tools v6.0.8

DeDRM_Macintosh_Application/DeDRM.app/Contents/Info.plist

@@ -24,19 +24,19 @@
 	<key>CFBundleExecutable</key>
 	<string>droplet</string>
 	<key>CFBundleGetInfoString</key>
-	<string>DeDRM AppleScript 6.0.7. Written 2010–2013 by Apprentice Alf and others.</string>
+	<string>DeDRM AppleScript 6.0.8. Written 2010–2013 by Apprentice Alf and others.</string>
 	<key>CFBundleIconFile</key>
 	<string>DeDRM</string>
 	<key>CFBundleIdentifier</key>
 	<string>com.apple.ScriptEditor.id.707CCCD5-0C6C-4BEB-B67C-B6E866ADE85A</string>
 	<key>CFBundleInfoDictionaryVersion</key>
-	<string>6.0.7</string>
+	<string>6.0.8</string>
 	<key>CFBundleName</key>
 	<string>DeDRM</string>
 	<key>CFBundlePackageType</key>
 	<string>APPL</string>
 	<key>CFBundleShortVersionString</key>
-	<string>6.0.7</string>
+	<string>6.0.8</string>
 	<key>CFBundleSignature</key>
 	<string>dplt</string>
 	<key>LSRequiresCarbon</key>

DeDRM_Macintosh_Application/DeDRM.app/Contents/Resources/__init__.py

@@ -31,14 +31,17 @@ __docformat__ = 'restructuredtext en'
 #   6.0.3 - Fixes for Kindle for Mac and Windows non-ascii user names
 #   6.0.4 - Fixes for stand-alone scripts and applications
 #           and pdb files in plugin and initial conversion of prefs.
+#   6.0.5 - Fix a key issue
 #   6.0.6 - Fix up an incorrect function call
+#   6.0.7 - Error handling for incomplete PDF metadata
+#   6.0.8 - Fixes a Wine key issue and topaz support
 
 """
 Decrypt DRMed ebooks.
 """
 
 PLUGIN_NAME = u"DeDRM"
-PLUGIN_VERSION_TUPLE = (6, 0, 7)
+PLUGIN_VERSION_TUPLE = (6, 0, 8)
 PLUGIN_VERSION = u".".join([unicode(str(x)) for x in PLUGIN_VERSION_TUPLE])
 # Include an html helpfile in the plugin's zipfile with the following name.
 RESOURCE_NAME = PLUGIN_NAME + '_Help.htm'
@@ -313,7 +316,7 @@ class DeDRM(FileTypePlugin):
                     from wineutils import WineGetKeys
 
                     scriptpath = os.path.join(self.alfdir,u"adobekey.py")
-                    defaultkeys = self.WineGetKeys(scriptpath, u".der",dedrmprefs['adobewineprefix'])
+                    defaultkeys = WineGetKeys(scriptpath, u".der",dedrmprefs['adobewineprefix'])
                 except:
                     pass
 
@@ -391,7 +394,7 @@ class DeDRM(FileTypePlugin):
                     from wineutils import WineGetKeys
 
                     scriptpath = os.path.join(self.alfdir,u"kindlekey.py")
-                    defaultkeys = self.WineGetKeys(scriptpath, u".k4i",dedrmprefs['kindlewineprefix'])
+                    defaultkeys = WineGetKeys(scriptpath, u".k4i",dedrmprefs['kindlewineprefix'])
             except:
                 pass
 

DeDRM_Macintosh_Application/DeDRM.app/Contents/Resources/android.py

@@ -0,0 +1,157 @@
+#!/usr/bin/env python
+#fileencoding: utf-8
+
+import os
+import sys
+import zlib
+import tarfile
+from hashlib import md5
+from cStringIO import StringIO
+from binascii import a2b_hex, b2a_hex
+
+STORAGE = 'AmazonSecureStorage.xml'
+
+class AndroidObfuscation(object):
+    '''AndroidObfuscation
+    For the key, it's written in java, and run in android dalvikvm
+    '''
+
+    key = a2b_hex('0176e04c9408b1702d90be333fd53523')
+
+    def encrypt(self, plaintext):
+        cipher = self._get_cipher()
+        padding = len(self.key) - len(plaintext) % len(self.key)
+        plaintext += chr(padding) * padding
+        return b2a_hex(cipher.encrypt(plaintext))
+
+    def decrypt(self, ciphertext):
+        cipher = self._get_cipher()
+        plaintext = cipher.decrypt(a2b_hex(ciphertext))
+        return plaintext[:-ord(plaintext[-1])]
+
+    def _get_cipher(self):
+        try:
+            from Crypto.Cipher import AES
+            return AES.new(self.key)
+        except ImportError:
+            from aescbc import AES, noPadding
+            return AES(self.key, padding=noPadding())
+
+class AndroidObfuscationV2(AndroidObfuscation):
+    '''AndroidObfuscationV2
+    '''
+
+    count = 503
+    password = 'Thomsun was here!'
+
+    def __init__(self, salt):
+        key = self.password + salt
+        for _ in range(self.count):
+            key = md5(key).digest()
+        self.key = key[:8]
+        self.iv = key[8:16]
+
+    def _get_cipher(self):
+        try :
+            from Crypto.Cipher import DES
+            return DES.new(self.key, DES.MODE_CBC, self.iv)
+        except ImportError:
+            from python_des import Des, CBC
+            return Des(self.key, CBC, self.iv)
+
+def parse_preference(path):
+    ''' parse android's shared preference xml '''
+    storage = {}
+    read = open(path)
+    for line in read:
+        line = line.strip()
+        # <string name="key">value</string>
+        if line.startswith('<string name="'):
+            index = line.find('"', 14)
+            key = line[14:index]
+            value = line[index+2:-9]
+            storage[key] = value
+    read.close()
+    return storage
+
+def get_serials(path=None):
+    ''' get serials from android's shared preference xml '''
+    if path is None:
+        if not os.path.isfile(STORAGE):
+            if os.path.isfile("backup.ab"):
+                get_storage()
+            else:
+                return []
+        path = STORAGE
+
+    if not os.path.isfile(path):
+        return []
+
+    storage = parse_preference(path)
+    salt = storage.get('AmazonSaltKey')
+    if salt and len(salt) == 16:
+        sys.stdout.write('Using AndroidObfuscationV2\n')
+        obfuscation = AndroidObfuscationV2(a2b_hex(salt))
+    else:
+        sys.stdout.write('Using AndroidObfuscation\n')
+        obfuscation = AndroidObfuscation()
+
+    def get_value(key):
+        encrypted_key = obfuscation.encrypt(key)
+        encrypted_value = storage.get(encrypted_key)
+        if encrypted_value:
+            return obfuscation.decrypt(encrypted_value)
+        return ''
+
+    # also see getK4Pids in kgenpids.py
+    try:
+        dsnid = get_value('DsnId')
+    except:
+        sys.stderr.write('cannot get DsnId\n')
+        return []
+
+    try:
+        tokens = set(get_value('kindle.account.tokens').split(','))
+    except:
+        return [dsnid]
+
+    serials = []
+    for token in tokens:
+        if token:
+            serials.append('%s%s' % (dsnid, token))
+    serials.append(dsnid)
+    for token in tokens:
+        if token:
+            serials.append(token)
+    return serials
+
+def get_storage(path='backup.ab'):
+    '''get AmazonSecureStorage.xml from android backup.ab
+    backup.ab can be get using adb command:
+    shell> adb backup com.amazon.kindle
+    '''
+    output = None
+    read = open(path, 'rb')
+    head = read.read(24)
+    if head == 'ANDROID BACKUP\n1\n1\nnone\n':
+        output = StringIO(zlib.decompress(read.read()))
+    read.close()
+
+    if not output:
+        return False
+
+    tar = tarfile.open(fileobj=output)
+    for member in tar.getmembers():
+        if member.name.strip().endswith(STORAGE):
+            write = open(STORAGE, 'w')
+            write.write(tar.extractfile(member).read())
+            write.close()
+            break
+
+    return True
+
+__all__ = [ 'get_storage', 'get_serials', 'parse_preference',
+            'AndroidObfuscation', 'AndroidObfuscationV2', 'STORAGE']
+
+if __name__ == '__main__':
+    print get_serials()

DeDRM_Macintosh_Application/DeDRM.app/Contents/Resources/android_readme.txt

@@ -0,0 +1,6 @@
+1.1 get AmazonSecureStorage.xml from /data/data/com.amazon.kindle/shared_prefs/AmazonSecureStorage.xml
+
+1.2 on android 4.0+, run `adb backup com.amazon.kindle` from PC will get backup.ab
+    now android.py can convert backup.ab to AmazonSecureStorage.xml
+
+2. run `k4mobidedrm.py -a AmazonSecureStorage.xml <infile> <outdir>'

