self.fo)
argres = self.decodeCMD(arg,argtype)
else :
# num_arg scalar arguments
for i in xrange(num_args):
argres.append(self.formatArg(readEncodedNumber(self.fo), argtype))
# build the return tag
result = []
tkn = self.get_tagpath(0)
result.append(tkn)
result.append(subtagres)
result.append(argtype)
result.append(argres)
self.tag_pop()
return result
# all tokens that need to be processed should be in the hash
# table if it may indicate a problem, either new token
# or an out of sync condition
else:
result = []
if (self.debug):
print 'Unknown Token:', token
self.tag_pop()
return result
# special loop used to process code snippets
# it is NEVER used to format arguments.
# builds the snippetList
def doLoop72(self, argtype):
cnt = readEncodedNumber(self.fo)
if self.debug :
result = 'Set of '+ str(cnt) + ' xml snippets. The overall structure \n'
result += 'of the document is indicated by snippet number sets at the\n'
result += 'end of each snippet. \n'
print result
for i in xrange(cnt):
if self.debug: print 'Snippet:',str(i)
snippet = []
snippet.append(i)
val = readEncodedNumber(self.fo)
snippet.append(self.procToken(self.dict.lookup(val)))
self.snippetList.append(snippet)
return
# general loop code gracisouly submitted by "skindle" - thank you!
def doLoop76Mode(self, argtype, cnt, mode):
result = []
adj = 0
if mode & 1:
adj = readEncodedNumber(self.fo)
mode = mode >> 1
x = []
for i in xrange(cnt):
x.append(readEncodedNumber(self.fo) - adj)
for i in xrange(mode):
for j in xrange(1, cnt):
x[j] = x[j] + x[j - 1]
for i in xrange(cnt):
result.append(self.formatArg(x[i],argtype))
return result
# dispatches loop commands bytes with various modes
# The 0x76 style loops are used to build vectors
# This was all derived by trial and error and
# new loop types may exist that are not handled here
# since they did not appear in the test cases
def decodeCMD(self, cmd, argtype):
if (cmd == 0x76):
# loop with cnt, and mode to control loop styles
cnt = readEncodedNumber(self.fo)
mode = readEncodedNumber(self.fo)
if self.debug : print 'Loop for', cnt, 'with mode', mode, ': '
return self.doLoop76Mode(argtype, cnt, mode)
if self.dbug: print "Unknown command", cmd
result = []
return result
# add full tag path to injected snippets
def updateName(self, tag, prefix):
name = tag[0]
subtagList = tag[1]
argtype = tag[2]
argList = tag[3]
nname = prefix + '.' + name
nsubtaglist = []
for j in subtagList:
nsubtaglist.append(self.updateName(j,prefix))
ntag = []
ntag.append(nname)
ntag.append(nsubtaglist)
ntag.append(argtype)
ntag.append(argList)
return ntag
# perform depth first injection of specified snippets into this one
def injectSnippets(self, snippet):
snipno, tag = snippet
name = tag[0]
subtagList = tag[1]
argtype = tag[2]
argList = tag[3]
nsubtagList = []
if len(argList) > 0 :
for j in argList:
asnip = self.snippetList[j]
aso, atag = self.injectSnippets(asnip)
atag = self.updateName(atag, name)
nsubtagList.append(atag)
argtype='number'
argList=[]
if len(nsubtagList) > 0 :
subtagList.extend(nsubtagList)
tag = []
tag.append(name)
tag.append(subtagList)
tag.append(argtype)
tag.append(argList)
snippet = []
snippet.append(snipno)
snippet.append(tag)
return snippet
# format the tag for output
def formatTag(self, node):
name = node[0]
subtagList = node[1]
argtype = node[2]
argList = node[3]
fullpathname = name.split('.')
nodename = fullpathname.pop()
ilvl = len(fullpathname)
indent = ' ' * (3 * ilvl)
rlst = []
rlst.append(indent + '<' + nodename + '>')
if len(argList) > 0:
alst = []
for j in argList:
if (argtype == 'text') or (argtype == 'scalar_text') :
alst.append(j + '|')
else :
alst.append(str(j) + ',')
argres = "".join(alst)
argres = argres[0:-1]
if argtype == 'snippets' :
rlst.append('snippets:' + argres)
else :
rlst.append(argres)
if len(subtagList) > 0 :
rlst.append('\n')
for j in subtagList:
if len(j) > 0 :
rlst.append(self.formatTag(j))
rlst.append(indent + '</' + nodename + '>\n')
else:
rlst.append('</' + nodename + '>\n')
return "".join(rlst)
# flatten tag
def flattenTag(self, node):
name = node[0]
subtagList = node[1]
argtype = node[2]
argList = node[3]
rlst = []
rlst.append(name)
if (len(argList) > 0):
alst = []
for j in argList:
if (argtype == 'text') or (argtype == 'scalar_text') :
alst.append(j + '|')
else :
alst.append(str(j) + '|')
argres = "".join(alst)
argres = argres[0:-1]
if argtype == 'snippets' :
rlst.append('.snippets=' + argres)
else :
rlst.append('=' + argres)
rlst.append('\n')
for j in subtagList:
if len(j) > 0 :
rlst.append(self.flattenTag(j))
return "".join(rlst)
# reduce create xml output
def formatDoc(self, flat_xml):
rlst = []
for j in self.doc :
if len(j) > 0:
if flat_xml:
rlst.append(self.flattenTag(j))
else:
rlst.append(self.formatTag(j))
result = "".join(rlst)
if self.debug : print result
return result
# main loop - parse the page.dat files
# to create structured document and snippets
# FIXME: value at end of magic appears to be a subtags count
# but for what? For now, inject an 'info" tag as it is in
# every dictionary and seems close to what is meant
# The alternative is to special case the last _ "0x5f" to mean something
def process(self):
# peek at the first bytes to see what type of file it is
magic = self.fo.read(9)
if (magic[0:1] == 'p') and (magic[2:9] == 'marker_'):
first_token = 'info'
elif (magic[0:1] == 'p') and (magic[2:9] == '__PAGE_'):
skip = self.fo.read(2)
first_token = 'info'
elif (magic[0:1] == 'p') and (magic[2:8] == '_PAGE_'):
first_token = 'info'
elif (magic[0:1] == 'g') and (magic[2:9] == '__GLYPH'):
skip = self.fo.read(3)
first_token = 'info'
else :
# other0.dat file
first_token = None
self.fo.seek(-9,1)
# main loop to read and build the document tree
while True:
if first_token != None :
# use "inserted" first token 'info' for page and glyph files
tag = self.procToken(first_token)
if len(tag) > 0 :
self.doc.append(tag)
first_token = None
v = self.getNext()
if (v == None):
break
if (v == 0x72):
self.doLoop72('number')
elif (v > 0) and (v < self.dict.getSize()) :
tag = self.procToken(self.dict.lookup(v))
if len(tag) > 0 :
self.doc.append(tag)
else:
if self.debug:
print "Main Loop: Unknown value: %x" % v
if (v == 0):
if (self.peek(1) == 0x5f):
skip = self.fo.read(1)
first_token = 'info'
# now do snippet injection
if len(self.snippetList) > 0 :
if self.debug : print 'Injecting Snippets:'
snippet = self.injectSnippets(self.snippetList[0])
snipno = snippet[0]
tag_add = snippet[1]
if self.debug : print self.formatTag(tag_add)
if len(tag_add) > 0:
self.doc.append(tag_add)
# handle generation of xml output
xmlpage = self.formatDoc(self.flat_xml)
return xmlpage
def fromData(dict, fname):
flat_xml = True
debug = False
pp = PageParser(fname, dict, debug, flat_xml)
xmlpage = pp.process()
return xmlpage
def getXML(dict, fname):
flat_xml = False
debug = False
pp = PageParser(fname, dict, debug, flat_xml)
xmlpage = pp.process()
return xmlpage
def usage():
print 'Usage: '
print ' convert2xml.py dict0000.dat infile.dat '
print ' '
print ' Options:'
print ' -h print this usage help message '
print ' -d turn on debug output to check for potential errors '
print ' --flat-xml output the flattened xml page description only '
print ' '
print ' This program will attempt to convert a page*.dat file or '
print ' glyphs*.dat file, using the dict0000.dat file, to its xml description. '
print ' '
print ' Use "cmbtc_dump.py" first to unencrypt, uncompress, and dump '
print ' the *.dat files from a Topaz format e-book.'
#
# Main
#
def main(argv):
dictFile = ""
pageFile = ""
debug = False
flat_xml = False
printOutput = False
if len(argv) == 0:
printOutput = True
argv = sys.argv
try:
opts, args = getopt.getopt(argv[1:], "hd", ["flat-xml"])
except getopt.GetoptError, err:
# print help information and exit:
print str(err) # will print something like "option -a not recognized"
usage()
sys.exit(2)
if len(opts) == 0 and len(args) == 0 :
usage()
sys.exit(2)
for o, a in opts:
if o =="-d":
debug=True
if o =="-h":
usage()
sys.exit(0)
if o =="--flat-xml":
flat_xml = True
dictFile, pageFile = args[0], args[1]
# read in the string table dictionary
dict = Dictionary(dictFile)
# dict.dumpDict()
# create a page parser
pp = PageParser(pageFile, dict, debug, flat_xml)
xmlpage = pp.process()
if printOutput:
print xmlpage
return 0
return xmlpage
if __name__ == '__main__':
sys.exit(main(''))

View File

@@ -0,0 +1,88 @@
#!/usr/bin/env python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
class Unbuffered:
def __init__(self, stream):
self.stream = stream
def write(self, data):
self.stream.write(data)
self.stream.flush()
def __getattr__(self, attr):
return getattr(self.stream, attr)
import sys
sys.stdout=Unbuffered(sys.stdout)
import os
import ineptepub
import ignobleepub
import zipfix
import re
def main(argv=sys.argv):
args = argv[1:]
if len(args) != 3:
return -1
infile = args[0]
outdir = args[1]
rscpath = args[2]
errlog = ''
# first fix the epub to make sure we do not get errors
name, ext = os.path.splitext(os.path.basename(infile))
bpath = os.path.dirname(infile)
zippath = os.path.join(bpath,name + '_temp.zip')
rv = zipfix.repairBook(infile, zippath)
if rv != 0:
print "Error while trying to fix epub"
return rv
# determine a good name for the output file
outfile = os.path.join(outdir, name + '_nodrm.epub')
rv = 1
# first try with the Adobe adept epub
# try with any keyfiles (*.der) in the rscpath
files = os.listdir(rscpath)
filefilter = re.compile("\.der$", re.IGNORECASE)
files = filter(filefilter.search, files)
if files:
for filename in files:
keypath = os.path.join(rscpath, filename)
try:
rv = ineptepub.decryptBook(keypath, zippath, outfile)
if rv == 0:
break
except Exception, e:
errlog += str(e)
rv = 1
pass
if rv == 0:
os.remove(zippath)
return 0
# still no luck
# now try with ignoble epub
# try with any keyfiles (*.b64) in the rscpath
files = os.listdir(rscpath)
filefilter = re.compile("\.b64$", re.IGNORECASE)
files = filter(filefilter.search, files)
if files:
for filename in files:
keypath = os.path.join(rscpath, filename)
try:
rv = ignobleepub.decryptBook(keypath, zippath, outfile)
if rv == 0:
break
except Exception, e:
errlog += str(e)
rv = 1
pass
os.remove(zippath)
if rv != 0:
print errlog
return rv
if __name__ == "__main__":
sys.exit(main())

View File

@@ -0,0 +1,45 @@
#!/usr/bin/env python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
class Unbuffered:
def __init__(self, stream):
self.stream = stream
def write(self, data):
self.stream.write(data)
self.stream.flush()
def __getattr__(self, attr):
return getattr(self.stream, attr)
import sys
sys.stdout=Unbuffered(sys.stdout)
import os
import erdr2pml
def main(argv=sys.argv):
args = argv[1:]
if len(args) != 3:
return -1
infile = args[0]
outdir = args[1]
rscpath = args[2]
rv = 1
socialpath = os.path.join(rscpath,'sdrmlist.txt')
if os.path.exists(socialpath):
keydata = file(socialpath,'r').read()
keydata = keydata.rstrip(os.linesep)
ar = keydata.split(',')
for i in ar:
try:
name, cc8 = i.split(':')
except ValueError:
print ' Error parsing user supplied social drm data.'
return 1
rv = erdr2pml.decryptBook(infile, outdir, name, cc8, True)
if rv == 0:
break
return rv
if __name__ == "__main__":
sys.exit(main())

View File

@@ -0,0 +1,54 @@
#!/usr/bin/env python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
class Unbuffered:
def __init__(self, stream):
self.stream = stream
def write(self, data):
self.stream.write(data)
self.stream.flush()
def __getattr__(self, attr):
return getattr(self.stream, attr)
import sys
sys.stdout=Unbuffered(sys.stdout)
import os
import re
import ineptpdf
def main(argv=sys.argv):
args = argv[1:]
if len(args) != 3:
return -1
infile = args[0]
outdir = args[1]
rscpath = args[2]
errlog = ''
rv = 1
# determine a good name for the output file
name, ext = os.path.splitext(os.path.basename(infile))
outfile = os.path.join(outdir, name + '_nodrm.pdf')
# try with any keyfiles (*.der) in the rscpath
files = os.listdir(rscpath)
filefilter = re.compile("\.der$", re.IGNORECASE)
files = filter(filefilter.search, files)
if files:
for filename in files:
keypath = os.path.join(rscpath, filename)
try:
rv = ineptpdf.decryptBook(keypath, infile, outfile)
if rv == 0:
break
except Exception, e:
errlog += str(e)
rv = 1
pass
if rv != 0:
print errlog
return rv
if __name__ == "__main__":
sys.exit(main())

View File

@@ -0,0 +1,526 @@
#!/usr/bin/env python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
#
# erdr2pml.py
#
# This is a python script. You need a Python interpreter to run it.
# For example, ActiveState Python, which exists for windows.
# Changelog
#
# Based on ereader2html version 0.08 plus some later small fixes
#
# 0.01 - Initial version
# 0.02 - Support more eReader files. Support bold text and links. Fix PML decoder parsing bug.
# 0.03 - Fix incorrect variable usage at one place.
# 0.03b - enhancement by DeBockle (version 259 support)
# Custom version 0.03 - no change to eReader support, only usability changes
# - start of pep-8 indentation (spaces not tab), fix trailing blanks
# - version variable, only one place to change
# - added main routine, now callable as a library/module,
# means tools can add optional support for ereader2html
# - outdir is no longer a mandatory parameter (defaults based on input name if missing)
# - time taken output to stdout
# - Psyco support - reduces runtime by a factor of (over) 3!
# E.g. (~600Kb file) 90 secs down to 24 secs
# - newstyle classes
# - changed map call to list comprehension
# may not work with python 2.3
# without Psyco this reduces runtime to 90%
# E.g. 90 secs down to 77 secs
# Psyco with map calls takes longer, do not run with map in Psyco JIT!
# - izip calls used instead of zip (if available), further reduction
# in run time (factor of 4.5).
# E.g. (~600Kb file) 90 secs down to 20 secs
# - Python 2.6+ support, avoid DeprecationWarning with sha/sha1
# 0.04 - Footnote support, PML output, correct charset in html, support more PML tags
# - Feature change, dump out PML file
# - Added supprt for footnote tags. NOTE footnote ids appear to be bad (not usable)
# in some pdb files :-( due to the same id being used multiple times
# - Added correct charset encoding (pml is based on cp1252)
# - Added logging support.
# 0.05 - Improved type 272 support for sidebars, links, chapters, metainfo, etc
# 0.06 - Merge of 0.04 and 0.05. Improved HTML output
# Placed images in subfolder, so that it's possible to just
# drop the book.pml file onto DropBook to make an unencrypted
# copy of the eReader file.
# Using that with Calibre works a lot better than the HTML
# conversion in this code.
# 0.07 - Further Improved type 272 support for sidebars with all earlier fixes
# 0.08 - fixed typos, removed extraneous things
# 0.09 - fixed typos in first_pages to first_page to again support older formats
# 0.10 - minor cleanups
# 0.11 - fixups for using correct xml for footnotes and sidebars for use with Dropbook
# 0.12 - Fix added to prevent lowercasing of image names when the pml code itself uses a different case in the link name.
# 0.13 - change to unbuffered stdout for use with gui front ends
# 0.14 - contributed enhancement to support --make-pmlz switch
# 0.15 - enabled high-ascii to pml character encoding. DropBook now works on Mac.
# 0.16 - convert to use openssl DES (very very fast) or pure python DES if openssl's libcrypto is not available
# 0.17 - added support for pycrypto's DES as well
# 0.18 - on Windows try PyCrypto first and OpenSSL next
# 0.19 - Modify the interface to allow use of import
# 0.20 - modify to allow use inside new interface for calibre plugins
# 0.21 - Support eReader (drm) version 11.
# - Don't reject dictionary format.
# - Ignore sidebars for dictionaries (different format?)
__version__='0.21'
class Unbuffered:
def __init__(self, stream):
self.stream = stream
def write(self, data):
self.stream.write(data)
self.stream.flush()
def __getattr__(self, attr):
return getattr(self.stream, attr)
import sys
import struct, binascii, getopt, zlib, os, os.path, urllib, tempfile
if 'calibre' in sys.modules:
inCalibre = True
else:
inCalibre = False
Des = None
if sys.platform.startswith('win'):
# first try with pycrypto
if inCalibre:
from calibre_plugins.erdrpdb2pml import pycrypto_des
else:
import pycrypto_des
Des = pycrypto_des.load_pycrypto()
if Des == None:
# they try with openssl
if inCalibre:
from calibre_plugins.erdrpdb2pml import openssl_des
else:
import openssl_des
Des = openssl_des.load_libcrypto()
else:
# first try with openssl
if inCalibre:
from calibre_plugins.erdrpdb2pml import openssl_des
else:
import openssl_des
Des = openssl_des.load_libcrypto()
if Des == None:
# then try with pycrypto
if inCalibre:
from calibre_plugins.erdrpdb2pml import pycrypto_des
else:
import pycrypto_des
Des = pycrypto_des.load_pycrypto()
# if that did not work then use pure python implementation
# of DES and try to speed it up with Psycho
if Des == None:
if inCalibre:
from calibre_plugins.erdrpdb2pml import python_des
else:
import python_des
Des = python_des.Des
# Import Psyco if available
try:
# http://psyco.sourceforge.net
import psyco
psyco.full()
except ImportError:
pass
try:
from hashlib import sha1
except ImportError:
# older Python release
import sha
sha1 = lambda s: sha.new(s)
import cgi
import logging
logging.basicConfig()
#logging.basicConfig(level=logging.DEBUG)
class Sectionizer(object):
bkType = "Book"
def __init__(self, filename, ident):
self.contents = file(filename, 'rb').read()
self.header = self.contents[0:72]
self.num_sections, = struct.unpack('>H', self.contents[76:78])
# Dictionary or normal content (TODO: Not hard-coded)
if self.header[0x3C:0x3C+8] != ident:
if self.header[0x3C:0x3C+8] == "PDctPPrs":
self.bkType = "Dict"
else:
raise ValueError('Invalid file format')
self.sections = []
for i in xrange(self.num_sections):
offset, a1,a2,a3,a4 = struct.unpack('>LBBBB', self.contents[78+i*8:78+i*8+8])
flags, val = a1, a2<<16|a3<<8|a4
self.sections.append( (offset, flags, val) )
def loadSection(self, section):
if section + 1 == self.num_sections:
end_off = len(self.contents)
else:
end_off = self.sections[section + 1][0]
off = self.sections[section][0]
return self.contents[off:end_off]
def sanitizeFileName(s):
r = ''
for c in s:
if c in "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789_.-":
r += c
return r
def fixKey(key):
def fixByte(b):
return b ^ ((b ^ (b<<1) ^ (b<<2) ^ (b<<3) ^ (b<<4) ^ (b<<5) ^ (b<<6) ^ (b<<7) ^ 0x80) & 0x80)
return "".join([chr(fixByte(ord(a))) for a in key])
def deXOR(text, sp, table):
r=''
j = sp
for i in xrange(len(text)):
r += chr(ord(table[j]) ^ ord(text[i]))
j = j + 1
if j == len(table):
j = 0
return r
class EreaderProcessor(object):
def __init__(self, sect, username, creditcard):
self.section_reader = sect.loadSection
data = self.section_reader(0)
version, = struct.unpack('>H', data[0:2])
self.version = version
logging.info('eReader file format version %s', version)
if version != 272 and version != 260 and version != 259:
raise ValueError('incorrect eReader version %d (error 1)' % version)
data = self.section_reader(1)
self.data = data
des = Des(fixKey(data[0:8]))
cookie_shuf, cookie_size = struct.unpack('>LL', des.decrypt(data[-8:]))
if cookie_shuf < 3 or cookie_shuf > 0x14 or cookie_size < 0xf0 or cookie_size > 0x200:
raise ValueError('incorrect eReader version (error 2)')
input = des.decrypt(data[-cookie_size:])
def unshuff(data, shuf):
r = [''] * len(data)
j = 0
for i in xrange(len(data)):
j = (j + shuf) % len(data)
r[j] = data[i]
assert len("".join(r)) == len(data)
return "".join(r)
r = unshuff(input[0:-8], cookie_shuf)
def fixUsername(s):
r = ''
for c in s.lower():
if (c >= 'a' and c <= 'z' or c >= '0' and c <= '9'):
r += c
return r
user_key = struct.pack('>LL', binascii.crc32(fixUsername(username)) & 0xffffffff, binascii.crc32(creditcard[-8:])& 0xffffffff)
drm_sub_version = struct.unpack('>H', r[0:2])[0]
self.num_text_pages = struct.unpack('>H', r[2:4])[0] - 1
self.num_image_pages = struct.unpack('>H', r[26:26+2])[0]
self.first_image_page = struct.unpack('>H', r[24:24+2])[0]
# Default values
self.num_footnote_pages = 0
self.num_sidebar_pages = 0
self.first_footnote_page = -1
self.first_sidebar_page = -1
if self.version == 272:
self.num_footnote_pages = struct.unpack('>H', r[46:46+2])[0]
self.first_footnote_page = struct.unpack('>H', r[44:44+2])[0]
if (sect.bkType == "Book"):
self.num_sidebar_pages = struct.unpack('>H', r[38:38+2])[0]
self.first_sidebar_page = struct.unpack('>H', r[36:36+2])[0]
# self.num_bookinfo_pages = struct.unpack('>H', r[34:34+2])[0]
# self.first_bookinfo_page = struct.unpack('>H', r[32:32+2])[0]
# self.num_chapter_pages = struct.unpack('>H', r[22:22+2])[0]
# self.first_chapter_page = struct.unpack('>H', r[20:20+2])[0]
# self.num_link_pages = struct.unpack('>H', r[30:30+2])[0]
# self.first_link_page = struct.unpack('>H', r[28:28+2])[0]
# self.num_xtextsize_pages = struct.unpack('>H', r[54:54+2])[0]
# self.first_xtextsize_page = struct.unpack('>H', r[52:52+2])[0]
# **before** data record 1 was decrypted and unshuffled, it contained data
# to create an XOR table and which is used to fix footnote record 0, link records, chapter records, etc
self.xortable_offset = struct.unpack('>H', r[40:40+2])[0]
self.xortable_size = struct.unpack('>H', r[42:42+2])[0]
self.xortable = self.data[self.xortable_offset:self.xortable_offset + self.xortable_size]
else:
# Nothing needs to be done
pass
# self.num_bookinfo_pages = 0
# self.num_chapter_pages = 0
# self.num_link_pages = 0
# self.num_xtextsize_pages = 0
# self.first_bookinfo_page = -1
# self.first_chapter_page = -1
# self.first_link_page = -1
# self.first_xtextsize_page = -1
logging.debug('self.num_text_pages %d', self.num_text_pages)
logging.debug('self.num_footnote_pages %d, self.first_footnote_page %d', self.num_footnote_pages , self.first_footnote_page)
logging.debug('self.num_sidebar_pages %d, self.first_sidebar_page %d', self.num_sidebar_pages , self.first_sidebar_page)
self.flags = struct.unpack('>L', r[4:8])[0]
reqd_flags = (1<<9) | (1<<7) | (1<<10)
if (self.flags & reqd_flags) != reqd_flags:
print "Flags: 0x%X" % self.flags
raise ValueError('incompatible eReader file')
des = Des(fixKey(user_key))
if version == 259:
if drm_sub_version != 7:
raise ValueError('incorrect eReader version %d (error 3)' % drm_sub_version)
encrypted_key_sha = r[44:44+20]
encrypted_key = r[64:64+8]
elif version == 260:
if drm_sub_version != 13 and drm_sub_version != 11:
raise ValueError('incorrect eReader version %d (error 3)' % drm_sub_version)
if drm_sub_version == 13:
encrypted_key = r[44:44+8]
encrypted_key_sha = r[52:52+20]
else:
encrypted_key = r[64:64+8]
encrypted_key_sha = r[44:44+20]
elif version == 272:
encrypted_key = r[172:172+8]
encrypted_key_sha = r[56:56+20]
self.content_key = des.decrypt(encrypted_key)
if sha1(self.content_key).digest() != encrypted_key_sha:
raise ValueError('Incorrect Name and/or Credit Card')
def getNumImages(self):
return self.num_image_pages
def getImage(self, i):
sect = self.section_reader(self.first_image_page + i)
name = sect[4:4+32].strip('\0')
data = sect[62:]
return sanitizeFileName(name), data
# def getChapterNamePMLOffsetData(self):
# cv = ''
# if self.num_chapter_pages > 0:
# for i in xrange(self.num_chapter_pages):
# chaps = self.section_reader(self.first_chapter_page + i)
# j = i % self.xortable_size
# offname = deXOR(chaps, j, self.xortable)
# offset = struct.unpack('>L', offname[0:4])[0]
# name = offname[4:].strip('\0')
# cv += '%d|%s\n' % (offset, name)
# return cv
# def getLinkNamePMLOffsetData(self):
# lv = ''
# if self.num_link_pages > 0:
# for i in xrange(self.num_link_pages):
# links = self.section_reader(self.first_link_page + i)
# j = i % self.xortable_size
# offname = deXOR(links, j, self.xortable)
# offset = struct.unpack('>L', offname[0:4])[0]
# name = offname[4:].strip('\0')
# lv += '%d|%s\n' % (offset, name)
# return lv
# def getExpandedTextSizesData(self):
# ts = ''
# if self.num_xtextsize_pages > 0:
# tsize = deXOR(self.section_reader(self.first_xtextsize_page), 0, self.xortable)
# for i in xrange(self.num_text_pages):
# xsize = struct.unpack('>H', tsize[0:2])[0]
# ts += "%d\n" % xsize
# tsize = tsize[2:]
# return ts
# def getBookInfo(self):
# bkinfo = ''
# if self.num_bookinfo_pages > 0:
# info = self.section_reader(self.first_bookinfo_page)
# bkinfo = deXOR(info, 0, self.xortable)
# bkinfo = bkinfo.replace('\0','|')
# bkinfo += '\n'
# return bkinfo
def getText(self):
des = Des(fixKey(self.content_key))
r = ''
for i in xrange(self.num_text_pages):
logging.debug('get page %d', i)
r += zlib.decompress(des.decrypt(self.section_reader(1 + i)))
# now handle footnotes pages
if self.num_footnote_pages > 0:
r += '\n'
# the record 0 of the footnote section must pass through the Xor Table to make it useful
sect = self.section_reader(self.first_footnote_page)
fnote_ids = deXOR(sect, 0, self.xortable)
# the remaining records of the footnote sections need to be decoded with the content_key and zlib inflated
des = Des(fixKey(self.content_key))
for i in xrange(1,self.num_footnote_pages):
logging.debug('get footnotepage %d', i)
id_len = ord(fnote_ids[2])
id = fnote_ids[3:3+id_len]
fmarker = '<footnote id="%s">\n' % id
fmarker += zlib.decompress(des.decrypt(self.section_reader(self.first_footnote_page + i)))
fmarker += '\n</footnote>\n'
r += fmarker
fnote_ids = fnote_ids[id_len+4:]
# TODO: Handle dictionary index (?) pages - which are also marked as
# sidebar_pages (?). For now dictionary sidebars are ignored
# For dictionaries - record 0 is null terminated strings, followed by
# blocks of around 62000 bytes and a final block. Not sure of the
# encoding
# now handle sidebar pages
if self.num_sidebar_pages > 0:
r += '\n'
# the record 0 of the sidebar section must pass through the Xor Table to make it useful
sect = self.section_reader(self.first_sidebar_page)
sbar_ids = deXOR(sect, 0, self.xortable)
# the remaining records of the sidebar sections need to be decoded with the content_key and zlib inflated
des = Des(fixKey(self.content_key))
for i in xrange(1,self.num_sidebar_pages):
id_len = ord(sbar_ids[2])
id = sbar_ids[3:3+id_len]
smarker = '<sidebar id="%s">\n' % id
smarker += zlib.decompress(des.decrypt(self.section_reader(self.first_sidebar_page + i)))
smarker += '\n</sidebar>\n'
r += smarker
sbar_ids = sbar_ids[id_len+4:]
return r
def cleanPML(pml):
# Convert special characters to proper PML code. High ASCII start at (\x80, \a128) and go up to (\xff, \a255)
pml2 = pml
for k in xrange(128,256):
badChar = chr(k)
pml2 = pml2.replace(badChar, '\\a%03d' % k)
return pml2
def convertEreaderToPml(infile, name, cc, outdir):
if not os.path.exists(outdir):
os.makedirs(outdir)
bookname = os.path.splitext(os.path.basename(infile))[0]
print " Decoding File"
sect = Sectionizer(infile, 'PNRdPPrs')
er = EreaderProcessor(sect, name, cc)
if er.getNumImages() > 0:
print " Extracting images"
imagedir = bookname + '_img/'
imagedirpath = os.path.join(outdir,imagedir)
if not os.path.exists(imagedirpath):
os.makedirs(imagedirpath)
for i in xrange(er.getNumImages()):
name, contents = er.getImage(i)
file(os.path.join(imagedirpath, name), 'wb').write(contents)
print " Extracting pml"
pml_string = er.getText()
pmlfilename = bookname + ".pml"
file(os.path.join(outdir, pmlfilename),'wb').write(cleanPML(pml_string))
# bkinfo = er.getBookInfo()
# if bkinfo != '':
# print " Extracting book meta information"
# file(os.path.join(outdir, 'bookinfo.txt'),'wb').write(bkinfo)
def decryptBook(infile, outdir, name, cc, make_pmlz):
if make_pmlz :
# ignore specified outdir, use tempdir instead
outdir = tempfile.mkdtemp()
try:
print "Processing..."
convertEreaderToPml(infile, name, cc, outdir)
if make_pmlz :
import zipfile
import shutil
print " Creating PMLZ file"
zipname = infile[:-4] + '.pmlz'
myZipFile = zipfile.ZipFile(zipname,'w',zipfile.ZIP_STORED, False)
list = os.listdir(outdir)
for file in list:
localname = file
filePath = os.path.join(outdir,file)
if os.path.isfile(filePath):
myZipFile.write(filePath, localname)
elif os.path.isdir(filePath):
imageList = os.listdir(filePath)
localimgdir = os.path.basename(filePath)
for image in imageList:
localname = os.path.join(localimgdir,image)
imagePath = os.path.join(filePath,image)
if os.path.isfile(imagePath):
myZipFile.write(imagePath, localname)
myZipFile.close()
# remove temporary directory
shutil.rmtree(outdir, True)
print 'output is %s' % zipname
else :
print 'output in %s' % outdir
print "done"
except ValueError, e:
print "Error: %s" % e
return 1
return 0
def usage():
print "Converts DRMed eReader books to PML Source"
print "Usage:"
print " erdr2pml [options] infile.pdb [outdir] \"your name\" credit_card_number "
print " "
print "Options: "
print " -h prints this message"
print " --make-pmlz create PMLZ instead of using output directory"
print " "
print "Note:"
print " if ommitted, outdir defaults based on 'infile.pdb'"
print " It's enough to enter the last 8 digits of the credit card number"
return
def main(argv=None):
try:
opts, args = getopt.getopt(sys.argv[1:], "h", ["make-pmlz"])
except getopt.GetoptError, err:
print str(err)
usage()
return 1
make_pmlz = False
for o, a in opts:
if o == "-h":
usage()
return 0
elif o == "--make-pmlz":
make_pmlz = True
print "eRdr2Pml v%s. Copyright (c) 2009 The Dark Reverser" % __version__
if len(args)!=3 and len(args)!=4:
usage()
return 1
if len(args)==3:
infile, name, cc = args[0], args[1], args[2]
outdir = infile[:-4] + '_Source'
elif len(args)==4:
infile, outdir, name, cc = args[0], args[1], args[2], args[3]
return decryptBook(infile, outdir, name, cc, make_pmlz)
if __name__ == "__main__":
sys.stdout=Unbuffered(sys.stdout)
sys.exit(main())

