francois/DeDRM_tools
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

tools v2.2

Browse Source
This commit is contained in:
Apprentice Alf
2010-11-11 22:11:36 +00:00
parent 5f0671db7f
commit c386ac6e6d
100 changed files with 3157 additions and 4807 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
Topaz_Tools/lib/changes.txt
View File

@@ -1,64 +0,0 @@
Changes in version 2.0
- gensvg.py now accepts two options
-x : output browseable XHTML+SVG pages (default)
-r : output raw SVG images (useful for later conversion to pdf)
- flatxml2html.py now understands page.groups of type graphic
and handles vertical regions as svg images
- genhtml.py now accepts an option
--fixed-image : which will force the conversion
of all fixed regions to svg images
- minor bug fixes and html conversion improvements
Changes in version 1.8
- gensvg.py now builds wonderful xhtml pages with embedded svg
that can be easily paged through as if reading a book!
(tested in Safari for Mac and Win and Firefox)
(requires javascript to be enabled)
- genhtml.py now REQUIRES that gensvg.py be run FIRST
this allows create of images on the fly from glyphs
- genhtml.py now automatically makes tables of words into svg
based images and will handle glyph based ornate first
letters of words
- cmbtc_dump_mac_linux.py has been renamed to be
cmbtc_dump_nonK4PC.py to make it clearer
when it needs to be used
Changes in version 1.7
- gensvg.py has been improved so that the glyphs render exactly (ClarkNova)
- gensvg.py has fixed a render order "bug" that allowed some images to cover or hide text. (ClarkNova)
- change generated html to use external stylesheet via a link to "style.css"
- add missing <title> tag
- make xhtml compliant doctype and minor changes to write correct xhtml
- make divs that act as anchors be hidden visually and to take up 0 height and 0 width to prevent any impact on layout
Changes in version 1.6
- support for books whose paragraphs have no styles
- support to run cmbtc_dump on Linux and Mac OSX provided you know your PID of your ipod or standalone Kindle
(contributed by DiapDealer)
Changes in version 1.5
- completely reworked generation of styles to use actual page heights and widths
- added new script getpagedim.py to support the above
- style names with underscores in them are now properly paired with their base class
- fixed hanging indents that did not ever set a left margin
- added support for a number of not previously known region types
- added support for a previously unknown snippet - <empty></empty>
- corrected a bug that caused unknown regions to abort the program
- added code to make the handling of unknown regions better in general
- corrected a bug that caused the last link on a page to be missing (if it was the last thing on the page)
Changes in version 1.3
- font generation by gensvg.py is now greatly improved with support for contour points added
- support for more region types
- support for inline images in paragraphs or text fields (ie. initial graphics for the first letter of a word)
- greatly improved dtd information used for the xml to prevent parsing mistakes
Version 1.0
- initial release