DeDRM_Macintosh_Application/DeDRM.app/Contents/Resources/flatxml2html.py

@@ -458,7 +458,11 @@ class DocParser(object):
             (wtype, num) = pdesc[j]
 
             if wtype == 'ocr' :
+                try:
                     word = self.ocrtext[num]
+                except:
+                    word = ""
+
                 sep = ' '
 
                 if handle_links:

DeDRM_Macintosh_Application/DeDRM.app/Contents/Resources/k4mobidedrm.py

@@ -80,10 +80,12 @@ if inCalibre:
     from calibre_plugins.dedrm import mobidedrm
     from calibre_plugins.dedrm import topazextract
     from calibre_plugins.dedrm import kgenpids
+    from calibre_plugins.dedrm import android
 else:
     import mobidedrm
     import topazextract
     import kgenpids
+    import android
 
 # Wrap a stream so that output gets flushed immediately
 # and also make sure that any unicode strings get
@@ -273,7 +275,7 @@ def decryptBook(infile, outdir, kDatabaseFiles, serials, pids):
 def usage(progname):
     print u"Removes DRM protection from Mobipocket, Amazon KF8, Amazon Print Replica and Amazon Topaz ebooks"
     print u"Usage:"
-    print u"    {0} [-k <kindle.k4i>] [-p <comma separated PIDs>] [-s <comma separated Kindle serial numbers>] <infile> <outdir>".format(progname)
+    print u"    {0} [-k <kindle.k4i>] [-p <comma separated PIDs>] [-s <comma separated Kindle serial numbers>] [ -a <AmazonSecureStorage.xml> ] <infile> <outdir>".format(progname)
 
 #
 # Main
@@ -284,7 +286,7 @@ def cli_main():
     print u"K4MobiDeDrm v{0}.\nCopyright © 2008-2013 The Dark Reverser et al.".format(__version__)
 
     try:
-        opts, args = getopt.getopt(argv[1:], "k:p:s:")
+        opts, args = getopt.getopt(argv[1:], "k:p:s:a:")
     except getopt.GetoptError, err:
         print u"Error in options or arguments: {0}".format(err.args[0])
         usage(progname)
@@ -312,6 +314,11 @@ def cli_main():
             if a == None :
                 raise DrmException("Invalid parameter for -s")
             serials = a.split(',')
+        if o == '-a':
+            if a == None:
+                continue
+            serials.extend(android.get_serials(a))
+    serials.extend(android.get_serials())
 
     # try with built in Kindle Info files if not on Linux
     k4 = not sys.platform.startswith('linux')

DeDRM_Macintosh_Application/DeDRM.app/Contents/Resources/kindlekey.py

@@ -19,6 +19,7 @@ from __future__ import with_statement
 #  1.6   - Fixed a problem getting the disk serial numbers
 #  1.7   - Work if TkInter is missing
 #  1.8   - Fixes for Kindle for Mac, and non-ascii in Windows user names
+#  1.9   - Fixes for Unicode in Windows user names
 
 
 """
@@ -26,7 +27,7 @@ Retrieve Kindle for PC/Mac user key.
 """
 
 __license__ = 'GPL v3'
-__version__ = '1.8'
+__version__ = '1.9'
 
 import sys, os, re
 from struct import pack, unpack, unpack_from
@@ -907,18 +908,34 @@ if iswindows:
         return CryptUnprotectData
     CryptUnprotectData = CryptUnprotectData()
 
+    # Returns Environmental Variables that contain unicode
+    def getEnvironmentVariable(name):
+        import ctypes
+        name = unicode(name) # make sure string argument is unicode
+        n = ctypes.windll.kernel32.GetEnvironmentVariableW(name, None, 0)
+        if n == 0:
+            return None
+        buf = ctypes.create_unicode_buffer(u'\0'*n)
+        ctypes.windll.kernel32.GetEnvironmentVariableW(name, buf, n)
+        return buf.value
 
     # Locate all of the kindle-info style files and return as list
     def getKindleInfoFiles():
         kInfoFiles = []
         # 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
+        # or the python interface to the 32 vs 64 bit registry is broken
         path = ""
         if 'LOCALAPPDATA' in os.environ.keys():
-            path = os.environ['LOCALAPPDATA']
+			# Python 2.x does not return unicode env. Use Python 3.x
+            path = winreg.ExpandEnvironmentStrings(u"%LOCALAPPDATA%")
+            # this is just another alternative.
+            # path = getEnvironmentVariable('LOCALAPPDATA')
+            if not os.path.isdir(path):
+                path = ""
         else:
             # User Shell Folders show take precedent over Shell Folders if present
             try:
+                # this will still break
                 regkey = winreg.OpenKey(winreg.HKEY_CURRENT_USER, "Software\\Microsoft\\Windows\\CurrentVersion\\Explorer\\User Shell Folders\\")
                 path = winreg.QueryValueEx(regkey, 'Local AppData')[0]
                 if not os.path.isdir(path):
@@ -937,13 +954,14 @@ if iswindows:
         if path == "":
             print ('Could not find the folder in which to look for kinfoFiles.')
         else:
-            print('searching for kinfoFiles in ' + path)
+            # Probably not the best. To Fix (shouldn't ignore in encoding) or use utf-8
+            print(u'searching for kinfoFiles in ' + path.encode('ascii', 'ignore'))
 
             # look for (K4PC 1.9.0 and later) .kinf2011 file
             kinfopath = path +'\\Amazon\\Kindle\\storage\\.kinf2011'
             if os.path.isfile(kinfopath):
                 found = True
-                print('Found K4PC 1.9+ kinf2011 file: ' + kinfopath)
+                print('Found K4PC 1.9+ kinf2011 file: ' + kinfopath.encode('ascii','ignore'))
                 kInfoFiles.append(kinfopath)
 
             # look for (K4PC 1.6.0 and later) rainier.2.1.1.kinf file
@@ -1142,7 +1160,7 @@ if iswindows:
                 cleartext = CryptUnprotectData(encryptedValue, entropy, 1)
                 DB[keyname] = cleartext
 
-        if 'MazamaRandomNumber' in DB and 'kindle.account.tokens' in DB:
+        if 'kindle.account.tokens' in DB:
             print u"Decrypted key file using IDString '{0:s}' and UserName '{1:s}'".format(GetIDString(), GetUserName().decode("latin-1"))
             # store values used in decryption
             DB['IDString'] = GetIDString()
@@ -1758,7 +1776,7 @@ elif isosx:
                         break
             except:
                 pass
-        if 'MazamaRandomNumber' in DB and 'kindle.account.tokens' in DB:
+        if 'kindle.account.tokens' in DB:
             # store values used in decryption
             print u"Decrypted key file using IDString '{0:s}' and UserName '{1:s}'".format(IDString, GetUserName())
             DB['IDString'] = IDString

DeDRM_Macintosh_Application/DeDRM.app/Contents/Resources/mobidedrm.py

@@ -156,6 +156,8 @@ def PC1(key, src, decryption=True):
         return Pukall_Cipher().PC1(key,src,decryption)
     except NameError:
         pass
+    except TypeError:
+        pass
 
     # use slow python version, since Pukall_Cipher didn't load
     sum1 = 0;

DeDRM_Macintosh_Application/DeDRM.app/Contents/Resources/stylexml2css.py

@@ -178,7 +178,12 @@ class DocParser(object):
                             if val == "":
                                 val = 0
 
-                            if not ((attr == 'hang') and (int(val) == 0)) :
+                            if not ((attr == 'hang') and (int(val) == 0)):
+                                try:
+                                    f = float(val)
+                                except:
+                                    print "Warning: unrecognised val, ignoring"
+                                    val = 0
                                 pv = float(val)/scale
                                 cssargs[attr] = (self.attr_val_map[attr], pv)
                                 keep = True

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/DeDRM_app.pyw

@@ -12,7 +12,7 @@
 #   6.0.4 - Fix for other potential unicode problems
 #   6.0.5 - Fix typo
 
-__version__ = '6.0.7'
+__version__ = '6.0.8'
 
 import sys
 import os, os.path

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/__init__.py

@@ -31,14 +31,17 @@ __docformat__ = 'restructuredtext en'
 #   6.0.3 - Fixes for Kindle for Mac and Windows non-ascii user names
 #   6.0.4 - Fixes for stand-alone scripts and applications
 #           and pdb files in plugin and initial conversion of prefs.
+#   6.0.5 - Fix a key issue
 #   6.0.6 - Fix up an incorrect function call
+#   6.0.7 - Error handling for incomplete PDF metadata
+#   6.0.8 - Fixes a Wine key issue and topaz support
 
 """
 Decrypt DRMed ebooks.
 """
 
 PLUGIN_NAME = u"DeDRM"
-PLUGIN_VERSION_TUPLE = (6, 0, 7)
+PLUGIN_VERSION_TUPLE = (6, 0, 8)
 PLUGIN_VERSION = u".".join([unicode(str(x)) for x in PLUGIN_VERSION_TUPLE])
 # Include an html helpfile in the plugin's zipfile with the following name.
 RESOURCE_NAME = PLUGIN_NAME + '_Help.htm'
@@ -313,7 +316,7 @@ class DeDRM(FileTypePlugin):
                     from wineutils import WineGetKeys
 
                     scriptpath = os.path.join(self.alfdir,u"adobekey.py")
-                    defaultkeys = self.WineGetKeys(scriptpath, u".der",dedrmprefs['adobewineprefix'])
+                    defaultkeys = WineGetKeys(scriptpath, u".der",dedrmprefs['adobewineprefix'])
                 except:
                     pass
 