View File

@@ -0,0 +1,793 @@
#! /usr/bin/python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
# For use with Topaz Scripts Version 2.6
import sys
import csv
import os
import math
import getopt
from struct import pack
from struct import unpack
class DocParser(object):
def __init__(self, flatxml, classlst, fileid, bookDir, gdict, fixedimage):
self.id = os.path.basename(fileid).replace('.dat','')
self.svgcount = 0
self.docList = flatxml.split('\n')
self.docSize = len(self.docList)
self.classList = {}
self.bookDir = bookDir
self.gdict = gdict
tmpList = classlst.split('\n')
for pclass in tmpList:
if pclass != '':
# remove the leading period from the css name
cname = pclass[1:]
self.classList[cname] = True
self.fixedimage = fixedimage
self.ocrtext = []
self.link_id = []
self.link_title = []
self.link_page = []
self.link_href = []
self.link_type = []
self.dehyphen_rootid = []
self.paracont_stemid = []
self.parastems_stemid = []
def getGlyph(self, gid):
result = ''
id='id="gl%d"' % gid
return self.gdict.lookup(id)
def glyphs_to_image(self, glyphList):
def extract(path, key):
b = path.find(key) + len(key)
e = path.find(' ',b)
return int(path[b:e])
svgDir = os.path.join(self.bookDir,'svg')
imgDir = os.path.join(self.bookDir,'img')
imgname = self.id + '_%04d.svg' % self.svgcount
imgfile = os.path.join(imgDir,imgname)
# get glyph information
gxList = self.getData('info.glyph.x',0,-1)
gyList = self.getData('info.glyph.y',0,-1)
gidList = self.getData('info.glyph.glyphID',0,-1)
gids = []
maxws = []
maxhs = []
xs = []
ys = []
gdefs = []
# get path defintions, positions, dimensions for each glyph
# that makes up the image, and find min x and min y to reposition origin
minx = -1
miny = -1
for j in glyphList:
gid = gidList[j]
gids.append(gid)
xs.append(gxList[j])
if minx == -1: minx = gxList[j]
else : minx = min(minx, gxList[j])
ys.append(gyList[j])
if miny == -1: miny = gyList[j]
else : miny = min(miny, gyList[j])
path = self.getGlyph(gid)
gdefs.append(path)
maxws.append(extract(path,'width='))
maxhs.append(extract(path,'height='))
# change the origin to minx, miny and calc max height and width
maxw = maxws[0] + xs[0] - minx
maxh = maxhs[0] + ys[0] - miny
for j in xrange(0, len(xs)):
xs[j] = xs[j] - minx
ys[j] = ys[j] - miny
maxw = max( maxw, (maxws[j] + xs[j]) )
maxh = max( maxh, (maxhs[j] + ys[j]) )
# open the image file for output
ifile = open(imgfile,'w')
ifile.write('<?xml version="1.0" standalone="no"?>\n')
ifile.write('<!DOCTYPE svg PUBLIC "-//W3C/DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">\n')
ifile.write('<svg width="%dpx" height="%dpx" viewBox="0 0 %d %d" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1">\n' % (math.floor(maxw/10), math.floor(maxh/10), maxw, maxh))
ifile.write('<defs>\n')
for j in xrange(0,len(gdefs)):
ifile.write(gdefs[j])
ifile.write('</defs>\n')
for j in xrange(0,len(gids)):
ifile.write('<use xlink:href="#gl%d" x="%d" y="%d" />\n' % (gids[j], xs[j], ys[j]))
ifile.write('</svg>')
ifile.close()
return 0
# return tag at line pos in document
def lineinDoc(self, pos) :
if (pos >= 0) and (pos < self.docSize) :
item = self.docList[pos]
if item.find('=') >= 0:
(name, argres) = item.split('=',1)
else :
name = item
argres = ''
return name, argres
# find tag in doc if within pos to end inclusive
def findinDoc(self, tagpath, pos, end) :
result = None
if end == -1 :
end = self.docSize
else:
end = min(self.docSize, end)
foundat = -1
for j in xrange(pos, end):
item = self.docList[j]
if item.find('=') >= 0:
(name, argres) = item.split('=',1)
else :
name = item
argres = ''
if name.endswith(tagpath) :
result = argres
foundat = j
break
return foundat, result
# return list of start positions for the tagpath
def posinDoc(self, tagpath):
startpos = []
pos = 0
res = ""
while res != None :
(foundpos, res) = self.findinDoc(tagpath, pos, -1)
if res != None :
startpos.append(foundpos)
pos = foundpos + 1
return startpos
# returns a vector of integers for the tagpath
def getData(self, tagpath, pos, end):
argres=[]
(foundat, argt) = self.findinDoc(tagpath, pos, end)
if (argt != None) and (len(argt) > 0) :
argList = argt.split('|')
argres = [ int(strval) for strval in argList]
return argres
# get the class
def getClass(self, pclass):
nclass = pclass
# class names are an issue given topaz may start them with numerals (not allowed),
# use a mix of cases (which cause some browsers problems), and actually
# attach numbers after "_reclustered*" to the end to deal classeses that inherit
# from a base class (but then not actually provide all of these _reclustereed
# classes in the stylesheet!
# so we clean this up by lowercasing, prepend 'cl-', and getting any baseclass
# that exists in the stylesheet first, and then adding this specific class
# after
# also some class names have spaces in them so need to convert to dashes
if nclass != None :
nclass = nclass.replace(' ','-')
classres = ''
nclass = nclass.lower()
nclass = 'cl-' + nclass
baseclass = ''
# graphic is the base class for captions
if nclass.find('cl-cap-') >=0 :
classres = 'graphic' + ' '
else :
# strip to find baseclass
p = nclass.find('_')
if p > 0 :
baseclass = nclass[0:p]
if baseclass in self.classList:
classres += baseclass + ' '
classres += nclass
nclass = classres
return nclass
# develop a sorted description of the starting positions of
# groups and regions on the page, as well as the page type
def PageDescription(self):
def compare(x, y):
(xtype, xval) = x
(ytype, yval) = y
if xval > yval:
return 1
if xval == yval:
return 0
return -1
result = []
(pos, pagetype) = self.findinDoc('page.type',0,-1)
groupList = self.posinDoc('page.group')
groupregionList = self.posinDoc('page.group.region')
pageregionList = self.posinDoc('page.region')
# integrate into one list
for j in groupList:
result.append(('grpbeg',j))
for j in groupregionList:
result.append(('gregion',j))
for j in pageregionList:
result.append(('pregion',j))
result.sort(compare)
# insert group end and page end indicators
inGroup = False
j = 0
while True:
if j == len(result): break
rtype = result[j][0]
rval = result[j][1]
if not inGroup and (rtype == 'grpbeg') :
inGroup = True
j = j + 1
elif inGroup and (rtype in ('grpbeg', 'pregion')):
result.insert(j,('grpend',rval))
inGroup = False
else:
j = j + 1
if inGroup:
result.append(('grpend',-1))
result.append(('pageend', -1))
return pagetype, result
# build a description of the paragraph
def getParaDescription(self, start, end, regtype):
result = []
# paragraph
(pos, pclass) = self.findinDoc('paragraph.class',start,end)
pclass = self.getClass(pclass)
# if paragraph uses extratokens (extra glyphs) then make it fixed
(pos, extraglyphs) = self.findinDoc('paragraph.extratokens',start,end)
# build up a description of the paragraph in result and return it
# first check for the basic - all words paragraph
(pos, sfirst) = self.findinDoc('paragraph.firstWord',start,end)
(pos, slast) = self.findinDoc('paragraph.lastWord',start,end)
if (sfirst != None) and (slast != None) :
first = int(sfirst)
last = int(slast)
makeImage = (regtype == 'vertical') or (regtype == 'table')
makeImage = makeImage or (extraglyphs != None)
if self.fixedimage:
makeImage = makeImage or (regtype == 'fixed')
if (pclass != None):
makeImage = makeImage or (pclass.find('.inverted') >= 0)
if self.fixedimage :
makeImage = makeImage or (pclass.find('cl-f-') >= 0)
# before creating an image make sure glyph info exists
gidList = self.getData('info.glyph.glyphID',0,-1)
makeImage = makeImage & (len(gidList) > 0)
if not makeImage :
# standard all word paragraph
for wordnum in xrange(first, last):
result.append(('ocr', wordnum))
return pclass, result
# convert paragraph to svg image
# translate first and last word into first and last glyphs
# and generate inline image and include it
glyphList = []
firstglyphList = self.getData('word.firstGlyph',0,-1)
gidList = self.getData('info.glyph.glyphID',0,-1)
firstGlyph = firstglyphList[first]
if last < len(firstglyphList):
lastGlyph = firstglyphList[last]
else :
lastGlyph = len(gidList)
# handle case of white sapce paragraphs with no actual glyphs in them
# by reverting to text based paragraph
if firstGlyph >= lastGlyph:
# revert to standard text based paragraph
for wordnum in xrange(first, last):
result.append(('ocr', wordnum))
return pclass, result
for glyphnum in xrange(firstGlyph, lastGlyph):
glyphList.append(glyphnum)
# include any extratokens if they exist
(pos, sfg) = self.findinDoc('extratokens.firstGlyph',start,end)
(pos, slg) = self.findinDoc('extratokens.lastGlyph',start,end)
if (sfg != None) and (slg != None):
for glyphnum in xrange(int(sfg), int(slg)):
glyphList.append(glyphnum)
num = self.svgcount
self.glyphs_to_image(glyphList)
self.svgcount += 1
result.append(('svg', num))
return pclass, result
# this type of paragraph may be made up of multiple spans, inline
# word monograms (images), and words with semantic meaning,
# plus glyphs used to form starting letter of first word
# need to parse this type line by line
line = start + 1
word_class = ''
# if end is -1 then we must search to end of document
if end == -1 :
end = self.docSize
# seems some xml has last* coming before first* so we have to
# handle any order
sp_first = -1
sp_last = -1
gl_first = -1
gl_last = -1
ws_first = -1
ws_last = -1
word_class = ''
word_semantic_type = ''
while (line < end) :
(name, argres) = self.lineinDoc(line)
if name.endswith('span.firstWord') :
sp_first = int(argres)
elif name.endswith('span.lastWord') :
sp_last = int(argres)
elif name.endswith('word.firstGlyph') :
gl_first = int(argres)
elif name.endswith('word.lastGlyph') :
gl_last = int(argres)
elif name.endswith('word_semantic.firstWord'):
ws_first = int(argres)
elif name.endswith('word_semantic.lastWord'):
ws_last = int(argres)
elif name.endswith('word.class'):
(cname, space) = argres.split('-',1)
if space == '' : space = '0'
if (cname == 'spaceafter') and (int(space) > 0) :
word_class = 'sa'
elif name.endswith('word.img.src'):
result.append(('img' + word_class, int(argres)))
word_class = ''
elif name.endswith('region.img.src'):
result.append(('img' + word_class, int(argres)))
if (sp_first != -1) and (sp_last != -1):
for wordnum in xrange(sp_first, sp_last):
result.append(('ocr', wordnum))
sp_first = -1
sp_last = -1
if (gl_first != -1) and (gl_last != -1):
glyphList = []
for glyphnum in xrange(gl_first, gl_last):
glyphList.append(glyphnum)
num = self.svgcount
self.glyphs_to_image(glyphList)
self.svgcount += 1
result.append(('svg', num))
gl_first = -1
gl_last = -1
if (ws_first != -1) and (ws_last != -1):
for wordnum in xrange(ws_first, ws_last):
result.append(('ocr', wordnum))
ws_first = -1
ws_last = -1
line += 1
return pclass, result
def buildParagraph(self, pclass, pdesc, type, regtype) :
parares = ''
sep =''
classres = ''
if pclass :
classres = ' class="' + pclass + '"'
br_lb = (regtype == 'fixed') or (regtype == 'chapterheading') or (regtype == 'vertical')
handle_links = len(self.link_id) > 0
if (type == 'full') or (type == 'begin') :
parares += '<p' + classres + '>'
if (type == 'end'):
parares += ' '
lstart = len(parares)
cnt = len(pdesc)
for j in xrange( 0, cnt) :
(wtype, num) = pdesc[j]
if wtype == 'ocr' :
word = self.ocrtext[num]
sep = ' '
if handle_links:
link = self.link_id[num]
if (link > 0):
linktype = self.link_type[link-1]
title = self.link_title[link-1]
if (title == "") or (parares.rfind(title) < 0):
title=parares[lstart:]
if linktype == 'external' :
linkhref = self.link_href[link-1]
linkhtml = '<a href="%s">' % linkhref
else :
if len(self.link_page) >= link :
ptarget = self.link_page[link-1] - 1
linkhtml = '<a href="#page%04d">' % ptarget
else :
# just link to the current page
linkhtml = '<a href="#' + self.id + '">'
linkhtml += title + '</a>'
pos = parares.rfind(title)
if pos >= 0:
parares = parares[0:pos] + linkhtml + parares[pos+len(title):]
else :
parares += linkhtml
lstart = len(parares)
if word == '_link_' : word = ''
elif (link < 0) :
if word == '_link_' : word = ''
if word == '_lb_':
if ((num-1) in self.dehyphen_rootid ) or handle_links:
word = ''
sep = ''
elif br_lb :
word = '<br />\n'
sep = ''
else :
word = '\n'
sep = ''
if num in self.dehyphen_rootid :
word = word[0:-1]
sep = ''
parares += word + sep
elif wtype == 'img' :
sep = ''
parares += '<img src="img/img%04d.jpg" alt="" />' % num
parares += sep
elif wtype == 'imgsa' :
sep = ' '
parares += '<img src="img/img%04d.jpg" alt="" />' % num
parares += sep
elif wtype == 'svg' :
sep = ''
parares += '<img src="img/' + self.id + '_%04d.svg" alt="" />' % num
parares += sep
if len(sep) > 0 : parares = parares[0:-1]
if (type == 'full') or (type == 'end') :
parares += '</p>'
return parares
def buildTOCEntry(self, pdesc) :
parares = ''
sep =''
tocentry = ''
handle_links = len(self.link_id) > 0
lstart = 0
cnt = len(pdesc)
for j in xrange( 0, cnt) :
(wtype, num) = pdesc[j]
if wtype == 'ocr' :
word = self.ocrtext[num]
sep = ' '
if handle_links:
link = self.link_id[num]
if (link > 0):
linktype = self.link_type[link-1]
title = self.link_title[link-1]
title = title.rstrip('. ')
alt_title = parares[lstart:]
alt_title = alt_title.strip()
# now strip off the actual printed page number
alt_title = alt_title.rstrip('01234567890ivxldIVXLD-.')
alt_title = alt_title.rstrip('. ')
# skip over any external links - can't have them in a books toc
if linktype == 'external' :
title = ''
alt_title = ''
linkpage = ''
else :
if len(self.link_page) >= link :
ptarget = self.link_page[link-1] - 1
linkpage = '%04d' % ptarget
else :
# just link to the current page
linkpage = self.id[4:]
if len(alt_title) >= len(title):
title = alt_title
if title != '' and linkpage != '':
tocentry += title + '|' + linkpage + '\n'
lstart = len(parares)
if word == '_link_' : word = ''
elif (link < 0) :
if word == '_link_' : word = ''
if word == '_lb_':
word = ''
sep = ''
if num in self.dehyphen_rootid :
word = word[0:-1]
sep = ''
parares += word + sep
else :
continue
return tocentry
# walk the document tree collecting the information needed
# to build an html page using the ocrText
def process(self):
tocinfo = ''
hlst = []
# get the ocr text
(pos, argres) = self.findinDoc('info.word.ocrText',0,-1)
if argres : self.ocrtext = argres.split('|')
# get information to dehyphenate the text
self.dehyphen_rootid = self.getData('info.dehyphen.rootID',0,-1)
# determine if first paragraph is continued from previous page
(pos, self.parastems_stemid) = self.findinDoc('info.paraStems.stemID',0,-1)
first_para_continued = (self.parastems_stemid != None)
# determine if last paragraph is continued onto the next page
(pos, self.paracont_stemid) = self.findinDoc('info.paraCont.stemID',0,-1)
last_para_continued = (self.paracont_stemid != None)
# collect link ids
self.link_id = self.getData('info.word.link_id',0,-1)
# collect link destination page numbers
self.link_page = self.getData('info.links.page',0,-1)
# collect link types (container versus external)
(pos, argres) = self.findinDoc('info.links.type',0,-1)
if argres : self.link_type = argres.split('|')
# collect link destinations
(pos, argres) = self.findinDoc('info.links.href',0,-1)
if argres : self.link_href = argres.split('|')
# collect link titles
(pos, argres) = self.findinDoc('info.links.title',0,-1)
if argres :
self.link_title = argres.split('|')
else:
self.link_title.append('')
# get a descriptions of the starting points of the regions
# and groups on the page
(pagetype, pageDesc) = self.PageDescription()
regcnt = len(pageDesc) - 1
anchorSet = False
breakSet = False
inGroup = False
# process each region on the page and convert what you can to html
for j in xrange(regcnt):
(etype, start) = pageDesc[j]
(ntype, end) = pageDesc[j+1]
# set anchor for link target on this page
if not anchorSet and not first_para_continued:
hlst.append('<div style="visibility: hidden; height: 0; width: 0;" id="')
hlst.append(self.id + '" title="pagetype_' + pagetype + '"></div>\n')
anchorSet = True
# handle groups of graphics with text captions
if (etype == 'grpbeg'):
(pos, grptype) = self.findinDoc('group.type', start, end)
if grptype != None:
if grptype == 'graphic':
gcstr = ' class="' + grptype + '"'
hlst.append('<div' + gcstr + '>')
inGroup = True
elif (etype == 'grpend'):
if inGroup:
hlst.append('</div>\n')
inGroup = False
else:
(pos, regtype) = self.findinDoc('region.type',start,end)
if regtype == 'graphic' :
(pos, simgsrc) = self.findinDoc('img.src',start,end)
if simgsrc:
if inGroup:
hlst.append('<img src="img/img%04d.jpg" alt="" />' % int(simgsrc))
else:
hlst.append('<div class="graphic"><img src="img/img%04d.jpg" alt="" /></div>' % int(simgsrc))
elif regtype == 'chapterheading' :
(pclass, pdesc) = self.getParaDescription(start,end, regtype)
if not breakSet:
hlst.append('<div style="page-break-after: always;">&nbsp;</div>\n')
breakSet = True
tag = 'h1'
if pclass and (len(pclass) >= 7):
if pclass[3:7] == 'ch1-' : tag = 'h1'
if pclass[3:7] == 'ch2-' : tag = 'h2'
if pclass[3:7] == 'ch3-' : tag = 'h3'
hlst.append('<' + tag + ' class="' + pclass + '">')
else:
hlst.append('<' + tag + '>')
hlst.append(self.buildParagraph(pclass, pdesc, 'middle', regtype))
hlst.append('</' + tag + '>')
elif (regtype == 'text') or (regtype == 'fixed') or (regtype == 'insert') or (regtype == 'listitem'):
ptype = 'full'
# check to see if this is a continution from the previous page
if first_para_continued :
ptype = 'end'
first_para_continued = False
(pclass, pdesc) = self.getParaDescription(start,end, regtype)
if pclass and (len(pclass) >= 6) and (ptype == 'full'):
tag = 'p'
if pclass[3:6] == 'h1-' : tag = 'h4'
if pclass[3:6] == 'h2-' : tag = 'h5'
if pclass[3:6] == 'h3-' : tag = 'h6'
hlst.append('<' + tag + ' class="' + pclass + '">')
hlst.append(self.buildParagraph(pclass, pdesc, 'middle', regtype))
hlst.append('</' + tag + '>')
else :
hlst.append(self.buildParagraph(pclass, pdesc, ptype, regtype))
elif (regtype == 'tocentry') :
ptype = 'full'
if first_para_continued :
ptype = 'end'
first_para_continued = False
(pclass, pdesc) = self.getParaDescription(start,end, regtype)
tocinfo += self.buildTOCEntry(pdesc)
hlst.append(self.buildParagraph(pclass, pdesc, ptype, regtype))
elif (regtype == 'vertical') or (regtype == 'table') :
ptype = 'full'
if inGroup:
ptype = 'middle'
if first_para_continued :
ptype = 'end'
first_para_continued = False
(pclass, pdesc) = self.getParaDescription(start, end, regtype)
hlst.append(self.buildParagraph(pclass, pdesc, ptype, regtype))
elif (regtype == 'synth_fcvr.center'):
(pos, simgsrc) = self.findinDoc('img.src',start,end)
if simgsrc:
hlst.append('<div class="graphic"><img src="img/img%04d.jpg" alt="" /></div>' % int(simgsrc))
else :
print ' Making region type', regtype,
(pos, temp) = self.findinDoc('paragraph',start,end)
(pos2, temp) = self.findinDoc('span',start,end)
if pos != -1 or pos2 != -1:
print ' a "text" region'
orig_regtype = regtype
regtype = 'fixed'
ptype = 'full'
# check to see if this is a continution from the previous page
if first_para_continued :
ptype = 'end'
first_para_continued = False
(pclass, pdesc) = self.getParaDescription(start,end, regtype)
if not pclass:
if orig_regtype.endswith('.right') : pclass = 'cl-right'
elif orig_regtype.endswith('.center') : pclass = 'cl-center'
elif orig_regtype.endswith('.left') : pclass = 'cl-left'
elif orig_regtype.endswith('.justify') : pclass = 'cl-justify'
if pclass and (ptype == 'full') and (len(pclass) >= 6):
tag = 'p'
if pclass[3:6] == 'h1-' : tag = 'h4'
if pclass[3:6] == 'h2-' : tag = 'h5'
if pclass[3:6] == 'h3-' : tag = 'h6'
hlst.append('<' + tag + ' class="' + pclass + '">')
hlst.append(self.buildParagraph(pclass, pdesc, 'middle', regtype))
hlst.append('</' + tag + '>')
else :
hlst.append(self.buildParagraph(pclass, pdesc, ptype, regtype))
else :
print ' a "graphic" region'
(pos, simgsrc) = self.findinDoc('img.src',start,end)
if simgsrc:
hlst.append('<div class="graphic"><img src="img/img%04d.jpg" alt="" /></div>' % int(simgsrc))
htmlpage = "".join(hlst)
if last_para_continued :
if htmlpage[-4:] == '</p>':
htmlpage = htmlpage[0:-4]
last_para_continued = False
return htmlpage, tocinfo
def convert2HTML(flatxml, classlst, fileid, bookDir, gdict, fixedimage):
# create a document parser
dp = DocParser(flatxml, classlst, fileid, bookDir, gdict, fixedimage)
htmlpage, tocinfo = dp.process()
return htmlpage, tocinfo

View File

@@ -0,0 +1,249 @@
#! /usr/bin/python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
import sys
import csv
import os
import getopt
from struct import pack
from struct import unpack
class PParser(object):
def __init__(self, gd, flatxml, meta_array):
self.gd = gd
self.flatdoc = flatxml.split('\n')
self.docSize = len(self.flatdoc)
self.temp = []
self.ph = -1
self.pw = -1
startpos = self.posinDoc('page.h') or self.posinDoc('book.h')
for p in startpos:
(name, argres) = self.lineinDoc(p)
self.ph = max(self.ph, int(argres))
startpos = self.posinDoc('page.w') or self.posinDoc('book.w')
for p in startpos:
(name, argres) = self.lineinDoc(p)
self.pw = max(self.pw, int(argres))
if self.ph <= 0:
self.ph = int(meta_array.get('pageHeight', '11000'))
if self.pw <= 0:
self.pw = int(meta_array.get('pageWidth', '8500'))
res = []
startpos = self.posinDoc('info.glyph.x')
for p in startpos:
argres = self.getDataatPos('info.glyph.x', p)
res.extend(argres)
self.gx = res
res = []
startpos = self.posinDoc('info.glyph.y')
for p in startpos:
argres = self.getDataatPos('info.glyph.y', p)
res.extend(argres)
self.gy = res
res = []
startpos = self.posinDoc('info.glyph.glyphID')
for p in startpos:
argres = self.getDataatPos('info.glyph.glyphID', p)
res.extend(argres)
self.gid = res
# return tag at line pos in document
def lineinDoc(self, pos) :
if (pos >= 0) and (pos < self.docSize) :
item = self.flatdoc[pos]
if item.find('=') >= 0:
(name, argres) = item.split('=',1)
else :
name = item
argres = ''
return name, argres
# find tag in doc if within pos to end inclusive
def findinDoc(self, tagpath, pos, end) :
result = None
if end == -1 :
end = self.docSize
else:
end = min(self.docSize, end)
foundat = -1
for j in xrange(pos, end):
item = self.flatdoc[j]
if item.find('=') >= 0:
(name, argres) = item.split('=',1)
else :
name = item
argres = ''
if name.endswith(tagpath) :
result = argres
foundat = j
break
return foundat, result
# return list of start positions for the tagpath
def posinDoc(self, tagpath):
startpos = []
pos = 0
res = ""
while res != None :
(foundpos, res) = self.findinDoc(tagpath, pos, -1)
if res != None :
startpos.append(foundpos)
pos = foundpos + 1
return startpos
def getData(self, path):
result = None
cnt = len(self.flatdoc)
for j in xrange(cnt):
item = self.flatdoc[j]
if item.find('=') >= 0:
(name, argt) = item.split('=')
argres = argt.split('|')
else:
name = item
argres = []
if (name.endswith(path)):
result = argres
break
if (len(argres) > 0) :
for j in xrange(0,len(argres)):
argres[j] = int(argres[j])
return result
def getDataatPos(self, path, pos):
result = None
item = self.flatdoc[pos]
if item.find('=') >= 0:
(name, argt) = item.split('=')
argres = argt.split('|')
else:
name = item
argres = []
if (len(argres) > 0) :
for j in xrange(0,len(argres)):
argres[j] = int(argres[j])
if (name.endswith(path)):
result = argres
return result
def getDataTemp(self, path):
result = None
cnt = len(self.temp)
for j in xrange(cnt):
item = self.temp[j]
if item.find('=') >= 0:
(name, argt) = item.split('=')
argres = argt.split('|')
else:
name = item
argres = []
if (name.endswith(path)):
result = argres
self.temp.pop(j)
break
if (len(argres) > 0) :
for j in xrange(0,len(argres)):
argres[j] = int(argres[j])
return result
def getImages(self):
result = []
self.temp = self.flatdoc
while (self.getDataTemp('img') != None):
h = self.getDataTemp('img.h')[0]
w = self.getDataTemp('img.w')[0]
x = self.getDataTemp('img.x')[0]
y = self.getDataTemp('img.y')[0]
src = self.getDataTemp('img.src')[0]
result.append('<image xlink:href="../img/img%04d.jpg" x="%d" y="%d" width="%d" height="%d" />\n' % (src, x, y, w, h))
return result
def getGlyphs(self):
result = []
if (self.gid != None) and (len(self.gid) > 0):
glyphs = []
for j in set(self.gid):
glyphs.append(j)
glyphs.sort()
for gid in glyphs:
id='id="gl%d"' % gid
path = self.gd.lookup(id)
if path:
result.append(id + ' ' + path)
return result
def convert2SVG(gdict, flat_xml, pageid, previd, nextid, svgDir, raw, meta_array, scaledpi):
mlst = []
pp = PParser(gdict, flat_xml, meta_array)
mlst.append('<?xml version="1.0" standalone="no"?>\n')
if (raw):
mlst.append('<!DOCTYPE svg PUBLIC "-//W3C/DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">\n')
mlst.append('<svg width="%fin" height="%fin" viewBox="0 0 %d %d" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1">\n' % (pp.pw / scaledpi, pp.ph / scaledpi, pp.pw -1, pp.ph -1))
mlst.append('<title>Page %d - %s by %s</title>\n' % (pageid, meta_array['Title'],meta_array['Authors']))
else:
mlst.append('<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">\n')
mlst.append('<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" ><head>\n')
mlst.append('<title>Page %d - %s by %s</title>\n' % (pageid, meta_array['Title'],meta_array['Authors']))
mlst.append('<script><![CDATA[\n')
mlst.append('function gd(){var p=window.location.href.replace(/^.*\?dpi=(\d+).*$/i,"$1");return p;}\n')
mlst.append('var dpi=%d;\n' % scaledpi)
if (previd) :
mlst.append('var prevpage="page%04d.xhtml";\n' % (previd))
if (nextid) :
mlst.append('var nextpage="page%04d.xhtml";\n' % (nextid))
mlst.append('var pw=%d;var ph=%d;' % (pp.pw, pp.ph))
mlst.append('function zoomin(){dpi=dpi*(0.8);setsize();}\n')
mlst.append('function zoomout(){dpi=dpi*1.25;setsize();}\n')
mlst.append('function setsize(){var svg=document.getElementById("svgimg");var prev=document.getElementById("prevsvg");var next=document.getElementById("nextsvg");var width=(pw/dpi)+"in";var height=(ph/dpi)+"in";svg.setAttribute("width",width);svg.setAttribute("height",height);prev.setAttribute("height",height);prev.setAttribute("width","50px");next.setAttribute("height",height);next.setAttribute("width","50px");}\n')
mlst.append('function ppage(){window.location.href=prevpage+"?dpi="+Math.round(dpi);}\n')
mlst.append('function npage(){window.location.href=nextpage+"?dpi="+Math.round(dpi);}\n')
mlst.append('var gt=gd();if(gt>0){dpi=gt;}\n')
mlst.append('window.onload=setsize;\n')
mlst.append(']]></script>\n')
mlst.append('</head>\n')
mlst.append('<body onLoad="setsize();" style="background-color:#777;text-align:center;">\n')
mlst.append('<div style="white-space:nowrap;">\n')
if previd == None:
mlst.append('<a href="javascript:ppage();"><svg id="prevsvg" viewBox="0 0 100 300" xmlns="http://www.w3.org/2000/svg" version="1.1" style="background-color:#777"></svg></a>\n')
else:
mlst.append('<a href="javascript:ppage();"><svg id="prevsvg" viewBox="0 0 100 300" xmlns="http://www.w3.org/2000/svg" version="1.1" style="background-color:#777"><polygon points="5,150,95,5,95,295" fill="#AAAAAA" /></svg></a>\n')
mlst.append('<a href="javascript:npage();"><svg id="svgimg" viewBox="0 0 %d %d" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" version="1.1" style="background-color:#FFF;border:1px solid black;">' % (pp.pw, pp.ph))
if (pp.gid != None):
mlst.append('<defs>\n')
gdefs = pp.getGlyphs()
for j in xrange(0,len(gdefs)):
mlst.append(gdefs[j])
mlst.append('</defs>\n')
img = pp.getImages()
if (img != None):
for j in xrange(0,len(img)):
mlst.append(img[j])
if (pp.gid != None):
for j in xrange(0,len(pp.gid)):
mlst.append('<use xlink:href="#gl%d" x="%d" y="%d" />\n' % (pp.gid[j], pp.gx[j], pp.gy[j]))
if (img == None or len(img) == 0) and (pp.gid == None or len(pp.gid) == 0):
xpos = "%d" % (pp.pw // 3)
ypos = "%d" % (pp.ph // 3)
mlst.append('<text x="' + xpos + '" y="' + ypos + '" font-size="' + meta_array['fontSize'] + '" font-family="Helvetica" stroke="black">This page intentionally left blank.</text>\n')
if (raw) :
mlst.append('</svg>')
else :
mlst.append('</svg></a>\n')
if nextid == None:
mlst.append('<a href="javascript:npage();"><svg id="nextsvg" viewBox="0 0 100 300" xmlns="http://www.w3.org/2000/svg" version="1.1" style="background-color:#777"></svg></a>\n')
else :
mlst.append('<a href="javascript:npage();"><svg id="nextsvg" viewBox="0 0 100 300" xmlns="http://www.w3.org/2000/svg" version="1.1" style="background-color:#777"><polygon points="5,5,5,295,95,150" fill="#AAAAAA" /></svg></a>\n')
mlst.append('</div>\n')
mlst.append('<div><a href="javascript:zoomin();">zoom in</a> - <a href="javascript:zoomout();">zoom out</a></div>\n')
mlst.append('</body>\n')
mlst.append('</html>\n')
return "".join(mlst)