604
Topaz_Tools/lib/cmbtc_dump.py
View File

@@ -1,19 +1,5 @@
#! /usr/bin/python
# For use in Topaz Scripts version 2.6
"""
Comprehensive Mazama Book DRM with Topaz Cryptography V2.0
-----BEGIN PUBLIC KEY-----
MIGfMA0GCSqGSIb3DQEBAQUAA4GNADCBiQKBgQDdBHJ4CNc6DNFCw4MRCw4SWAK6
M8hYfnNEI0yQmn5Ti+W8biT7EatpauE/5jgQMPBmdNrDr1hbHyHBSP7xeC2qlRWC
B62UCxeu/fpfnvNHDN/wPWWH4jynZ2M6cdcnE5LQ+FfeKqZn7gnG2No1U9h7oOHx
y2/pHuYme7U1TsgSjwIDAQAB
-----END PUBLIC KEY-----
"""
from __future__ import with_statement
#!/usr/bin/env python
# For use with Topaz Scripts Version 2.6
class Unbuffered:
def __init__(self, stream):
@@ -27,155 +13,64 @@ class Unbuffered:
import sys
sys.stdout=Unbuffered(sys.stdout)
import csv
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
global kindleDatabase
global verbose
global PIDs
if sys.platform.startswith('win'):
from k4pcutils import openKindleInfo, CryptUnprotectData, GetUserName, GetVolumeSerialNumber, charMap1, charMap2, charMap3, charMap4
if sys.platform.startswith('darwin'):
from k4mutils import openKindleInfo, CryptUnprotectData, GetUserName, GetVolumeSerialNumber, charMap1, charMap2, charMap3, charMap4
#
# 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"
#
# Returns the MD5 digest of "message"
def SHA1(message):
ctx = hashlib.sha1()
ctx.update(message)
return ctx.digest()
#
# Open the book file at path
#
# 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
#
# Encode the bytes in data with the characters in map
def encode(data, map):
result = ""
for char in data:
@@ -186,55 +81,52 @@ def encode(data, map):
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):
for i in range (0,len(data)-1,2):
high = map.find(data[i])
low = map.find(data[i+1])
value = (((high * 0x40) ^ 0x80) & 0xFF) + low
if (high == -1) or (low == -1) :
break
value = (((high * len(map)) ^ 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():
def parseKindleInfo(kInfoFile):
DB = {}
infoReader = openKindleInfo()
infoReader = openKindleInfo(kInfoFile)
infoReader.read(1)
data = infoReader.read()
items = data.split('{')
if sys.platform.startswith('win'):
items = data.split('{')
else :
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)
#
# 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)
if sys.platform.startswith('win'):
return CryptUnprotectData(encryptedValue,"")
else:
cleartext = CryptUnprotectData(encryptedValue)
return decode(cleartext, charMap1)
# 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))
# Find if the original string for a hashed/encoded string is known. If so return the original string othwise return an empty string.
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 = ""
@@ -242,42 +134,81 @@ def findNameForHash(hash):
if hash == encodeHash(name, charMap2):
result = name
break
return name
return result
#
# 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")
print ("--------------------------")
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
# PID generation routines
#
def getKindleInfoValueForKey(key):
return getKindleInfoValueForHash(encodeHash(key,charMap2))
#
# Get a 7 bit encoded number from the book file
#
# 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):
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
# Get a 7 bit encoded number from the book file
def bookReadEncodedNumber():
flag = False
data = ord(bookFile.read(1))
@@ -297,14 +228,12 @@ def bookReadEncodedNumber():
data = -data
return data
#
# Encode a number in 7 bit format
#
def encodeNumber(number):
result = ""
negative = False
flag = 0
print("Using encodeNumber routine")
if number < 0 :
number = -number + 1
@@ -326,26 +255,17 @@ def encodeNumber(number):
return result[::-1]
#
# Get a length prefixed string from the file
#
# 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,decompressedLength,compressedLength],...]
#
def bookReadHeaderRecordData():
nbValues = bookReadEncodedNumber()
values = []
@@ -353,10 +273,7 @@ def bookReadHeaderRecordData():
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,decompressedLength,compressedLength],...]
#
def parseTopazHeaderRecord():
if ord(bookFile.read(1)) != 0x63:
raise CMBDTCFatal("Parse Error : Invalid Header")
@@ -365,10 +282,7 @@ def parseTopazHeaderRecord():
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
@@ -382,7 +296,7 @@ def parseTopazHeader():
for i in range (0,nbRecords):
result = parseTopazHeaderRecord()
print result[0], result[1]
#print result[0], result[1]
bookHeaderRecords[result[0]] = result[1]
if ord(bookFile.read(1)) != 0x64 :
@@ -390,11 +304,8 @@ def parseTopazHeader():
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
# Correction, the record is correctly decompressed too
#
def getBookPayloadRecord(name, index):
encrypted = False
compressed = False
@@ -434,10 +345,7 @@ def getBookPayloadRecord(name, index):
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