@@ -391,7 +394,7 @@ class DeDRM(FileTypePlugin):
                     from wineutils import WineGetKeys
 
                     scriptpath = os.path.join(self.alfdir,u"kindlekey.py")
-                    defaultkeys = self.WineGetKeys(scriptpath, u".k4i",dedrmprefs['kindlewineprefix'])
+                    defaultkeys = WineGetKeys(scriptpath, u".k4i",dedrmprefs['kindlewineprefix'])
             except:
                 pass
 

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/android.py

@@ -0,0 +1,157 @@
+#!/usr/bin/env python
+#fileencoding: utf-8
+
+import os
+import sys
+import zlib
+import tarfile
+from hashlib import md5
+from cStringIO import StringIO
+from binascii import a2b_hex, b2a_hex
+
+STORAGE = 'AmazonSecureStorage.xml'
+
+class AndroidObfuscation(object):
+    '''AndroidObfuscation
+    For the key, it's written in java, and run in android dalvikvm
+    '''
+
+    key = a2b_hex('0176e04c9408b1702d90be333fd53523')
+
+    def encrypt(self, plaintext):
+        cipher = self._get_cipher()
+        padding = len(self.key) - len(plaintext) % len(self.key)
+        plaintext += chr(padding) * padding
+        return b2a_hex(cipher.encrypt(plaintext))
+
+    def decrypt(self, ciphertext):
+        cipher = self._get_cipher()
+        plaintext = cipher.decrypt(a2b_hex(ciphertext))
+        return plaintext[:-ord(plaintext[-1])]
+
+    def _get_cipher(self):
+        try:
+            from Crypto.Cipher import AES
+            return AES.new(self.key)
+        except ImportError:
+            from aescbc import AES, noPadding
+            return AES(self.key, padding=noPadding())
+
+class AndroidObfuscationV2(AndroidObfuscation):
+    '''AndroidObfuscationV2
+    '''
+
+    count = 503
+    password = 'Thomsun was here!'
+
+    def __init__(self, salt):
+        key = self.password + salt
+        for _ in range(self.count):
+            key = md5(key).digest()
+        self.key = key[:8]
+        self.iv = key[8:16]
+
+    def _get_cipher(self):
+        try :
+            from Crypto.Cipher import DES
+            return DES.new(self.key, DES.MODE_CBC, self.iv)
+        except ImportError:
+            from python_des import Des, CBC
+            return Des(self.key, CBC, self.iv)
+
+def parse_preference(path):
+    ''' parse android's shared preference xml '''
+    storage = {}
+    read = open(path)
+    for line in read:
+        line = line.strip()
+        # <string name="key">value</string>
+        if line.startswith('<string name="'):
+            index = line.find('"', 14)
+            key = line[14:index]
+            value = line[index+2:-9]
+            storage[key] = value
+    read.close()
+    return storage
+
+def get_serials(path=None):
+    ''' get serials from android's shared preference xml '''
+    if path is None:
+        if not os.path.isfile(STORAGE):
+            if os.path.isfile("backup.ab"):
+                get_storage()
+            else:
+                return []
+        path = STORAGE
+
+    if not os.path.isfile(path):
+        return []
+
+    storage = parse_preference(path)
+    salt = storage.get('AmazonSaltKey')
+    if salt and len(salt) == 16:
+        sys.stdout.write('Using AndroidObfuscationV2\n')
+        obfuscation = AndroidObfuscationV2(a2b_hex(salt))
+    else:
+        sys.stdout.write('Using AndroidObfuscation\n')
+        obfuscation = AndroidObfuscation()
+
+    def get_value(key):
+        encrypted_key = obfuscation.encrypt(key)
+        encrypted_value = storage.get(encrypted_key)
+        if encrypted_value:
+            return obfuscation.decrypt(encrypted_value)
+        return ''
+
+    # also see getK4Pids in kgenpids.py
+    try:
+        dsnid = get_value('DsnId')
+    except:
+        sys.stderr.write('cannot get DsnId\n')
+        return []
+
+    try:
+        tokens = set(get_value('kindle.account.tokens').split(','))
+    except:
+        return [dsnid]
+
+    serials = []
+    for token in tokens:
+        if token:
+            serials.append('%s%s' % (dsnid, token))
+    serials.append(dsnid)
+    for token in tokens:
+        if token:
+            serials.append(token)
+    return serials
+
+def get_storage(path='backup.ab'):
+    '''get AmazonSecureStorage.xml from android backup.ab
+    backup.ab can be get using adb command:
+    shell> adb backup com.amazon.kindle
+    '''
+    output = None
+    read = open(path, 'rb')
+    head = read.read(24)
+    if head == 'ANDROID BACKUP\n1\n1\nnone\n':
+        output = StringIO(zlib.decompress(read.read()))
+    read.close()
+
+    if not output:
+        return False
+
+    tar = tarfile.open(fileobj=output)
+    for member in tar.getmembers():
+        if member.name.strip().endswith(STORAGE):
+            write = open(STORAGE, 'w')
+            write.write(tar.extractfile(member).read())
+            write.close()
+            break
+
+    return True
+
+__all__ = [ 'get_storage', 'get_serials', 'parse_preference',
+            'AndroidObfuscation', 'AndroidObfuscationV2', 'STORAGE']
+
+if __name__ == '__main__':
+    print get_serials()

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/android_readme.txt

@@ -0,0 +1,6 @@
+1.1 get AmazonSecureStorage.xml from /data/data/com.amazon.kindle/shared_prefs/AmazonSecureStorage.xml
+
+1.2 on android 4.0+, run `adb backup com.amazon.kindle` from PC will get backup.ab
+    now android.py can convert backup.ab to AmazonSecureStorage.xml
+
+2. run `k4mobidedrm.py -a AmazonSecureStorage.xml <infile> <outdir>'

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/flatxml2html.py

@@ -458,7 +458,11 @@ class DocParser(object):
             (wtype, num) = pdesc[j]
 
             if wtype == 'ocr' :
+                try:
                     word = self.ocrtext[num]
+                except:
+                    word = ""
+
                 sep = ' '
 
                 if handle_links:

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/k4mobidedrm.py

@@ -80,10 +80,12 @@ if inCalibre:
     from calibre_plugins.dedrm import mobidedrm
     from calibre_plugins.dedrm import topazextract
     from calibre_plugins.dedrm import kgenpids
+    from calibre_plugins.dedrm import android
 else:
     import mobidedrm
     import topazextract
     import kgenpids
+    import android
 
 # Wrap a stream so that output gets flushed immediately
 # and also make sure that any unicode strings get
@@ -273,7 +275,7 @@ def decryptBook(infile, outdir, kDatabaseFiles, serials, pids):
 def usage(progname):
     print u"Removes DRM protection from Mobipocket, Amazon KF8, Amazon Print Replica and Amazon Topaz ebooks"
     print u"Usage:"
-    print u"    {0} [-k <kindle.k4i>] [-p <comma separated PIDs>] [-s <comma separated Kindle serial numbers>] <infile> <outdir>".format(progname)
+    print u"    {0} [-k <kindle.k4i>] [-p <comma separated PIDs>] [-s <comma separated Kindle serial numbers>] [ -a <AmazonSecureStorage.xml> ] <infile> <outdir>".format(progname)
 
 #
 # Main
@@ -284,7 +286,7 @@ def cli_main():
     print u"K4MobiDeDrm v{0}.\nCopyright © 2008-2013 The Dark Reverser et al.".format(__version__)
 
     try:
-        opts, args = getopt.getopt(argv[1:], "k:p:s:")
+        opts, args = getopt.getopt(argv[1:], "k:p:s:a:")
     except getopt.GetoptError, err:
         print u"Error in options or arguments: {0}".format(err.args[0])
         usage(progname)
@@ -312,6 +314,11 @@ def cli_main():
             if a == None :
                 raise DrmException("Invalid parameter for -s")
             serials = a.split(',')
+        if o == '-a':
+            if a == None:
+                continue
+            serials.extend(android.get_serials(a))
+    serials.extend(android.get_serials())
 
     # try with built in Kindle Info files if not on Linux
     k4 = not sys.platform.startswith('linux')

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/kindlekey.py

@@ -19,6 +19,7 @@ from __future__ import with_statement
 #  1.6   - Fixed a problem getting the disk serial numbers
 #  1.7   - Work if TkInter is missing
 #  1.8   - Fixes for Kindle for Mac, and non-ascii in Windows user names
+#  1.9   - Fixes for Unicode in Windows user names
 