View File

@@ -0,0 +1,721 @@
#! /usr/bin/python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
class Unbuffered:
def __init__(self, stream):
self.stream = stream
def write(self, data):
self.stream.write(data)
self.stream.flush()
def __getattr__(self, attr):
return getattr(self.stream, attr)
import sys
sys.stdout=Unbuffered(sys.stdout)
import csv
import os
import getopt
from struct import pack
from struct import unpack
class TpzDRMError(Exception):
pass
# local support routines
if 'calibre' in sys.modules:
inCalibre = True
else:
inCalibre = False
if inCalibre :
from calibre_plugins.k4mobidedrm import convert2xml
from calibre_plugins.k4mobidedrm import flatxml2html
from calibre_plugins.k4mobidedrm import flatxml2svg
from calibre_plugins.k4mobidedrm import stylexml2css
else :
import convert2xml
import flatxml2html
import flatxml2svg
import stylexml2css
# global switch
buildXML = False
# Get a 7 bit encoded number from a file
def readEncodedNumber(file):
flag = False
c = file.read(1)
if (len(c) == 0):
return None
data = ord(c)
if data == 0xFF:
flag = True
c = file.read(1)
if (len(c) == 0):
return None
data = ord(c)
if data >= 0x80:
datax = (data & 0x7F)
while data >= 0x80 :
c = file.read(1)
if (len(c) == 0):
return None
data = ord(c)
datax = (datax <<7) + (data & 0x7F)
data = datax
if flag:
data = -data
return data
# Get a length prefixed string from the file
def lengthPrefixString(data):
return encodeNumber(len(data))+data
def readString(file):
stringLength = readEncodedNumber(file)
if (stringLength == None):
return None
sv = file.read(stringLength)
if (len(sv) != stringLength):
return ""
return unpack(str(stringLength)+"s",sv)[0]
def getMetaArray(metaFile):
# parse the meta file
result = {}
fo = file(metaFile,'rb')
size = readEncodedNumber(fo)
for i in xrange(size):
tag = readString(fo)
value = readString(fo)
result[tag] = value
# print tag, value
fo.close()
return result
# dictionary of all text strings by index value
class Dictionary(object):
def __init__(self, dictFile):
self.filename = dictFile
self.size = 0
self.fo = file(dictFile,'rb')
self.stable = []
self.size = readEncodedNumber(self.fo)
for i in xrange(self.size):
self.stable.append(self.escapestr(readString(self.fo)))
self.pos = 0
def escapestr(self, str):
str = str.replace('&','&amp;')
str = str.replace('<','&lt;')
str = str.replace('>','&gt;')
str = str.replace('=','&#61;')
return str
def lookup(self,val):
if ((val >= 0) and (val < self.size)) :
self.pos = val
return self.stable[self.pos]
else:
print "Error - %d outside of string table limits" % val
raise TpzDRMError('outside or string table limits')
# sys.exit(-1)
def getSize(self):
return self.size
def getPos(self):
return self.pos
class PageDimParser(object):
def __init__(self, flatxml):
self.flatdoc = flatxml.split('\n')
# find tag if within pos to end inclusive
def findinDoc(self, tagpath, pos, end) :
result = None
docList = self.flatdoc
cnt = len(docList)
if end == -1 :
end = cnt
else:
end = min(cnt,end)
foundat = -1
for j in xrange(pos, end):
item = docList[j]
if item.find('=') >= 0:
(name, argres) = item.split('=')
else :
name = item
argres = ''
if name.endswith(tagpath) :
result = argres
foundat = j
break
return foundat, result
def process(self):
(pos, sph) = self.findinDoc('page.h',0,-1)
(pos, spw) = self.findinDoc('page.w',0,-1)
if (sph == None): sph = '-1'
if (spw == None): spw = '-1'
return sph, spw
def getPageDim(flatxml):
# create a document parser
dp = PageDimParser(flatxml)
(ph, pw) = dp.process()
return ph, pw
class GParser(object):
def __init__(self, flatxml):
self.flatdoc = flatxml.split('\n')
self.dpi = 1440
self.gh = self.getData('info.glyph.h')
self.gw = self.getData('info.glyph.w')
self.guse = self.getData('info.glyph.use')
if self.guse :
self.count = len(self.guse)
else :
self.count = 0
self.gvtx = self.getData('info.glyph.vtx')
self.glen = self.getData('info.glyph.len')
self.gdpi = self.getData('info.glyph.dpi')
self.vx = self.getData('info.vtx.x')
self.vy = self.getData('info.vtx.y')
self.vlen = self.getData('info.len.n')
if self.vlen :
self.glen.append(len(self.vlen))
elif self.glen:
self.glen.append(0)
if self.vx :
self.gvtx.append(len(self.vx))
elif self.gvtx :
self.gvtx.append(0)
def getData(self, path):
result = None
cnt = len(self.flatdoc)
for j in xrange(cnt):
item = self.flatdoc[j]
if item.find('=') >= 0:
(name, argt) = item.split('=')
argres = argt.split('|')
else:
name = item
argres = []
if (name == path):
result = argres
break
if (len(argres) > 0) :
for j in xrange(0,len(argres)):
argres[j] = int(argres[j])
return result
def getGlyphDim(self, gly):
if self.gdpi[gly] == 0:
return 0, 0
maxh = (self.gh[gly] * self.dpi) / self.gdpi[gly]
maxw = (self.gw[gly] * self.dpi) / self.gdpi[gly]
return maxh, maxw
def getPath(self, gly):
path = ''
if (gly < 0) or (gly >= self.count):
return path
tx = self.vx[self.gvtx[gly]:self.gvtx[gly+1]]
ty = self.vy[self.gvtx[gly]:self.gvtx[gly+1]]
p = 0
for k in xrange(self.glen[gly], self.glen[gly+1]):
if (p == 0):
zx = tx[0:self.vlen[k]+1]
zy = ty[0:self.vlen[k]+1]
else:
zx = tx[self.vlen[k-1]+1:self.vlen[k]+1]
zy = ty[self.vlen[k-1]+1:self.vlen[k]+1]
p += 1
j = 0
while ( j < len(zx) ):
if (j == 0):
# Start Position.
path += 'M %d %d ' % (zx[j] * self.dpi / self.gdpi[gly], zy[j] * self.dpi / self.gdpi[gly])
elif (j <= len(zx)-3):
# Cubic Bezier Curve
path += 'C %d %d %d %d %d %d ' % (zx[j] * self.dpi / self.gdpi[gly], zy[j] * self.dpi / self.gdpi[gly], zx[j+1] * self.dpi / self.gdpi[gly], zy[j+1] * self.dpi / self.gdpi[gly], zx[j+2] * self.dpi / self.gdpi[gly], zy[j+2] * self.dpi / self.gdpi[gly])
j += 2
elif (j == len(zx)-2):
# Cubic Bezier Curve to Start Position
path += 'C %d %d %d %d %d %d ' % (zx[j] * self.dpi / self.gdpi[gly], zy[j] * self.dpi / self.gdpi[gly], zx[j+1] * self.dpi / self.gdpi[gly], zy[j+1] * self.dpi / self.gdpi[gly], zx[0] * self.dpi / self.gdpi[gly], zy[0] * self.dpi / self.gdpi[gly])
j += 1
elif (j == len(zx)-1):
# Quadratic Bezier Curve to Start Position
path += 'Q %d %d %d %d ' % (zx[j] * self.dpi / self.gdpi[gly], zy[j] * self.dpi / self.gdpi[gly], zx[0] * self.dpi / self.gdpi[gly], zy[0] * self.dpi / self.gdpi[gly])
j += 1
path += 'z'
return path
# dictionary of all text strings by index value
class GlyphDict(object):
def __init__(self):
self.gdict = {}
def lookup(self, id):
# id='id="gl%d"' % val
if id in self.gdict:
return self.gdict[id]
return None
def addGlyph(self, val, path):
id='id="gl%d"' % val
self.gdict[id] = path
def generateBook(bookDir, raw, fixedimage):
# sanity check Topaz file extraction
if not os.path.exists(bookDir) :
print "Can not find directory with unencrypted book"
return 1
dictFile = os.path.join(bookDir,'dict0000.dat')
if not os.path.exists(dictFile) :
print "Can not find dict0000.dat file"
return 1
pageDir = os.path.join(bookDir,'page')
if not os.path.exists(pageDir) :
print "Can not find page directory in unencrypted book"
return 1
imgDir = os.path.join(bookDir,'img')
if not os.path.exists(imgDir) :
print "Can not find image directory in unencrypted book"
return 1
glyphsDir = os.path.join(bookDir,'glyphs')
if not os.path.exists(glyphsDir) :
print "Can not find glyphs directory in unencrypted book"
return 1
metaFile = os.path.join(bookDir,'metadata0000.dat')
if not os.path.exists(metaFile) :
print "Can not find metadata0000.dat in unencrypted book"
return 1
svgDir = os.path.join(bookDir,'svg')
if not os.path.exists(svgDir) :
os.makedirs(svgDir)
if buildXML:
xmlDir = os.path.join(bookDir,'xml')
if not os.path.exists(xmlDir) :
os.makedirs(xmlDir)
otherFile = os.path.join(bookDir,'other0000.dat')
if not os.path.exists(otherFile) :
print "Can not find other0000.dat in unencrypted book"
return 1
print "Updating to color images if available"
spath = os.path.join(bookDir,'color_img')
dpath = os.path.join(bookDir,'img')
filenames = os.listdir(spath)
filenames = sorted(filenames)
for filename in filenames:
imgname = filename.replace('color','img')
sfile = os.path.join(spath,filename)
dfile = os.path.join(dpath,imgname)
imgdata = file(sfile,'rb').read()
file(dfile,'wb').write(imgdata)
print "Creating cover.jpg"
isCover = False
cpath = os.path.join(bookDir,'img')
cpath = os.path.join(cpath,'img0000.jpg')
if os.path.isfile(cpath):
cover = file(cpath, 'rb').read()
cpath = os.path.join(bookDir,'cover.jpg')
file(cpath, 'wb').write(cover)
isCover = True
print 'Processing Dictionary'
dict = Dictionary(dictFile)
print 'Processing Meta Data and creating OPF'
meta_array = getMetaArray(metaFile)
# replace special chars in title and authors like & < >
title = meta_array.get('Title','No Title Provided')
title = title.replace('&','&amp;')
title = title.replace('<','&lt;')
title = title.replace('>','&gt;')
meta_array['Title'] = title
authors = meta_array.get('Authors','No Authors Provided')
authors = authors.replace('&','&amp;')
authors = authors.replace('<','&lt;')
authors = authors.replace('>','&gt;')
meta_array['Authors'] = authors
if buildXML:
xname = os.path.join(xmlDir, 'metadata.xml')
mlst = []
for key in meta_array:
mlst.append('<meta name="' + key + '" content="' + meta_array[key] + '" />\n')
metastr = "".join(mlst)
mlst = None
file(xname, 'wb').write(metastr)
print 'Processing StyleSheet'
# get some scaling info from metadata to use while processing styles
# and first page info
fontsize = '135'
if 'fontSize' in meta_array:
fontsize = meta_array['fontSize']
# also get the size of a normal text page
# get the total number of pages unpacked as a safety check
filenames = os.listdir(pageDir)
numfiles = len(filenames)
spage = '1'
if 'firstTextPage' in meta_array:
spage = meta_array['firstTextPage']
pnum = int(spage)
if pnum >= numfiles or pnum < 0:
# metadata is wrong so just select a page near the front
# 10% of the book to get a normal text page
pnum = int(0.10 * numfiles)
# print "first normal text page is", spage
# get page height and width from first text page for use in stylesheet scaling
pname = 'page%04d.dat' % (pnum + 1)
fname = os.path.join(pageDir,pname)
flat_xml = convert2xml.fromData(dict, fname)
(ph, pw) = getPageDim(flat_xml)
if (ph == '-1') or (ph == '0') : ph = '11000'
if (pw == '-1') or (pw == '0') : pw = '8500'
meta_array['pageHeight'] = ph
meta_array['pageWidth'] = pw
if 'fontSize' not in meta_array.keys():
meta_array['fontSize'] = fontsize
# process other.dat for css info and for map of page files to svg images
# this map is needed because some pages actually are made up of multiple
# pageXXXX.xml files
xname = os.path.join(bookDir, 'style.css')
flat_xml = convert2xml.fromData(dict, otherFile)
# extract info.original.pid to get original page information
pageIDMap = {}
pageidnums = stylexml2css.getpageIDMap(flat_xml)
if len(pageidnums) == 0:
filenames = os.listdir(pageDir)
numfiles = len(filenames)
for k in range(numfiles):
pageidnums.append(k)
# create a map from page ids to list of page file nums to process for that page
for i in range(len(pageidnums)):
id = pageidnums[i]
if id in pageIDMap.keys():
pageIDMap[id].append(i)
else:
pageIDMap[id] = [i]
# now get the css info
cssstr , classlst = stylexml2css.convert2CSS(flat_xml, fontsize, ph, pw)
file(xname, 'wb').write(cssstr)
if buildXML:
xname = os.path.join(xmlDir, 'other0000.xml')
file(xname, 'wb').write(convert2xml.getXML(dict, otherFile))
print 'Processing Glyphs'
gd = GlyphDict()
filenames = os.listdir(glyphsDir)
filenames = sorted(filenames)
glyfname = os.path.join(svgDir,'glyphs.svg')
glyfile = open(glyfname, 'w')
glyfile.write('<?xml version="1.0" standalone="no"?>\n')
glyfile.write('<!DOCTYPE svg PUBLIC "-//W3C/DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">\n')
glyfile.write('<svg width="512" height="512" viewBox="0 0 511 511" xmlns="http://www.w3.org/2000/svg" version="1.1">\n')
glyfile.write('<title>Glyphs for %s</title>\n' % meta_array['Title'])
glyfile.write('<defs>\n')
counter = 0
for filename in filenames:
# print ' ', filename
print '.',
fname = os.path.join(glyphsDir,filename)
flat_xml = convert2xml.fromData(dict, fname)
if buildXML:
xname = os.path.join(xmlDir, filename.replace('.dat','.xml'))
file(xname, 'wb').write(convert2xml.getXML(dict, fname))
gp = GParser(flat_xml)
for i in xrange(0, gp.count):
path = gp.getPath(i)
maxh, maxw = gp.getGlyphDim(i)
fullpath = '<path id="gl%d" d="%s" fill="black" /><!-- width=%d height=%d -->\n' % (counter * 256 + i, path, maxw, maxh)
glyfile.write(fullpath)
gd.addGlyph(counter * 256 + i, fullpath)
counter += 1
glyfile.write('</defs>\n')
glyfile.write('</svg>\n')
glyfile.close()
print " "
# start up the html
# also build up tocentries while processing html
htmlFileName = "book.html"
hlst = []
hlst.append('<?xml version="1.0" encoding="utf-8"?>\n')
hlst.append('<!DOCTYPE HTML PUBLIC "-//W3C//DTD XHTML 1.1 Strict//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11-strict.dtd">\n')
hlst.append('<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">\n')
hlst.append('<head>\n')
hlst.append('<meta http-equiv="content-type" content="text/html; charset=utf-8"/>\n')
hlst.append('<title>' + meta_array['Title'] + ' by ' + meta_array['Authors'] + '</title>\n')
hlst.append('<meta name="Author" content="' + meta_array['Authors'] + '" />\n')
hlst.append('<meta name="Title" content="' + meta_array['Title'] + '" />\n')
if 'ASIN' in meta_array:
hlst.append('<meta name="ASIN" content="' + meta_array['ASIN'] + '" />\n')
if 'GUID' in meta_array:
hlst.append('<meta name="GUID" content="' + meta_array['GUID'] + '" />\n')
hlst.append('<link href="style.css" rel="stylesheet" type="text/css" />\n')
hlst.append('</head>\n<body>\n')
print 'Processing Pages'
# Books are at 1440 DPI. This is rendering at twice that size for
# readability when rendering to the screen.
scaledpi = 1440.0
filenames = os.listdir(pageDir)
filenames = sorted(filenames)
numfiles = len(filenames)
xmllst = []
elst = []
for filename in filenames:
# print ' ', filename
print ".",
fname = os.path.join(pageDir,filename)
flat_xml = convert2xml.fromData(dict, fname)
# keep flat_xml for later svg processing
xmllst.append(flat_xml)
if buildXML:
xname = os.path.join(xmlDir, filename.replace('.dat','.xml'))
file(xname, 'wb').write(convert2xml.getXML(dict, fname))
# first get the html
pagehtml, tocinfo = flatxml2html.convert2HTML(flat_xml, classlst, fname, bookDir, gd, fixedimage)
elst.append(tocinfo)
hlst.append(pagehtml)
# finish up the html string and output it
hlst.append('</body>\n</html>\n')
htmlstr = "".join(hlst)
hlst = None
file(os.path.join(bookDir, htmlFileName), 'wb').write(htmlstr)
print " "
print 'Extracting Table of Contents from Amazon OCR'
# first create a table of contents file for the svg images
tlst = []
tlst.append('<?xml version="1.0" encoding="utf-8"?>\n')
tlst.append('<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">\n')
tlst.append('<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" >')
tlst.append('<head>\n')
tlst.append('<title>' + meta_array['Title'] + '</title>\n')
tlst.append('<meta name="Author" content="' + meta_array['Authors'] + '" />\n')
tlst.append('<meta name="Title" content="' + meta_array['Title'] + '" />\n')
if 'ASIN' in meta_array:
tlst.append('<meta name="ASIN" content="' + meta_array['ASIN'] + '" />\n')
if 'GUID' in meta_array:
tlst.append('<meta name="GUID" content="' + meta_array['GUID'] + '" />\n')
tlst.append('</head>\n')
tlst.append('<body>\n')
tlst.append('<h2>Table of Contents</h2>\n')
start = pageidnums[0]
if (raw):
startname = 'page%04d.svg' % start
else:
startname = 'page%04d.xhtml' % start
tlst.append('<h3><a href="' + startname + '">Start of Book</a></h3>\n')
# build up a table of contents for the svg xhtml output
tocentries = "".join(elst)
elst = None
toclst = tocentries.split('\n')
toclst.pop()
for entry in toclst:
print entry
title, pagenum = entry.split('|')
id = pageidnums[int(pagenum)]
if (raw):
fname = 'page%04d.svg' % id
else:
fname = 'page%04d.xhtml' % id
tlst.append('<h3><a href="'+ fname + '">' + title + '</a></h3>\n')
tlst.append('</body>\n')
tlst.append('</html>\n')
tochtml = "".join(tlst)
file(os.path.join(svgDir, 'toc.xhtml'), 'wb').write(tochtml)
# now create index_svg.xhtml that points to all required files
slst = []
slst.append('<?xml version="1.0" encoding="utf-8"?>\n')
slst.append('<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">\n')
slst.append('<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" >')
slst.append('<head>\n')
slst.append('<title>' + meta_array['Title'] + '</title>\n')
slst.append('<meta name="Author" content="' + meta_array['Authors'] + '" />\n')
slst.append('<meta name="Title" content="' + meta_array['Title'] + '" />\n')
if 'ASIN' in meta_array:
slst.append('<meta name="ASIN" content="' + meta_array['ASIN'] + '" />\n')
if 'GUID' in meta_array:
slst.append('<meta name="GUID" content="' + meta_array['GUID'] + '" />\n')
slst.append('</head>\n')
slst.append('<body>\n')
print "Building svg images of each book page"
slst.append('<h2>List of Pages</h2>\n')
slst.append('<div>\n')
idlst = sorted(pageIDMap.keys())
numids = len(idlst)
cnt = len(idlst)
previd = None
for j in range(cnt):
pageid = idlst[j]
if j < cnt - 1:
nextid = idlst[j+1]
else:
nextid = None
print '.',
pagelst = pageIDMap[pageid]
flst = []
for page in pagelst:
flst.append(xmllst[page])
flat_svg = "".join(flst)
flst=None
svgxml = flatxml2svg.convert2SVG(gd, flat_svg, pageid, previd, nextid, svgDir, raw, meta_array, scaledpi)
if (raw) :
pfile = open(os.path.join(svgDir,'page%04d.svg' % pageid),'w')
slst.append('<a href="svg/page%04d.svg">Page %d</a>\n' % (pageid, pageid))
else :
pfile = open(os.path.join(svgDir,'page%04d.xhtml' % pageid), 'w')
slst.append('<a href="svg/page%04d.xhtml">Page %d</a>\n' % (pageid, pageid))
previd = pageid
pfile.write(svgxml)
pfile.close()
counter += 1
slst.append('</div>\n')
slst.append('<h2><a href="svg/toc.xhtml">Table of Contents</a></h2>\n')
slst.append('</body>\n</html>\n')
svgindex = "".join(slst)
slst = None
file(os.path.join(bookDir, 'index_svg.xhtml'), 'wb').write(svgindex)
print " "
# build the opf file
opfname = os.path.join(bookDir, 'book.opf')
olst = []
olst.append('<?xml version="1.0" encoding="utf-8"?>\n')
olst.append('<package xmlns="http://www.idpf.org/2007/opf" unique-identifier="guid_id">\n')
# adding metadata
olst.append(' <metadata xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:opf="http://www.idpf.org/2007/opf">\n')
if 'GUID' in meta_array:
olst.append(' <dc:identifier opf:scheme="GUID" id="guid_id">' + meta_array['GUID'] + '</dc:identifier>\n')
if 'ASIN' in meta_array:
olst.append(' <dc:identifier opf:scheme="ASIN">' + meta_array['ASIN'] + '</dc:identifier>\n')
if 'oASIN' in meta_array:
olst.append(' <dc:identifier opf:scheme="oASIN">' + meta_array['oASIN'] + '</dc:identifier>\n')
olst.append(' <dc:title>' + meta_array['Title'] + '</dc:title>\n')
olst.append(' <dc:creator opf:role="aut">' + meta_array['Authors'] + '</dc:creator>\n')
olst.append(' <dc:language>en</dc:language>\n')
olst.append(' <dc:date>' + meta_array['UpdateTime'] + '</dc:date>\n')
if isCover:
olst.append(' <meta name="cover" content="bookcover"/>\n')
olst.append(' </metadata>\n')
olst.append('<manifest>\n')
olst.append(' <item id="book" href="book.html" media-type="application/xhtml+xml"/>\n')
olst.append(' <item id="stylesheet" href="style.css" media-type="text/css"/>\n')
# adding image files to manifest
filenames = os.listdir(imgDir)
filenames = sorted(filenames)
for filename in filenames:
imgname, imgext = os.path.splitext(filename)
if imgext == '.jpg':
imgext = 'jpeg'
if imgext == '.svg':
imgext = 'svg+xml'
olst.append(' <item id="' + imgname + '" href="img/' + filename + '" media-type="image/' + imgext + '"/>\n')
if isCover:
olst.append(' <item id="bookcover" href="cover.jpg" media-type="image/jpeg" />\n')
olst.append('</manifest>\n')
# adding spine
olst.append('<spine>\n <itemref idref="book" />\n</spine>\n')
if isCover:
olst.append(' <guide>\n')
olst.append(' <reference href="cover.jpg" type="cover" title="Cover"/>\n')
olst.append(' </guide>\n')
olst.append('</package>\n')
opfstr = "".join(olst)
olst = None
file(opfname, 'wb').write(opfstr)
print 'Processing Complete'
return 0
def usage():
print "genbook.py generates a book from the extract Topaz Files"
print "Usage:"
print " genbook.py [-r] [-h [--fixed-image] <bookDir> "
print " "
print "Options:"
print " -h : help - print this usage message"
print " -r : generate raw svg files (not wrapped in xhtml)"
print " --fixed-image : genearate any Fixed Area as an svg image in the html"
print " "
def main(argv):
bookDir = ''
if len(argv) == 0:
argv = sys.argv
try:
opts, args = getopt.getopt(argv[1:], "rh:",["fixed-image"])
except getopt.GetoptError, err:
print str(err)
usage()
return 1
if len(opts) == 0 and len(args) == 0 :
usage()
return 1
raw = 0
fixedimage = True
for o, a in opts:
if o =="-h":
usage()
return 0
if o =="-r":
raw = 1
if o =="--fixed-image":
fixedimage = True
bookDir = args[0]
rv = generateBook(bookDir, raw, fixedimage)
return rv
if __name__ == '__main__':
sys.exit(main(''))

View File

@@ -0,0 +1,77 @@
#!/usr/bin/python
#
# This is a python script. You need a Python interpreter to run it.
# For example, ActiveState Python, which exists for windows.
#
# Changelog
# 1.00 - Initial version
__version__ = '1.00'
import sys
class Unbuffered:
def __init__(self, stream):
self.stream = stream
def write(self, data):
self.stream.write(data)
self.stream.flush()
def __getattr__(self, attr):
return getattr(self.stream, attr)
sys.stdout=Unbuffered(sys.stdout)
import os
import struct
import binascii
import kgenpids
import topazextract
import mobidedrm
from alfcrypto import Pukall_Cipher
class DrmException(Exception):
pass
def getK4PCpids(path_to_ebook):
# Return Kindle4PC PIDs. Assumes that the caller checked that we are not on Linux, which will raise an exception
mobi = True
magic3 = file(path_to_ebook,'rb').read(3)
if magic3 == 'TPZ':
mobi = False
if mobi:
mb = mobidedrm.MobiBook(path_to_ebook,False)
else:
mb = topazextract.TopazBook(path_to_ebook)
md1, md2 = mb.getPIDMetaInfo()
return kgenpids.getPidList(md1, md2, True, [], [], [])
def main(argv=sys.argv):
print ('getk4pcpids.py v%(__version__)s. '
'Copyright 2012 Apprentic Alf' % globals())
if len(argv)<2 or len(argv)>3:
print "Gets the possible book-specific PIDs from K4PC for a particular book"
print "Usage:"
print " %s <bookfile> [<outfile>]" % sys.argv[0]
return 1
else:
infile = argv[1]
try:
pidlist = getK4PCpids(infile)
except DrmException, e:
print "Error: %s" % e
return 1
pidstring = ','.join(pidlist)
print "Possible PIDs are: ", pidstring
if len(argv) is 3:
outfile = argv[2]
file(outfile, 'w').write(pidstring)
return 0
if __name__ == "__main__":
sys.exit(main())

View File

@@ -0,0 +1,336 @@
#! /usr/bin/python
from __future__ import with_statement
# ignobleepub.pyw, version 3.4
# To run this program install Python 2.6 from <http://www.python.org/download/>
# and OpenSSL or PyCrypto from http://www.voidspace.org.uk/python/modules.shtml#pycrypto
# (make sure to install the version for Python 2.6). Save this script file as
# ignobleepub.pyw and double-click on it to run it.
# Revision history:
# 1 - Initial release
# 2 - Added OS X support by using OpenSSL when available
# 3 - screen out improper key lengths to prevent segfaults on Linux
# 3.1 - Allow Windows versions of libcrypto to be found
# 3.2 - add support for encoding to 'utf-8' when building up list of files to cecrypt from encryption.xml
# 3.3 - On Windows try PyCrypto first and OpenSSL next
# 3.4 - Modify interace to allow use with import
__license__ = 'GPL v3'
import sys
import os
import zlib
import zipfile
from zipfile import ZipFile, ZIP_STORED, ZIP_DEFLATED
from contextlib import closing
import xml.etree.ElementTree as etree
import Tkinter
import Tkconstants
import tkFileDialog
import tkMessageBox
class IGNOBLEError(Exception):
pass
def _load_crypto_libcrypto():
from ctypes import CDLL, POINTER, c_void_p, c_char_p, c_int, c_long, \
Structure, c_ulong, create_string_buffer, cast
from ctypes.util import find_library
if sys.platform.startswith('win'):
libcrypto = find_library('libeay32')
else:
libcrypto = find_library('crypto')
if libcrypto is None:
raise IGNOBLEError('libcrypto not found')
libcrypto = CDLL(libcrypto)
AES_MAXNR = 14
c_char_pp = POINTER(c_char_p)
c_int_p = POINTER(c_int)
class AES_KEY(Structure):
_fields_ = [('rd_key', c_long * (4 * (AES_MAXNR + 1))),
('rounds', c_int)]
AES_KEY_p = POINTER(AES_KEY)
def F(restype, name, argtypes):
func = getattr(libcrypto, name)
func.restype = restype
func.argtypes = argtypes
return func
AES_cbc_encrypt = F(None, 'AES_cbc_encrypt',
[c_char_p, c_char_p, c_ulong, AES_KEY_p, c_char_p,
c_int])
AES_set_decrypt_key = F(c_int, 'AES_set_decrypt_key',
[c_char_p, c_int, AES_KEY_p])
AES_cbc_encrypt = F(None, 'AES_cbc_encrypt',
[c_char_p, c_char_p, c_ulong, AES_KEY_p, c_char_p,
c_int])
class AES(object):
def __init__(self, userkey):
self._blocksize = len(userkey)
if (self._blocksize != 16) and (self._blocksize != 24) and (self._blocksize != 32) :
raise IGNOBLEError('AES improper key used')
return
key = self._key = AES_KEY()
rv = AES_set_decrypt_key(userkey, len(userkey) * 8, key)
if rv < 0:
raise IGNOBLEError('Failed to initialize AES key')
def decrypt(self, data):
out = create_string_buffer(len(data))
iv = ("\x00" * self._blocksize)
rv = AES_cbc_encrypt(data, out, len(data), self._key, iv, 0)
if rv == 0:
raise IGNOBLEError('AES decryption failed')
return out.raw
return AES
def _load_crypto_pycrypto():
from Crypto.Cipher import AES as _AES
class AES(object):
def __init__(self, key):
self._aes = _AES.new(key, _AES.MODE_CBC)
def decrypt(self, data):
return self._aes.decrypt(data)
return AES
def _load_crypto():
AES = None
cryptolist = (_load_crypto_libcrypto, _load_crypto_pycrypto)
if sys.platform.startswith('win'):
cryptolist = (_load_crypto_pycrypto, _load_crypto_libcrypto)
for loader in cryptolist:
try:
AES = loader()
break
except (ImportError, IGNOBLEError):
pass
return AES
AES = _load_crypto()
"""
Decrypt Barnes & Noble ADEPT encrypted EPUB books.
"""
META_NAMES = ('mimetype', 'META-INF/rights.xml', 'META-INF/encryption.xml')
NSMAP = {'adept': 'http://ns.adobe.com/adept',
'enc': 'http://www.w3.org/2001/04/xmlenc#'}
class ZipInfo(zipfile.ZipInfo):
def __init__(self, *args, **kwargs):
if 'compress_type' in kwargs:
compress_type = kwargs.pop('compress_type')
super(ZipInfo, self).__init__(*args, **kwargs)
self.compress_type = compress_type
class Decryptor(object):
def __init__(self, bookkey, encryption):
enc = lambda tag: '{%s}%s' % (NSMAP['enc'], tag)
# self._aes = AES.new(bookkey, AES.MODE_CBC)
self._aes = AES(bookkey)
encryption = etree.fromstring(encryption)
self._encrypted = encrypted = set()
expr = './%s/%s/%s' % (enc('EncryptedData'), enc('CipherData'),
enc('CipherReference'))
for elem in encryption.findall(expr):
path = elem.get('URI', None)
path = path.encode('utf-8')
if path is not None:
encrypted.add(path)
def decompress(self, bytes):
dc = zlib.decompressobj(-15)
bytes = dc.decompress(bytes)
ex = dc.decompress('Z') + dc.flush()
if ex:
bytes = bytes + ex
return bytes
def decrypt(self, path, data):
if path in self._encrypted:
data = self._aes.decrypt(data)[16:]
data = data[:-ord(data[-1])]
data = self.decompress(data)
return data
class DecryptionDialog(Tkinter.Frame):
def __init__(self, root):
Tkinter.Frame.__init__(self, root, border=5)
self.status = Tkinter.Label(self, text='Select files for decryption')
self.status.pack(fill=Tkconstants.X, expand=1)
body = Tkinter.Frame(self)
body.pack(fill=Tkconstants.X, expand=1)
sticky = Tkconstants.E + Tkconstants.W
body.grid_columnconfigure(1, weight=2)
Tkinter.Label(body, text='Key file').grid(row=0)
self.keypath = Tkinter.Entry(body, width=30)
self.keypath.grid(row=0, column=1, sticky=sticky)
if os.path.exists('bnepubkey.b64'):
self.keypath.insert(0, 'bnepubkey.b64')
button = Tkinter.Button(body, text="...", command=self.get_keypath)
button.grid(row=0, column=2)
Tkinter.Label(body, text='Input file').grid(row=1)
self.inpath = Tkinter.Entry(body, width=30)
self.inpath.grid(row=1, column=1, sticky=sticky)
button = Tkinter.Button(body, text="...", command=self.get_inpath)
button.grid(row=1, column=2)
Tkinter.Label(body, text='Output file').grid(row=2)
self.outpath = Tkinter.Entry(body, width=30)
self.outpath.grid(row=2, column=1, sticky=sticky)
button = Tkinter.Button(body, text="...", command=self.get_outpath)
button.grid(row=2, column=2)
buttons = Tkinter.Frame(self)
buttons.pack()
botton = Tkinter.Button(
buttons, text="Decrypt", width=10, command=self.decrypt)
botton.pack(side=Tkconstants.LEFT)
Tkinter.Frame(buttons, width=10).pack(side=Tkconstants.LEFT)
button = Tkinter.Button(
buttons, text="Quit", width=10, command=self.quit)
button.pack(side=Tkconstants.RIGHT)
def get_keypath(self):
keypath = tkFileDialog.askopenfilename(
parent=None, title='Select B&N EPUB key file',
defaultextension='.b64',
filetypes=[('base64-encoded files', '.b64'),
('All Files', '.*')])
if keypath:
keypath = os.path.normpath(keypath)
self.keypath.delete(0, Tkconstants.END)
self.keypath.insert(0, keypath)
return
def get_inpath(self):
inpath = tkFileDialog.askopenfilename(
parent=None, title='Select B&N-encrypted EPUB file to decrypt',
defaultextension='.epub', filetypes=[('EPUB files', '.epub'),
('All files', '.*')])
if inpath:
inpath = os.path.normpath(inpath)
self.inpath.delete(0, Tkconstants.END)
self.inpath.insert(0, inpath)
return
def get_outpath(self):
outpath = tkFileDialog.asksaveasfilename(
parent=None, title='Select unencrypted EPUB file to produce',
defaultextension='.epub', filetypes=[('EPUB files', '.epub'),
('All files', '.*')])
if outpath:
outpath = os.path.normpath(outpath)
self.outpath.delete(0, Tkconstants.END)
self.outpath.insert(0, outpath)
return
def decrypt(self):
keypath = self.keypath.get()
inpath = self.inpath.get()
outpath = self.outpath.get()
if not keypath or not os.path.exists(keypath):
self.status['text'] = 'Specified key file does not exist'
return
if not inpath or not os.path.exists(inpath):
self.status['text'] = 'Specified input file does not exist'
return
if not outpath:
self.status['text'] = 'Output file not specified'
return
if inpath == outpath:
self.status['text'] = 'Must have different input and output files'
return
argv = [sys.argv[0], keypath, inpath, outpath]
self.status['text'] = 'Decrypting...'
try:
cli_main(argv)
except Exception, e:
self.status['text'] = 'Error: ' + str(e)
return
self.status['text'] = 'File successfully decrypted'
def decryptBook(keypath, inpath, outpath):
with open(keypath, 'rb') as f:
keyb64 = f.read()
key = keyb64.decode('base64')[:16]
# aes = AES.new(key, AES.MODE_CBC)
aes = AES(key)
with closing(ZipFile(open(inpath, 'rb'))) as inf:
namelist = set(inf.namelist())
if 'META-INF/rights.xml' not in namelist or \
'META-INF/encryption.xml' not in namelist:
raise IGNOBLEError('%s: not an B&N ADEPT EPUB' % (inpath,))
for name in META_NAMES:
namelist.remove(name)
rights = etree.fromstring(inf.read('META-INF/rights.xml'))
adept = lambda tag: '{%s}%s' % (NSMAP['adept'], tag)
expr = './/%s' % (adept('encryptedKey'),)
bookkey = ''.join(rights.findtext(expr))
bookkey = aes.decrypt(bookkey.decode('base64'))
bookkey = bookkey[:-ord(bookkey[-1])]
encryption = inf.read('META-INF/encryption.xml')
decryptor = Decryptor(bookkey[-16:], encryption)
kwds = dict(compression=ZIP_DEFLATED, allowZip64=False)
with closing(ZipFile(open(outpath, 'wb'), 'w', **kwds)) as outf:
zi = ZipInfo('mimetype', compress_type=ZIP_STORED)
outf.writestr(zi, inf.read('mimetype'))
for path in namelist:
data = inf.read(path)
outf.writestr(path, decryptor.decrypt(path, data))
return 0
def cli_main(argv=sys.argv):
progname = os.path.basename(argv[0])
if AES is None:
print "%s: This script requires OpenSSL or PyCrypto, which must be installed " \
"separately. Read the top-of-script comment for details." % \
(progname,)
return 1
if len(argv) != 4:
print "usage: %s KEYFILE INBOOK OUTBOOK" % (progname,)
return 1
keypath, inpath, outpath = argv[1:]
return decryptBook(keypath, inpath, outpath)
def gui_main():
root = Tkinter.Tk()
if AES is None:
root.withdraw()
tkMessageBox.showerror(
"Ignoble EPUB Decrypter",
"This script requires OpenSSL or PyCrypto, which must be installed "
"separately. Read the top-of-script comment for details.")
return 1
root.title('Ignoble EPUB Decrypter')
root.resizable(True, False)
root.minsize(300, 0)
DecryptionDialog(root).pack(fill=Tkconstants.X, expand=1)
root.mainloop()
return 0
if __name__ == '__main__':
if len(sys.argv) > 1:
sys.exit(cli_main())
sys.exit(gui_main())