@@ -463,17 +371,11 @@ def extractBookPayloadRecord(name, index, filename):
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
@@ -491,40 +393,7 @@ def parseMetadata():
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
@@ -534,10 +403,7 @@ def topazCryptoInit(key):
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]
@@ -553,18 +419,12 @@ def topazCryptoDecrypt(data, ctx):
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)
@@ -577,11 +437,8 @@ def decryptDkeyRecord(data,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 = []
@@ -597,57 +454,7 @@ def decryptDkeyRecords(data,PID):
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):
for headerRecord in bookHeaderRecords:
name = headerRecord
@@ -670,10 +477,7 @@ def createDecryptedPayload(payload):
outputFile = os.path.join(destdir,fname)
file(outputFile, 'wb').write(getBookPayloadRecord(name, index))
# Create decrypted book
#
def createDecryptedBook(outdir):
if not os.path.exists(outdir):
os.makedirs(outdir)
@@ -696,11 +500,7 @@ def createDecryptedBook(outdir):
createDecryptedPayload(outdir)
#
# Set the command to execute by the programm according to cmdLine parameters
#
def setCommand(name) :
global command
if command != "" :
@@ -708,28 +508,85 @@ def setCommand(name) :
else :
command = name
#
# Program usage
#
def usage():
print("\nUsage:")
print("\ncmbtc_dump.py [options] bookFileName\n")
print("\ncmbtc_dump_linux.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 Dumps the unencrypted book as files to outdir")
print("-o Output directory to save book files to")
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
print("-k Adds the path to an alternate kindle.info file")
def prepTopazBook(bookPath):
global bookFile
bookFile = openBook(bookPath)
parseTopazHeader()
parseMetadata()
# Get Pids
def getK4Pids(kInfoFile=None):
global kindleDatabase
global PIDs
# Read the encrypted database
kindleDatabase = None
try:
kindleDatabase = parseKindleInfo(kInfoFile)
except Exception as message:
#if verbose > 0:
# print(message)
pass
if kindleDatabase != None :
# Compute the DSN
# Get the Mazama Random number
MazamaRandomNumber = getKindleInfoValueForKey("MazamaRandomNumber")
# Get the HDD serial
encodedSystemVolumeSerialNumber = encodeHash(GetVolumeSerialNumber(),charMap1)
# Get the current user name
encodedUsername = encodeHash(GetUserName(),charMap1)
# concat, hash and encode
DSN = encode(SHA1(MazamaRandomNumber+encodedSystemVolumeSerialNumber+encodedUsername),charMap1)
if verbose > 0:
print("DSN: " + DSN)
# Compute the device PID
table = generatePidEncryptionTable()
devicePID = generateDevicePID(table,DSN,4)
PIDs.append(devicePID)
if verbose > 0:
print("Device PID: " + devicePID)
# Compute book PID
# Get the account token
kindleAccountToken = getKindleInfoValueForKey("kindle.account.tokens")
if verbose > 0:
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 )
# Main
def main(argv=sys.argv):
global verbose
global PIDs
global bookKey
global command
progname = os.path.basename(argv[0])
@@ -739,12 +596,12 @@ def main(argv=sys.argv):
recordIndex = 0
outdir = ""
PIDs = []
kindleDatabase = None
command = ""
kInfoFiles = []
try:
opts, args = getopt.getopt(sys.argv[1:], "vi:o:p:d")
opts, args = getopt.getopt(sys.argv[1:], "vi:k:o:p:d")
except getopt.GetoptError, err:
# print help information and exit:
print str(err) # will print something like "option -a not recognized"
@@ -760,103 +617,50 @@ def main(argv=sys.argv):
verbose+=1
if o == "-i":
setCommand("printInfo")
if o == "-k":
if a == None :
raise CMBDTCFatal("Invalid parameter for -k")
kInfoFiles.append(a)
if o =="-o":
if a == None :
raise CMBDTCFatal("Invalid parameter for -o")
outdir = a
if o =="-p":
if a == None :
raise CMBDTCFatal("Invalid parameter for -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 as 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
#
raise Exception("No action supplied on command line")
# 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)
# Open the ebook
prepTopazBook(args[0])
# Always try to get the default Kindle installation info.
getK4Pids()
# If Alternate kindle.info files were supplied, parse them too.
if kInfoFiles:
for infoFile in kInfoFiles:
getK4Pids(infoFile)
# Print the kindle info if requested.
if kindleDatabase != None :
if command == "printInfo" :
printKindleInfo()
# Remove any duplicates that may occur from the PIDs List
PIDs = list(set(PIDs))
# Decrypt book key
dkey = getBookPayloadRecord('dkey', 0)
bookKeys = []
for PID in PIDs :
print PID
bookKeys+=decryptDkeyRecords(dkey,PID)
if len(bookKeys) == 0 :
@@ -867,7 +671,7 @@ def main(argv=sys.argv):
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 :
@@ -875,7 +679,7 @@ def main(argv=sys.argv):
elif command == "doit" :
if outdir != "" :
createDecryptedBook(outdir)
if verbose >0 :
if verbose > 0 :
print ("Decrypted book saved. Don't pirate!")
elif verbose > 0:
print("Output directory name was not supplied.")