 
 """
@@ -26,7 +27,7 @@ Retrieve Kindle for PC/Mac user key.
 """
 
 __license__ = 'GPL v3'
-__version__ = '1.8'
+__version__ = '1.9'
 
 import sys, os, re
 from struct import pack, unpack, unpack_from
@@ -907,18 +908,34 @@ if iswindows:
         return CryptUnprotectData
     CryptUnprotectData = CryptUnprotectData()
 
+    # Returns Environmental Variables that contain unicode
+    def getEnvironmentVariable(name):
+        import ctypes
+        name = unicode(name) # make sure string argument is unicode
+        n = ctypes.windll.kernel32.GetEnvironmentVariableW(name, None, 0)
+        if n == 0:
+            return None
+        buf = ctypes.create_unicode_buffer(u'\0'*n)
+        ctypes.windll.kernel32.GetEnvironmentVariableW(name, buf, n)
+        return buf.value
 
     # Locate all of the kindle-info style files and return as list
     def getKindleInfoFiles():
         kInfoFiles = []
         # 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
+        # or the python interface to the 32 vs 64 bit registry is broken
         path = ""
         if 'LOCALAPPDATA' in os.environ.keys():
-            path = os.environ['LOCALAPPDATA']
+			# Python 2.x does not return unicode env. Use Python 3.x
+            path = winreg.ExpandEnvironmentStrings(u"%LOCALAPPDATA%")
+            # this is just another alternative.
+            # path = getEnvironmentVariable('LOCALAPPDATA')
+            if not os.path.isdir(path):
+                path = ""
         else:
             # User Shell Folders show take precedent over Shell Folders if present
             try:
+                # this will still break
                 regkey = winreg.OpenKey(winreg.HKEY_CURRENT_USER, "Software\\Microsoft\\Windows\\CurrentVersion\\Explorer\\User Shell Folders\\")
                 path = winreg.QueryValueEx(regkey, 'Local AppData')[0]
                 if not os.path.isdir(path):
@@ -937,13 +954,14 @@ if iswindows:
         if path == "":
             print ('Could not find the folder in which to look for kinfoFiles.')
         else:
-            print('searching for kinfoFiles in ' + path)
+            # Probably not the best. To Fix (shouldn't ignore in encoding) or use utf-8
+            print(u'searching for kinfoFiles in ' + path.encode('ascii', 'ignore'))
 
             # look for (K4PC 1.9.0 and later) .kinf2011 file
             kinfopath = path +'\\Amazon\\Kindle\\storage\\.kinf2011'
             if os.path.isfile(kinfopath):
                 found = True
-                print('Found K4PC 1.9+ kinf2011 file: ' + kinfopath)
+                print('Found K4PC 1.9+ kinf2011 file: ' + kinfopath.encode('ascii','ignore'))
                 kInfoFiles.append(kinfopath)
 
             # look for (K4PC 1.6.0 and later) rainier.2.1.1.kinf file
@@ -1142,7 +1160,7 @@ if iswindows:
                 cleartext = CryptUnprotectData(encryptedValue, entropy, 1)
                 DB[keyname] = cleartext
 
-        if 'MazamaRandomNumber' in DB and 'kindle.account.tokens' in DB:
+        if 'kindle.account.tokens' in DB:
             print u"Decrypted key file using IDString '{0:s}' and UserName '{1:s}'".format(GetIDString(), GetUserName().decode("latin-1"))
             # store values used in decryption
             DB['IDString'] = GetIDString()
@@ -1758,7 +1776,7 @@ elif isosx:
                         break
             except:
                 pass
-        if 'MazamaRandomNumber' in DB and 'kindle.account.tokens' in DB:
+        if 'kindle.account.tokens' in DB:
             # store values used in decryption
             print u"Decrypted key file using IDString '{0:s}' and UserName '{1:s}'".format(IDString, GetUserName())
             DB['IDString'] = IDString

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/mobidedrm.py

@@ -156,6 +156,8 @@ def PC1(key, src, decryption=True):
         return Pukall_Cipher().PC1(key,src,decryption)
     except NameError:
         pass
+    except TypeError:
+        pass
 
     # use slow python version, since Pukall_Cipher didn't load
     sum1 = 0;

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/stylexml2css.py

@@ -178,7 +178,12 @@ class DocParser(object):
                             if val == "":
                                 val = 0
 
-                            if not ((attr == 'hang') and (int(val) == 0)) :
+                            if not ((attr == 'hang') and (int(val) == 0)):
+                                try:
+                                    f = float(val)
+                                except:
+                                    print "Warning: unrecognised val, ignoring"
+                                    val = 0
                                 pv = float(val)/scale
                                 cssargs[attr] = (self.attr_val_map[attr], pv)
                                 keep = True

DeDRM_calibre_plugin/DeDRM_plugin/__init__.py

@@ -31,14 +31,17 @@ __docformat__ = 'restructuredtext en'
 #   6.0.3 - Fixes for Kindle for Mac and Windows non-ascii user names
 #   6.0.4 - Fixes for stand-alone scripts and applications
 #           and pdb files in plugin and initial conversion of prefs.
+#   6.0.5 - Fix a key issue
 #   6.0.6 - Fix up an incorrect function call
+#   6.0.7 - Error handling for incomplete PDF metadata
+#   6.0.8 - Fixes a Wine key issue and topaz support
 
 """
 Decrypt DRMed ebooks.
 """
 
 PLUGIN_NAME = u"DeDRM"
-PLUGIN_VERSION_TUPLE = (6, 0, 7)
+PLUGIN_VERSION_TUPLE = (6, 0, 8)
 PLUGIN_VERSION = u".".join([unicode(str(x)) for x in PLUGIN_VERSION_TUPLE])
 # Include an html helpfile in the plugin's zipfile with the following name.
 RESOURCE_NAME = PLUGIN_NAME + '_Help.htm'
@@ -313,7 +316,7 @@ class DeDRM(FileTypePlugin):
                     from wineutils import WineGetKeys
 
                     scriptpath = os.path.join(self.alfdir,u"adobekey.py")
-                    defaultkeys = self.WineGetKeys(scriptpath, u".der",dedrmprefs['adobewineprefix'])
+                    defaultkeys = WineGetKeys(scriptpath, u".der",dedrmprefs['adobewineprefix'])
                 except:
                     pass
 
@@ -391,7 +394,7 @@ class DeDRM(FileTypePlugin):
                     from wineutils import WineGetKeys
 
                     scriptpath = os.path.join(self.alfdir,u"kindlekey.py")
-                    defaultkeys = self.WineGetKeys(scriptpath, u".k4i",dedrmprefs['kindlewineprefix'])
+                    defaultkeys = WineGetKeys(scriptpath, u".k4i",dedrmprefs['kindlewineprefix'])
             except:
                 pass
 