View File

@@ -0,0 +1,239 @@
#! /usr/bin/python
from __future__ import with_statement
# ignoblekeygen.pyw, version 2.3
# To run this program install Python 2.6 from <http://www.python.org/download/>
# and OpenSSL or PyCrypto from http://www.voidspace.org.uk/python/modules.shtml#pycrypto
# (make sure to install the version for Python 2.6). Save this script file as
# ignoblekeygen.pyw and double-click on it to run it.
# Revision history:
# 1 - Initial release
# 2 - Add OS X support by using OpenSSL when available (taken/modified from ineptepub v5)
# 2.1 - Allow Windows versions of libcrypto to be found
# 2.2 - On Windows try PyCrypto first and then OpenSSL next
# 2.3 - Modify interface to allow use of import
"""
Generate Barnes & Noble EPUB user key from name and credit card number.
"""
__license__ = 'GPL v3'
import sys
import os
import hashlib
import Tkinter
import Tkconstants
import tkFileDialog
import tkMessageBox
# use openssl's libcrypt if it exists in place of pycrypto
# code extracted from the Adobe Adept DRM removal code also by I HeartCabbages
class IGNOBLEError(Exception):
pass
def _load_crypto_libcrypto():
from ctypes import CDLL, POINTER, c_void_p, c_char_p, c_int, c_long, \
Structure, c_ulong, create_string_buffer, cast
from ctypes.util import find_library
if sys.platform.startswith('win'):
libcrypto = find_library('libeay32')
else:
libcrypto = find_library('crypto')
if libcrypto is None:
print 'libcrypto not found'
raise IGNOBLEError('libcrypto not found')
libcrypto = CDLL(libcrypto)
AES_MAXNR = 14
c_char_pp = POINTER(c_char_p)
c_int_p = POINTER(c_int)
class AES_KEY(Structure):
_fields_ = [('rd_key', c_long * (4 * (AES_MAXNR + 1))),
('rounds', c_int)]
AES_KEY_p = POINTER(AES_KEY)
def F(restype, name, argtypes):
func = getattr(libcrypto, name)
func.restype = restype
func.argtypes = argtypes
return func
AES_set_encrypt_key = F(c_int, 'AES_set_encrypt_key',
[c_char_p, c_int, AES_KEY_p])
AES_cbc_encrypt = F(None, 'AES_cbc_encrypt',
[c_char_p, c_char_p, c_ulong, AES_KEY_p, c_char_p,
c_int])
class AES(object):
def __init__(self, userkey, iv):
self._blocksize = len(userkey)
self._iv = iv
key = self._key = AES_KEY()
rv = AES_set_encrypt_key(userkey, len(userkey) * 8, key)
if rv < 0:
raise IGNOBLEError('Failed to initialize AES Encrypt key')
def encrypt(self, data):
out = create_string_buffer(len(data))
rv = AES_cbc_encrypt(data, out, len(data), self._key, self._iv, 1)
if rv == 0:
raise IGNOBLEError('AES encryption failed')
return out.raw
return AES
def _load_crypto_pycrypto():
from Crypto.Cipher import AES as _AES
class AES(object):
def __init__(self, key, iv):
self._aes = _AES.new(key, _AES.MODE_CBC, iv)
def encrypt(self, data):
return self._aes.encrypt(data)
return AES
def _load_crypto():
AES = None
cryptolist = (_load_crypto_libcrypto, _load_crypto_pycrypto)
if sys.platform.startswith('win'):
cryptolist = (_load_crypto_pycrypto, _load_crypto_libcrypto)
for loader in cryptolist:
try:
AES = loader()
break
except (ImportError, IGNOBLEError):
pass
return AES
AES = _load_crypto()
def normalize_name(name):
return ''.join(x for x in name.lower() if x != ' ')
def generate_keyfile(name, ccn, outpath):
name = normalize_name(name) + '\x00'
ccn = ccn + '\x00'
name_sha = hashlib.sha1(name).digest()[:16]
ccn_sha = hashlib.sha1(ccn).digest()[:16]
both_sha = hashlib.sha1(name + ccn).digest()
aes = AES(ccn_sha, name_sha)
crypt = aes.encrypt(both_sha + ('\x0c' * 0x0c))
userkey = hashlib.sha1(crypt).digest()
with open(outpath, 'wb') as f:
f.write(userkey.encode('base64'))
return userkey
class DecryptionDialog(Tkinter.Frame):
def __init__(self, root):
Tkinter.Frame.__init__(self, root, border=5)
self.status = Tkinter.Label(self, text='Enter parameters')
self.status.pack(fill=Tkconstants.X, expand=1)
body = Tkinter.Frame(self)
body.pack(fill=Tkconstants.X, expand=1)
sticky = Tkconstants.E + Tkconstants.W
body.grid_columnconfigure(1, weight=2)
Tkinter.Label(body, text='Name').grid(row=1)
self.name = Tkinter.Entry(body, width=30)
self.name.grid(row=1, column=1, sticky=sticky)
Tkinter.Label(body, text='CC#').grid(row=2)
self.ccn = Tkinter.Entry(body, width=30)
self.ccn.grid(row=2, column=1, sticky=sticky)
Tkinter.Label(body, text='Output file').grid(row=0)
self.keypath = Tkinter.Entry(body, width=30)
self.keypath.grid(row=0, column=1, sticky=sticky)
self.keypath.insert(0, 'bnepubkey.b64')
button = Tkinter.Button(body, text="...", command=self.get_keypath)
button.grid(row=0, column=2)
buttons = Tkinter.Frame(self)
buttons.pack()
botton = Tkinter.Button(
buttons, text="Generate", width=10, command=self.generate)
botton.pack(side=Tkconstants.LEFT)
Tkinter.Frame(buttons, width=10).pack(side=Tkconstants.LEFT)
button = Tkinter.Button(
buttons, text="Quit", width=10, command=self.quit)
button.pack(side=Tkconstants.RIGHT)
def get_keypath(self):
keypath = tkFileDialog.asksaveasfilename(
parent=None, title='Select B&N EPUB key file to produce',
defaultextension='.b64',
filetypes=[('base64-encoded files', '.b64'),
('All Files', '.*')])
if keypath:
keypath = os.path.normpath(keypath)
self.keypath.delete(0, Tkconstants.END)
self.keypath.insert(0, keypath)
return
def generate(self):
name = self.name.get()
ccn = self.ccn.get()
keypath = self.keypath.get()
if not name:
self.status['text'] = 'Name not specified'
return
if not ccn:
self.status['text'] = 'Credit card number not specified'
return
if not keypath:
self.status['text'] = 'Output keyfile path not specified'
return
self.status['text'] = 'Generating...'
try:
generate_keyfile(name, ccn, keypath)
except Exception, e:
self.status['text'] = 'Error: ' + str(e)
return
self.status['text'] = 'Keyfile successfully generated'
def cli_main(argv=sys.argv):
progname = os.path.basename(argv[0])
if AES is None:
print "%s: This script requires OpenSSL or PyCrypto, which must be installed " \
"separately. Read the top-of-script comment for details." % \
(progname,)
return 1
if len(argv) != 4:
print "usage: %s NAME CC# OUTFILE" % (progname,)
return 1
name, ccn, outpath = argv[1:]
generate_keyfile(name, ccn, outpath)
return 0
def gui_main():
root = Tkinter.Tk()
if AES is None:
root.withdraw()
tkMessageBox.showerror(
"Ignoble EPUB Keyfile Generator",
"This script requires OpenSSL or PyCrypto, which must be installed "
"separately. Read the top-of-script comment for details.")
return 1
root.title('Ignoble EPUB Keyfile Generator')
root.resizable(True, False)
root.minsize(300, 0)
DecryptionDialog(root).pack(fill=Tkconstants.X, expand=1)
root.mainloop()
return 0
if __name__ == '__main__':
if len(sys.argv) > 1:
sys.exit(cli_main())
sys.exit(gui_main())

View File

@@ -0,0 +1,476 @@
#! /usr/bin/python
# -*- coding: utf-8 -*-
from __future__ import with_statement
# ineptepub.pyw, version 5.6
# Copyright © 2009-2010 i♥cabbages
# Released under the terms of the GNU General Public Licence, version 3 or
# later. <http://www.gnu.org/licenses/>
# Windows users: Before running this program, you must first install Python 2.6
# from <http://www.python.org/download/> and PyCrypto from
# <http://www.voidspace.org.uk/python/modules.shtml#pycrypto> (make sure to
# install the version for Python 2.6). Save this script file as
# ineptepub.pyw and double-click on it to run it.
#
# Mac OS X users: Save this script file as ineptepub.pyw. You can run this
# program from the command line (pythonw ineptepub.pyw) or by double-clicking
# it when it has been associated with PythonLauncher.
# Revision history:
# 1 - Initial release
# 2 - Rename to INEPT, fix exit code
# 5 - Version bump to avoid (?) confusion;
# Improve OS X support by using OpenSSL when available
# 5.1 - Improve OpenSSL error checking
# 5.2 - Fix ctypes error causing segfaults on some systems
# 5.3 - add support for OpenSSL on Windows, fix bug with some versions of libcrypto 0.9.8 prior to path level o
# 5.4 - add support for encoding to 'utf-8' when building up list of files to decrypt from encryption.xml
# 5.5 - On Windows try PyCrypto first, OpenSSL next
# 5.6 - Modify interface to allow use with import
"""
Decrypt Adobe ADEPT-encrypted EPUB books.
"""
__license__ = 'GPL v3'
import sys
import os
import zlib
import zipfile
from zipfile import ZipFile, ZIP_STORED, ZIP_DEFLATED
from contextlib import closing
import xml.etree.ElementTree as etree
import Tkinter
import Tkconstants
import tkFileDialog
import tkMessageBox
class ADEPTError(Exception):
pass
def _load_crypto_libcrypto():
from ctypes import CDLL, POINTER, c_void_p, c_char_p, c_int, c_long, \
Structure, c_ulong, create_string_buffer, cast
from ctypes.util import find_library
if sys.platform.startswith('win'):
libcrypto = find_library('libeay32')
else:
libcrypto = find_library('crypto')
if libcrypto is None:
raise ADEPTError('libcrypto not found')
libcrypto = CDLL(libcrypto)
RSA_NO_PADDING = 3
AES_MAXNR = 14
c_char_pp = POINTER(c_char_p)
c_int_p = POINTER(c_int)
class RSA(Structure):
pass
RSA_p = POINTER(RSA)
class AES_KEY(Structure):
_fields_ = [('rd_key', c_long * (4 * (AES_MAXNR + 1))),
('rounds', c_int)]
AES_KEY_p = POINTER(AES_KEY)
def F(restype, name, argtypes):
func = getattr(libcrypto, name)
func.restype = restype
func.argtypes = argtypes
return func
d2i_RSAPrivateKey = F(RSA_p, 'd2i_RSAPrivateKey',
[RSA_p, c_char_pp, c_long])
RSA_size = F(c_int, 'RSA_size', [RSA_p])
RSA_private_decrypt = F(c_int, 'RSA_private_decrypt',
[c_int, c_char_p, c_char_p, RSA_p, c_int])
RSA_free = F(None, 'RSA_free', [RSA_p])
AES_set_decrypt_key = F(c_int, 'AES_set_decrypt_key',
[c_char_p, c_int, AES_KEY_p])
AES_cbc_encrypt = F(None, 'AES_cbc_encrypt',
[c_char_p, c_char_p, c_ulong, AES_KEY_p, c_char_p,
c_int])
class RSA(object):
def __init__(self, der):
buf = create_string_buffer(der)
pp = c_char_pp(cast(buf, c_char_p))
rsa = self._rsa = d2i_RSAPrivateKey(None, pp, len(der))
if rsa is None:
raise ADEPTError('Error parsing ADEPT user key DER')
def decrypt(self, from_):
rsa = self._rsa
to = create_string_buffer(RSA_size(rsa))
dlen = RSA_private_decrypt(len(from_), from_, to, rsa,
RSA_NO_PADDING)
if dlen < 0:
raise ADEPTError('RSA decryption failed')
return to[:dlen]
def __del__(self):
if self._rsa is not None:
RSA_free(self._rsa)
self._rsa = None
class AES(object):
def __init__(self, userkey):
self._blocksize = len(userkey)
if (self._blocksize != 16) and (self._blocksize != 24) and (self._blocksize != 32) :
raise ADEPTError('AES improper key used')
return
key = self._key = AES_KEY()
rv = AES_set_decrypt_key(userkey, len(userkey) * 8, key)
if rv < 0:
raise ADEPTError('Failed to initialize AES key')
def decrypt(self, data):
out = create_string_buffer(len(data))
iv = ("\x00" * self._blocksize)
rv = AES_cbc_encrypt(data, out, len(data), self._key, iv, 0)
if rv == 0:
raise ADEPTError('AES decryption failed')
return out.raw
return (AES, RSA)
def _load_crypto_pycrypto():
from Crypto.Cipher import AES as _AES
from Crypto.PublicKey import RSA as _RSA
# ASN.1 parsing code from tlslite
class ASN1Error(Exception):
pass
class ASN1Parser(object):
class Parser(object):
def __init__(self, bytes):
self.bytes = bytes
self.index = 0
def get(self, length):
if self.index + length > len(self.bytes):
raise ASN1Error("Error decoding ASN.1")
x = 0
for count in range(length):
x <<= 8
x |= self.bytes[self.index]
self.index += 1
return x
def getFixBytes(self, lengthBytes):
bytes = self.bytes[self.index : self.index+lengthBytes]
self.index += lengthBytes
return bytes
def getVarBytes(self, lengthLength):
lengthBytes = self.get(lengthLength)
return self.getFixBytes(lengthBytes)
def getFixList(self, length, lengthList):
l = [0] * lengthList
for x in range(lengthList):
l[x] = self.get(length)
return l
def getVarList(self, length, lengthLength):
lengthList = self.get(lengthLength)
if lengthList % length != 0:
raise ASN1Error("Error decoding ASN.1")
lengthList = int(lengthList/length)
l = [0] * lengthList
for x in range(lengthList):
l[x] = self.get(length)
return l
def startLengthCheck(self, lengthLength):
self.lengthCheck = self.get(lengthLength)
self.indexCheck = self.index
def setLengthCheck(self, length):
self.lengthCheck = length
self.indexCheck = self.index
def stopLengthCheck(self):
if (self.index - self.indexCheck) != self.lengthCheck:
raise ASN1Error("Error decoding ASN.1")
def atLengthCheck(self):
if (self.index - self.indexCheck) < self.lengthCheck:
return False
elif (self.index - self.indexCheck) == self.lengthCheck:
return True
else:
raise ASN1Error("Error decoding ASN.1")
def __init__(self, bytes):
p = self.Parser(bytes)
p.get(1)
self.length = self._getASN1Length(p)
self.value = p.getFixBytes(self.length)
def getChild(self, which):
p = self.Parser(self.value)
for x in range(which+1):
markIndex = p.index
p.get(1)
length = self._getASN1Length(p)
p.getFixBytes(length)
return ASN1Parser(p.bytes[markIndex:p.index])
def _getASN1Length(self, p):
firstLength = p.get(1)
if firstLength<=127:
return firstLength
else:
lengthLength = firstLength & 0x7F
return p.get(lengthLength)
class AES(object):
def __init__(self, key):
self._aes = _AES.new(key, _AES.MODE_CBC)
def decrypt(self, data):
return self._aes.decrypt(data)
class RSA(object):
def __init__(self, der):
key = ASN1Parser([ord(x) for x in der])
key = [key.getChild(x).value for x in xrange(1, 4)]
key = [self.bytesToNumber(v) for v in key]
self._rsa = _RSA.construct(key)
def bytesToNumber(self, bytes):
total = 0L
for byte in bytes:
total = (total << 8) + byte
return total
def decrypt(self, data):
return self._rsa.decrypt(data)
return (AES, RSA)
def _load_crypto():
AES = RSA = None
cryptolist = (_load_crypto_libcrypto, _load_crypto_pycrypto)
if sys.platform.startswith('win'):
cryptolist = (_load_crypto_pycrypto, _load_crypto_libcrypto)
for loader in cryptolist:
try:
AES, RSA = loader()
break
except (ImportError, ADEPTError):
pass
return (AES, RSA)
AES, RSA = _load_crypto()
META_NAMES = ('mimetype', 'META-INF/rights.xml', 'META-INF/encryption.xml')
NSMAP = {'adept': 'http://ns.adobe.com/adept',
'enc': 'http://www.w3.org/2001/04/xmlenc#'}
class ZipInfo(zipfile.ZipInfo):
def __init__(self, *args, **kwargs):
if 'compress_type' in kwargs:
compress_type = kwargs.pop('compress_type')
super(ZipInfo, self).__init__(*args, **kwargs)
self.compress_type = compress_type
class Decryptor(object):
def __init__(self, bookkey, encryption):
enc = lambda tag: '{%s}%s' % (NSMAP['enc'], tag)
self._aes = AES(bookkey)
encryption = etree.fromstring(encryption)
self._encrypted = encrypted = set()
expr = './%s/%s/%s' % (enc('EncryptedData'), enc('CipherData'),
enc('CipherReference'))
for elem in encryption.findall(expr):
path = elem.get('URI', None)
if path is not None:
path = path.encode('utf-8')
encrypted.add(path)
def decompress(self, bytes):
dc = zlib.decompressobj(-15)
bytes = dc.decompress(bytes)
ex = dc.decompress('Z') + dc.flush()
if ex:
bytes = bytes + ex
return bytes
def decrypt(self, path, data):
if path in self._encrypted:
data = self._aes.decrypt(data)[16:]
data = data[:-ord(data[-1])]
data = self.decompress(data)
return data
class DecryptionDialog(Tkinter.Frame):
def __init__(self, root):
Tkinter.Frame.__init__(self, root, border=5)
self.status = Tkinter.Label(self, text='Select files for decryption')
self.status.pack(fill=Tkconstants.X, expand=1)
body = Tkinter.Frame(self)
body.pack(fill=Tkconstants.X, expand=1)
sticky = Tkconstants.E + Tkconstants.W
body.grid_columnconfigure(1, weight=2)
Tkinter.Label(body, text='Key file').grid(row=0)
self.keypath = Tkinter.Entry(body, width=30)
self.keypath.grid(row=0, column=1, sticky=sticky)
if os.path.exists('adeptkey.der'):
self.keypath.insert(0, 'adeptkey.der')
button = Tkinter.Button(body, text="...", command=self.get_keypath)
button.grid(row=0, column=2)
Tkinter.Label(body, text='Input file').grid(row=1)
self.inpath = Tkinter.Entry(body, width=30)
self.inpath.grid(row=1, column=1, sticky=sticky)
button = Tkinter.Button(body, text="...", command=self.get_inpath)
button.grid(row=1, column=2)
Tkinter.Label(body, text='Output file').grid(row=2)
self.outpath = Tkinter.Entry(body, width=30)
self.outpath.grid(row=2, column=1, sticky=sticky)
button = Tkinter.Button(body, text="...", command=self.get_outpath)
button.grid(row=2, column=2)
buttons = Tkinter.Frame(self)
buttons.pack()
botton = Tkinter.Button(
buttons, text="Decrypt", width=10, command=self.decrypt)
botton.pack(side=Tkconstants.LEFT)
Tkinter.Frame(buttons, width=10).pack(side=Tkconstants.LEFT)
button = Tkinter.Button(
buttons, text="Quit", width=10, command=self.quit)
button.pack(side=Tkconstants.RIGHT)
def get_keypath(self):
keypath = tkFileDialog.askopenfilename(
parent=None, title='Select ADEPT key file',
defaultextension='.der', filetypes=[('DER-encoded files', '.der'),
('All Files', '.*')])
if keypath:
keypath = os.path.normpath(keypath)
self.keypath.delete(0, Tkconstants.END)
self.keypath.insert(0, keypath)
return
def get_inpath(self):
inpath = tkFileDialog.askopenfilename(
parent=None, title='Select ADEPT-encrypted EPUB file to decrypt',
defaultextension='.epub', filetypes=[('EPUB files', '.epub'),
('All files', '.*')])
if inpath:
inpath = os.path.normpath(inpath)
self.inpath.delete(0, Tkconstants.END)
self.inpath.insert(0, inpath)
return
def get_outpath(self):
outpath = tkFileDialog.asksaveasfilename(
parent=None, title='Select unencrypted EPUB file to produce',
defaultextension='.epub', filetypes=[('EPUB files', '.epub'),
('All files', '.*')])
if outpath:
outpath = os.path.normpath(outpath)
self.outpath.delete(0, Tkconstants.END)
self.outpath.insert(0, outpath)
return
def decrypt(self):
keypath = self.keypath.get()
inpath = self.inpath.get()
outpath = self.outpath.get()
if not keypath or not os.path.exists(keypath):
self.status['text'] = 'Specified key file does not exist'
return
if not inpath or not os.path.exists(inpath):
self.status['text'] = 'Specified input file does not exist'
return
if not outpath:
self.status['text'] = 'Output file not specified'
return
if inpath == outpath:
self.status['text'] = 'Must have different input and output files'
return
argv = [sys.argv[0], keypath, inpath, outpath]
self.status['text'] = 'Decrypting...'
try:
cli_main(argv)
except Exception, e:
self.status['text'] = 'Error: ' + str(e)
return
self.status['text'] = 'File successfully decrypted'
def decryptBook(keypath, inpath, outpath):
with open(keypath, 'rb') as f:
keyder = f.read()
rsa = RSA(keyder)
with closing(ZipFile(open(inpath, 'rb'))) as inf:
namelist = set(inf.namelist())
if 'META-INF/rights.xml' not in namelist or \
'META-INF/encryption.xml' not in namelist:
raise ADEPTError('%s: not an ADEPT EPUB' % (inpath,))
for name in META_NAMES:
namelist.remove(name)
rights = etree.fromstring(inf.read('META-INF/rights.xml'))
adept = lambda tag: '{%s}%s' % (NSMAP['adept'], tag)
expr = './/%s' % (adept('encryptedKey'),)
bookkey = ''.join(rights.findtext(expr))
bookkey = rsa.decrypt(bookkey.decode('base64'))
# Padded as per RSAES-PKCS1-v1_5
if bookkey[-17] != '\x00':
raise ADEPTError('problem decrypting session key')
encryption = inf.read('META-INF/encryption.xml')
decryptor = Decryptor(bookkey[-16:], encryption)
kwds = dict(compression=ZIP_DEFLATED, allowZip64=False)
with closing(ZipFile(open(outpath, 'wb'), 'w', **kwds)) as outf:
zi = ZipInfo('mimetype', compress_type=ZIP_STORED)
outf.writestr(zi, inf.read('mimetype'))
for path in namelist:
data = inf.read(path)
outf.writestr(path, decryptor.decrypt(path, data))
return 0
def cli_main(argv=sys.argv):
progname = os.path.basename(argv[0])
if AES is None:
print "%s: This script requires OpenSSL or PyCrypto, which must be" \
" installed separately. Read the top-of-script comment for" \
" details." % (progname,)
return 1
if len(argv) != 4:
print "usage: %s KEYFILE INBOOK OUTBOOK" % (progname,)
return 1
keypath, inpath, outpath = argv[1:]
return decryptBook(keypath, inpath, outpath)
def gui_main():
root = Tkinter.Tk()
if AES is None:
root.withdraw()
tkMessageBox.showerror(
"INEPT EPUB Decrypter",
"This script requires OpenSSL or PyCrypto, which must be"
" installed separately. Read the top-of-script comment for"
" details.")
return 1
root.title('INEPT EPUB Decrypter')
root.resizable(True, False)
root.minsize(300, 0)
DecryptionDialog(root).pack(fill=Tkconstants.X, expand=1)
root.mainloop()
return 0
if __name__ == '__main__':
if len(sys.argv) > 1:
sys.exit(cli_main())
sys.exit(gui_main())