524
Topaz_Tools/lib/cmbtc_dump_nonK4PC.py
View File

@@ -1,524 +0,0 @@
#!/usr/bin/python
# 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
import zlib
from struct import pack
from struct import unpack
MAX_PATH = 255
# Put the first 8 characters of your Kindle PID here
# or supply it with the -p option in the command line
####################################################
kindlePID = "12345678"
####################################################
global bookFile
global bookPayloadOffset
global bookHeaderRecords
global bookMetadata
global bookKey
global command
#
# Exceptions for all the problems that might happen during the script
#
class CMBDTCError(Exception):
pass
class CMBDTCFatal(Exception):
pass
#
# Open the book file at path
#
def openBook(path):
try:
return open(path,'rb')
except:
raise CMBDTCFatal("Could not open book file: " + path)
#
# 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
print("Using encodeNumber routine")
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,decompressedLength,compressedLength],...]
#
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,decompressedLength,compressedLength],...]
#
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()
print result[0], result[1]
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
# Correction, the record is correctly decompressed too
#
def getBookPayloadRecord(name, index):
encrypted = False
compressed = 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):
compressed = True
record = bookFile.read(bookHeaderRecords[name][index][2])
else:
record = bookFile.read(bookHeaderRecords[name][index][1])
if encrypted:
ctx = topazCryptoInit(bookKey)
record = topazCryptoDecrypt(record,ctx)
if compressed:
record = zlib.decompress(record)
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]
#
# 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
#
# Create decrypted book payload
#
def createDecryptedPayload(payload):
for headerRecord in bookHeaderRecords:
name = headerRecord
if name != "dkey" :
ext = '.dat'
if name == 'img' : ext = '.jpg'
if name == 'color' : ext = '.jpg'
for index in range (0,len(bookHeaderRecords[name])) :
fnum = "%04d" % index
fname = name + fnum + ext
destdir = payload
if name == 'img':
destdir = os.path.join(payload,'img')
if name == 'color':
destdir = os.path.join(payload,'color_img')
if name == 'page':
destdir = os.path.join(payload,'page')
if name == 'glyphs':
destdir = os.path.join(payload,'glyphs')
outputFile = os.path.join(destdir,fname)
file(outputFile, 'wb').write(getBookPayloadRecord(name, index))
# Create decrypted book
#
def createDecryptedBook(outdir):
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)
createDecryptedPayload(outdir)
#
# 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("\ncmbtc_dump_linux.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 Dumps the unencrypted book as files to outdir")
print("-o Output directory to save book files to")
print("-v Verbose (can be used several times)")
#
# Main
#
def main(argv=sys.argv):
global bookMetadata
global bookKey
global bookFile
global command
progname = os.path.basename(argv[0])
verbose = 0
recordName = ""
recordIndex = 0
outdir = ""
PIDs = []
command = ""
# Preloads your Kindle pid from the top of the program.
PIDs.append(kindlePID)
try:
opts, args = getopt.getopt(sys.argv[1:], "vo:p:d")
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 =="-o":
if a == None :
raise CMBDTCFatal("Invalid parameter for -o")
outdir = a
if o =="-p":
PIDs.append(a)
if o =="-d":
setCommand("doit")
if command == "" :
raise CMBDTCFatal("No action supplied on command line")
#
# Open book and parse metadata
#
if len(args) == 1:
bookFile = openBook(args[0])
parseTopazHeader()
parseMetadata()
#
# 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.")
return 1
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 outdir != "" :
createDecryptedBook(outdir)
if verbose >0 :
print ("Decrypted book saved. Don't pirate!")
elif verbose > 0:
print("Output directory name was not supplied.")
return 1
return 0
if __name__ == '__main__':
sys.exit(main())