DeDRM_calibre_plugin/DeDRM_plugin/android.py

@@ -0,0 +1,157 @@
+#!/usr/bin/env python
+#fileencoding: utf-8
+
+import os
+import sys
+import zlib
+import tarfile
+from hashlib import md5
+from cStringIO import StringIO
+from binascii import a2b_hex, b2a_hex
+
+STORAGE = 'AmazonSecureStorage.xml'
+
+class AndroidObfuscation(object):
+    '''AndroidObfuscation
+    For the key, it's written in java, and run in android dalvikvm
+    '''
+
+    key = a2b_hex('0176e04c9408b1702d90be333fd53523')
+
+    def encrypt(self, plaintext):
+        cipher = self._get_cipher()
+        padding = len(self.key) - len(plaintext) % len(self.key)
+        plaintext += chr(padding) * padding
+        return b2a_hex(cipher.encrypt(plaintext))
+
+    def decrypt(self, ciphertext):
+        cipher = self._get_cipher()
+        plaintext = cipher.decrypt(a2b_hex(ciphertext))
+        return plaintext[:-ord(plaintext[-1])]
+
+    def _get_cipher(self):
+        try:
+            from Crypto.Cipher import AES
+            return AES.new(self.key)
+        except ImportError:
+            from aescbc import AES, noPadding
+            return AES(self.key, padding=noPadding())
+
+class AndroidObfuscationV2(AndroidObfuscation):
+    '''AndroidObfuscationV2
+    '''
+
+    count = 503
+    password = 'Thomsun was here!'
+
+    def __init__(self, salt):
+        key = self.password + salt
+        for _ in range(self.count):
+            key = md5(key).digest()
+        self.key = key[:8]
+        self.iv = key[8:16]
+
+    def _get_cipher(self):
+        try :
+            from Crypto.Cipher import DES
+            return DES.new(self.key, DES.MODE_CBC, self.iv)
+        except ImportError:
+            from python_des import Des, CBC
+            return Des(self.key, CBC, self.iv)
+
+def parse_preference(path):
+    ''' parse android's shared preference xml '''
+    storage = {}
+    read = open(path)
+    for line in read:
+        line = line.strip()
+        # <string name="key">value</string>
+        if line.startswith('<string name="'):
+            index = line.find('"', 14)
+            key = line[14:index]
+            value = line[index+2:-9]
+            storage[key] = value
+    read.close()
+    return storage
+
+def get_serials(path=None):
+    ''' get serials from android's shared preference xml '''
+    if path is None:
+        if not os.path.isfile(STORAGE):
+            if os.path.isfile("backup.ab"):
+                get_storage()
+            else:
+                return []
+        path = STORAGE
+
+    if not os.path.isfile(path):
+        return []
+
+    storage = parse_preference(path)
+    salt = storage.get('AmazonSaltKey')
+    if salt and len(salt) == 16:
+        sys.stdout.write('Using AndroidObfuscationV2\n')
+        obfuscation = AndroidObfuscationV2(a2b_hex(salt))
+    else:
+        sys.stdout.write('Using AndroidObfuscation\n')
+        obfuscation = AndroidObfuscation()
+
+    def get_value(key):
+        encrypted_key = obfuscation.encrypt(key)
+        encrypted_value = storage.get(encrypted_key)
+        if encrypted_value:
+            return obfuscation.decrypt(encrypted_value)
+        return ''
+
+    # also see getK4Pids in kgenpids.py
+    try:
+        dsnid = get_value('DsnId')
+    except:
+        sys.stderr.write('cannot get DsnId\n')
+        return []
+
+    try:
+        tokens = set(get_value('kindle.account.tokens').split(','))
+    except:
+        return [dsnid]
+
+    serials = []
+    for token in tokens:
+        if token:
+            serials.append('%s%s' % (dsnid, token))
+    serials.append(dsnid)
+    for token in tokens:
+        if token:
+            serials.append(token)
+    return serials
+
+def get_storage(path='backup.ab'):
+    '''get AmazonSecureStorage.xml from android backup.ab
+    backup.ab can be get using adb command:
+    shell> adb backup com.amazon.kindle
+    '''
+    output = None
+    read = open(path, 'rb')
+    head = read.read(24)
+    if head == 'ANDROID BACKUP\n1\n1\nnone\n':
+        output = StringIO(zlib.decompress(read.read()))
+    read.close()
+
+    if not output:
+        return False
+
+    tar = tarfile.open(fileobj=output)
+    for member in tar.getmembers():
+        if member.name.strip().endswith(STORAGE):
+            write = open(STORAGE, 'w')
+            write.write(tar.extractfile(member).read())
+            write.close()
+            break
+
+    return True
+
+__all__ = [ 'get_storage', 'get_serials', 'parse_preference',
+            'AndroidObfuscation', 'AndroidObfuscationV2', 'STORAGE']
+
+if __name__ == '__main__':
+    print get_serials()

DeDRM_calibre_plugin/DeDRM_plugin/android_readme.txt

@@ -0,0 +1,6 @@
+1.1 get AmazonSecureStorage.xml from /data/data/com.amazon.kindle/shared_prefs/AmazonSecureStorage.xml
+
+1.2 on android 4.0+, run `adb backup com.amazon.kindle` from PC will get backup.ab
+    now android.py can convert backup.ab to AmazonSecureStorage.xml
+
+2. run `k4mobidedrm.py -a AmazonSecureStorage.xml <infile> <outdir>'

DeDRM_calibre_plugin/DeDRM_plugin/dialogs.py

@@ -1,45 +0,0 @@
-#!/usr/bin/env python
-# -*- coding: utf-8 -*-
-
-# base64.py, version 1.0
-# Copyright © 2010 Apprentice Alf
-
-# Released under the terms of the GNU General Public Licence, version 3 or
-# later.  <http://www.gnu.org/licenses/>
-
-# Revision history:
-#   1 - Initial release. To allow Applescript to do base64 encoding
-
-"""
-Provide base64 encoding.
-"""
-
-from __future__ import with_statement
-
-__license__ = 'GPL v3'
-
-import sys
-import os
-import base64
-
-def usage(progname):
-    print "Applies base64 encoding to the supplied file, sending to standard output"
-    print "Usage:"
-    print "    %s <infile>" % progname
-
-def cli_main(argv=sys.argv):
-    progname = os.path.basename(argv[0])
-
-    if len(argv)<2:
-        usage(progname)
-        sys.exit(2)
-
-    keypath = argv[1]
-    with open(keypath, 'rb') as f:
-        keyder = f.read()
-        print keyder.encode('base64')
-    return 0
-
-
-if __name__ == '__main__':
-    sys.exit(cli_main())

DeDRM_calibre_plugin/DeDRM_plugin/encodebase64.py

@@ -1,208 +1,45 @@
-#!/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 drmcheck
-#  1.00 - Initial version, with code from various other scripts
-#  1.01 - Moved authorship announcement to usage section.
-#
-# Changelog epubtest
-#  1.00 - Cut to epubtest.py, testing ePub files only by Apprentice Alf
-#  1.01 - Added routine for use by Windows DeDRM
-#
-# Written in 2011 by Paul Durrant
-# Released with unlicense. See http://unlicense.org/
-#
-#############################################################################
-#
-# This is free and unencumbered software released into the public domain.
-#
-# Anyone is free to copy, modify, publish, use, compile, sell, or
-# distribute this software, either in source code form or as a compiled
-# binary, for any purpose, commercial or non-commercial, and by any
-# means.
-#
-# In jurisdictions that recognize copyright laws, the author or authors
-# of this software dedicate any and all copyright interest in the
-# software to the public domain. We make this dedication for the benefit
-# of the public at large and to the detriment of our heirs and
-# successors. We intend this dedication to be an overt act of
-# relinquishment in perpetuity of all present and future rights to this
-# software under copyright law.
-#
-# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
-# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
-# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
-# IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR
-# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
-# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
-# OTHER DEALINGS IN THE SOFTWARE.
-#
-#############################################################################
-#
-# It's still polite to give attribution if you do reuse this code.
-#
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+# base64.py, version 1.0
+# Copyright © 2010 Apprentice Alf
+
+# Released under the terms of the GNU General Public Licence, version 3 or
+# later.  <http://www.gnu.org/licenses/>
+
+# Revision history:
+#   1 - Initial release. To allow Applescript to do base64 encoding
+
+"""
+Provide base64 encoding.
+"""
 
 from __future__ import with_statement
 
-__version__ = '1.01'
+__license__ = 'GPL v3'
 
-import sys, struct, os
-import zlib
-import zipfile
-import xml.etree.ElementTree as etree
+import sys
+import os
+import base64
 
-NSMAP = {'adept': 'http://ns.adobe.com/adept',
-         'enc': 'http://www.w3.org/2001/04/xmlenc#'}
+def usage(progname):
+    print "Applies base64 encoding to the supplied file, sending to standard output"
+    print "Usage:"
+    print "    %s <infile>" % progname
 
-# Wrap a stream so that output gets flushed immediately
-# and also make sure that any unicode strings get
-# encoded using "replace" before writing them.
-class SafeUnbuffered:
-    def __init__(self, stream):
-        self.stream = stream
-        self.encoding = stream.encoding
-        if self.encoding == None:
-            self.encoding = "utf-8"
-    def write(self, data):
-        if isinstance(data,unicode):
-            data = data.encode(self.encoding,"replace")
-        self.stream.write(data)
-        self.stream.flush()
-    def __getattr__(self, attr):
-        return getattr(self.stream, attr)
+def cli_main(argv=sys.argv):
+    progname = os.path.basename(argv[0])
 
-try:
-    from calibre.constants import iswindows, isosx
-except:
-    iswindows = sys.platform.startswith('win')
-    isosx = sys.platform.startswith('darwin')
+    if len(argv)<2:
+        usage(progname)
+        sys.exit(2)
 