View File

@@ -0,0 +1,467 @@
#! /usr/bin/python
# -*- coding: utf-8 -*-
from __future__ import with_statement
# ineptkey.pyw, version 5.4
# Copyright © 2009-2010 i♥cabbages
# Released under the terms of the GNU General Public Licence, version 3 or
# later. <http://www.gnu.org/licenses/>
# Windows users: Before running this program, you must first install Python 2.6
# from <http://www.python.org/download/> and PyCrypto from
# <http://www.voidspace.org.uk/python/modules.shtml#pycrypto> (make certain
# to install the version for Python 2.6). Then save this script file as
# ineptkey.pyw and double-click on it to run it. It will create a file named
# adeptkey.der in the same directory. This is your ADEPT user key.
#
# Mac OS X users: Save this script file as ineptkey.pyw. You can run this
# program from the command line (pythonw ineptkey.pyw) or by double-clicking
# it when it has been associated with PythonLauncher. It will create a file
# named adeptkey.der in the same directory. This is your ADEPT user key.
# Revision history:
# 1 - Initial release, for Adobe Digital Editions 1.7
# 2 - Better algorithm for finding pLK; improved error handling
# 3 - Rename to INEPT
# 4 - Series of changes by joblack (and others?) --
# 4.1 - quick beta fix for ADE 1.7.2 (anon)
# 4.2 - added old 1.7.1 processing
# 4.3 - better key search
# 4.4 - Make it working on 64-bit Python
# 5 - Clean up and improve 4.x changes;
# Clean up and merge OS X support by unknown
# 5.1 - add support for using OpenSSL on Windows in place of PyCrypto
# 5.2 - added support for output of key to a particular file
# 5.3 - On Windows try PyCrypto first, OpenSSL next
# 5.4 - Modify interface to allow use of import
"""
Retrieve Adobe ADEPT user key.
"""
__license__ = 'GPL v3'
import sys
import os
import struct
import Tkinter
import Tkconstants
import tkMessageBox
import traceback
class ADEPTError(Exception):
pass
if sys.platform.startswith('win'):
from ctypes import windll, c_char_p, c_wchar_p, c_uint, POINTER, byref, \
create_unicode_buffer, create_string_buffer, CFUNCTYPE, addressof, \
string_at, Structure, c_void_p, cast, c_size_t, memmove, CDLL, c_int, \
c_long, c_ulong
from ctypes.wintypes import LPVOID, DWORD, BOOL
import _winreg as winreg
def _load_crypto_libcrypto():
from ctypes.util import find_library
libcrypto = find_library('libeay32')
if libcrypto is None:
raise ADEPTError('libcrypto not found')
libcrypto = CDLL(libcrypto)
AES_MAXNR = 14
c_char_pp = POINTER(c_char_p)
c_int_p = POINTER(c_int)
class AES_KEY(Structure):
_fields_ = [('rd_key', c_long * (4 * (AES_MAXNR + 1))),
('rounds', c_int)]
AES_KEY_p = POINTER(AES_KEY)
def F(restype, name, argtypes):
func = getattr(libcrypto, name)
func.restype = restype
func.argtypes = argtypes
return func
AES_set_decrypt_key = F(c_int, 'AES_set_decrypt_key',
[c_char_p, c_int, AES_KEY_p])
AES_cbc_encrypt = F(None, 'AES_cbc_encrypt',
[c_char_p, c_char_p, c_ulong, AES_KEY_p, c_char_p,
c_int])
class AES(object):
def __init__(self, userkey):
self._blocksize = len(userkey)
if (self._blocksize != 16) and (self._blocksize != 24) and (self._blocksize != 32) :
raise ADEPTError('AES improper key used')
key = self._key = AES_KEY()
rv = AES_set_decrypt_key(userkey, len(userkey) * 8, key)
if rv < 0:
raise ADEPTError('Failed to initialize AES key')
def decrypt(self, data):
out = create_string_buffer(len(data))
iv = ("\x00" * self._blocksize)
rv = AES_cbc_encrypt(data, out, len(data), self._key, iv, 0)
if rv == 0:
raise ADEPTError('AES decryption failed')
return out.raw
return AES
def _load_crypto_pycrypto():
from Crypto.Cipher import AES as _AES
class AES(object):
def __init__(self, key):
self._aes = _AES.new(key, _AES.MODE_CBC)
def decrypt(self, data):
return self._aes.decrypt(data)
return AES
def _load_crypto():
AES = None
for loader in (_load_crypto_pycrypto, _load_crypto_libcrypto):
try:
AES = loader()
break
except (ImportError, ADEPTError):
pass
return AES
AES = _load_crypto()
DEVICE_KEY_PATH = r'Software\Adobe\Adept\Device'
PRIVATE_LICENCE_KEY_PATH = r'Software\Adobe\Adept\Activation'
MAX_PATH = 255
kernel32 = windll.kernel32
advapi32 = windll.advapi32
crypt32 = windll.crypt32
def GetSystemDirectory():
GetSystemDirectoryW = kernel32.GetSystemDirectoryW
GetSystemDirectoryW.argtypes = [c_wchar_p, c_uint]
GetSystemDirectoryW.restype = c_uint
def GetSystemDirectory():
buffer = create_unicode_buffer(MAX_PATH + 1)
GetSystemDirectoryW(buffer, len(buffer))
return buffer.value
return GetSystemDirectory
GetSystemDirectory = GetSystemDirectory()
def GetVolumeSerialNumber():
GetVolumeInformationW = kernel32.GetVolumeInformationW
GetVolumeInformationW.argtypes = [c_wchar_p, c_wchar_p, c_uint,
POINTER(c_uint), POINTER(c_uint),
POINTER(c_uint), c_wchar_p, c_uint]
GetVolumeInformationW.restype = c_uint
def GetVolumeSerialNumber(path):
vsn = c_uint(0)
GetVolumeInformationW(
path, None, 0, byref(vsn), None, None, None, 0)
return vsn.value
return GetVolumeSerialNumber
GetVolumeSerialNumber = GetVolumeSerialNumber()
def GetUserName():
GetUserNameW = advapi32.GetUserNameW
GetUserNameW.argtypes = [c_wchar_p, POINTER(c_uint)]
GetUserNameW.restype = c_uint
def GetUserName():
buffer = create_unicode_buffer(32)
size = c_uint(len(buffer))
while not GetUserNameW(buffer, byref(size)):
buffer = create_unicode_buffer(len(buffer) * 2)
size.value = len(buffer)
return buffer.value.encode('utf-16-le')[::2]
return GetUserName
GetUserName = GetUserName()
PAGE_EXECUTE_READWRITE = 0x40
MEM_COMMIT = 0x1000
MEM_RESERVE = 0x2000
def VirtualAlloc():
_VirtualAlloc = kernel32.VirtualAlloc
_VirtualAlloc.argtypes = [LPVOID, c_size_t, DWORD, DWORD]
_VirtualAlloc.restype = LPVOID
def VirtualAlloc(addr, size, alloctype=(MEM_COMMIT | MEM_RESERVE),
protect=PAGE_EXECUTE_READWRITE):
return _VirtualAlloc(addr, size, alloctype, protect)
return VirtualAlloc
VirtualAlloc = VirtualAlloc()
MEM_RELEASE = 0x8000
def VirtualFree():
_VirtualFree = kernel32.VirtualFree
_VirtualFree.argtypes = [LPVOID, c_size_t, DWORD]
_VirtualFree.restype = BOOL
def VirtualFree(addr, size=0, freetype=MEM_RELEASE):
return _VirtualFree(addr, size, freetype)
return VirtualFree
VirtualFree = VirtualFree()
class NativeFunction(object):
def __init__(self, restype, argtypes, insns):
self._buf = buf = VirtualAlloc(None, len(insns))
memmove(buf, insns, len(insns))
ftype = CFUNCTYPE(restype, *argtypes)
self._native = ftype(buf)
def __call__(self, *args):
return self._native(*args)
def __del__(self):
if self._buf is not None:
VirtualFree(self._buf)
self._buf = None
if struct.calcsize("P") == 4:
CPUID0_INSNS = (
"\x53" # push %ebx
"\x31\xc0" # xor %eax,%eax
"\x0f\xa2" # cpuid
"\x8b\x44\x24\x08" # mov 0x8(%esp),%eax
"\x89\x18" # mov %ebx,0x0(%eax)
"\x89\x50\x04" # mov %edx,0x4(%eax)
"\x89\x48\x08" # mov %ecx,0x8(%eax)
"\x5b" # pop %ebx
"\xc3" # ret
)
CPUID1_INSNS = (
"\x53" # push %ebx
"\x31\xc0" # xor %eax,%eax
"\x40" # inc %eax
"\x0f\xa2" # cpuid
"\x5b" # pop %ebx
"\xc3" # ret
)
else:
CPUID0_INSNS = (
"\x49\x89\xd8" # mov %rbx,%r8
"\x49\x89\xc9" # mov %rcx,%r9
"\x48\x31\xc0" # xor %rax,%rax
"\x0f\xa2" # cpuid
"\x4c\x89\xc8" # mov %r9,%rax
"\x89\x18" # mov %ebx,0x0(%rax)
"\x89\x50\x04" # mov %edx,0x4(%rax)
"\x89\x48\x08" # mov %ecx,0x8(%rax)
"\x4c\x89\xc3" # mov %r8,%rbx
"\xc3" # retq
)
CPUID1_INSNS = (
"\x53" # push %rbx
"\x48\x31\xc0" # xor %rax,%rax
"\x48\xff\xc0" # inc %rax
"\x0f\xa2" # cpuid
"\x5b" # pop %rbx
"\xc3" # retq
)
def cpuid0():
_cpuid0 = NativeFunction(None, [c_char_p], CPUID0_INSNS)
buf = create_string_buffer(12)
def cpuid0():
_cpuid0(buf)
return buf.raw
return cpuid0
cpuid0 = cpuid0()
cpuid1 = NativeFunction(c_uint, [], CPUID1_INSNS)
class DataBlob(Structure):
_fields_ = [('cbData', c_uint),
('pbData', c_void_p)]
DataBlob_p = POINTER(DataBlob)
def CryptUnprotectData():
_CryptUnprotectData = crypt32.CryptUnprotectData
_CryptUnprotectData.argtypes = [DataBlob_p, c_wchar_p, DataBlob_p,
c_void_p, c_void_p, c_uint, DataBlob_p]
_CryptUnprotectData.restype = c_uint
def CryptUnprotectData(indata, entropy):
indatab = create_string_buffer(indata)
indata = DataBlob(len(indata), cast(indatab, c_void_p))
entropyb = create_string_buffer(entropy)
entropy = DataBlob(len(entropy), cast(entropyb, c_void_p))
outdata = DataBlob()
if not _CryptUnprotectData(byref(indata), None, byref(entropy),
None, None, 0, byref(outdata)):
raise ADEPTError("Failed to decrypt user key key (sic)")
return string_at(outdata.pbData, outdata.cbData)
return CryptUnprotectData
CryptUnprotectData = CryptUnprotectData()
def retrieve_key(keypath):
if AES is None:
tkMessageBox.showerror(
"ADEPT Key",
"This script requires PyCrypto or OpenSSL which must be installed "
"separately. Read the top-of-script comment for details.")
return False
root = GetSystemDirectory().split('\\')[0] + '\\'
serial = GetVolumeSerialNumber(root)
vendor = cpuid0()
signature = struct.pack('>I', cpuid1())[1:]
user = GetUserName()
entropy = struct.pack('>I12s3s13s', serial, vendor, signature, user)
cuser = winreg.HKEY_CURRENT_USER
try:
regkey = winreg.OpenKey(cuser, DEVICE_KEY_PATH)
except WindowsError:
raise ADEPTError("Adobe Digital Editions not activated")
device = winreg.QueryValueEx(regkey, 'key')[0]
keykey = CryptUnprotectData(device, entropy)
userkey = None
try:
plkroot = winreg.OpenKey(cuser, PRIVATE_LICENCE_KEY_PATH)
except WindowsError:
raise ADEPTError("Could not locate ADE activation")
for i in xrange(0, 16):
try:
plkparent = winreg.OpenKey(plkroot, "%04d" % (i,))
except WindowsError:
break
ktype = winreg.QueryValueEx(plkparent, None)[0]
if ktype != 'credentials':
continue
for j in xrange(0, 16):
try:
plkkey = winreg.OpenKey(plkparent, "%04d" % (j,))
except WindowsError:
break
ktype = winreg.QueryValueEx(plkkey, None)[0]
if ktype != 'privateLicenseKey':
continue
userkey = winreg.QueryValueEx(plkkey, 'value')[0]
break
if userkey is not None:
break
if userkey is None:
raise ADEPTError('Could not locate privateLicenseKey')
userkey = userkey.decode('base64')
aes = AES(keykey)
userkey = aes.decrypt(userkey)
userkey = userkey[26:-ord(userkey[-1])]
with open(keypath, 'wb') as f:
f.write(userkey)
return True
elif sys.platform.startswith('darwin'):
import xml.etree.ElementTree as etree
import Carbon.File
import Carbon.Folder
import Carbon.Folders
import MacOS
ACTIVATION_PATH = 'Adobe/Digital Editions/activation.dat'
NSMAP = {'adept': 'http://ns.adobe.com/adept',
'enc': 'http://www.w3.org/2001/04/xmlenc#'}
def find_folder(domain, dtype):
try:
fsref = Carbon.Folder.FSFindFolder(domain, dtype, False)
return Carbon.File.pathname(fsref)
except MacOS.Error:
return None
def find_app_support_file(subpath):
dtype = Carbon.Folders.kApplicationSupportFolderType
for domain in Carbon.Folders.kUserDomain, Carbon.Folders.kLocalDomain:
path = find_folder(domain, dtype)
if path is None:
continue
path = os.path.join(path, subpath)
if os.path.isfile(path):
return path
return None
def retrieve_key(keypath):
actpath = find_app_support_file(ACTIVATION_PATH)
if actpath is None:
raise ADEPTError("Could not locate ADE activation")
tree = etree.parse(actpath)
adept = lambda tag: '{%s}%s' % (NSMAP['adept'], tag)
expr = '//%s/%s' % (adept('credentials'), adept('privateLicenseKey'))
userkey = tree.findtext(expr)
userkey = userkey.decode('base64')
userkey = userkey[26:]
with open(keypath, 'wb') as f:
f.write(userkey)
return True
elif sys.platform.startswith('cygwin'):
def retrieve_key(keypath):
tkMessageBox.showerror(
"ADEPT Key",
"This script requires a Windows-native Python, and cannot be run "
"under Cygwin. Please install a Windows-native Python and/or "
"check your file associations.")
return False
else:
def retrieve_key(keypath):
tkMessageBox.showerror(
"ADEPT Key",
"This script only supports Windows and Mac OS X. For Linux "
"you should be able to run ADE and this script under Wine (with "
"an appropriate version of Windows Python installed).")
return False
class ExceptionDialog(Tkinter.Frame):
def __init__(self, root, text):
Tkinter.Frame.__init__(self, root, border=5)
label = Tkinter.Label(self, text="Unexpected error:",
anchor=Tkconstants.W, justify=Tkconstants.LEFT)
label.pack(fill=Tkconstants.X, expand=0)
self.text = Tkinter.Text(self)
self.text.pack(fill=Tkconstants.BOTH, expand=1)
self.text.insert(Tkconstants.END, text)
def extractKeyfile(keypath):
try:
success = retrieve_key(keypath)
except ADEPTError, e:
print "Key generation Error: " + str(e)
return 1
except Exception, e:
print "General Error: " + str(e)
return 1
if not success:
return 1
return 0
def cli_main(argv=sys.argv):
keypath = argv[1]
return extractKeyfile(keypath)
def main(argv=sys.argv):
root = Tkinter.Tk()
root.withdraw()
progname = os.path.basename(argv[0])
keypath = 'adeptkey.der'
success = False
try:
success = retrieve_key(keypath)
except ADEPTError, e:
tkMessageBox.showerror("ADEPT Key", "Error: " + str(e))
except Exception:
root.wm_state('normal')
root.title('ADEPT Key')
text = traceback.format_exc()
ExceptionDialog(root, text).pack(fill=Tkconstants.BOTH, expand=1)
root.mainloop()
if not success:
return 1
tkMessageBox.showinfo(
"ADEPT Key", "Key successfully retrieved to %s" % (keypath))
return 0
if __name__ == '__main__':
if len(sys.argv) > 1:
sys.exit(cli_main())
sys.exit(main())

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,223 @@
#!/usr/bin/env python
from __future__ import with_statement
# engine to remove drm from Kindle for Mac and Kindle for PC books
# for personal use for archiving and converting your ebooks
# PLEASE DO NOT PIRATE EBOOKS!
# We want all authors and publishers, and eBook stores to live
# long and prosperous lives but at the same time we just want to
# be able to read OUR books on whatever device we want and to keep
# readable for a long, long time
# This borrows very heavily from works by CMBDTC, IHeartCabbages, skindle,
# unswindle, DarkReverser, ApprenticeAlf, DiapDealer, some_updates
# and many many others
__version__ = '4.4'
class Unbuffered:
def __init__(self, stream):
self.stream = stream
def write(self, data):
self.stream.write(data)
self.stream.flush()
def __getattr__(self, attr):
return getattr(self.stream, attr)
import sys
import os, csv, getopt
import string
import re
import traceback
buildXML = False
class DrmException(Exception):
pass
if 'calibre' in sys.modules:
inCalibre = True
else:
inCalibre = False
if inCalibre:
from calibre_plugins.k4mobidedrm import mobidedrm
from calibre_plugins.k4mobidedrm import topazextract
from calibre_plugins.k4mobidedrm import kgenpids
else:
import mobidedrm
import topazextract
import kgenpids
# cleanup bytestring filenames
# borrowed from calibre from calibre/src/calibre/__init__.py
# added in removal of non-printing chars
# and removal of . at start
# convert underscores to spaces (we're OK with spaces in file names)
def cleanup_name(name):
_filename_sanitize = re.compile(r'[\xae\0\\|\?\*<":>\+/]')
substitute='_'
one = ''.join(char for char in name if char in string.printable)
one = _filename_sanitize.sub(substitute, one)
one = re.sub(r'\s', ' ', one).strip()
one = re.sub(r'^\.+$', '_', one)
one = one.replace('..', substitute)
# Windows doesn't like path components that end with a period
if one.endswith('.'):
one = one[:-1]+substitute
# Mac and Unix don't like file names that begin with a full stop
if len(one) > 0 and one[0] == '.':
one = substitute+one[1:]
one = one.replace('_',' ')
return one
def decryptBook(infile, outdir, k4, kInfoFiles, serials, pids):
global buildXML
# handle the obvious cases at the beginning
if not os.path.isfile(infile):
print >>sys.stderr, ('K4MobiDeDrm v%(__version__)s\n' % globals()) + "Error: Input file does not exist"
return 1
mobi = True
magic3 = file(infile,'rb').read(3)
if magic3 == 'TPZ':
mobi = False
bookname = os.path.splitext(os.path.basename(infile))[0]
if mobi:
mb = mobidedrm.MobiBook(infile)
else:
mb = topazextract.TopazBook(infile)
title = mb.getBookTitle()
print "Processing Book: ", title
filenametitle = cleanup_name(title)
outfilename = cleanup_name(bookname)
# generate 'sensible' filename, that will sort with the original name,
# but is close to the name from the file.
outlength = len(outfilename)
comparelength = min(8,min(outlength,len(filenametitle)))
copylength = min(max(outfilename.find(' '),8),len(outfilename))
if outlength==0:
outfilename = filenametitle
elif comparelength > 0:
if outfilename[:comparelength] == filenametitle[:comparelength]:
outfilename = filenametitle
else:
outfilename = outfilename[:copylength] + " " + filenametitle
# avoid excessively long file names
if len(outfilename)>150:
outfilename = outfilename[:150]
# build pid list
md1, md2 = mb.getPIDMetaInfo()
pidlst = kgenpids.getPidList(md1, md2, k4, pids, serials, kInfoFiles)
try:
mb.processBook(pidlst)
except mobidedrm.DrmException, e:
print >>sys.stderr, ('K4MobiDeDrm v%(__version__)s\n' % globals()) + "Error: " + str(e) + "\nDRM Removal Failed.\n"
return 1
except topazextract.TpzDRMError, e:
print >>sys.stderr, ('K4MobiDeDrm v%(__version__)s\n' % globals()) + "Error: " + str(e) + "\nDRM Removal Failed.\n"
return 1
except Exception, e:
print >>sys.stderr, ('K4MobiDeDrm v%(__version__)s\n' % globals()) + "Error: " + str(e) + "\nDRM Removal Failed.\n"
return 1
if mobi:
if mb.getPrintReplica():
outfile = os.path.join(outdir, outfilename + '_nodrm' + '.azw4')
elif mb.getMobiVersion() >= 8:
outfile = os.path.join(outdir, outfilename + '_nodrm' + '.azw3')
else:
outfile = os.path.join(outdir, outfilename + '_nodrm' + '.mobi')
mb.getMobiFile(outfile)
return 0
# topaz:
print " Creating NoDRM HTMLZ Archive"
zipname = os.path.join(outdir, outfilename + '_nodrm' + '.htmlz')
mb.getHTMLZip(zipname)
print " Creating SVG ZIP Archive"
zipname = os.path.join(outdir, outfilename + '_SVG' + '.zip')
mb.getSVGZip(zipname)
if buildXML:
print " Creating XML ZIP Archive"
zipname = os.path.join(outdir, outfilename + '_XML' + '.zip')
mb.getXMLZip(zipname)
# remove internal temporary directory of Topaz pieces
mb.cleanup()
return 0
def usage(progname):
print "Removes DRM protection from K4PC/M, Kindle, Mobi and Topaz ebooks"
print "Usage:"
print " %s [-k <kindle.info>] [-p <pidnums>] [-s <kindleSerialNumbers>] <infile> <outdir> " % progname
#
# Main
#
def main(argv=sys.argv):
progname = os.path.basename(argv[0])
k4 = False
kInfoFiles = []
serials = []
pids = []
print ('K4MobiDeDrm v%(__version__)s '
'provided by the work of many including DiapDealer, SomeUpdates, IHeartCabbages, CMBDTC, Skindle, DarkReverser, ApprenticeAlf, etc .' % globals())
try:
opts, args = getopt.getopt(sys.argv[1:], "k:p:s:")
except getopt.GetoptError, err:
print str(err)
usage(progname)
sys.exit(2)
if len(args)<2:
usage(progname)
sys.exit(2)
for o, a in opts:
if o == "-k":
if a == None :
raise DrmException("Invalid parameter for -k")
kInfoFiles.append(a)
if o == "-p":
if a == None :
raise DrmException("Invalid parameter for -p")
pids = a.split(',')
if o == "-s":
if a == None :
raise DrmException("Invalid parameter for -s")
serials = a.split(',')
# try with built in Kindle Info files
k4 = True
if sys.platform.startswith('linux'):
k4 = False
kInfoFiles = None
infile = args[0]
outdir = args[1]
return decryptBook(infile, outdir, k4, kInfoFiles, serials, pids)
if __name__ == '__main__':
sys.stdout=Unbuffered(sys.stdout)
sys.exit(main())

View File

@@ -0,0 +1,734 @@
# standlone set of Mac OSX specific routines needed for KindleBooks
from __future__ import with_statement
import sys
import os
import os.path
import re
import copy
import subprocess
from struct import pack, unpack, unpack_from
class DrmException(Exception):
pass
# interface to needed routines in openssl's libcrypto
def _load_crypto_libcrypto():
from ctypes import CDLL, byref, POINTER, c_void_p, c_char_p, c_int, c_long, \
Structure, c_ulong, create_string_buffer, addressof, string_at, cast
from ctypes.util import find_library
libcrypto = find_library('crypto')
if libcrypto is None:
raise DrmException('libcrypto not found')
libcrypto = CDLL(libcrypto)
# From OpenSSL's crypto aes header
#
# AES_ENCRYPT 1
# AES_DECRYPT 0
# AES_MAXNR 14 (in bytes)
# AES_BLOCK_SIZE 16 (in bytes)
#
# struct aes_key_st {
# unsigned long rd_key[4 *(AES_MAXNR + 1)];
# int rounds;
# };
# typedef struct aes_key_st AES_KEY;
#
# int AES_set_decrypt_key(const unsigned char *userKey, const int bits, AES_KEY *key);
#
# note: the ivec string, and output buffer are both mutable
# void AES_cbc_encrypt(const unsigned char *in, unsigned char *out,
# const unsigned long length, const AES_KEY *key, unsigned char *ivec, const int enc);
AES_MAXNR = 14
c_char_pp = POINTER(c_char_p)
c_int_p = POINTER(c_int)
class AES_KEY(Structure):
_fields_ = [('rd_key', c_long * (4 * (AES_MAXNR + 1))), ('rounds', c_int)]
AES_KEY_p = POINTER(AES_KEY)
def F(restype, name, argtypes):
func = getattr(libcrypto, name)
func.restype = restype
func.argtypes = argtypes
return func
AES_cbc_encrypt = F(None, 'AES_cbc_encrypt',[c_char_p, c_char_p, c_ulong, AES_KEY_p, c_char_p,c_int])
AES_set_decrypt_key = F(c_int, 'AES_set_decrypt_key',[c_char_p, c_int, AES_KEY_p])
# From OpenSSL's Crypto evp/p5_crpt2.c
#
# int PKCS5_PBKDF2_HMAC_SHA1(const char *pass, int passlen,
# const unsigned char *salt, int saltlen, int iter,
# int keylen, unsigned char *out);
PKCS5_PBKDF2_HMAC_SHA1 = F(c_int, 'PKCS5_PBKDF2_HMAC_SHA1',
[c_char_p, c_ulong, c_char_p, c_ulong, c_ulong, c_ulong, c_char_p])
class LibCrypto(object):
def __init__(self):
self._blocksize = 0
self._keyctx = None
self._iv = 0
def set_decrypt_key(self, userkey, iv):
self._blocksize = len(userkey)
if (self._blocksize != 16) and (self._blocksize != 24) and (self._blocksize != 32) :
raise DrmException('AES improper key used')
return
keyctx = self._keyctx = AES_KEY()
self._iv = iv
self._userkey = userkey
rv = AES_set_decrypt_key(userkey, len(userkey) * 8, keyctx)
if rv < 0:
raise DrmException('Failed to initialize AES key')
def decrypt(self, data):
out = create_string_buffer(len(data))
mutable_iv = create_string_buffer(self._iv, len(self._iv))
keyctx = self._keyctx
rv = AES_cbc_encrypt(data, out, len(data), keyctx, mutable_iv, 0)
if rv == 0:
raise DrmException('AES decryption failed')
return out.raw
def keyivgen(self, passwd, salt, iter, keylen):
saltlen = len(salt)
passlen = len(passwd)
out = create_string_buffer(keylen)
rv = PKCS5_PBKDF2_HMAC_SHA1(passwd, passlen, salt, saltlen, iter, keylen, out)
return out.raw
return LibCrypto
def _load_crypto():
LibCrypto = None
try:
LibCrypto = _load_crypto_libcrypto()
except (ImportError, DrmException):
pass
return LibCrypto
LibCrypto = _load_crypto()
#
# Utility Routines
#
# crypto digestroutines
import hashlib
def MD5(message):
ctx = hashlib.md5()
ctx.update(message)
return ctx.digest()
def SHA1(message):
ctx = hashlib.sha1()
ctx.update(message)
return ctx.digest()
def SHA256(message):
ctx = hashlib.sha256()
ctx.update(message)
return ctx.digest()
# Various character maps used to decrypt books. Probably supposed to act as obfuscation
charMap1 = "n5Pr6St7Uv8Wx9YzAb0Cd1Ef2Gh3Jk4M"
charMap2 = "ZB0bYyc1xDdW2wEV3Ff7KkPpL8UuGA4gz-Tme9Nn_tHh5SvXCsIiR6rJjQaqlOoM"
# For kinf approach of K4Mac 1.6.X or later
# On K4PC charMap5 = "AzB0bYyCeVvaZ3FfUuG4g-TtHh5SsIiR6rJjQq7KkPpL8lOoMm9Nn_c1XxDdW2wE"
# For Mac they seem to re-use charMap2 here
charMap5 = charMap2
# new in K4M 1.9.X
testMap8 = "YvaZ3FfUm9Nn_c1XuG4yCAzB0beVg-TtHh5SsIiR6rJjQdW2wEq7KkPpL8lOoMxD"
def encode(data, map):
result = ""
for char in data:
value = ord(char)
Q = (value ^ 0x80) // len(map)
R = value % len(map)
result += map[Q]
result += map[R]
return result
# Hash the bytes in data and then encode the digest with the characters in map
def encodeHash(data,map):
return encode(MD5(data),map)
# Decode the string in data with the characters in map. Returns the decoded bytes
def decode(data,map):
result = ""
for i in range (0,len(data)-1,2):
high = map.find(data[i])
low = map.find(data[i+1])
if (high == -1) or (low == -1) :
break
value = (((high * len(map)) ^ 0x80) & 0xFF) + low
result += pack("B",value)
return result
# For K4M 1.6.X and later
# generate table of prime number less than or equal to int n
def primes(n):
if n==2: return [2]
elif n<2: return []
s=range(3,n+1,2)
mroot = n ** 0.5
half=(n+1)/2-1
i=0
m=3
while m <= mroot:
if s[i]:
j=(m*m-3)/2
s[j]=0
while j<half:
s[j]=0
j+=m
i=i+1
m=2*i+3
return [2]+[x for x in s if x]
# uses a sub process to get the Hard Drive Serial Number using ioreg
# returns with the serial number of drive whose BSD Name is "disk0"
def GetVolumeSerialNumber():
sernum = os.getenv('MYSERIALNUMBER')
if sernum != None:
return sernum
cmdline = '/usr/sbin/ioreg -l -S -w 0 -r -c AppleAHCIDiskDriver'
cmdline = cmdline.encode(sys.getfilesystemencoding())
p = subprocess.Popen(cmdline, shell=True, stdin=None, stdout=subprocess.PIPE, stderr=subprocess.PIPE, close_fds=False)
out1, out2 = p.communicate()
reslst = out1.split('\n')
cnt = len(reslst)
bsdname = None
sernum = None
foundIt = False
for j in xrange(cnt):
resline = reslst[j]
pp = resline.find('"Serial Number" = "')
if pp >= 0:
sernum = resline[pp+19:-1]
sernum = sernum.strip()
bb = resline.find('"BSD Name" = "')
if bb >= 0:
bsdname = resline[bb+14:-1]
bsdname = bsdname.strip()
if (bsdname == 'disk0') and (sernum != None):
foundIt = True
break
if not foundIt:
sernum = ''
return sernum
def GetUserHomeAppSupKindleDirParitionName():
home = os.getenv('HOME')
dpath = home + '/Library/Application Support/Kindle'
cmdline = '/sbin/mount'
cmdline = cmdline.encode(sys.getfilesystemencoding())
p = subprocess.Popen(cmdline, shell=True, stdin=None, stdout=subprocess.PIPE, stderr=subprocess.PIPE, close_fds=False)
out1, out2 = p.communicate()
reslst = out1.split('\n')
cnt = len(reslst)
disk = ''
foundIt = False
for j in xrange(cnt):
resline = reslst[j]
if resline.startswith('/dev'):
(devpart, mpath) = resline.split(' on ')
dpart = devpart[5:]
pp = mpath.find('(')
if pp >= 0:
mpath = mpath[:pp-1]
if dpath.startswith(mpath):
disk = dpart
return disk
# uses a sub process to get the UUID of the specified disk partition using ioreg
def GetDiskPartitionUUID(diskpart):
uuidnum = os.getenv('MYUUIDNUMBER')
if uuidnum != None:
return uuidnum
cmdline = '/usr/sbin/ioreg -l -S -w 0 -r -c AppleAHCIDiskDriver'
cmdline = cmdline.encode(sys.getfilesystemencoding())
p = subprocess.Popen(cmdline, shell=True, stdin=None, stdout=subprocess.PIPE, stderr=subprocess.PIPE, close_fds=False)
out1, out2 = p.communicate()
reslst = out1.split('\n')
cnt = len(reslst)
bsdname = None
uuidnum = None
foundIt = False
nest = 0
uuidnest = -1
partnest = -2
for j in xrange(cnt):
resline = reslst[j]
if resline.find('{') >= 0:
nest += 1
if resline.find('}') >= 0:
nest -= 1
pp = resline.find('"UUID" = "')
if pp >= 0:
uuidnum = resline[pp+10:-1]
uuidnum = uuidnum.strip()
uuidnest = nest
if partnest == uuidnest and uuidnest > 0:
foundIt = True
break
bb = resline.find('"BSD Name" = "')
if bb >= 0:
bsdname = resline[bb+14:-1]
bsdname = bsdname.strip()
if (bsdname == diskpart):
partnest = nest
else :
partnest = -2
if partnest == uuidnest and partnest > 0:
foundIt = True
break
if nest == 0:
partnest = -2
uuidnest = -1
uuidnum = None
bsdname = None
if not foundIt:
uuidnum = ''
return uuidnum
def GetMACAddressMunged():
macnum = os.getenv('MYMACNUM')
if macnum != None:
return macnum
cmdline = '/sbin/ifconfig en0'
cmdline = cmdline.encode(sys.getfilesystemencoding())
p = subprocess.Popen(cmdline, shell=True, stdin=None, stdout=subprocess.PIPE, stderr=subprocess.PIPE, close_fds=False)
out1, out2 = p.communicate()
reslst = out1.split('\n')
cnt = len(reslst)
macnum = None
foundIt = False
for j in xrange(cnt):
resline = reslst[j]
pp = resline.find('ether ')
if pp >= 0:
macnum = resline[pp+6:-1]
macnum = macnum.strip()
# print "original mac", macnum
# now munge it up the way Kindle app does
# by xoring it with 0xa5 and swapping elements 3 and 4
maclst = macnum.split(':')
n = len(maclst)
if n != 6:
fountIt = False
break
for i in range(6):
maclst[i] = int('0x' + maclst[i], 0)
mlst = [0x00, 0x00, 0x00, 0x00, 0x00, 0x00]
mlst[5] = maclst[5] ^ 0xa5
mlst[4] = maclst[3] ^ 0xa5
mlst[3] = maclst[4] ^ 0xa5
mlst[2] = maclst[2] ^ 0xa5
mlst[1] = maclst[1] ^ 0xa5
mlst[0] = maclst[0] ^ 0xa5
macnum = "%0.2x%0.2x%0.2x%0.2x%0.2x%0.2x" % (mlst[0], mlst[1], mlst[2], mlst[3], mlst[4], mlst[5])
foundIt = True
break
if not foundIt:
macnum = ''
return macnum
# uses unix env to get username instead of using sysctlbyname
def GetUserName():
username = os.getenv('USER')
return username
def isNewInstall():
home = os.getenv('HOME')
# soccer game fan anyone
dpath = home + '/Library/Application Support/Kindle/storage/.pes2011'
# print dpath, os.path.exists(dpath)
if os.path.exists(dpath):
return True
return False
def GetIDString():
# K4Mac now has an extensive set of ids strings it uses
# in encoding pids and in creating unique passwords
# for use in its own version of CryptUnprotectDataV2
# BUT Amazon has now become nasty enough to detect when its app
# is being run under a debugger and actually changes code paths
# including which one of these strings is chosen, all to try
# to prevent reverse engineering
# Sad really ... they will only hurt their own sales ...
# true book lovers really want to keep their books forever
# and move them to their devices and DRM prevents that so they
# will just buy from someplace else that they can remove
# the DRM from
# Amazon should know by now that true book lover's are not like
# penniless kids that pirate music, we do not pirate books
if isNewInstall():
mungedmac = GetMACAddressMunged()
if len(mungedmac) > 7:
print('Using Munged MAC Address for ID: '+mungedmac)
return mungedmac
sernum = GetVolumeSerialNumber()
if len(sernum) > 7:
print('Using Volume Serial Number for ID: '+sernum)
return sernum
diskpart = GetUserHomeAppSupKindleDirParitionName()
uuidnum = GetDiskPartitionUUID(diskpart)
if len(uuidnum) > 7:
print('Using Disk Partition UUID for ID: '+uuidnum)
return uuidnum
mungedmac = GetMACAddressMunged()
if len(mungedmac) > 7:
print('Using Munged MAC Address for ID: '+mungedmac)
return mungedmac
print('Using Fixed constant 9999999999 for ID.')
return '9999999999'
# implements an Pseudo Mac Version of Windows built-in Crypto routine
# used by Kindle for Mac versions < 1.6.0
class CryptUnprotectData(object):
def __init__(self):
sernum = GetVolumeSerialNumber()
if sernum == '':
sernum = '9999999999'
sp = sernum + '!@#' + GetUserName()
passwdData = encode(SHA256(sp),charMap1)
salt = '16743'
self.crp = LibCrypto()
iter = 0x3e8
keylen = 0x80
key_iv = self.crp.keyivgen(passwdData, salt, iter, keylen)
self.key = key_iv[0:32]
self.iv = key_iv[32:48]
self.crp.set_decrypt_key(self.key, self.iv)
def decrypt(self, encryptedData):
cleartext = self.crp.decrypt(encryptedData)
cleartext = decode(cleartext,charMap1)
return cleartext
# implements an Pseudo Mac Version of Windows built-in Crypto routine
# used for Kindle for Mac Versions >= 1.6.0
class CryptUnprotectDataV2(object):
def __init__(self):
sp = GetUserName() + ':&%:' + GetIDString()
passwdData = encode(SHA256(sp),charMap5)
# salt generation as per the code
salt = 0x0512981d * 2 * 1 * 1
salt = str(salt) + GetUserName()
salt = encode(salt,charMap5)
self.crp = LibCrypto()
iter = 0x800
keylen = 0x400
key_iv = self.crp.keyivgen(passwdData, salt, iter, keylen)
self.key = key_iv[0:32]
self.iv = key_iv[32:48]
self.crp.set_decrypt_key(self.key, self.iv)
def decrypt(self, encryptedData):
cleartext = self.crp.decrypt(encryptedData)
cleartext = decode(cleartext, charMap5)
return cleartext
# unprotect the new header blob in .kinf2011
# used in Kindle for Mac Version >= 1.9.0
def UnprotectHeaderData(encryptedData):
passwdData = 'header_key_data'
salt = 'HEADER.2011'
iter = 0x80
keylen = 0x100
crp = LibCrypto()
key_iv = crp.keyivgen(passwdData, salt, iter, keylen)
key = key_iv[0:32]
iv = key_iv[32:48]
crp.set_decrypt_key(key,iv)
cleartext = crp.decrypt(encryptedData)
return cleartext
# implements an Pseudo Mac Version of Windows built-in Crypto routine
# used for Kindle for Mac Versions >= 1.9.0
class CryptUnprotectDataV3(object):
def __init__(self, entropy):
sp = GetUserName() + '+@#$%+' + GetIDString()
passwdData = encode(SHA256(sp),charMap2)
salt = entropy
self.crp = LibCrypto()
iter = 0x800
keylen = 0x400
key_iv = self.crp.keyivgen(passwdData, salt, iter, keylen)
self.key = key_iv[0:32]
self.iv = key_iv[32:48]
self.crp.set_decrypt_key(self.key, self.iv)
def decrypt(self, encryptedData):
cleartext = self.crp.decrypt(encryptedData)
cleartext = decode(cleartext, charMap2)
return cleartext
# Locate the .kindle-info files
def getKindleInfoFiles(kInfoFiles):
# first search for current .kindle-info files
home = os.getenv('HOME')
cmdline = 'find "' + home + '/Library/Application Support" -name ".kindle-info"'
cmdline = cmdline.encode(sys.getfilesystemencoding())
p1 = subprocess.Popen(cmdline, shell=True, stdin=None, stdout=subprocess.PIPE, stderr=subprocess.PIPE, close_fds=False)
out1, out2 = p1.communicate()
reslst = out1.split('\n')
kinfopath = 'NONE'
found = False
for resline in reslst:
if os.path.isfile(resline):
kInfoFiles.append(resline)
print('Found K4Mac kindle-info file: ' + resline)
found = True
# add any .rainier*-kinf files
cmdline = 'find "' + home + '/Library/Application Support" -name ".rainier*-kinf"'
cmdline = cmdline.encode(sys.getfilesystemencoding())
p1 = subprocess.Popen(cmdline, shell=True, stdin=None, stdout=subprocess.PIPE, stderr=subprocess.PIPE, close_fds=False)
out1, out2 = p1.communicate()
reslst = out1.split('\n')
for resline in reslst:
if os.path.isfile(resline):
kInfoFiles.append(resline)
print('Found k4Mac kinf file: ' + resline)
found = True
# add any .kinf2011 files
cmdline = 'find "' + home + '/Library/Application Support" -name ".kinf2011"'
cmdline = cmdline.encode(sys.getfilesystemencoding())
p1 = subprocess.Popen(cmdline, shell=True, stdin=None, stdout=subprocess.PIPE, stderr=subprocess.PIPE, close_fds=False)
out1, out2 = p1.communicate()
reslst = out1.split('\n')
for resline in reslst:
if os.path.isfile(resline):
kInfoFiles.append(resline)
print('Found k4Mac kinf2011 file: ' + resline)
found = True
if not found:
print('No k4Mac kindle-info/kinf/kinf2011 files have been found.')
return kInfoFiles
# determine type of kindle info provided and return a
# database of keynames and values
def getDBfromFile(kInfoFile):
names = ["kindle.account.tokens","kindle.cookie.item","eulaVersionAccepted","login_date","kindle.token.item","login","kindle.key.item","kindle.name.info","kindle.device.info", "MazamaRandomNumber", "max_date", "SIGVERIF"]
DB = {}
cnt = 0
infoReader = open(kInfoFile, 'r')
hdr = infoReader.read(1)
data = infoReader.read()
if data.find('[') != -1 :
# older style kindle-info file
cud = CryptUnprotectData()
items = data.split('[')
for item in items:
if item != '':
keyhash, rawdata = item.split(':')
keyname = "unknown"
for name in names:
if encodeHash(name,charMap2) == keyhash:
keyname = name
break
if keyname == "unknown":
keyname = keyhash
encryptedValue = decode(rawdata,charMap2)
cleartext = cud.decrypt(encryptedValue)
DB[keyname] = cleartext
cnt = cnt + 1
if cnt == 0:
DB = None
return DB
if hdr == '/':
# else newer style .kinf file used by K4Mac >= 1.6.0
# the .kinf file uses "/" to separate it into records
# so remove the trailing "/" to make it easy to use split
data = data[:-1]
items = data.split('/')
cud = CryptUnprotectDataV2()
# loop through the item records until all are processed
while len(items) > 0:
# get the first item record
item = items.pop(0)
# the first 32 chars of the first record of a group
# is the MD5 hash of the key name encoded by charMap5
keyhash = item[0:32]
keyname = "unknown"
# the raw keyhash string is also used to create entropy for the actual
# CryptProtectData Blob that represents that keys contents
# "entropy" not used for K4Mac only K4PC
# entropy = SHA1(keyhash)
# the remainder of the first record when decoded with charMap5
# has the ':' split char followed by the string representation
# of the number of records that follow
# and make up the contents
srcnt = decode(item[34:],charMap5)
rcnt = int(srcnt)
# read and store in rcnt records of data
# that make up the contents value
edlst = []
for i in xrange(rcnt):
item = items.pop(0)
edlst.append(item)
keyname = "unknown"
for name in names:
if encodeHash(name,charMap5) == keyhash:
keyname = name
break
if keyname == "unknown":
keyname = keyhash
# the charMap5 encoded contents data has had a length
# of chars (always odd) cut off of the front and moved
# to the end to prevent decoding using charMap5 from
# working properly, and thereby preventing the ensuing
# CryptUnprotectData call from succeeding.
# The offset into the charMap5 encoded contents seems to be:
# len(contents) - largest prime number less than or equal to int(len(content)/3)
# (in other words split "about" 2/3rds of the way through)
# move first offsets chars to end to align for decode by charMap5
encdata = "".join(edlst)
contlen = len(encdata)
# now properly split and recombine
# by moving noffset chars from the start of the
# string to the end of the string
noffset = contlen - primes(int(contlen/3))[-1]
pfx = encdata[0:noffset]
encdata = encdata[noffset:]
encdata = encdata + pfx
# decode using charMap5 to get the CryptProtect Data
encryptedValue = decode(encdata,charMap5)
cleartext = cud.decrypt(encryptedValue)
DB[keyname] = cleartext
cnt = cnt + 1
if cnt == 0:
DB = None
return DB
# the latest .kinf2011 version for K4M 1.9.1
# put back the hdr char, it is needed
data = hdr + data
data = data[:-1]
items = data.split('/')
# the headerblob is the encrypted information needed to build the entropy string
headerblob = items.pop(0)
encryptedValue = decode(headerblob, charMap1)
cleartext = UnprotectHeaderData(encryptedValue)
# now extract the pieces in the same way
# this version is different from K4PC it scales the build number by multipying by 735
pattern = re.compile(r'''\[Version:(\d+)\]\[Build:(\d+)\]\[Cksum:([^\]]+)\]\[Guid:([\{\}a-z0-9\-]+)\]''', re.IGNORECASE)
for m in re.finditer(pattern, cleartext):
entropy = str(int(m.group(2)) * 0x2df) + m.group(4)
cud = CryptUnprotectDataV3(entropy)
# loop through the item records until all are processed
while len(items) > 0:
# get the first item record
item = items.pop(0)
# the first 32 chars of the first record of a group
# is the MD5 hash of the key name encoded by charMap5
keyhash = item[0:32]
keyname = "unknown"
# unlike K4PC the keyhash is not used in generating entropy
# entropy = SHA1(keyhash) + added_entropy
# entropy = added_entropy
# the remainder of the first record when decoded with charMap5
# has the ':' split char followed by the string representation
# of the number of records that follow
# and make up the contents
srcnt = decode(item[34:],charMap5)
rcnt = int(srcnt)
# read and store in rcnt records of data
# that make up the contents value
edlst = []
for i in xrange(rcnt):
item = items.pop(0)
edlst.append(item)
keyname = "unknown"
for name in names:
if encodeHash(name,testMap8) == keyhash:
keyname = name
break
if keyname == "unknown":
keyname = keyhash
# the testMap8 encoded contents data has had a length
# of chars (always odd) cut off of the front and moved
# to the end to prevent decoding using testMap8 from
# working properly, and thereby preventing the ensuing
# CryptUnprotectData call from succeeding.
# The offset into the testMap8 encoded contents seems to be:
# len(contents) - largest prime number less than or equal to int(len(content)/3)
# (in other words split "about" 2/3rds of the way through)
# move first offsets chars to end to align for decode by testMap8
encdata = "".join(edlst)
contlen = len(encdata)
# now properly split and recombine
# by moving noffset chars from the start of the
# string to the end of the string
noffset = contlen - primes(int(contlen/3))[-1]
pfx = encdata[0:noffset]
encdata = encdata[noffset:]
encdata = encdata + pfx
# decode using testMap8 to get the CryptProtect Data
encryptedValue = decode(encdata,testMap8)
cleartext = cud.decrypt(encryptedValue)
# print keyname
# print cleartext
DB[keyname] = cleartext
cnt = cnt + 1
if cnt == 0:
DB = None
return DB