322
Topaz_Tools/lib/k4mutils.py Normal file
View File

@@ -0,0 +1,322 @@
# standlone set of Mac OSX specific routines needed for K4DeDRM
from __future__ import with_statement
import sys
import os
#Exception Handling
class K4MDrmException(Exception):
pass
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()
# 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 K4MDrmException('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])
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 K4MDrmException('AES 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 K4MDrmException('Failed to initialize AES key')
def decrypt(self, data):
out = create_string_buffer(len(data))
rv = AES_cbc_encrypt(data, out, len(data), self._keyctx, self.iv, 0)
if rv == 0:
raise K4MDrmException('AES decryption failed')
return out.raw
def keyivgen(self, passwd):
salt = '16743'
saltlen = 5
passlen = len(passwd)
iter = 0x3e8
keylen = 80
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, K4MDrmException):
pass
return LibCrypto
LibCrypto = _load_crypto()
#
# Utility Routines
#
# uses a sub process to get the Hard Drive Serial Number using ioreg
# returns with the first found serial number in that class
def GetVolumeSerialNumber():
cmdline = '/usr/sbin/ioreg -r -c AppleAHCIDiskDriver'
cmdline = cmdline.encode(sys.getfilesystemencoding())
p = Process(cmdline, shell=True, bufsize=1, stdin=None, stdout=PIPE, stderr=PIPE, close_fds=False)
poll = p.wait('wait')
results = p.read()
reslst = results.split('\n')
sernum = '9999999999'
cnt = len(reslst)
for j in xrange(cnt):
resline = reslst[j]
pp = resline.find('"Serial Number" = "')
if pp >= 0:
sernum = resline[pp+19:]
sernum = sernum[:-1]
sernum = sernum.lstrip()
break
return sernum
# uses unix env to get username instead of using sysctlbyname
def GetUserName():
username = os.getenv('USER')
return username
# Various character maps used to decrypt books. Probably supposed to act as obfuscation
charMap1 = "n5Pr6St7Uv8Wx9YzAb0Cd1Ef2Gh3Jk4M"
charMap2 = "ZB0bYyc1xDdW2wEV3Ff7KkPpL8UuGA4gz-Tme9Nn_tHh5SvXCsIiR6rJjQaqlOoM"
charMap3 = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"
charMap4 = "ABCDEFGHIJKLMNPQRSTUVWXYZ123456789"
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
import hashlib
def SHA256(message):
ctx = hashlib.sha256()
ctx.update(message)
return ctx.digest()
# implements an Pseudo Mac Version of Windows built-in Crypto routine
def CryptUnprotectData(encryptedData):
sp = GetVolumeSerialNumber() + '!@#' + GetUserName()
passwdData = encode(SHA256(sp),charMap1)
crp = LibCrypto()
key_iv = crp.keyivgen(passwdData)
key = key_iv[0:32]
iv = key_iv[32:48]
crp.set_decrypt_key(key,iv)
cleartext = crp.decrypt(encryptedData)
return cleartext
# Locate and open the .kindle-info file
def openKindleInfo(kInfoFile=None):
if kInfoFile == None:
home = os.getenv('HOME')
cmdline = 'find "' + home + '/Library/Application Support" -name ".kindle-info"'
cmdline = cmdline.encode(sys.getfilesystemencoding())
p1 = Process(cmdline, shell=True, bufsize=1, stdin=None, stdout=PIPE, stderr=PIPE, close_fds=False)
poll = p1.wait('wait')
results = p1.read()
reslst = results.split('\n')
kinfopath = 'NONE'
cnt = len(reslst)
for j in xrange(cnt):
resline = reslst[j]
pp = resline.find('.kindle-info')
if pp >= 0:
kinfopath = resline
break
if not os.path.exists(kinfopath):
raise K4MDrmException('Error: .kindle-info file can not be found')
return open(kinfopath,'r')
else:
return open(kInfoFile, 'r')