-def unicode_argv():
-    if iswindows:
-        # Uses shell32.GetCommandLineArgvW to get sys.argv as a list of Unicode
-        # strings.
-
-        # Versions 2.x of Python don't support Unicode in sys.argv on
-        # Windows, with the underlying Windows API instead replacing multi-byte
-        # characters with '?'.  So use shell32.GetCommandLineArgvW to get sys.argv
-        # as a list of Unicode strings and encode them as utf-8
-
-        from ctypes import POINTER, byref, cdll, c_int, windll
-        from ctypes.wintypes import LPCWSTR, LPWSTR
-
-        GetCommandLineW = cdll.kernel32.GetCommandLineW
-        GetCommandLineW.argtypes = []
-        GetCommandLineW.restype = LPCWSTR
-
-        CommandLineToArgvW = windll.shell32.CommandLineToArgvW
-        CommandLineToArgvW.argtypes = [LPCWSTR, POINTER(c_int)]
-        CommandLineToArgvW.restype = POINTER(LPWSTR)
-
-        cmd = GetCommandLineW()
-        argc = c_int(0)
-        argv = CommandLineToArgvW(cmd, byref(argc))
-        if argc.value > 0:
-            # Remove Python executable and commands if present
-            start = argc.value - len(sys.argv)
-            return [argv[i] for i in
-                    xrange(start, argc.value)]
-        # if we don't have any arguments at all, just pass back script name
-        # this should never happen
-        return [u"epubtest.py"]
-    else:
-        argvencoding = sys.stdin.encoding
-        if argvencoding == None:
-            argvencoding = "utf-8"
-        return [arg if (type(arg) == unicode) else unicode(arg,argvencoding) for arg in sys.argv]
-
-_FILENAME_LEN_OFFSET = 26
-_EXTRA_LEN_OFFSET = 28
-_FILENAME_OFFSET = 30
-_MAX_SIZE = 64 * 1024
-
-
-def uncompress(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(file, zi):
-    # get file name length and exta data length to find start of file data
-    local_header_offset = zi.header_offset
-
-    file.seek(local_header_offset + _FILENAME_LEN_OFFSET)
-    leninfo = file.read(2)
-    local_name_length, = struct.unpack('<H', leninfo)
-
-    file.seek(local_header_offset + _EXTRA_LEN_OFFSET)
-    exinfo = file.read(2)
-    extra_field_length, = struct.unpack('<H', exinfo)
-
-    file.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 = file.read(zi.file_size)
-
-    # if compressed we must decompress it using zlib
-    if zi.compress_type == zipfile.ZIP_DEFLATED:
-        cmpdata = file.read(zi.compress_size)
-        data = uncompress(cmpdata)
-
-    return data
-
-def encryption(infile):
-    # returns encryption: one of Unencrypted, Adobe, B&N and Unknown
-    encryption = "Unknown"
-    try:
-        with open(infile,'rb') as infileobject:
-            bookdata = infileobject.read(58)
-            # Check for Zip
-            if bookdata[0:0+2] == "PK":
-                foundrights = False
-                foundencryption = False
-                inzip = zipfile.ZipFile(infile,'r')
-                namelist = set(inzip.namelist())
-                if 'META-INF/rights.xml' not in namelist or 'META-INF/encryption.xml' not in namelist:
-                    encryption = "Unencrypted"
-                else:
-                    rights = etree.fromstring(inzip.read('META-INF/rights.xml'))
-                    adept = lambda tag: '{%s}%s' % (NSMAP['adept'], tag)
-                    expr = './/%s' % (adept('encryptedKey'),)
-                    bookkey = ''.join(rights.findtext(expr))
-                    if len(bookkey) == 172:
-                        encryption = "Adobe"
-                    elif len(bookkey) == 64:
-                        encryption = "B&N"
-                    else:
-                        encryption = "Unknown"
-    except:
-        traceback.print_exc()
-    return encryption
-
-def main():
-    argv=unicode_argv()
-    print encryption(argv[1])
+    keypath = argv[1]
+    with open(keypath, 'rb') as f:
+        keyder = f.read()
+        print keyder.encode('base64')
     return 0
 
-if __name__ == "__main__":
-    sys.stdout=SafeUnbuffered(sys.stdout)
-    sys.stderr=SafeUnbuffered(sys.stderr)
-    sys.exit(main())
+
+if __name__ == '__main__':
+    sys.exit(cli_main())

DeDRM_calibre_plugin/DeDRM_plugin/epubtest.py

@@ -1,82 +1,60 @@
-#!/usr/bin/env python
-# -*- coding: utf-8 -*-
-
-# erdr2pml.py
-# Copyright © 2008 The Dark Reverser
+#!/usr/bin/python
 #
-# Modified 2008–2012 by some_updates, DiapDealer and Apprentice Alf
-
 # 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
+# Changelog drmcheck
+#  1.00 - Initial version, with code from various other scripts
+#  1.01 - Moved authorship announcement to usage section.
+#
+# Changelog epubtest
+#  1.00 - Cut to epubtest.py, testing ePub files only by Apprentice Alf
+#  1.01 - Added routine for use by Windows DeDRM
+#
+# Written in 2011 by Paul Durrant
+# Released with unlicense. See http://unlicense.org/
+#
+#############################################################################
+#
+# This is free and unencumbered software released into the public domain.
+#
+# Anyone is free to copy, modify, publish, use, compile, sell, or
+# distribute this software, either in source code form or as a compiled
+# binary, for any purpose, commercial or non-commercial, and by any
+# means.
+#
+# In jurisdictions that recognize copyright laws, the author or authors
+# of this software dedicate any and all copyright interest in the
+# software to the public domain. We make this dedication for the benefit
+# of the public at large and to the detriment of our heirs and
+# successors. We intend this dedication to be an overt act of
+# relinquishment in perpetuity of all present and future rights to this
+# software under copyright law.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+# IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY CLAIM, DAMAGES OR
+# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+# OTHER DEALINGS IN THE SOFTWARE.
+#
+#############################################################################
+#
+# It's still polite to give attribution if you do reuse this code.
 #
-#  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?)
-#  0.22 - Unicode and plugin support, different image folders for PMLZ and source
-#  0.23 - moved unicode_argv call inside main for Windows DeDRM compatibility
 
-__version__='0.23'
+from __future__ import with_statement
 
-import sys, re
-import struct, binascii, getopt, zlib, os, os.path, urllib, tempfile, traceback
+__version__ = '1.01'
 
-if 'calibre' in sys.modules:
-    inCalibre = True
-else:
-    inCalibre = False
+import sys, struct, os
+import zlib
+import zipfile
+import xml.etree.ElementTree as etree
+
+NSMAP = {'adept': 'http://ns.adobe.com/adept',
+         'enc': 'http://www.w3.org/2001/04/xmlenc#'}
 
 # Wrap a stream so that output gets flushed immediately
 # and also make sure that any unicode strings get
@@ -95,8 +73,11 @@ class SafeUnbuffered:
     def __getattr__(self, attr):
         return getattr(self.stream, attr)
 
-iswindows = sys.platform.startswith('win')
-isosx = sys.platform.startswith('darwin')
+try:
+    from calibre.constants import iswindows, isosx
+except:
+    iswindows = sys.platform.startswith('win')
+    isosx = sys.platform.startswith('darwin')
 
 def unicode_argv():
     if iswindows:
@@ -105,8 +86,8 @@ def unicode_argv():
 
         # Versions 2.x of Python don't support Unicode in sys.argv on
         # Windows, with the underlying Windows API instead replacing multi-byte
-        # characters with '?'.
-
+        # characters with '?'.  So use shell32.GetCommandLineArgvW to get sys.argv
+        # as a list of Unicode strings and encode them as utf-8
 
         from ctypes import POINTER, byref, cdll, c_int, windll
         from ctypes.wintypes import LPCWSTR, LPWSTR
@@ -129,469 +110,99 @@ def unicode_argv():
                     xrange(start, argc.value)]
         # if we don't have any arguments at all, just pass back script name
         # this should never happen
-        return [u"mobidedrm.py"]
+        return [u"epubtest.py"]
     else:
         argvencoding = sys.stdin.encoding
         if argvencoding == None:
             argvencoding = "utf-8"
         return [arg if (type(arg) == unicode) else unicode(arg,argvencoding) for arg in sys.argv]
 
-Des = None
-if iswindows:
-    # first try with pycrypto
-    if inCalibre:
-        from calibre_plugins.dedrm import pycrypto_des
-    else:
-        import pycrypto_des
-    Des = pycrypto_des.load_pycrypto()
-    if Des == None:
-        # they try with openssl
-        if inCalibre:
-            from calibre_plugins.dedrm import openssl_des
-        else:
-            import openssl_des
-        Des = openssl_des.load_libcrypto()
-else:
-    # first try with openssl
-    if inCalibre:
-        from calibre_plugins.dedrm import openssl_des
-    else:
-        import openssl_des
-    Des = openssl_des.load_libcrypto()
-    if Des == None:
-        # then try with pycrypto
-        if inCalibre:
-            from calibre_plugins.dedrm import pycrypto_des
-        else:
-            import pycrypto_des
-        Des = pycrypto_des.load_pycrypto()
+_FILENAME_LEN_OFFSET = 26
+_EXTRA_LEN_OFFSET = 28
+_FILENAME_OFFSET = 30
+_MAX_SIZE = 64 * 1024
 