View File

@@ -0,0 +1,437 @@
#!/usr/bin/env python
# K4PC Windows specific routines
from __future__ import with_statement
import sys, os, re
from struct import pack, unpack, unpack_from
from ctypes import windll, c_char_p, c_wchar_p, c_uint, POINTER, byref, \
create_unicode_buffer, create_string_buffer, CFUNCTYPE, addressof, \
string_at, Structure, c_void_p, cast
import _winreg as winreg
MAX_PATH = 255
kernel32 = windll.kernel32
advapi32 = windll.advapi32
crypt32 = windll.crypt32
import traceback
# crypto digestroutines
import hashlib
def MD5(message):
ctx = hashlib.md5()
ctx.update(message)
return ctx.digest()
def SHA1(message):
ctx = hashlib.sha1()
ctx.update(message)
return ctx.digest()
def SHA256(message):
ctx = hashlib.sha256()
ctx.update(message)
return ctx.digest()
# For K4PC 1.9.X
# use routines in alfcrypto:
# AES_cbc_encrypt
# AES_set_decrypt_key
# PKCS5_PBKDF2_HMAC_SHA1
from alfcrypto import AES_CBC, KeyIVGen
def UnprotectHeaderData(encryptedData):
passwdData = 'header_key_data'
salt = 'HEADER.2011'
iter = 0x80
keylen = 0x100
key_iv = KeyIVGen().pbkdf2(passwdData, salt, iter, keylen)
key = key_iv[0:32]
iv = key_iv[32:48]
aes=AES_CBC()
aes.set_decrypt_key(key, iv)
cleartext = aes.decrypt(encryptedData)
return cleartext
# simple primes table (<= n) calculator
def primes(n):
if n==2: return [2]
elif n<2: return []
s=range(3,n+1,2)
mroot = n ** 0.5
half=(n+1)/2-1
i=0
m=3
while m <= mroot:
if s[i]:
j=(m*m-3)/2
s[j]=0
while j<half:
s[j]=0
j+=m
i=i+1
m=2*i+3
return [2]+[x for x in s if x]
# Various character maps used to decrypt kindle info values.
# Probably supposed to act as obfuscation
charMap2 = "AaZzB0bYyCc1XxDdW2wEeVv3FfUuG4g-TtHh5SsIiR6rJjQq7KkPpL8lOoMm9Nn_"
charMap5 = "AzB0bYyCeVvaZ3FfUuG4g-TtHh5SsIiR6rJjQq7KkPpL8lOoMm9Nn_c1XxDdW2wE"
# New maps in K4PC 1.9.0
testMap1 = "n5Pr6St7Uv8Wx9YzAb0Cd1Ef2Gh3Jk4M"
testMap6 = "9YzAb0Cd1Ef2n5Pr6St7Uvh3Jk4M8WxG"
testMap8 = "YvaZ3FfUm9Nn_c1XuG4yCAzB0beVg-TtHh5SsIiR6rJjQdW2wEq7KkPpL8lOoMxD"
class DrmException(Exception):
pass
# Encode the bytes in data with the characters in map
def encode(data, map):
result = ""
for char in data:
value = ord(char)
Q = (value ^ 0x80) // len(map)
R = value % len(map)
result += map[Q]
result += map[R]
return result
# Hash the bytes in data and then encode the digest with the characters in map
def encodeHash(data,map):
return encode(MD5(data),map)
# Decode the string in data with the characters in map. Returns the decoded bytes
def decode(data,map):
result = ""
for i in range (0,len(data)-1,2):
high = map.find(data[i])
low = map.find(data[i+1])
if (high == -1) or (low == -1) :
break
value = (((high * len(map)) ^ 0x80) & 0xFF) + low
result += pack("B",value)
return result
# interface with Windows OS Routines
class DataBlob(Structure):
_fields_ = [('cbData', c_uint),
('pbData', c_void_p)]
DataBlob_p = POINTER(DataBlob)
def GetSystemDirectory():
GetSystemDirectoryW = kernel32.GetSystemDirectoryW
GetSystemDirectoryW.argtypes = [c_wchar_p, c_uint]
GetSystemDirectoryW.restype = c_uint
def GetSystemDirectory():
buffer = create_unicode_buffer(MAX_PATH + 1)
GetSystemDirectoryW(buffer, len(buffer))
return buffer.value
return GetSystemDirectory
GetSystemDirectory = GetSystemDirectory()
def GetVolumeSerialNumber():
GetVolumeInformationW = kernel32.GetVolumeInformationW
GetVolumeInformationW.argtypes = [c_wchar_p, c_wchar_p, c_uint,
POINTER(c_uint), POINTER(c_uint),
POINTER(c_uint), c_wchar_p, c_uint]
GetVolumeInformationW.restype = c_uint
def GetVolumeSerialNumber(path = GetSystemDirectory().split('\\')[0] + '\\'):
vsn = c_uint(0)
GetVolumeInformationW(path, None, 0, byref(vsn), None, None, None, 0)
return str(vsn.value)
return GetVolumeSerialNumber
GetVolumeSerialNumber = GetVolumeSerialNumber()
def GetIDString():
vsn = GetVolumeSerialNumber()
print('Using Volume Serial Number for ID: '+vsn)
return vsn
def getLastError():
GetLastError = kernel32.GetLastError
GetLastError.argtypes = None
GetLastError.restype = c_uint
def getLastError():
return GetLastError()
return getLastError
getLastError = getLastError()
def GetUserName():
GetUserNameW = advapi32.GetUserNameW
GetUserNameW.argtypes = [c_wchar_p, POINTER(c_uint)]
GetUserNameW.restype = c_uint
def GetUserName():
buffer = create_unicode_buffer(2)
size = c_uint(len(buffer))
while not GetUserNameW(buffer, byref(size)):
errcd = getLastError()
if errcd == 234:
# bad wine implementation up through wine 1.3.21
return "AlternateUserName"
buffer = create_unicode_buffer(len(buffer) * 2)
size.value = len(buffer)
return buffer.value.encode('utf-16-le')[::2]
return GetUserName
GetUserName = GetUserName()
def CryptUnprotectData():
_CryptUnprotectData = crypt32.CryptUnprotectData
_CryptUnprotectData.argtypes = [DataBlob_p, c_wchar_p, DataBlob_p,
c_void_p, c_void_p, c_uint, DataBlob_p]
_CryptUnprotectData.restype = c_uint
def CryptUnprotectData(indata, entropy, flags):
indatab = create_string_buffer(indata)
indata = DataBlob(len(indata), cast(indatab, c_void_p))
entropyb = create_string_buffer(entropy)
entropy = DataBlob(len(entropy), cast(entropyb, c_void_p))
outdata = DataBlob()
if not _CryptUnprotectData(byref(indata), None, byref(entropy),
None, None, flags, byref(outdata)):
# raise DrmException("Failed to Unprotect Data")
return 'failed'
return string_at(outdata.pbData, outdata.cbData)
return CryptUnprotectData
CryptUnprotectData = CryptUnprotectData()
# Locate all of the kindle-info style files and return as list
def getKindleInfoFiles(kInfoFiles):
regkey = winreg.OpenKey(winreg.HKEY_CURRENT_USER, "Software\\Microsoft\\Windows\\CurrentVersion\\Explorer\\Shell Folders\\")
path = winreg.QueryValueEx(regkey, 'Local AppData')[0]
# some 64 bit machines do not have the proper registry key for some reason
# or the pythonn interface to the 32 vs 64 bit registry is broken
if 'LOCALAPPDATA' in os.environ.keys():
path = os.environ['LOCALAPPDATA']
print('searching for kinfoFiles in ' + path)
found = False
# first look for older kindle-info files
kinfopath = path +'\\Amazon\\Kindle For PC\\{AMAwzsaPaaZAzmZzZQzgZCAkZ3AjA_AY}\\kindle.info'
if os.path.isfile(kinfopath):
found = True
print('Found K4PC kindle.info file: ' + kinfopath)
kInfoFiles.append(kinfopath)
# now look for newer (K4PC 1.5.0 and later rainier.2.1.1.kinf file
kinfopath = path +'\\Amazon\\Kindle For PC\\storage\\rainier.2.1.1.kinf'
if os.path.isfile(kinfopath):
found = True
print('Found K4PC 1.5.X kinf file: ' + kinfopath)
kInfoFiles.append(kinfopath)
# now look for even newer (K4PC 1.6.0 and later) rainier.2.1.1.kinf file
kinfopath = path +'\\Amazon\\Kindle\\storage\\rainier.2.1.1.kinf'
if os.path.isfile(kinfopath):
found = True
print('Found K4PC 1.6.X kinf file: ' + kinfopath)
kInfoFiles.append(kinfopath)
# now look for even newer (K4PC 1.9.0 and later) .kinf2011 file
kinfopath = path +'\\Amazon\\Kindle\\storage\\.kinf2011'
if os.path.isfile(kinfopath):
found = True
print('Found K4PC kinf2011 file: ' + kinfopath)
kInfoFiles.append(kinfopath)
if not found:
print('No K4PC kindle.info/kinf/kinf2011 files have been found.')
return kInfoFiles
# determine type of kindle info provided and return a
# database of keynames and values
def getDBfromFile(kInfoFile):
names = ["kindle.account.tokens","kindle.cookie.item","eulaVersionAccepted","login_date","kindle.token.item","login","kindle.key.item","kindle.name.info","kindle.device.info", "MazamaRandomNumber", "max_date", "SIGVERIF"]
DB = {}
cnt = 0
infoReader = open(kInfoFile, 'r')
hdr = infoReader.read(1)
data = infoReader.read()
if data.find('{') != -1 :
# older style kindle-info file
items = data.split('{')
for item in items:
if item != '':
keyhash, rawdata = item.split(':')
keyname = "unknown"
for name in names:
if encodeHash(name,charMap2) == keyhash:
keyname = name
break
if keyname == "unknown":
keyname = keyhash
encryptedValue = decode(rawdata,charMap2)
DB[keyname] = CryptUnprotectData(encryptedValue, "", 0)
cnt = cnt + 1
if cnt == 0:
DB = None
return DB
if hdr == '/':
# else rainier-2-1-1 .kinf file
# the .kinf file uses "/" to separate it into records
# so remove the trailing "/" to make it easy to use split
data = data[:-1]
items = data.split('/')
# loop through the item records until all are processed
while len(items) > 0:
# get the first item record
item = items.pop(0)
# the first 32 chars of the first record of a group
# is the MD5 hash of the key name encoded by charMap5
keyhash = item[0:32]
# the raw keyhash string is used to create entropy for the actual
# CryptProtectData Blob that represents that keys contents
entropy = SHA1(keyhash)
# the remainder of the first record when decoded with charMap5
# has the ':' split char followed by the string representation
# of the number of records that follow
# and make up the contents
srcnt = decode(item[34:],charMap5)
rcnt = int(srcnt)
# read and store in rcnt records of data
# that make up the contents value
edlst = []
for i in xrange(rcnt):
item = items.pop(0)
edlst.append(item)
keyname = "unknown"
for name in names:
if encodeHash(name,charMap5) == keyhash:
keyname = name
break
if keyname == "unknown":
keyname = keyhash
# the charMap5 encoded contents data has had a length
# of chars (always odd) cut off of the front and moved
# to the end to prevent decoding using charMap5 from
# working properly, and thereby preventing the ensuing
# CryptUnprotectData call from succeeding.
# The offset into the charMap5 encoded contents seems to be:
# len(contents)-largest prime number <= int(len(content)/3)
# (in other words split "about" 2/3rds of the way through)
# move first offsets chars to end to align for decode by charMap5
encdata = "".join(edlst)
contlen = len(encdata)
noffset = contlen - primes(int(contlen/3))[-1]
# now properly split and recombine
# by moving noffset chars from the start of the
# string to the end of the string
pfx = encdata[0:noffset]
encdata = encdata[noffset:]
encdata = encdata + pfx
# decode using Map5 to get the CryptProtect Data
encryptedValue = decode(encdata,charMap5)
DB[keyname] = CryptUnprotectData(encryptedValue, entropy, 1)
cnt = cnt + 1
if cnt == 0:
DB = None
return DB
# else newest .kinf2011 style .kinf file
# the .kinf file uses "/" to separate it into records
# so remove the trailing "/" to make it easy to use split
# need to put back the first char read because it it part
# of the added entropy blob
data = hdr + data[:-1]
items = data.split('/')
# starts with and encoded and encrypted header blob
headerblob = items.pop(0)
encryptedValue = decode(headerblob, testMap1)
cleartext = UnprotectHeaderData(encryptedValue)
# now extract the pieces that form the added entropy
pattern = re.compile(r'''\[Version:(\d+)\]\[Build:(\d+)\]\[Cksum:([^\]]+)\]\[Guid:([\{\}a-z0-9\-]+)\]''', re.IGNORECASE)
for m in re.finditer(pattern, cleartext):
added_entropy = m.group(2) + m.group(4)
# loop through the item records until all are processed
while len(items) > 0:
# get the first item record
item = items.pop(0)
# the first 32 chars of the first record of a group
# is the MD5 hash of the key name encoded by charMap5
keyhash = item[0:32]
# the sha1 of raw keyhash string is used to create entropy along
# with the added entropy provided above from the headerblob
entropy = SHA1(keyhash) + added_entropy
# the remainder of the first record when decoded with charMap5
# has the ':' split char followed by the string representation
# of the number of records that follow
# and make up the contents
srcnt = decode(item[34:],charMap5)
rcnt = int(srcnt)
# read and store in rcnt records of data
# that make up the contents value
edlst = []
for i in xrange(rcnt):
item = items.pop(0)
edlst.append(item)
# key names now use the new testMap8 encoding
keyname = "unknown"
for name in names:
if encodeHash(name,testMap8) == keyhash:
keyname = name
break
# the testMap8 encoded contents data has had a length
# of chars (always odd) cut off of the front and moved
# to the end to prevent decoding using testMap8 from
# working properly, and thereby preventing the ensuing
# CryptUnprotectData call from succeeding.
# The offset into the testMap8 encoded contents seems to be:
# len(contents)-largest prime number <= int(len(content)/3)
# (in other words split "about" 2/3rds of the way through)
# move first offsets chars to end to align for decode by testMap8
# by moving noffset chars from the start of the
# string to the end of the string
encdata = "".join(edlst)
contlen = len(encdata)
noffset = contlen - primes(int(contlen/3))[-1]
pfx = encdata[0:noffset]
encdata = encdata[noffset:]
encdata = encdata + pfx
# decode using new testMap8 to get the original CryptProtect Data
encryptedValue = decode(encdata,testMap8)
cleartext = CryptUnprotectData(encryptedValue, entropy, 1)
DB[keyname] = cleartext
cnt = cnt + 1
if cnt == 0:
DB = None
return DB

View File

@@ -0,0 +1,276 @@
#!/usr/bin/env python
from __future__ import with_statement
import sys
import os, csv
import binascii
import zlib
import re
from struct import pack, unpack, unpack_from
class DrmException(Exception):
pass
global charMap1
global charMap3
global charMap4
if 'calibre' in sys.modules:
inCalibre = True
else:
inCalibre = False
if inCalibre:
if sys.platform.startswith('win'):
from calibre_plugins.k4mobidedrm.k4pcutils import getKindleInfoFiles, getDBfromFile, GetUserName, GetIDString
if sys.platform.startswith('darwin'):
from calibre_plugins.k4mobidedrm.k4mutils import getKindleInfoFiles, getDBfromFile, GetUserName, GetIDString
else:
if sys.platform.startswith('win'):
from k4pcutils import getKindleInfoFiles, getDBfromFile, GetUserName, GetIDString
if sys.platform.startswith('darwin'):
from k4mutils import getKindleInfoFiles, getDBfromFile, GetUserName, GetIDString
charMap1 = "n5Pr6St7Uv8Wx9YzAb0Cd1Ef2Gh3Jk4M"
charMap3 = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"
charMap4 = "ABCDEFGHIJKLMNPQRSTUVWXYZ123456789"
# crypto digestroutines
import hashlib
def MD5(message):
ctx = hashlib.md5()
ctx.update(message)
return ctx.digest()
def SHA1(message):
ctx = hashlib.sha1()
ctx.update(message)
return ctx.digest()
# Encode the bytes in data with the characters in map
def encode(data, map):
result = ""
for char in data:
value = ord(char)
Q = (value ^ 0x80) // len(map)
R = value % len(map)
result += map[Q]
result += map[R]
return result
# Hash the bytes in data and then encode the digest with the characters in map
def encodeHash(data,map):
return encode(MD5(data),map)
# Decode the string in data with the characters in map. Returns the decoded bytes
def decode(data,map):
result = ""
for i in range (0,len(data)-1,2):
high = map.find(data[i])
low = map.find(data[i+1])
if (high == -1) or (low == -1) :
break
value = (((high * len(map)) ^ 0x80) & 0xFF) + low
result += pack("B",value)
return result
#
# PID generation routines
#
# Returns two bit at offset from a bit field
def getTwoBitsFromBitField(bitField,offset):
byteNumber = offset // 4
bitPosition = 6 - 2*(offset % 4)
return ord(bitField[byteNumber]) >> bitPosition & 3
# Returns the six bits at offset from a bit field
def getSixBitsFromBitField(bitField,offset):
offset *= 3
value = (getTwoBitsFromBitField(bitField,offset) <<4) + (getTwoBitsFromBitField(bitField,offset+1) << 2) +getTwoBitsFromBitField(bitField,offset+2)
return value
# 8 bits to six bits encoding from hash to generate PID string
def encodePID(hash):
global charMap3
PID = ""
for position in range (0,8):
PID += charMap3[getSixBitsFromBitField(hash,position)]
return PID
# Encryption table used to generate the device PID
def generatePidEncryptionTable() :
table = []
for counter1 in range (0,0x100):
value = counter1
for counter2 in range (0,8):
if (value & 1 == 0) :
value = value >> 1
else :
value = value >> 1
value = value ^ 0xEDB88320
table.append(value)
return table
# Seed value used to generate the device PID
def generatePidSeed(table,dsn) :
value = 0
for counter in range (0,4) :
index = (ord(dsn[counter]) ^ value) &0xFF
value = (value >> 8) ^ table[index]
return value
# Generate the device PID
def generateDevicePID(table,dsn,nbRoll):
global charMap4
seed = generatePidSeed(table,dsn)
pidAscii = ""
pid = [(seed >>24) &0xFF,(seed >> 16) &0xff,(seed >> 8) &0xFF,(seed) & 0xFF,(seed>>24) & 0xFF,(seed >> 16) &0xff,(seed >> 8) &0xFF,(seed) & 0xFF]
index = 0
for counter in range (0,nbRoll):
pid[index] = pid[index] ^ ord(dsn[counter])
index = (index+1) %8
for counter in range (0,8):
index = ((((pid[counter] >>5) & 3) ^ pid[counter]) & 0x1f) + (pid[counter] >> 7)
pidAscii += charMap4[index]
return pidAscii
def crc32(s):
return (~binascii.crc32(s,-1))&0xFFFFFFFF
# convert from 8 digit PID to 10 digit PID with checksum
def checksumPid(s):
global charMap4
crc = crc32(s)
crc = crc ^ (crc >> 16)
res = s
l = len(charMap4)
for i in (0,1):
b = crc & 0xff
pos = (b // l) ^ (b % l)
res += charMap4[pos%l]
crc >>= 8
return res
# old kindle serial number to fixed pid
def pidFromSerial(s, l):
global charMap4
crc = crc32(s)
arr1 = [0]*l
for i in xrange(len(s)):
arr1[i%l] ^= ord(s[i])
crc_bytes = [crc >> 24 & 0xff, crc >> 16 & 0xff, crc >> 8 & 0xff, crc & 0xff]
for i in xrange(l):
arr1[i] ^= crc_bytes[i&3]
pid = ""
for i in xrange(l):
b = arr1[i] & 0xff
pid+=charMap4[(b >> 7) + ((b >> 5 & 3) ^ (b & 0x1f))]
return pid
# Parse the EXTH header records and use the Kindle serial number to calculate the book pid.
def getKindlePid(pidlst, rec209, token, serialnum):
# Compute book PID
pidHash = SHA1(serialnum+rec209+token)
bookPID = encodePID(pidHash)
bookPID = checksumPid(bookPID)
pidlst.append(bookPID)
# compute fixed pid for old pre 2.5 firmware update pid as well
bookPID = pidFromSerial(serialnum, 7) + "*"
bookPID = checksumPid(bookPID)
pidlst.append(bookPID)
return pidlst
# parse the Kindleinfo file to calculate the book pid.
keynames = ["kindle.account.tokens","kindle.cookie.item","eulaVersionAccepted","login_date","kindle.token.item","login","kindle.key.item","kindle.name.info","kindle.device.info", "MazamaRandomNumber"]
def getK4Pids(pidlst, rec209, token, kInfoFile):
global charMap1
kindleDatabase = None
try:
kindleDatabase = getDBfromFile(kInfoFile)
except Exception, message:
print(message)
kindleDatabase = None
pass
if kindleDatabase == None :
return pidlst
try:
# Get the Mazama Random number
MazamaRandomNumber = kindleDatabase["MazamaRandomNumber"]
# Get the kindle account token
kindleAccountToken = kindleDatabase["kindle.account.tokens"]
except KeyError:
print "Keys not found in " + kInfoFile
return pidlst
# Get the ID string used
encodedIDString = encodeHash(GetIDString(),charMap1)
# Get the current user name
encodedUsername = encodeHash(GetUserName(),charMap1)
# concat, hash and encode to calculate the DSN
DSN = encode(SHA1(MazamaRandomNumber+encodedIDString+encodedUsername),charMap1)
# Compute the device PID (for which I can tell, is used for nothing).
table = generatePidEncryptionTable()
devicePID = generateDevicePID(table,DSN,4)
devicePID = checksumPid(devicePID)
pidlst.append(devicePID)
# Compute book PIDs
# book pid
pidHash = SHA1(DSN+kindleAccountToken+rec209+token)
bookPID = encodePID(pidHash)
bookPID = checksumPid(bookPID)
pidlst.append(bookPID)
# variant 1
pidHash = SHA1(kindleAccountToken+rec209+token)
bookPID = encodePID(pidHash)
bookPID = checksumPid(bookPID)
pidlst.append(bookPID)
# variant 2
pidHash = SHA1(DSN+rec209+token)
bookPID = encodePID(pidHash)
bookPID = checksumPid(bookPID)
pidlst.append(bookPID)
return pidlst
def getPidList(md1, md2, k4, pids, serials, kInfoFiles):
pidlst = []
if kInfoFiles is None:
kInfoFiles = []
if k4:
kInfoFiles = getKindleInfoFiles(kInfoFiles)
for infoFile in kInfoFiles:
try:
pidlst = getK4Pids(pidlst, md1, md2, infoFile)
except Exception, message:
print("Error getting PIDs from " + infoFile + ": " + message)
for serialnum in serials:
try:
pidlst = getKindlePid(pidlst, md1, md2, serialnum)
except Exception, message:
print("Error getting PIDs from " + serialnum + ": " + message)
for pid in pids:
pidlst.append(pid)
return pidlst