110
Topaz_Tools/lib/k4pcutils.py Normal file
View File

@@ -0,0 +1,110 @@
# K4PC Windows specific routines
from __future__ import with_statement
import sys, os
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 traceback
MAX_PATH = 255
kernel32 = windll.kernel32
advapi32 = windll.advapi32
crypt32 = windll.crypt32
#
# 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 DrmException(Exception):
pass
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 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 DrmException("Failed to Unprotect Data")
return string_at(outdata.pbData, outdata.cbData)
return CryptUnprotectData
CryptUnprotectData = CryptUnprotectData()
#
# Locate and open the Kindle.info file.
#
def openKindleInfo(kInfoFile=None):
if kInfoFile == None:
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')
else:
return open(kInfoFile, 'r')

93
Topaz_Tools/lib/topaz-changes.txt
View File

@@ -1,93 +0,0 @@
Changes in this Version
- bug fix to prevent problems with sample books
modified version of patch submitted by that-guy
Changes in 2.6
- fix for many additional version tags
- fixes to generate better links
- fixes to handle external links
- now handles new "marker" page .dat files
- improved special region handling
- properly handle class names with spaces
- handle default alignment for synthetic regions
Changes in 2.3
- fix for use with non-latin1 based systems (thank you Tedd)
- fixes for out of order tokens in xml
Changes in 2.2
- fix for minor bug in encode_Number from clark nova
- more fixes to handle paths with spaces in them
- updates to work better with the gui front end
Changes in 2.1
- extremely minor changes to support a gui frontend
- no changes to functionality
Changes in version 2.0
- gensvg.py now accepts two options
-x : output browseable XHTML+SVG pages (default)
-r : output raw SVG images (useful for later conversion to pdf)
- flatxml2html.py now understands page.groups of type graphic
and handles vertical regions as svg images
- genhtml.py now accepts an option
--fixed-image : which will force the conversion
of all fixed regions to svg images
- minor bug fixes and html conversion improvements
Changes in version 1.8
- gensvg.py now builds wonderful xhtml pages with embedded svg
that can be easily paged through as if reading a book!
(tested in Safari for Mac and Win and Firefox)
(requires javascript to be enabled)
- genhtml.py now REQUIRES that gensvg.py be run FIRST
this allows create of images on the fly from glyphs
- genhtml.py now automatically makes tables of words into svg
based images and will handle glyph based ornate first
letters of words
- cmbtc_dump_mac_linux.py has been renamed to be
cmbtc_dump_nonK4PC.py to make it clearer
when it needs to be used
Changes in version 1.7
- gensvg.py has been improved so that the glyphs render exactly (ClarkNova)
- gensvg.py has fixed a render order "bug" that allowed some images to cover or hide text. (ClarkNova)
- change generated html to use external stylesheet via a link to "style.css"
- add missing <title> tag
- make xhtml compliant doctype and minor changes to write correct xhtml
- make divs that act as anchors be hidden visually and to take up 0 height and 0 width to prevent any impact on layout
Changes in version 1.6
- support for books whose paragraphs have no styles
- support to run cmbtc_dump on Linux and Mac OSX provided you know your PID of your ipod or standalone Kindle
(contributed by DiapDealer)
Changes in version 1.5
- completely reworked generation of styles to use actual page heights and widths
- added new script getpagedim.py to support the above
- style names with underscores in them are now properly paired with their base class
- fixed hanging indents that did not ever set a left margin
- added support for a number of not previously known region types
- added support for a previously unknown snippet - <empty></empty>
- corrected a bug that caused unknown regions to abort the program
- added code to make the handling of unknown regions better in general
- corrected a bug that caused the last link on a page to be missing (if it was the last thing on the page)
Changes in version 1.3
- font generation by gensvg.py is now greatly improved with support for contour points added
- support for more region types
- support for inline images in paragraphs or text fields (ie. initial graphics for the first letter of a word)
- greatly improved dtd information used for the xml to prevent parsing mistakes
Version 1.0
- initial release