-# 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.dedrm import python_des
+
+def uncompress(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:
-        import python_des
-    Des = python_des.Des
-    # Import Psyco if available
+            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(file, zi):
+    # get file name length and exta data length to find start of file data
+    local_header_offset = zi.header_offset
+
+    file.seek(local_header_offset + _FILENAME_LEN_OFFSET)
+    leninfo = file.read(2)
+    local_name_length, = struct.unpack('<H', leninfo)
+
+    file.seek(local_header_offset + _EXTRA_LEN_OFFSET)
+    exinfo = file.read(2)
+    extra_field_length, = struct.unpack('<H', exinfo)
+
+    file.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 = file.read(zi.file_size)
+
+    # if compressed we must decompress it using zlib
+    if zi.compress_type == zipfile.ZIP_DEFLATED:
+        cmpdata = file.read(zi.compress_size)
+        data = uncompress(cmpdata)
+
+    return data
+
+def encryption(infile):
+    # returns encryption: one of Unencrypted, Adobe, B&N and Unknown
+    encryption = "Unknown"
     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"
+        with open(infile,'rb') as infileobject:
+            bookdata = infileobject.read(58)
+            # Check for Zip
+            if bookdata[0:0+2] == "PK":
+                foundrights = False
+                foundencryption = False
+                inzip = zipfile.ZipFile(infile,'r')
+                namelist = set(inzip.namelist())
+                if 'META-INF/rights.xml' not in namelist or 'META-INF/encryption.xml' not in namelist:
+                    encryption = "Unencrypted"
                 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)
+                    rights = etree.fromstring(inzip.read('META-INF/rights.xml'))
+                    adept = lambda tag: '{%s}%s' % (NSMAP['adept'], tag)
+                    expr = './/%s' % (adept('encryptedKey'),)
+                    bookkey = ''.join(rights.findtext(expr))
+                    if len(bookkey) == 172:
+                        encryption = "Adobe"
+                    elif len(bookkey) == 64:
+                        encryption = "B&N"
                     else:
-            end_off = self.sections[section + 1][0]
-        off = self.sections[section][0]
-        return self.contents[off:end_off]
-
-# cleanup unicode filenames
-# borrowed from calibre from calibre/src/calibre/__init__.py
-# added in removal of control (<32) chars
-# and removal of . at start and end
-# and with some (heavily edited) code from Paul Durrant's kindlenamer.py
-def sanitizeFileName(name):
-    # substitute filename unfriendly characters
-    name = name.replace(u"<",u"[").replace(u">",u"]").replace(u" : ",u" – ").replace(u": ",u" – ").replace(u":",u"—").replace(u"/",u"_").replace(u"\\",u"_").replace(u"|",u"_").replace(u"\"",u"\'")
-    # delete control characters
-    name = u"".join(char for char in name if ord(char)>=32)
-    # white space to single space, delete leading and trailing while space
-    name = re.sub(ur"\s", u" ", name).strip()
-    # remove leading dots
-    while len(name)>0 and name[0] == u".":
-        name = name[1:]
-    # remove trailing dots (Windows doesn't like them)
-    if name.endswith(u'.'):
-        name = name[:-1]
-    return name
-
-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, user_key):
-        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)
-
-        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(unicode(name,'windows-1252')), 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 decryptBook(infile, outpath, make_pmlz, user_key):
-    bookname = os.path.splitext(os.path.basename(infile))[0]
-    if make_pmlz:
-        # outpath is actually pmlz name
-        pmlzname = outpath
-        outdir = tempfile.mkdtemp()
-        imagedirpath = os.path.join(outdir,u"images")
-    else:
-        pmlzname = None
-        outdir = outpath
-        imagedirpath = os.path.join(outdir,bookname + u"_img")
-
-    try:
-        if not os.path.exists(outdir):
-            os.makedirs(outdir)
-        print u"Decoding File"
-        sect = Sectionizer(infile, 'PNRdPPrs')
-        er = EreaderProcessor(sect, user_key)
-
-        if er.getNumImages() > 0:
-            print u"Extracting images"
-            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 u"Extracting pml"
-        pml_string = er.getText()
-        pmlfilename = bookname + ".pml"
-        file(os.path.join(outdir, pmlfilename),'wb').write(cleanPML(pml_string))
-        if pmlzname is not None:
-            import zipfile
-            import shutil
-            print u"Creating PMLZ file {0}".format(os.path.basename(pmlzname))
-            myZipFile = zipfile.ZipFile(pmlzname,'w',zipfile.ZIP_STORED, False)
-            list = os.listdir(outdir)
-            for filename in list:
-                localname = filename
-                filePath = os.path.join(outdir,filename)
-                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 u"Output is {0}".format(pmlzname)
-        else :
-            print u"Output is in {0}".format(outdir)
-        print "done"
-    except ValueError, e:
-        print u"Error: {0}".format(e)
+                        encryption = "Unknown"
+    except:
         traceback.print_exc()
-        return 1
-    return 0
-
-
-def usage():
-    print u"Converts DRMed eReader books to PML Source"
-    print u"Usage:"
-    print u"  erdr2pml [options] infile.pdb [outpath] \"your name\" credit_card_number"
-    print u" "
-    print u"Options: "
-    print u"  -h             prints this message"
-    print u"  -p             create PMLZ instead of source folder"
-    print u"  --make-pmlz    create PMLZ instead of source folder"
-    print u" "
-    print u"Note:"
-    print u"  if outpath is ommitted, creates source in 'infile_Source' folder"
-    print u"  if outpath is ommitted and pmlz option, creates PMLZ 'infile.pmlz'"
-    print u"  if source folder created, images are in infile_img folder"
-    print u"  if pmlz file created, images are in images folder"
-    print u"  It's enough to enter the last 8 digits of the credit card number"
-    return
-
-def getuser_key(name,cc):
-    newname = "".join(c for c in name.lower() if c >= 'a' and c <= 'z' or c >= '0' and c <= '9')
-    cc = cc.replace(" ","")
-    return struct.pack('>LL', binascii.crc32(newname) & 0xffffffff,binascii.crc32(cc[-8:])& 0xffffffff)
-
-def cli_main():
-    print u"eRdr2Pml v{0}. Copyright © 2009–2012 The Dark Reverser et al.".format(__version__)
+    return encryption
 
+def main():
     argv=unicode_argv()
-    try:
-        opts, args = getopt.getopt(argv[1:], "hp", ["make-pmlz"])
-    except getopt.GetoptError, err:
-        print err.args[0]
-        usage()
-        return 1
-    make_pmlz = False
-    for o, a in opts:
-        if o == "-h":
-            usage()
+    print encryption(argv[1])
     return 0
-        elif o == "-p":
-            make_pmlz = True
-        elif o == "--make-pmlz":
-            make_pmlz = True
-
-    if len(args)!=3 and len(args)!=4:
-        usage()
-        return 1
-
-    if len(args)==3:
-        infile, name, cc = args
-        if make_pmlz:
-            outpath = os.path.splitext(infile)[0] + u".pmlz"
-        else:
-            outpath = os.path.splitext(infile)[0] + u"_Source"
-    elif len(args)==4:
-        infile, outpath, name, cc = args
-
-    print getuser_key(name,cc).encode('hex')
-
-    return decryptBook(infile, outpath, make_pmlz, getuser_key(name,cc))
-
 
 if __name__ == "__main__":
     sys.stdout=SafeUnbuffered(sys.stdout)
     sys.stderr=SafeUnbuffered(sys.stderr)
-    sys.exit(cli_main())
-
+    sys.exit(main())

DeDRM_calibre_plugin/DeDRM_plugin/flatxml2html.py

@@ -1,63 +1,127 @@
 #! /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 PParser(object):
-    def __init__(self, gd, flatxml, meta_array):
-        self.gd = gd
-        self.flatdoc = flatxml.split('\n')
-        self.docSize = len(self.flatdoc)
-        self.temp = []
+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 = []
 
-        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'))
+    def getGlyph(self, gid):
+        result = ''
+        id='id="gl%d"' % gid
+        return self.gdict.lookup(id)
 
-        res = []
-        startpos = self.posinDoc('info.glyph.x')
-        for p in startpos:
-            argres = self.getDataatPos('info.glyph.x', p)
-            res.extend(argres)
-        self.gx = res
+    def glyphs_to_image(self, glyphList):
 
-        res = []
-        startpos = self.posinDoc('info.glyph.y')
-        for p in startpos:
-            argres = self.getDataatPos('info.glyph.y', p)
-            res.extend(argres)
-        self.gy = res
+        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
 
-        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]
+            item = self.docList[pos]
             if item.find('=') >= 0:
                 (name, argres) = item.split('=',1)
             else :
@@ -65,6 +129,7 @@ class PParser(object):
                 argres = ''
         return name, argres
 
+
     # find tag in doc if within pos to end inclusive
     def findinDoc(self, tagpath, pos, end) :
         result = None
@@ -74,7 +139,7 @@ class PParser(object):
             end = min(self.docSize, end)
         foundat = -1
         for j in xrange(pos, end):
-            item = self.flatdoc[j]
+            item = self.docList[j]
             if item.find('=') >= 0:
                 (name, argres) = item.split('=',1)
             else :
@@ -86,6 +151,7 @@ class PParser(object):
                 break
         return foundat, result
 
+
     # return list of start positions for the tagpath
     def posinDoc(self, tagpath):
         startpos = []
@@ -98,152 +164,638 @@ class PParser(object):
             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
+    # 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
 
 
-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')
+    # get the class
+    def getClass(self, pclass):
+        nclass = pclass
 
-        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>')
+        # 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 :
-        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')
+                # 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 :
-            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)
+                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'):
+                # we only handle spaceafter word class
+                try:
+                    (cname, space) = argres.split('-',1)
+                    if space == '' : space = '0'
+                    if (cname == 'spaceafter') and (int(space) > 0) :
+                        word_class = 'sa'
+                except:
+                    pass
+
+            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' :
+                try:
+                    word = self.ocrtext[num]
+                except:
+                    word = ""
+
+                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