View File

@@ -0,0 +1,460 @@
#!/usr/bin/python
#
# This is a python script. You need a Python interpreter to run it.
# For example, ActiveState Python, which exists for windows.
#
# Changelog
# 0.01 - Initial version
# 0.02 - Huffdic compressed books were not properly decrypted
# 0.03 - Wasn't checking MOBI header length
# 0.04 - Wasn't sanity checking size of data record
# 0.05 - It seems that the extra data flags take two bytes not four
# 0.06 - And that low bit does mean something after all :-)
# 0.07 - The extra data flags aren't present in MOBI header < 0xE8 in size
# 0.08 - ...and also not in Mobi header version < 6
# 0.09 - ...but they are there with Mobi header version 6, header size 0xE4!
# 0.10 - Outputs unencrypted files as-is, so that when run as a Calibre
# import filter it works when importing unencrypted files.
# Also now handles encrypted files that don't need a specific PID.
# 0.11 - use autoflushed stdout and proper return values
# 0.12 - Fix for problems with metadata import as Calibre plugin, report errors
# 0.13 - Formatting fixes: retabbed file, removed trailing whitespace
# and extra blank lines, converted CR/LF pairs at ends of each line,
# and other cosmetic fixes.
# 0.14 - Working out when the extra data flags are present has been problematic
# Versions 7 through 9 have tried to tweak the conditions, but have been
# only partially successful. Closer examination of lots of sample
# files reveals that a confusion has arisen because trailing data entries
# are not encrypted, but it turns out that the multibyte entries
# in utf8 file are encrypted. (Although neither kind gets compressed.)
# This knowledge leads to a simplification of the test for the
# trailing data byte flags - version 5 and higher AND header size >= 0xE4.
# 0.15 - Now outputs 'heartbeat', and is also quicker for long files.
# 0.16 - And reverts to 'done' not 'done.' at the end for unswindle compatibility.
# 0.17 - added modifications to support its use as an imported python module
# both inside calibre and also in other places (ie K4DeDRM tools)
# 0.17a- disabled the standalone plugin feature since a plugin can not import
# a plugin
# 0.18 - It seems that multibyte entries aren't encrypted in a v7 file...
# Removed the disabled Calibre plug-in code
# Permit use of 8-digit PIDs
# 0.19 - It seems that multibyte entries aren't encrypted in a v6 file either.
# 0.20 - Correction: It seems that multibyte entries are encrypted in a v6 file.
# 0.21 - Added support for multiple pids
# 0.22 - revised structure to hold MobiBook as a class to allow an extended interface
# 0.23 - fixed problem with older files with no EXTH section
# 0.24 - add support for type 1 encryption and 'TEXtREAd' books as well
# 0.25 - Fixed support for 'BOOKMOBI' type 1 encryption
# 0.26 - Now enables Text-To-Speech flag and sets clipping limit to 100%
# 0.27 - Correct pid metadata token generation to match that used by skindle (Thank You Bart!)
# 0.28 - slight additional changes to metadata token generation (None -> '')
# 0.29 - It seems that the ideas about when multibyte trailing characters were
# included in the encryption were wrong. They are for DOC compressed
# files, but they are not for HUFF/CDIC compress files!
# 0.30 - Modified interface slightly to work better with new calibre plugin style
# 0.31 - The multibyte encrytion info is true for version 7 files too.
# 0.32 - Added support for "Print Replica" Kindle ebooks
# 0.33 - Performance improvements for large files (concatenation)
# 0.34 - Performance improvements in decryption (libalfcrypto)
# 0.35 - add interface to get mobi_version
# 0.36 - fixed problem with TEXtREAd and getBookTitle interface
# 0.37 - Fixed double announcement for stand-alone operation
__version__ = '0.37'
import sys
class Unbuffered:
def __init__(self, stream):
self.stream = stream
def write(self, data):
self.stream.write(data)
self.stream.flush()
def __getattr__(self, attr):
return getattr(self.stream, attr)
sys.stdout=Unbuffered(sys.stdout)
import os
import struct
import binascii
from alfcrypto import Pukall_Cipher
class DrmException(Exception):
pass
#
# MobiBook Utility Routines
#
# Implementation of Pukall Cipher 1
def PC1(key, src, decryption=True):
return Pukall_Cipher().PC1(key,src,decryption)
# sum1 = 0;
# sum2 = 0;
# keyXorVal = 0;
# if len(key)!=16:
# print "Bad key length!"
# return None
# wkey = []
# for i in xrange(8):
# wkey.append(ord(key[i*2])<<8 | ord(key[i*2+1]))
# dst = ""
# for i in xrange(len(src)):
# temp1 = 0;
# byteXorVal = 0;
# for j in xrange(8):
# temp1 ^= wkey[j]
# sum2 = (sum2+j)*20021 + sum1
# sum1 = (temp1*346)&0xFFFF
# sum2 = (sum2+sum1)&0xFFFF
# temp1 = (temp1*20021+1)&0xFFFF
# byteXorVal ^= temp1 ^ sum2
# curByte = ord(src[i])
# if not decryption:
# keyXorVal = curByte * 257;
# curByte = ((curByte ^ (byteXorVal >> 8)) ^ byteXorVal) & 0xFF
# if decryption:
# keyXorVal = curByte * 257;
# for j in xrange(8):
# wkey[j] ^= keyXorVal;
# dst+=chr(curByte)
# return dst
def checksumPid(s):
letters = "ABCDEFGHIJKLMNPQRSTUVWXYZ123456789"
crc = (~binascii.crc32(s,-1))&0xFFFFFFFF
crc = crc ^ (crc >> 16)
res = s
l = len(letters)
for i in (0,1):
b = crc & 0xff
pos = (b // l) ^ (b % l)
res += letters[pos%l]
crc >>= 8
return res
def getSizeOfTrailingDataEntries(ptr, size, flags):
def getSizeOfTrailingDataEntry(ptr, size):
bitpos, result = 0, 0
if size <= 0:
return result
while True:
v = ord(ptr[size-1])
result |= (v & 0x7F) << bitpos
bitpos += 7
size -= 1
if (v & 0x80) != 0 or (bitpos >= 28) or (size == 0):
return result
num = 0
testflags = flags >> 1
while testflags:
if testflags & 1:
num += getSizeOfTrailingDataEntry(ptr, size - num)
testflags >>= 1
# Check the low bit to see if there's multibyte data present.
# if multibyte data is included in the encryped data, we'll
# have already cleared this flag.
if flags & 1:
num += (ord(ptr[size - num - 1]) & 0x3) + 1
return num
class MobiBook:
def loadSection(self, section):
if (section + 1 == self.num_sections):
endoff = len(self.data_file)
else:
endoff = self.sections[section + 1][0]
off = self.sections[section][0]
return self.data_file[off:endoff]
def __init__(self, infile, announce = True):
if announce:
print ('MobiDeDrm v%(__version__)s. '
'Copyright 2008-2012 The Dark Reverser et al.' % globals())
# initial sanity check on file
self.data_file = file(infile, 'rb').read()
self.mobi_data = ''
self.header = self.data_file[0:78]
if self.header[0x3C:0x3C+8] != 'BOOKMOBI' and self.header[0x3C:0x3C+8] != 'TEXtREAd':
raise DrmException("invalid file format")
self.magic = self.header[0x3C:0x3C+8]
self.crypto_type = -1
# build up section offset and flag info
self.num_sections, = struct.unpack('>H', self.header[76:78])
self.sections = []
for i in xrange(self.num_sections):
offset, a1,a2,a3,a4 = struct.unpack('>LBBBB', self.data_file[78+i*8:78+i*8+8])
flags, val = a1, a2<<16|a3<<8|a4
self.sections.append( (offset, flags, val) )
# parse information from section 0
self.sect = self.loadSection(0)
self.records, = struct.unpack('>H', self.sect[0x8:0x8+2])
self.compression, = struct.unpack('>H', self.sect[0x0:0x0+2])
if self.magic == 'TEXtREAd':
print "Book has format: ", self.magic
self.extra_data_flags = 0
self.mobi_length = 0
self.mobi_codepage = 1252
self.mobi_version = -1
self.meta_array = {}
return
self.mobi_length, = struct.unpack('>L',self.sect[0x14:0x18])
self.mobi_codepage, = struct.unpack('>L',self.sect[0x1c:0x20])
self.mobi_version, = struct.unpack('>L',self.sect[0x68:0x6C])
print "MOBI header version = %d, length = %d" %(self.mobi_version, self.mobi_length)
self.extra_data_flags = 0
if (self.mobi_length >= 0xE4) and (self.mobi_version >= 5):
self.extra_data_flags, = struct.unpack('>H', self.sect[0xF2:0xF4])
print "Extra Data Flags = %d" % self.extra_data_flags
if (self.compression != 17480):
# multibyte utf8 data is included in the encryption for PalmDoc compression
# so clear that byte so that we leave it to be decrypted.
self.extra_data_flags &= 0xFFFE
# if exth region exists parse it for metadata array
self.meta_array = {}
try:
exth_flag, = struct.unpack('>L', self.sect[0x80:0x84])
exth = 'NONE'
if exth_flag & 0x40:
exth = self.sect[16 + self.mobi_length:]
if (len(exth) >= 4) and (exth[:4] == 'EXTH'):
nitems, = struct.unpack('>I', exth[8:12])
pos = 12
for i in xrange(nitems):
type, size = struct.unpack('>II', exth[pos: pos + 8])
content = exth[pos + 8: pos + size]
self.meta_array[type] = content
# reset the text to speech flag and clipping limit, if present
if type == 401 and size == 9:
# set clipping limit to 100%
self.patchSection(0, "\144", 16 + self.mobi_length + pos + 8)
elif type == 404 and size == 9:
# make sure text to speech is enabled
self.patchSection(0, "\0", 16 + self.mobi_length + pos + 8)
# print type, size, content, content.encode('hex')
pos += size
except:
self.meta_array = {}
pass
self.print_replica = False
def getBookTitle(self):
codec_map = {
1252 : 'windows-1252',
65001 : 'utf-8',
}
title = ''
codec = 'windows-1252'
if self.magic == 'BOOKMOBI':
if 503 in self.meta_array:
title = self.meta_array[503]
else:
toff, tlen = struct.unpack('>II', self.sect[0x54:0x5c])
tend = toff + tlen
title = self.sect[toff:tend]
if self.mobi_codepage in codec_map.keys():
codec = codec_map[self.mobi_codepage]
if title == '':
title = self.header[:32]
title = title.split("\0")[0]
return unicode(title, codec).encode('utf-8')
def getPIDMetaInfo(self):
rec209 = ''
token = ''
if 209 in self.meta_array:
rec209 = self.meta_array[209]
data = rec209
# The 209 data comes in five byte groups. Interpret the last four bytes
# of each group as a big endian unsigned integer to get a key value
# if that key exists in the meta_array, append its contents to the token
for i in xrange(0,len(data),5):
val, = struct.unpack('>I',data[i+1:i+5])
sval = self.meta_array.get(val,'')
token += sval
return rec209, token
def patch(self, off, new):
self.data_file = self.data_file[:off] + new + self.data_file[off+len(new):]
def patchSection(self, section, new, in_off = 0):
if (section + 1 == self.num_sections):
endoff = len(self.data_file)
else:
endoff = self.sections[section + 1][0]
off = self.sections[section][0]
assert off + in_off + len(new) <= endoff
self.patch(off + in_off, new)
def parseDRM(self, data, count, pidlist):
found_key = None
keyvec1 = "\x72\x38\x33\xB0\xB4\xF2\xE3\xCA\xDF\x09\x01\xD6\xE2\xE0\x3F\x96"
for pid in pidlist:
bigpid = pid.ljust(16,'\0')
temp_key = PC1(keyvec1, bigpid, False)
temp_key_sum = sum(map(ord,temp_key)) & 0xff
found_key = None
for i in xrange(count):
verification, size, type, cksum, cookie = struct.unpack('>LLLBxxx32s', data[i*0x30:i*0x30+0x30])
if cksum == temp_key_sum:
cookie = PC1(temp_key, cookie)
ver,flags,finalkey,expiry,expiry2 = struct.unpack('>LL16sLL', cookie)
if verification == ver and (flags & 0x1F) == 1:
found_key = finalkey
break
if found_key != None:
break
if not found_key:
# Then try the default encoding that doesn't require a PID
pid = "00000000"
temp_key = keyvec1
temp_key_sum = sum(map(ord,temp_key)) & 0xff
for i in xrange(count):
verification, size, type, cksum, cookie = struct.unpack('>LLLBxxx32s', data[i*0x30:i*0x30+0x30])
if cksum == temp_key_sum:
cookie = PC1(temp_key, cookie)
ver,flags,finalkey,expiry,expiry2 = struct.unpack('>LL16sLL', cookie)
if verification == ver:
found_key = finalkey
break
return [found_key,pid]
def getMobiFile(self, outpath):
file(outpath,'wb').write(self.mobi_data)
def getMobiVersion(self):
return self.mobi_version
def getPrintReplica(self):
return self.print_replica
def processBook(self, pidlist):
crypto_type, = struct.unpack('>H', self.sect[0xC:0xC+2])
print 'Crypto Type is: ', crypto_type
self.crypto_type = crypto_type
if crypto_type == 0:
print "This book is not encrypted."
# we must still check for Print Replica
self.print_replica = (self.loadSection(1)[0:4] == '%MOP')
self.mobi_data = self.data_file
return
if crypto_type != 2 and crypto_type != 1:
raise DrmException("Cannot decode unknown Mobipocket encryption type %d" % crypto_type)
if 406 in self.meta_array:
data406 = self.meta_array[406]
val406, = struct.unpack('>Q',data406)
if val406 != 0:
raise DrmException("Cannot decode library or rented ebooks.")
goodpids = []
for pid in pidlist:
if len(pid)==10:
if checksumPid(pid[0:-2]) != pid:
print "Warning: PID " + pid + " has incorrect checksum, should have been "+checksumPid(pid[0:-2])
goodpids.append(pid[0:-2])
elif len(pid)==8:
goodpids.append(pid)
if self.crypto_type == 1:
t1_keyvec = "QDCVEPMU675RUBSZ"
if self.magic == 'TEXtREAd':
bookkey_data = self.sect[0x0E:0x0E+16]
elif self.mobi_version < 0:
bookkey_data = self.sect[0x90:0x90+16]
else:
bookkey_data = self.sect[self.mobi_length+16:self.mobi_length+32]
pid = "00000000"
found_key = PC1(t1_keyvec, bookkey_data)
else :
# calculate the keys
drm_ptr, drm_count, drm_size, drm_flags = struct.unpack('>LLLL', self.sect[0xA8:0xA8+16])
if drm_count == 0:
raise DrmException("Not yet initialised with PID. Must be opened with Mobipocket Reader first.")
found_key, pid = self.parseDRM(self.sect[drm_ptr:drm_ptr+drm_size], drm_count, goodpids)
if not found_key:
raise DrmException("No key found in " + str(len(goodpids)) + " keys tried. Please report this failure for help.")
# kill the drm keys
self.patchSection(0, "\0" * drm_size, drm_ptr)
# kill the drm pointers
self.patchSection(0, "\xff" * 4 + "\0" * 12, 0xA8)
if pid=="00000000":
print "File has default encryption, no specific PID."
else:
print "File is encoded with PID "+checksumPid(pid)+"."
# clear the crypto type
self.patchSection(0, "\0" * 2, 0xC)
# decrypt sections
print "Decrypting. Please wait . . .",
mobidataList = []
mobidataList.append(self.data_file[:self.sections[1][0]])
for i in xrange(1, self.records+1):
data = self.loadSection(i)
extra_size = getSizeOfTrailingDataEntries(data, len(data), self.extra_data_flags)
if i%100 == 0:
print ".",
# print "record %d, extra_size %d" %(i,extra_size)
decoded_data = PC1(found_key, data[0:len(data) - extra_size])
if i==1:
self.print_replica = (decoded_data[0:4] == '%MOP')
mobidataList.append(decoded_data)
if extra_size > 0:
mobidataList.append(data[-extra_size:])
if self.num_sections > self.records+1:
mobidataList.append(self.data_file[self.sections[self.records+1][0]:])
self.mobi_data = "".join(mobidataList)
print "done"
return
def getUnencryptedBook(infile,pid,announce=True):
if not os.path.isfile(infile):
raise DrmException('Input File Not Found')
book = MobiBook(infile,announce)
book.processBook([pid])
return book.mobi_data
def getUnencryptedBookWithList(infile,pidlist,announce=True):
if not os.path.isfile(infile):
raise DrmException('Input File Not Found')
book = MobiBook(infile, announce)
book.processBook(pidlist)
return book.mobi_data
def main(argv=sys.argv):
print ('MobiDeDrm v%(__version__)s. '
'Copyright 2008-2012 The Dark Reverser et al.' % globals())
if len(argv)<3 or len(argv)>4:
print "Removes protection from Kindle/Mobipocket, Kindle/KF8 and Kindle/Print Replica ebooks"
print "Usage:"
print " %s <infile> <outfile> [<Comma separated list of PIDs to try>]" % sys.argv[0]
return 1
else:
infile = argv[1]
outfile = argv[2]
if len(argv) is 4:
pidlist = argv[3].split(',')
else:
pidlist = {}
try:
stripped_file = getUnencryptedBookWithList(infile, pidlist, False)
file(outfile, 'wb').write(stripped_file)
except DrmException, e:
print "Error: %s" % e
return 1
return 0
if __name__ == "__main__":
sys.exit(main())

View File

@@ -0,0 +1,89 @@
#!/usr/bin/env python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
# implement just enough of des from openssl to make erdr2pml.py happy
def load_libcrypto():
from ctypes import CDLL, POINTER, c_void_p, c_char_p, c_char, c_int, c_long, \
Structure, c_ulong, create_string_buffer, cast
from ctypes.util import find_library
import sys
if sys.platform.startswith('win'):
libcrypto = find_library('libeay32')
else:
libcrypto = find_library('crypto')
if libcrypto is None:
return None
libcrypto = CDLL(libcrypto)
# typedef struct DES_ks
# {
# union
# {
# DES_cblock cblock;
# /* make sure things are correct size on machines with
# * 8 byte longs */
# DES_LONG deslong[2];
# } ks[16];
# } DES_key_schedule;
# just create a big enough place to hold everything
# it will have alignment of structure so we should be okay (16 byte aligned?)
class DES_KEY_SCHEDULE(Structure):
_fields_ = [('DES_cblock1', c_char * 16),
('DES_cblock2', c_char * 16),
('DES_cblock3', c_char * 16),
('DES_cblock4', c_char * 16),
('DES_cblock5', c_char * 16),
('DES_cblock6', c_char * 16),
('DES_cblock7', c_char * 16),
('DES_cblock8', c_char * 16),
('DES_cblock9', c_char * 16),
('DES_cblock10', c_char * 16),
('DES_cblock11', c_char * 16),
('DES_cblock12', c_char * 16),
('DES_cblock13', c_char * 16),
('DES_cblock14', c_char * 16),
('DES_cblock15', c_char * 16),
('DES_cblock16', c_char * 16)]
DES_KEY_SCHEDULE_p = POINTER(DES_KEY_SCHEDULE)
def F(restype, name, argtypes):
func = getattr(libcrypto, name)
func.restype = restype
func.argtypes = argtypes
return func
DES_set_key = F(None, 'DES_set_key',[c_char_p, DES_KEY_SCHEDULE_p])
DES_ecb_encrypt = F(None, 'DES_ecb_encrypt',[c_char_p, c_char_p, DES_KEY_SCHEDULE_p, c_int])
class DES(object):
def __init__(self, key):
if len(key) != 8 :
raise Error('DES improper key used')
return
self.key = key
self.keyschedule = DES_KEY_SCHEDULE()
DES_set_key(self.key, self.keyschedule)
def desdecrypt(self, data):
ob = create_string_buffer(len(data))
DES_ecb_encrypt(data, ob, self.keyschedule, 0)
return ob.raw
def decrypt(self, data):
if not data:
return ''
i = 0
result = []
while i < len(data):
block = data[i:i+8]
processed_block = self.desdecrypt(block)
result.append(processed_block)
i += 8
return ''.join(result)
return DES

View File

@@ -0,0 +1,68 @@
# A simple implementation of pbkdf2 using stock python modules. See RFC2898
# for details. Basically, it derives a key from a password and salt.
# Copyright 2004 Matt Johnston <matt @ ucc asn au>
# Copyright 2009 Daniel Holth <dholth@fastmail.fm>
# This code may be freely used and modified for any purpose.
# Revision history
# v0.1 October 2004 - Initial release
# v0.2 8 March 2007 - Make usable with hashlib in Python 2.5 and use
# v0.3 "" the correct digest_size rather than always 20
# v0.4 Oct 2009 - Rescue from chandler svn, test and optimize.
import sys
import hmac
from struct import pack
try:
# only in python 2.5
import hashlib
sha = hashlib.sha1
md5 = hashlib.md5
sha256 = hashlib.sha256
except ImportError: # pragma: NO COVERAGE
# fallback
import sha
import md5
# this is what you want to call.
def pbkdf2( password, salt, itercount, keylen, hashfn = sha ):
try:
# depending whether the hashfn is from hashlib or sha/md5
digest_size = hashfn().digest_size
except TypeError: # pragma: NO COVERAGE
digest_size = hashfn.digest_size
# l - number of output blocks to produce
l = keylen / digest_size
if keylen % digest_size != 0:
l += 1
h = hmac.new( password, None, hashfn )
T = ""
for i in range(1, l+1):
T += pbkdf2_F( h, salt, itercount, i )
return T[0: keylen]
def xorstr( a, b ):
if len(a) != len(b):
raise ValueError("xorstr(): lengths differ")
return ''.join((chr(ord(x)^ord(y)) for x, y in zip(a, b)))
def prf( h, data ):
hm = h.copy()
hm.update( data )
return hm.digest()
# Helper as per the spec. h is a hmac which has been created seeded with the
# password, it will be copy()ed and not modified.
def pbkdf2_F( h, salt, itercount, blocknum ):
U = prf( h, salt + pack('>i',blocknum ) )
T = U
for i in range(2, itercount+1):
U = prf( h, U )
T = xorstr( T, U )
return T

View File

@@ -0,0 +1,30 @@
#!/usr/bin/env python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
def load_pycrypto():
try :
from Crypto.Cipher import DES as _DES
except:
return None
class DES(object):
def __init__(self, key):
if len(key) != 8 :
raise Error('DES improper key used')
self.key = key
self._des = _DES.new(key,_DES.MODE_ECB)
def desdecrypt(self, data):
return self._des.decrypt(data)
def decrypt(self, data):
if not data:
return ''
i = 0
result = []
while i < len(data):
block = data[i:i+8]
processed_block = self.desdecrypt(block)
result.append(processed_block)
i += 8
return ''.join(result)
return DES

View File

@@ -0,0 +1,220 @@
#!/usr/bin/env python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
import sys
ECB = 0
CBC = 1
class Des(object):
__pc1 = [56, 48, 40, 32, 24, 16, 8, 0, 57, 49, 41, 33, 25, 17,
9, 1, 58, 50, 42, 34, 26, 18, 10, 2, 59, 51, 43, 35,
62, 54, 46, 38, 30, 22, 14, 6, 61, 53, 45, 37, 29, 21,
13, 5, 60, 52, 44, 36, 28, 20, 12, 4, 27, 19, 11, 3]
__left_rotations = [1, 1, 2, 2, 2, 2, 2, 2, 1, 2, 2, 2, 2, 2, 2, 1]
__pc2 = [13, 16, 10, 23, 0, 4,2, 27, 14, 5, 20, 9,
22, 18, 11, 3, 25, 7, 15, 6, 26, 19, 12, 1,
40, 51, 30, 36, 46, 54, 29, 39, 50, 44, 32, 47,
43, 48, 38, 55, 33, 52, 45, 41, 49, 35, 28, 31]
__ip = [57, 49, 41, 33, 25, 17, 9, 1, 59, 51, 43, 35, 27, 19, 11, 3,
61, 53, 45, 37, 29, 21, 13, 5, 63, 55, 47, 39, 31, 23, 15, 7,
56, 48, 40, 32, 24, 16, 8, 0, 58, 50, 42, 34, 26, 18, 10, 2,
60, 52, 44, 36, 28, 20, 12, 4, 62, 54, 46, 38, 30, 22, 14, 6]
__expansion_table = [31, 0, 1, 2, 3, 4, 3, 4, 5, 6, 7, 8,
7, 8, 9, 10, 11, 12,11, 12, 13, 14, 15, 16,
15, 16, 17, 18, 19, 20,19, 20, 21, 22, 23, 24,
23, 24, 25, 26, 27, 28,27, 28, 29, 30, 31, 0]
__sbox = [[14, 4, 13, 1, 2, 15, 11, 8, 3, 10, 6, 12, 5, 9, 0, 7,
0, 15, 7, 4, 14, 2, 13, 1, 10, 6, 12, 11, 9, 5, 3, 8,
4, 1, 14, 8, 13, 6, 2, 11, 15, 12, 9, 7, 3, 10, 5, 0,
15, 12, 8, 2, 4, 9, 1, 7, 5, 11, 3, 14, 10, 0, 6, 13],
[15, 1, 8, 14, 6, 11, 3, 4, 9, 7, 2, 13, 12, 0, 5, 10,
3, 13, 4, 7, 15, 2, 8, 14, 12, 0, 1, 10, 6, 9, 11, 5,
0, 14, 7, 11, 10, 4, 13, 1, 5, 8, 12, 6, 9, 3, 2, 15,
13, 8, 10, 1, 3, 15, 4, 2, 11, 6, 7, 12, 0, 5, 14, 9],
[10, 0, 9, 14, 6, 3, 15, 5, 1, 13, 12, 7, 11, 4, 2, 8,
13, 7, 0, 9, 3, 4, 6, 10, 2, 8, 5, 14, 12, 11, 15, 1,
13, 6, 4, 9, 8, 15, 3, 0, 11, 1, 2, 12, 5, 10, 14, 7,
1, 10, 13, 0, 6, 9, 8, 7, 4, 15, 14, 3, 11, 5, 2, 12],
[7, 13, 14, 3, 0, 6, 9, 10, 1, 2, 8, 5, 11, 12, 4, 15,
13, 8, 11, 5, 6, 15, 0, 3, 4, 7, 2, 12, 1, 10, 14, 9,
10, 6, 9, 0, 12, 11, 7, 13, 15, 1, 3, 14, 5, 2, 8, 4,
3, 15, 0, 6, 10, 1, 13, 8, 9, 4, 5, 11, 12, 7, 2, 14],
[2, 12, 4, 1, 7, 10, 11, 6, 8, 5, 3, 15, 13, 0, 14, 9,
14, 11, 2, 12, 4, 7, 13, 1, 5, 0, 15, 10, 3, 9, 8, 6,
4, 2, 1, 11, 10, 13, 7, 8, 15, 9, 12, 5, 6, 3, 0, 14,
11, 8, 12, 7, 1, 14, 2, 13, 6, 15, 0, 9, 10, 4, 5, 3],
[12, 1, 10, 15, 9, 2, 6, 8, 0, 13, 3, 4, 14, 7, 5, 11,
10, 15, 4, 2, 7, 12, 9, 5, 6, 1, 13, 14, 0, 11, 3, 8,
9, 14, 15, 5, 2, 8, 12, 3, 7, 0, 4, 10, 1, 13, 11, 6,
4, 3, 2, 12, 9, 5, 15, 10, 11, 14, 1, 7, 6, 0, 8, 13],
[4, 11, 2, 14, 15, 0, 8, 13, 3, 12, 9, 7, 5, 10, 6, 1,
13, 0, 11, 7, 4, 9, 1, 10, 14, 3, 5, 12, 2, 15, 8, 6,
1, 4, 11, 13, 12, 3, 7, 14, 10, 15, 6, 8, 0, 5, 9, 2,
6, 11, 13, 8, 1, 4, 10, 7, 9, 5, 0, 15, 14, 2, 3, 12],
[13, 2, 8, 4, 6, 15, 11, 1, 10, 9, 3, 14, 5, 0, 12, 7,
1, 15, 13, 8, 10, 3, 7, 4, 12, 5, 6, 11, 0, 14, 9, 2,
7, 11, 4, 1, 9, 12, 14, 2, 0, 6, 10, 13, 15, 3, 5, 8,
2, 1, 14, 7, 4, 10, 8, 13, 15, 12, 9, 0, 3, 5, 6, 11],]
__p = [15, 6, 19, 20, 28, 11,27, 16, 0, 14, 22, 25,
4, 17, 30, 9, 1, 7,23,13, 31, 26, 2, 8,18, 12, 29, 5, 21, 10,3, 24]
__fp = [39, 7, 47, 15, 55, 23, 63, 31,38, 6, 46, 14, 54, 22, 62, 30,
37, 5, 45, 13, 53, 21, 61, 29,36, 4, 44, 12, 52, 20, 60, 28,
35, 3, 43, 11, 51, 19, 59, 27,34, 2, 42, 10, 50, 18, 58, 26,
33, 1, 41, 9, 49, 17, 57, 25,32, 0, 40, 8, 48, 16, 56, 24]
# Type of crypting being done
ENCRYPT = 0x00
DECRYPT = 0x01
def __init__(self, key, mode=ECB, IV=None):
if len(key) != 8:
raise ValueError("Invalid DES key size. Key must be exactly 8 bytes long.")
self.block_size = 8
self.key_size = 8
self.__padding = ''
self.setMode(mode)
if IV:
self.setIV(IV)
self.L = []
self.R = []
self.Kn = [ [0] * 48 ] * 16 # 16 48-bit keys (K1 - K16)
self.final = []
self.setKey(key)
def getKey(self):
return self.__key
def setKey(self, key):
self.__key = key
self.__create_sub_keys()
def getMode(self):
return self.__mode
def setMode(self, mode):
self.__mode = mode
def getIV(self):
return self.__iv
def setIV(self, IV):
if not IV or len(IV) != self.block_size:
raise ValueError("Invalid Initial Value (IV), must be a multiple of " + str(self.block_size) + " bytes")
self.__iv = IV
def getPadding(self):
return self.__padding
def __String_to_BitList(self, data):
l = len(data) * 8
result = [0] * l
pos = 0
for c in data:
i = 7
ch = ord(c)
while i >= 0:
if ch & (1 << i) != 0:
result[pos] = 1
else:
result[pos] = 0
pos += 1
i -= 1
return result
def __BitList_to_String(self, data):
result = ''
pos = 0
c = 0
while pos < len(data):
c += data[pos] << (7 - (pos % 8))
if (pos % 8) == 7:
result += chr(c)
c = 0
pos += 1
return result
def __permutate(self, table, block):
return [block[x] for x in table]
def __create_sub_keys(self):
key = self.__permutate(Des.__pc1, self.__String_to_BitList(self.getKey()))
i = 0
self.L = key[:28]
self.R = key[28:]
while i < 16:
j = 0
while j < Des.__left_rotations[i]:
self.L.append(self.L[0])
del self.L[0]
self.R.append(self.R[0])
del self.R[0]
j += 1
self.Kn[i] = self.__permutate(Des.__pc2, self.L + self.R)
i += 1
def __des_crypt(self, block, crypt_type):
block = self.__permutate(Des.__ip, block)
self.L = block[:32]
self.R = block[32:]
if crypt_type == Des.ENCRYPT:
iteration = 0
iteration_adjustment = 1
else:
iteration = 15
iteration_adjustment = -1
i = 0
while i < 16:
tempR = self.R[:]
self.R = self.__permutate(Des.__expansion_table, self.R)
self.R = [x ^ y for x,y in zip(self.R, self.Kn[iteration])]
B = [self.R[:6], self.R[6:12], self.R[12:18], self.R[18:24], self.R[24:30], self.R[30:36], self.R[36:42], self.R[42:]]
j = 0
Bn = [0] * 32
pos = 0
while j < 8:
m = (B[j][0] << 1) + B[j][5]
n = (B[j][1] << 3) + (B[j][2] << 2) + (B[j][3] << 1) + B[j][4]
v = Des.__sbox[j][(m << 4) + n]
Bn[pos] = (v & 8) >> 3
Bn[pos + 1] = (v & 4) >> 2
Bn[pos + 2] = (v & 2) >> 1
Bn[pos + 3] = v & 1
pos += 4
j += 1
self.R = self.__permutate(Des.__p, Bn)
self.R = [x ^ y for x, y in zip(self.R, self.L)]
self.L = tempR
i += 1
iteration += iteration_adjustment
self.final = self.__permutate(Des.__fp, self.R + self.L)
return self.final
def crypt(self, data, crypt_type):
if not data:
return ''
if len(data) % self.block_size != 0:
if crypt_type == Des.DECRYPT: # Decryption must work on 8 byte blocks
raise ValueError("Invalid data length, data must be a multiple of " + str(self.block_size) + " bytes\n.")
if not self.getPadding():
raise ValueError("Invalid data length, data must be a multiple of " + str(self.block_size) + " bytes\n. Try setting the optional padding character")
else:
data += (self.block_size - (len(data) % self.block_size)) * self.getPadding()
if self.getMode() == CBC:
if self.getIV():
iv = self.__String_to_BitList(self.getIV())
else:
raise ValueError("For CBC mode, you must supply the Initial Value (IV) for ciphering")
i = 0
dict = {}
result = []
while i < len(data):
block = self.__String_to_BitList(data[i:i+8])
if self.getMode() == CBC:
if crypt_type == Des.ENCRYPT:
block = [x ^ y for x, y in zip(block, iv)]
processed_block = self.__des_crypt(block, crypt_type)
if crypt_type == Des.DECRYPT:
processed_block = [x ^ y for x, y in zip(processed_block, iv)]
iv = block
else:
iv = processed_block
else:
processed_block = self.__des_crypt(block, crypt_type)
result.append(self.__BitList_to_String(processed_block))
i += 8
if crypt_type == Des.DECRYPT and self.getPadding():
s = result[-1]
while s[-1] == self.getPadding():
s = s[:-1]
result[-1] = s
return ''.join(result)
def encrypt(self, data, pad=''):
self.__padding = pad
return self.crypt(data, Des.ENCRYPT)
def decrypt(self, data, pad=''):
self.__padding = pad
return self.crypt(data, Des.DECRYPT)