20
Topaz_Tools/lib/topaz-readme.txt
View File

@@ -19,8 +19,7 @@ Here are the steps:
1. Unzip the topazscripts.zip file to get the full set of python scripts.
The files you should have after unzipping are:
cmbtc_dump.py - (author: cmbtc) unencrypts and dumps sections into separate files for Kindle for PC
cmbtc_dump_nonK4PC.py - (author - DiapDealer) for use with standalone Kindle and ipod/iphone topaz books
cmbtc_dump.py - (author: cmbtc) unencrypts and dumps sections into separate files for Kindle for PC and Mac
decode_meta.py - converts metadata0000.dat to make it available
convert2xml.py - converts page*.dat, other*.dat, and glyphs*.dat files to pseudo xml descriptions
flatxml2html.py - converts a "flattened" xml description to html using the ocrtext
@@ -29,6 +28,9 @@ getpagedim.py - reads page0000.dat to get the book height and width parameters
genxml.py - main program to convert everything to xml
genhtml.py - main program to generate "book.html"
gensvg.py - (author: clarknova) main program to create an xhmtl page with embedded svg graphics
k4mutils.py - Mac OSX support routines for cmbtc_dump.py
k4pcutils.py - Windows support routines for cmbtc_dump.py
Please note, these scripts all import code from each other so please
@@ -42,18 +44,15 @@ of its contents as files
All Thanks go to CMBTC who broke the DRM for Topaz - without it nothing else
would be possible
If you purchased the book for Kindle For PC, you must do the following:
If you purchased the book for Kindle for PC or Kindle for Mac, you must do the following:
cmbtc_dump.py -d -o TARGETDIR [-p pid] YOURTOPAZBOOKNAMEHERE
However, if you purchased the book for a standalone Kindle or ipod/iphone
If you purchased the book for a standalone Kindle 1 or ipod/iphone/ipad
and you know your pid (at least the first 8 characters) then you should
instead do the following
cmbtc_dump_nonK4PC.py -d -o TARGETDIR -p 12345678 YOURTOPAZBOOKNAMEHERE
where 12345678 should be replaced by the first 8 characters of your PID
add that using -p 12345678 switch as indicated above, replacing the
12345678 with the 8 characters of your pid
This should create a directory called "TARGETDIR" in your current directory.
@@ -64,7 +63,8 @@ other0000.dat - information used to create a style sheet
dict0000.dat - dictionary of words used to build page descriptions
page - directory filled with page*.dat files
glyphs - directory filled with glyphs*.dat files
img - directory filled with images
color_img - directory used for color images
3. REQUIRED: Create xhtml page descriptions with embedded svg
that show the exact representation of each page as an image