View File

@@ -0,0 +1,27 @@
#!/usr/bin/env python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
import Tkinter
import Tkconstants
# basic scrolled text widget
class ScrolledText(Tkinter.Text):
def __init__(self, master=None, **kw):
self.frame = Tkinter.Frame(master)
self.vbar = Tkinter.Scrollbar(self.frame)
self.vbar.pack(side=Tkconstants.RIGHT, fill=Tkconstants.Y)
kw.update({'yscrollcommand': self.vbar.set})
Tkinter.Text.__init__(self, self.frame, **kw)
self.pack(side=Tkconstants.LEFT, fill=Tkconstants.BOTH, expand=True)
self.vbar['command'] = self.yview
# Copy geometry methods of self.frame without overriding Text
# methods = hack!
text_meths = vars(Tkinter.Text).keys()
methods = vars(Tkinter.Pack).keys() + vars(Tkinter.Grid).keys() + vars(Tkinter.Place).keys()
methods = set(methods).difference(text_meths)
for m in methods:
if m[0] != '_' and m != 'config' and m != 'configure':
setattr(self, m, getattr(self.frame, m))
def __str__(self):
return str(self.frame)

View File

@@ -0,0 +1,77 @@
#!/usr/bin/env python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
import sys
import os, os.path
import shutil
class SimplePrefsError(Exception):
pass
class SimplePrefs(object):
def __init__(self, target, description):
self.prefs = {}
self.key2file={}
self.file2key={}
for keyfilemap in description:
[key, filename] = keyfilemap
self.key2file[key] = filename
self.file2key[filename] = key
self.target = target + 'Prefs'
if sys.platform.startswith('win'):
import _winreg as winreg
regkey = winreg.OpenKey(winreg.HKEY_CURRENT_USER, "Software\\Microsoft\\Windows\\CurrentVersion\\Explorer\\Shell Folders\\")
path = winreg.QueryValueEx(regkey, 'Local AppData')[0]
prefdir = path + os.sep + self.target
elif sys.platform.startswith('darwin'):
home = os.getenv('HOME')
prefdir = os.path.join(home,'Library','Preferences','org.' + self.target)
else:
# linux and various flavors of unix
home = os.getenv('HOME')
prefdir = os.path.join(home,'.' + self.target)
if not os.path.exists(prefdir):
os.makedirs(prefdir)
self.prefdir = prefdir
self.prefs['dir'] = self.prefdir
self._loadPreferences()
def _loadPreferences(self):
filenames = os.listdir(self.prefdir)
for filename in filenames:
if filename in self.file2key:
key = self.file2key[filename]
filepath = os.path.join(self.prefdir,filename)
if os.path.isfile(filepath):
try :
data = file(filepath,'rb').read()
self.prefs[key] = data
except Exception, e:
pass
def getPreferences(self):
return self.prefs
def setPreferences(self, newprefs={}):
if 'dir' not in newprefs:
raise SimplePrefsError('Error: Attempt to Set Preferences in unspecified directory')
if newprefs['dir'] != self.prefs['dir']:
raise SimplePrefsError('Error: Attempt to Set Preferences in unspecified directory')
for key in newprefs:
if key != 'dir':
if key in self.key2file:
filename = self.key2file[key]
filepath = os.path.join(self.prefdir,filename)
data = newprefs[key]
if data != None:
data = str(data)
if data == None or data == '':
if os.path.exists(filepath):
os.remove(filepath)
else:
try:
file(filepath,'wb').write(data)
except Exception, e:
pass
self.prefs = newprefs
return

View File

@@ -0,0 +1,266 @@
#! /usr/bin/python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
# For use with Topaz Scripts Version 2.6
import csv
import sys
import os
import getopt
import re
from struct import pack
from struct import unpack
class DocParser(object):
def __init__(self, flatxml, fontsize, ph, pw):
self.flatdoc = flatxml.split('\n')
self.fontsize = int(fontsize)
self.ph = int(ph) * 1.0
self.pw = int(pw) * 1.0
stags = {
'paragraph' : 'p',
'graphic' : '.graphic'
}
attr_val_map = {
'hang' : 'text-indent: ',
'indent' : 'text-indent: ',
'line-space' : 'line-height: ',
'margin-bottom' : 'margin-bottom: ',
'margin-left' : 'margin-left: ',
'margin-right' : 'margin-right: ',
'margin-top' : 'margin-top: ',
'space-after' : 'padding-bottom: ',
}
attr_str_map = {
'align-center' : 'text-align: center; margin-left: auto; margin-right: auto;',
'align-left' : 'text-align: left;',
'align-right' : 'text-align: right;',
'align-justify' : 'text-align: justify;',
'display-inline' : 'display: inline;',
'pos-left' : 'text-align: left;',
'pos-right' : 'text-align: right;',
'pos-center' : 'text-align: center; margin-left: auto; margin-right: auto;',
}
# find tag if within pos to end inclusive
def findinDoc(self, tagpath, pos, end) :
result = None
docList = self.flatdoc
cnt = len(docList)
if end == -1 :
end = cnt
else:
end = min(cnt,end)
foundat = -1
for j in xrange(pos, end):
item = docList[j]
if item.find('=') >= 0:
(name, argres) = item.split('=',1)
else :
name = item
argres = ''
if name.endswith(tagpath) :
result = argres
foundat = j
break
return foundat, result
# return list of start positions for the tagpath
def posinDoc(self, tagpath):
startpos = []
pos = 0
res = ""
while res != None :
(foundpos, res) = self.findinDoc(tagpath, pos, -1)
if res != None :
startpos.append(foundpos)
pos = foundpos + 1
return startpos
# returns a vector of integers for the tagpath
def getData(self, tagpath, pos, end, clean=False):
if clean:
digits_only = re.compile(r'''([0-9]+)''')
argres=[]
(foundat, argt) = self.findinDoc(tagpath, pos, end)
if (argt != None) and (len(argt) > 0) :
argList = argt.split('|')
for strval in argList:
if clean:
m = re.search(digits_only, strval)
if m != None:
strval = m.group()
argres.append(int(strval))
return argres
def process(self):
classlst = ''
csspage = '.cl-center { text-align: center; margin-left: auto; margin-right: auto; }\n'
csspage += '.cl-right { text-align: right; }\n'
csspage += '.cl-left { text-align: left; }\n'
csspage += '.cl-justify { text-align: justify; }\n'
# generate a list of each <style> starting point in the stylesheet
styleList= self.posinDoc('book.stylesheet.style')
stylecnt = len(styleList)
styleList.append(-1)
# process each style converting what you can
for j in xrange(stylecnt):
start = styleList[j]
end = styleList[j+1]
(pos, tag) = self.findinDoc('style._tag',start,end)
if tag == None :
(pos, tag) = self.findinDoc('style.type',start,end)
# Is this something we know how to convert to css
if tag in self.stags :
# get the style class
(pos, sclass) = self.findinDoc('style.class',start,end)
if sclass != None:
sclass = sclass.replace(' ','-')
sclass = '.cl-' + sclass.lower()
else :
sclass = ''
# check for any "after class" specifiers
(pos, aftclass) = self.findinDoc('style._after_class',start,end)
if aftclass != None:
aftclass = aftclass.replace(' ','-')
aftclass = '.cl-' + aftclass.lower()
else :
aftclass = ''
cssargs = {}
while True :
(pos1, attr) = self.findinDoc('style.rule.attr', start, end)
(pos2, val) = self.findinDoc('style.rule.value', start, end)
if attr == None : break
if (attr == 'display') or (attr == 'pos') or (attr == 'align'):
# handle text based attributess
attr = attr + '-' + val
if attr in self.attr_str_map :
cssargs[attr] = (self.attr_str_map[attr], '')
else :
# handle value based attributes
if attr in self.attr_val_map :
name = self.attr_val_map[attr]
if attr in ('margin-bottom', 'margin-top', 'space-after') :
scale = self.ph
elif attr in ('margin-right', 'indent', 'margin-left', 'hang') :
scale = self.pw
elif attr == 'line-space':
scale = self.fontsize * 2.0
if val == "":
val = 0
if not ((attr == 'hang') and (int(val) == 0)) :
pv = float(val)/scale
cssargs[attr] = (self.attr_val_map[attr], pv)
keep = True
start = max(pos1, pos2) + 1
# disable all of the after class tags until I figure out how to handle them
if aftclass != "" : keep = False
if keep :
# make sure line-space does not go below 100% or above 300% since
# it can be wacky in some styles
if 'line-space' in cssargs:
seg = cssargs['line-space'][0]
val = cssargs['line-space'][1]
if val < 1.0: val = 1.0
if val > 3.0: val = 3.0
del cssargs['line-space']
cssargs['line-space'] = (self.attr_val_map['line-space'], val)
# handle modifications for css style hanging indents
if 'hang' in cssargs:
hseg = cssargs['hang'][0]
hval = cssargs['hang'][1]
del cssargs['hang']
cssargs['hang'] = (self.attr_val_map['hang'], -hval)
mval = 0
mseg = 'margin-left: '
mval = hval
if 'margin-left' in cssargs:
mseg = cssargs['margin-left'][0]
mval = cssargs['margin-left'][1]
if mval < 0: mval = 0
mval = hval + mval
cssargs['margin-left'] = (mseg, mval)
if 'indent' in cssargs:
del cssargs['indent']
cssline = sclass + ' { '
for key in iter(cssargs):
mseg = cssargs[key][0]
mval = cssargs[key][1]
if mval == '':
cssline += mseg + ' '
else :
aseg = mseg + '%.1f%%;' % (mval * 100.0)
cssline += aseg + ' '
cssline += '}'
if sclass != '' :
classlst += sclass + '\n'
# handle special case of paragraph class used inside chapter heading
# and non-chapter headings
if sclass != '' :
ctype = sclass[4:7]
if ctype == 'ch1' :
csspage += 'h1' + cssline + '\n'
if ctype == 'ch2' :
csspage += 'h2' + cssline + '\n'
if ctype == 'ch3' :
csspage += 'h3' + cssline + '\n'
if ctype == 'h1-' :
csspage += 'h4' + cssline + '\n'
if ctype == 'h2-' :
csspage += 'h5' + cssline + '\n'
if ctype == 'h3_' :
csspage += 'h6' + cssline + '\n'
if cssline != ' { }':
csspage += self.stags[tag] + cssline + '\n'
return csspage, classlst
def convert2CSS(flatxml, fontsize, ph, pw):
print ' ', 'Using font size:',fontsize
print ' ', 'Using page height:', ph
print ' ', 'Using page width:', pw
# create a document parser
dp = DocParser(flatxml, fontsize, ph, pw)
csspage = dp.process()
return csspage
def getpageIDMap(flatxml):
dp = DocParser(flatxml, 0, 0, 0)
pageidnumbers = dp.getData('info.original.pid', 0, -1, True)
return pageidnumbers

View File

@@ -0,0 +1,148 @@
#!/usr/bin/env python
# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
import os, sys
import signal
import threading
import subprocess
from subprocess import Popen, PIPE, STDOUT
# **heavily** chopped up and modfied version of asyncproc.py
# to make it actually work on Windows as well as Mac/Linux
# For the original see:
# "http://www.lysator.liu.se/~bellman/download/"
# author is "Thomas Bellman <bellman@lysator.liu.se>"
# available under GPL version 3 or Later
# create an asynchronous subprocess whose output can be collected in
# a non-blocking manner
# What a mess! Have to use threads just to get non-blocking io
# in a cross-platform manner
# luckily all thread use is hidden within this class
class Process(object):
def __init__(self, *params, **kwparams):
if len(params) <= 3:
kwparams.setdefault('stdin', subprocess.PIPE)
if len(params) <= 4:
kwparams.setdefault('stdout', subprocess.PIPE)
if len(params) <= 5:
kwparams.setdefault('stderr', subprocess.PIPE)
self.__pending_input = []
self.__collected_outdata = []
self.__collected_errdata = []
self.__exitstatus = None
self.__lock = threading.Lock()
self.__inputsem = threading.Semaphore(0)
self.__quit = False
self.__process = subprocess.Popen(*params, **kwparams)
if self.__process.stdin:
self.__stdin_thread = threading.Thread(
name="stdin-thread",
target=self.__feeder, args=(self.__pending_input,
self.__process.stdin))
self.__stdin_thread.setDaemon(True)
self.__stdin_thread.start()
if self.__process.stdout:
self.__stdout_thread = threading.Thread(
name="stdout-thread",
target=self.__reader, args=(self.__collected_outdata,
self.__process.stdout))
self.__stdout_thread.setDaemon(True)
self.__stdout_thread.start()
if self.__process.stderr:
self.__stderr_thread = threading.Thread(
name="stderr-thread",
target=self.__reader, args=(self.__collected_errdata,
self.__process.stderr))
self.__stderr_thread.setDaemon(True)
self.__stderr_thread.start()
def pid(self):
return self.__process.pid
def kill(self, signal):
self.__process.send_signal(signal)
# check on subprocess (pass in 'nowait') to act like poll
def wait(self, flag):
if flag.lower() == 'nowait':
rc = self.__process.poll()
else:
rc = self.__process.wait()
if rc != None:
if self.__process.stdin:
self.closeinput()
if self.__process.stdout:
self.__stdout_thread.join()
if self.__process.stderr:
self.__stderr_thread.join()
return self.__process.returncode
def terminate(self):
if self.__process.stdin:
self.closeinput()
self.__process.terminate()
# thread gets data from subprocess stdout
def __reader(self, collector, source):
while True:
data = os.read(source.fileno(), 65536)
self.__lock.acquire()
collector.append(data)
self.__lock.release()
if data == "":
source.close()
break
return
# thread feeds data to subprocess stdin
def __feeder(self, pending, drain):
while True:
self.__inputsem.acquire()
self.__lock.acquire()
if not pending and self.__quit:
drain.close()
self.__lock.release()
break
data = pending.pop(0)
self.__lock.release()
drain.write(data)
# non-blocking read of data from subprocess stdout
def read(self):
self.__lock.acquire()
outdata = "".join(self.__collected_outdata)
del self.__collected_outdata[:]
self.__lock.release()
return outdata
# non-blocking read of data from subprocess stderr
def readerr(self):
self.__lock.acquire()
errdata = "".join(self.__collected_errdata)
del self.__collected_errdata[:]
self.__lock.release()
return errdata
# non-blocking write to stdin of subprocess
def write(self, data):
if self.__process.stdin is None:
raise ValueError("Writing to process with stdin not a pipe")
self.__lock.acquire()
self.__pending_input.append(data)
self.__inputsem.release()
self.__lock.release()
# close stdinput of subprocess
def closeinput(self):
self.__lock.acquire()
self.__quit = True
self.__inputsem.release()
self.__lock.release()

View File

@@ -0,0 +1,482 @@
#!/usr/bin/env python
class Unbuffered:
def __init__(self, stream):
self.stream = stream
def write(self, data):
self.stream.write(data)
self.stream.flush()
def __getattr__(self, attr):
return getattr(self.stream, attr)
import sys
if 'calibre' in sys.modules:
inCalibre = True
else:
inCalibre = False
buildXML = False
import os, csv, getopt
import zlib, zipfile, tempfile, shutil
from struct import pack
from struct import unpack
from alfcrypto import Topaz_Cipher
class TpzDRMError(Exception):
pass
# local support routines
if inCalibre:
from calibre_plugins.k4mobidedrm import kgenpids
else:
import kgenpids
# recursive zip creation support routine
def zipUpDir(myzip, tdir, localname):
currentdir = tdir
if localname != "":
currentdir = os.path.join(currentdir,localname)
list = os.listdir(currentdir)
for file in list:
afilename = file
localfilePath = os.path.join(localname, afilename)
realfilePath = os.path.join(currentdir,file)
if os.path.isfile(realfilePath):
myzip.write(realfilePath, localfilePath)
elif os.path.isdir(realfilePath):
zipUpDir(myzip, tdir, localfilePath)
#
# Utility routines
#
# Get a 7 bit encoded number from file
def bookReadEncodedNumber(fo):
flag = False
data = ord(fo.read(1))
if data == 0xFF:
flag = True
data = ord(fo.read(1))
if data >= 0x80:
datax = (data & 0x7F)
while data >= 0x80 :
data = ord(fo.read(1))
datax = (datax <<7) + (data & 0x7F)
data = datax
if flag:
data = -data
return data
# Get a length prefixed string from file
def bookReadString(fo):
stringLength = bookReadEncodedNumber(fo)
return unpack(str(stringLength)+"s",fo.read(stringLength))[0]
#
# crypto routines
#
# Context initialisation for the Topaz Crypto
def topazCryptoInit(key):
return Topaz_Cipher().ctx_init(key)
# ctx1 = 0x0CAFFE19E
# for keyChar in key:
# keyByte = ord(keyChar)
# ctx2 = ctx1
# ctx1 = ((((ctx1 >>2) * (ctx1 >>7))&0xFFFFFFFF) ^ (keyByte * keyByte * 0x0F902007)& 0xFFFFFFFF )
# return [ctx1,ctx2]
# decrypt data with the context prepared by topazCryptoInit()
def topazCryptoDecrypt(data, ctx):
return Topaz_Cipher().decrypt(data, ctx)
# ctx1 = ctx[0]
# ctx2 = ctx[1]
# plainText = ""
# for dataChar in data:
# dataByte = ord(dataChar)
# m = (dataByte ^ ((ctx1 >> 3) &0xFF) ^ ((ctx2<<3) & 0xFF)) &0xFF
# ctx2 = ctx1
# ctx1 = (((ctx1 >> 2) * (ctx1 >> 7)) &0xFFFFFFFF) ^((m * m * 0x0F902007) &0xFFFFFFFF)
# plainText += chr(m)
# return plainText
# Decrypt data with the PID
def decryptRecord(data,PID):
ctx = topazCryptoInit(PID)
return topazCryptoDecrypt(data, ctx)
# Try to decrypt a dkey record (contains the bookPID)
def decryptDkeyRecord(data,PID):
record = decryptRecord(data,PID)
fields = unpack("3sB8sB8s3s",record)
if fields[0] != "PID" or fields[5] != "pid" :
raise TpzDRMError("Didn't find PID magic numbers in record")
elif fields[1] != 8 or fields[3] != 8 :
raise TpzDRMError("Record didn't contain correct length fields")
elif fields[2] != PID :
raise TpzDRMError("Record didn't contain PID")
return fields[4]
# Decrypt all dkey records (contain the book PID)
def decryptDkeyRecords(data,PID):
nbKeyRecords = ord(data[0])
records = []
data = data[1:]
for i in range (0,nbKeyRecords):
length = ord(data[0])
try:
key = decryptDkeyRecord(data[1:length+1],PID)
records.append(key)
except TpzDRMError:
pass
data = data[1+length:]
if len(records) == 0:
raise TpzDRMError("BookKey Not Found")
return records
class TopazBook:
def __init__(self, filename):
self.fo = file(filename, 'rb')
self.outdir = tempfile.mkdtemp()
# self.outdir = 'rawdat'
self.bookPayloadOffset = 0
self.bookHeaderRecords = {}
self.bookMetadata = {}
self.bookKey = None
magic = unpack("4s",self.fo.read(4))[0]
if magic != 'TPZ0':
raise TpzDRMError("Parse Error : Invalid Header, not a Topaz file")
self.parseTopazHeaders()
self.parseMetadata()
def parseTopazHeaders(self):
def bookReadHeaderRecordData():
# Read and return the data of one header record at the current book file position
# [[offset,decompressedLength,compressedLength],...]
nbValues = bookReadEncodedNumber(self.fo)
values = []
for i in range (0,nbValues):
values.append([bookReadEncodedNumber(self.fo),bookReadEncodedNumber(self.fo),bookReadEncodedNumber(self.fo)])
return values
def parseTopazHeaderRecord():
# Read and parse one header record at the current book file position and return the associated data
# [[offset,decompressedLength,compressedLength],...]
if ord(self.fo.read(1)) != 0x63:
raise TpzDRMError("Parse Error : Invalid Header")
tag = bookReadString(self.fo)
record = bookReadHeaderRecordData()
return [tag,record]
nbRecords = bookReadEncodedNumber(self.fo)
for i in range (0,nbRecords):
result = parseTopazHeaderRecord()
# print result[0], result[1]
self.bookHeaderRecords[result[0]] = result[1]
if ord(self.fo.read(1)) != 0x64 :
raise TpzDRMError("Parse Error : Invalid Header")
self.bookPayloadOffset = self.fo.tell()
def parseMetadata(self):
# Parse the metadata record from the book payload and return a list of [key,values]
self.fo.seek(self.bookPayloadOffset + self.bookHeaderRecords["metadata"][0][0])
tag = bookReadString(self.fo)
if tag != "metadata" :
raise TpzDRMError("Parse Error : Record Names Don't Match")
flags = ord(self.fo.read(1))
nbRecords = ord(self.fo.read(1))
# print nbRecords
for i in range (0,nbRecords) :
keyval = bookReadString(self.fo)
content = bookReadString(self.fo)
# print keyval
# print content
self.bookMetadata[keyval] = content
return self.bookMetadata
def getPIDMetaInfo(self):
keysRecord = self.bookMetadata.get('keys','')
keysRecordRecord = ''
if keysRecord != '':
keylst = keysRecord.split(',')
for keyval in keylst:
keysRecordRecord += self.bookMetadata.get(keyval,'')
return keysRecord, keysRecordRecord
def getBookTitle(self):
title = ''
if 'Title' in self.bookMetadata:
title = self.bookMetadata['Title']
return title
def setBookKey(self, key):
self.bookKey = key
def getBookPayloadRecord(self, name, index):
# Get a record in the book payload, given its name and index.
# decrypted and decompressed if necessary
encrypted = False
compressed = False
try:
recordOffset = self.bookHeaderRecords[name][index][0]
except:
raise TpzDRMError("Parse Error : Invalid Record, record not found")
self.fo.seek(self.bookPayloadOffset + recordOffset)
tag = bookReadString(self.fo)
if tag != name :
raise TpzDRMError("Parse Error : Invalid Record, record name doesn't match")
recordIndex = bookReadEncodedNumber(self.fo)
if recordIndex < 0 :
encrypted = True
recordIndex = -recordIndex -1
if recordIndex != index :
raise TpzDRMError("Parse Error : Invalid Record, index doesn't match")
if (self.bookHeaderRecords[name][index][2] > 0):
compressed = True
record = self.fo.read(self.bookHeaderRecords[name][index][2])
else:
record = self.fo.read(self.bookHeaderRecords[name][index][1])
if encrypted:
if self.bookKey:
ctx = topazCryptoInit(self.bookKey)
record = topazCryptoDecrypt(record,ctx)
else :
raise TpzDRMError("Error: Attempt to decrypt without bookKey")
if compressed:
record = zlib.decompress(record)
return record
def processBook(self, pidlst):
raw = 0
fixedimage=True
try:
keydata = self.getBookPayloadRecord('dkey', 0)
except TpzDRMError, e:
print "no dkey record found, book may not be encrypted"
print "attempting to extrct files without a book key"
self.createBookDirectory()
self.extractFiles()
print "Successfully Extracted Topaz contents"
if inCalibre:
from calibre_plugins.k4mobidedrm import genbook
else:
import genbook
rv = genbook.generateBook(self.outdir, raw, fixedimage)
if rv == 0:
print "\nBook Successfully generated"
return rv
# try each pid to decode the file
bookKey = None
for pid in pidlst:
# use 8 digit pids here
pid = pid[0:8]
print "\nTrying: ", pid
bookKeys = []
data = keydata
try:
bookKeys+=decryptDkeyRecords(data,pid)
except TpzDRMError, e:
pass
else:
bookKey = bookKeys[0]
print "Book Key Found!"
break
if not bookKey:
raise TpzDRMError('Decryption Unsucessful; No valid pid found')
self.setBookKey(bookKey)
self.createBookDirectory()
self.extractFiles()
print "Successfully Extracted Topaz contents"
if inCalibre:
from calibre_plugins.k4mobidedrm import genbook
else:
import genbook
rv = genbook.generateBook(self.outdir, raw, fixedimage)
if rv == 0:
print "\nBook Successfully generated"
return rv
def createBookDirectory(self):
outdir = self.outdir
# create output directory structure
if not os.path.exists(outdir):
os.makedirs(outdir)
destdir = os.path.join(outdir,'img')
if not os.path.exists(destdir):
os.makedirs(destdir)
destdir = os.path.join(outdir,'color_img')
if not os.path.exists(destdir):
os.makedirs(destdir)
destdir = os.path.join(outdir,'page')
if not os.path.exists(destdir):
os.makedirs(destdir)
destdir = os.path.join(outdir,'glyphs')
if not os.path.exists(destdir):
os.makedirs(destdir)
def extractFiles(self):
outdir = self.outdir
for headerRecord in self.bookHeaderRecords:
name = headerRecord
if name != "dkey" :
ext = '.dat'
if name == 'img' : ext = '.jpg'
if name == 'color' : ext = '.jpg'
print "\nProcessing Section: %s " % name
for index in range (0,len(self.bookHeaderRecords[name])) :
fnum = "%04d" % index
fname = name + fnum + ext
destdir = outdir
if name == 'img':
destdir = os.path.join(outdir,'img')
if name == 'color':
destdir = os.path.join(outdir,'color_img')
if name == 'page':
destdir = os.path.join(outdir,'page')
if name == 'glyphs':
destdir = os.path.join(outdir,'glyphs')
outputFile = os.path.join(destdir,fname)
print ".",
record = self.getBookPayloadRecord(name,index)
if record != '':
file(outputFile, 'wb').write(record)
print " "
def getHTMLZip(self, zipname):
htmlzip = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
htmlzip.write(os.path.join(self.outdir,'book.html'),'book.html')
htmlzip.write(os.path.join(self.outdir,'book.opf'),'book.opf')
if os.path.isfile(os.path.join(self.outdir,'cover.jpg')):
htmlzip.write(os.path.join(self.outdir,'cover.jpg'),'cover.jpg')
htmlzip.write(os.path.join(self.outdir,'style.css'),'style.css')
zipUpDir(htmlzip, self.outdir, 'img')
htmlzip.close()
def getSVGZip(self, zipname):
svgzip = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
svgzip.write(os.path.join(self.outdir,'index_svg.xhtml'),'index_svg.xhtml')
zipUpDir(svgzip, self.outdir, 'svg')
zipUpDir(svgzip, self.outdir, 'img')
svgzip.close()
def getXMLZip(self, zipname):
xmlzip = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
targetdir = os.path.join(self.outdir,'xml')
zipUpDir(xmlzip, targetdir, '')
zipUpDir(xmlzip, self.outdir, 'img')
xmlzip.close()
def cleanup(self):
if os.path.isdir(self.outdir):
shutil.rmtree(self.outdir, True)
def usage(progname):
print "Removes DRM protection from Topaz ebooks and extract the contents"
print "Usage:"
print " %s [-k <kindle.info>] [-p <pidnums>] [-s <kindleSerialNumbers>] <infile> <outdir> " % progname
# Main
def main(argv=sys.argv):
global buildXML
progname = os.path.basename(argv[0])
k4 = False
pids = []
serials = []
kInfoFiles = []
try:
opts, args = getopt.getopt(sys.argv[1:], "k:p:s:")
except getopt.GetoptError, err:
print str(err)
usage(progname)
return 1
if len(args)<2:
usage(progname)
return 1
for o, a in opts:
if o == "-k":
if a == None :
print "Invalid parameter for -k"
return 1
kInfoFiles.append(a)
if o == "-p":
if a == None :
print "Invalid parameter for -p"
return 1
pids = a.split(',')
if o == "-s":
if a == None :
print "Invalid parameter for -s"
return 1
serials = a.split(',')
k4 = True
infile = args[0]
outdir = args[1]
if not os.path.isfile(infile):
print "Input File Does Not Exist"
return 1
bookname = os.path.splitext(os.path.basename(infile))[0]
tb = TopazBook(infile)
title = tb.getBookTitle()
print "Processing Book: ", title
keysRecord, keysRecordRecord = tb.getPIDMetaInfo()
pidlst = kgenpids.getPidList(keysRecord, keysRecordRecord, k4, pids, serials, kInfoFiles)
try:
print "Decrypting Book"
tb.processBook(pidlst)
print " Creating HTML ZIP Archive"
zipname = os.path.join(outdir, bookname + '_nodrm' + '.htmlz')
tb.getHTMLZip(zipname)
print " Creating SVG ZIP Archive"
zipname = os.path.join(outdir, bookname + '_SVG' + '.zip')
tb.getSVGZip(zipname)
if buildXML:
print " Creating XML ZIP Archive"
zipname = os.path.join(outdir, bookname + '_XML' + '.zip')
tb.getXMLZip(zipname)
# removing internal temporary directory of pieces
tb.cleanup()
except TpzDRMError, e:
print str(e)
# tb.cleanup()
return 1
except Exception, e:
print str(e)
# tb.cleanup
return 1
return 0
if __name__ == '__main__':
sys.stdout=Unbuffered(sys.stdout)
sys.exit(main())

View File

@@ -0,0 +1,154 @@
#!/usr/bin/env python
import sys
import zlib
import zipfile
import os
import os.path
import getopt
from struct import unpack
_FILENAME_LEN_OFFSET = 26
_EXTRA_LEN_OFFSET = 28
_FILENAME_OFFSET = 30
_MAX_SIZE = 64 * 1024
_MIMETYPE = 'application/epub+zip'
class ZipInfo(zipfile.ZipInfo):
def __init__(self, *args, **kwargs):
if 'compress_type' in kwargs:
compress_type = kwargs.pop('compress_type')
super(ZipInfo, self).__init__(*args, **kwargs)
self.compress_type = compress_type
class fixZip:
def __init__(self, zinput, zoutput):
self.ztype = 'zip'
if zinput.lower().find('.epub') >= 0 :
self.ztype = 'epub'
self.inzip = zipfile.ZipFile(zinput,'r')
self.outzip = zipfile.ZipFile(zoutput,'w')
# open the input zip for reading only as a raw file
self.bzf = file(zinput,'rb')
def getlocalname(self, zi):
local_header_offset = zi.header_offset
self.bzf.seek(local_header_offset + _FILENAME_LEN_OFFSET)
leninfo = self.bzf.read(2)
local_name_length, = unpack('<H', leninfo)
self.bzf.seek(local_header_offset + _FILENAME_OFFSET)
local_name = self.bzf.read(local_name_length)
return local_name
def uncompress(self, cmpdata):
dc = zlib.decompressobj(-15)
data = ''
while len(cmpdata) > 0:
if len(cmpdata) > _MAX_SIZE :
newdata = cmpdata[0:_MAX_SIZE]
cmpdata = cmpdata[_MAX_SIZE:]
else:
newdata = cmpdata
cmpdata = ''
newdata = dc.decompress(newdata)
unprocessed = dc.unconsumed_tail
if len(unprocessed) == 0:
newdata += dc.flush()
data += newdata
cmpdata += unprocessed
unprocessed = ''
return data
def getfiledata(self, zi):
# get file name length and exta data length to find start of file data
local_header_offset = zi.header_offset
self.bzf.seek(local_header_offset + _FILENAME_LEN_OFFSET)
leninfo = self.bzf.read(2)
local_name_length, = unpack('<H', leninfo)
self.bzf.seek(local_header_offset + _EXTRA_LEN_OFFSET)
exinfo = self.bzf.read(2)
extra_field_length, = unpack('<H', exinfo)
self.bzf.seek(local_header_offset + _FILENAME_OFFSET + local_name_length + extra_field_length)
data = None
# if not compressed we are good to go
if zi.compress_type == zipfile.ZIP_STORED:
data = self.bzf.read(zi.file_size)
# if compressed we must decompress it using zlib
if zi.compress_type == zipfile.ZIP_DEFLATED:
cmpdata = self.bzf.read(zi.compress_size)
data = self.uncompress(cmpdata)
return data
def fix(self):
# get the zipinfo for each member of the input archive
# and copy member over to output archive
# if problems exist with local vs central filename, fix them
# if epub write mimetype file first, with no compression
if self.ztype == 'epub':
nzinfo = ZipInfo('mimetype', compress_type=zipfile.ZIP_STORED)
self.outzip.writestr(nzinfo, _MIMETYPE)
# write the rest of the files
for zinfo in self.inzip.infolist():
if zinfo.filename != "mimetype" or self.ztype == '.zip':
data = None
nzinfo = zinfo
try:
data = self.inzip.read(zinfo.filename)
except zipfile.BadZipfile or zipfile.error:
local_name = self.getlocalname(zinfo)
data = self.getfiledata(zinfo)
nzinfo.filename = local_name
nzinfo.date_time = zinfo.date_time
nzinfo.compress_type = zinfo.compress_type
nzinfo.flag_bits = 0
nzinfo.internal_attr = 0
self.outzip.writestr(nzinfo,data)
self.bzf.close()
self.inzip.close()
self.outzip.close()
def usage():
print """usage: zipfix.py inputzip outputzip
inputzip is the source zipfile to fix
outputzip is the fixed zip archive
"""
def repairBook(infile, outfile):
if not os.path.exists(infile):
print "Error: Input Zip File does not exist"
return 1
try:
fr = fixZip(infile, outfile)
fr.fix()
return 0
except Exception, e:
print "Error Occurred ", e
return 2
def main(argv=sys.argv):
if len(argv)!=3:
usage()
return 1
infile = argv[1]
outfile = argv[2]
return repairBook(infile, outfile)
if __name__ == '__main__' :
sys.exit(main())