DeDRM_calibre_plugin/DeDRM_plugin/flatxml2svg.py

@@ -1,148 +1,82 @@
 #! /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.dedrm import convert2xml
-    from calibre_plugins.dedrm import flatxml2html
-    from calibre_plugins.dedrm import flatxml2svg
-    from calibre_plugins.dedrm 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):
+class PParser(object):
+    def __init__(self, gd, flatxml, meta_array):
+        self.gd = gd
         self.flatdoc = flatxml.split('\n')
-    # find tag if within pos to end inclusive
+        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
-        docList = self.flatdoc
-        cnt = len(docList)
         if end == -1 :
-            end = cnt
+            end = self.docSize
         else:
-            end = min(cnt,end)
+            end = min(self.docSize, end)
         foundat = -1
         for j in xrange(pos, end):
-            item = docList[j]
+            item = self.flatdoc[j]
             if item.find('=') >= 0:
-                (name, argres) = item.split('=')
+                (name, argres) = item.split('=',1)
             else :
                 name = item
                 argres = ''
@@ -151,44 +85,19 @@ class PageDimParser(object):
                 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
+    # 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
 
-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)
@@ -200,522 +109,141 @@ class GParser(object):
             else:
                 name = item
                 argres = []
-            if (name == path):
+            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 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]
+
+    def getDataatPos(self, path, pos):
+        result = None
+        item = self.flatdoc[pos]
+        if item.find('=') >= 0:
+            (name, argt) = item.split('=')
+            argres = argt.split('|')
         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])
+            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
 
-                j += 1
-        path += 'z'
-        return path
+    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
 
 
-
-# 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')
+def convert2SVG(gdict, flat_xml, pageid, previd, nextid, svgDir, raw, meta_array, scaledpi):
     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]
+    pp = PParser(gdict, flat_xml, meta_array)
+    mlst.append('<?xml version="1.0" standalone="no"?>\n')
     if (raw):
-        startname = 'page%04d.svg' % start
+        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:
-        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
+        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:
-            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)
+            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')
 
-
-    # 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)
+        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) :
-            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))
+        mlst.append('</svg>')
     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(''))
+        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)

DeDRM_calibre_plugin/DeDRM_plugin/ignobleepub.py

@@ -3,47 +3,54 @@
 
 from __future__ import with_statement
 
-# ignoblekeygen.pyw, version 2.5
-# Copyright © 2009-2010 i♥cabbages
+# ignobleepub.pyw, version 3.8
+# Copyright © 2009-2010 by i♥cabbages
 
 # Released under the terms of the GNU General Public Licence, version 3
 # <http://www.gnu.org/licenses/>
 
 # Modified 2010–2013 by some_updates, DiapDealer and Apprentice Alf
 
-# Windows users: Before running this program, you must first install Python.
-#   We recommend ActiveState Python 2.7.X for Windows (x86) from
-#   http://www.activestate.com/activepython/downloads.
-#   You must also install PyCrypto from
-#   http://www.voidspace.org.uk/python/modules.shtml#pycrypto
-#   (make certain to install the version for Python 2.7).
-#   Then save this script file as ignoblekeygen.pyw and double-click on it to run it.
+# 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 ignoblekeygen.pyw.  You can run this
-#   program from the command line (python ignoblekeygen.pyw) or by double-clicking
+# 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 - 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
-#   2.4 - Improvements to UI and now works in plugins
-#   2.5 - Additional improvement for unicode and plugin support
-#   2.6 - moved unicode_argv call inside main for Windows DeDRM compatibility
-#   2.7 - Work if TkInter is missing
+#   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 decrypt from encryption.xml
+#   3.3 - On Windows try PyCrypto first, OpenSSL next
+#   3.4 - Modify interface to allow use with import
+#   3.5 - Fix for potential problem with PyCrypto
+#   3.6 - Revised to allow use in calibre plugins to eliminate need for duplicate code
+#   3.7 - Tweaked to match ineptepub more closely
+#   3.8 - Fixed to retain zip file metadata (e.g. file modification date)
+#   3.9 - moved unicode_argv call inside main for Windows DeDRM compatibility
+#   4.0 - Work if TkInter is missing
 
 """
-Generate Barnes & Noble EPUB user key from name and credit card number.
+Decrypt Barnes & Noble encrypted ePub books.
 """
 
 __license__ = 'GPL v3'
-__version__ = "2.7"
+__version__ = "4.0"
 
 import sys
 import os
-import hashlib
+import traceback
+import zlib
+import zipfile
+from zipfile import ZipInfo, ZipFile, ZIP_STORED, ZIP_DEFLATED
+from contextlib import closing
+import xml.etree.ElementTree as etree
 
 # Wrap a stream so that output gets flushed immediately
 # and also make sure that any unicode strings get
@@ -75,8 +82,8 @@ def unicode_argv():
 
         # Versions 2.x of Python don't support Unicode in sys.argv on
         # Windows, with the underlying Windows API instead replacing multi-byte
-        # characters with '?'.  So use shell32.GetCommandLineArgvW to get sys.argv
-        # as a list of Unicode strings and encode them as utf-8
+        # characters with '?'.
+
 
         from ctypes import POINTER, byref, cdll, c_int, windll
         from ctypes.wintypes import LPCWSTR, LPWSTR
@@ -97,9 +104,7 @@ def unicode_argv():
             start = argc.value - len(sys.argv)
             return [argv[i] for i in
                     xrange(start, argc.value)]
-        # if we don't have any arguments at all, just pass back script name
-        # this should never happen
-        return [u"ignoblekeygen.py"]
+        return [u"ineptepub.py"]
     else:
         argvencoding = sys.stdin.encoding
         if argvencoding == None:
@@ -140,26 +145,29 @@ def _load_crypto_libcrypto():
         func.argtypes = argtypes
         return func
 
-    AES_set_encrypt_key = F(c_int, 'AES_set_encrypt_key',
+    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, iv):
+        def __init__(self, userkey):
             self._blocksize = len(userkey)
-            self._iv = iv
+            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_encrypt_key(userkey, len(userkey) * 8, key)
+            rv = AES_set_decrypt_key(userkey, len(userkey) * 8, key)
             if rv < 0:
-                raise IGNOBLEError('Failed to initialize AES Encrypt key')
+                raise IGNOBLEError('Failed to initialize AES key')
 
-        def encrypt(self, data):
+        def decrypt(self, data):
             out = create_string_buffer(len(data))
-            rv = AES_cbc_encrypt(data, out, len(data), self._key, self._iv, 1)
+            iv = ("\x00" * self._blocksize)
+            rv = AES_cbc_encrypt(data, out, len(data), self._key, iv, 0)
             if rv == 0:
-                raise IGNOBLEError('AES encryption failed')
+                raise IGNOBLEError('AES decryption failed')
             return out.raw
 
     return AES
@@ -168,11 +176,11 @@ 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 __init__(self, key):
+            self._aes = _AES.new(key, _AES.MODE_CBC, '\x00'*16)
 
-        def encrypt(self, data):
-            return self._aes.encrypt(data)
+        def decrypt(self, data):
+            return self._aes.decrypt(data)
 
     return AES
 
@@ -191,29 +199,123 @@ def _load_crypto():
 
 AES = _load_crypto()
 
-def normalize_name(name):
-    return ''.join(x for x in name.lower() if x != ' ')
+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 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 generate_key(name, ccn):
-    # remove spaces and case from name and CC numbers.
-    if type(name)==unicode:
-        name = name.encode('utf-8')
-    if type(ccn)==unicode:
-        ccn = ccn.encode('utf-8')
+    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
 
-    name = normalize_name(name) + '\x00'
-    ccn = normalize_name(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()
-    return userkey.encode('base64')
+    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
 
+# check file to make check whether it's probably an Adobe Adept encrypted ePub
+def ignobleBook(inpath):
+    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:
+            return False
+        try:
+            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))
+            if len(bookkey) == 64:
+                return True
+        except:
+            # if we couldn't check, assume it is
+            return True
+    return False
 
+def decryptBook(keyb64, inpath, outpath):
+    if AES is None:
+        raise IGNOBLEError(u"PyCrypto or OpenSSL must be installed.")
+    key = keyb64.decode('base64')[:16]
+    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:
+            print u"{0:s} is DRM-free.".format(os.path.basename(inpath))
+            return 1
+        for name in META_NAMES:
+            namelist.remove(name)
+        try:
+            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))
+            if len(bookkey) != 64:
+                print u"{0:s} is not a secure Barnes & Noble ePub.".format(os.path.basename(inpath))
+                return 1
+            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')
+                zi.compress_type=ZIP_STORED
+                try:
+                    # if the mimetype is present, get its info, including time-stamp
+                    oldzi = inf.getinfo('mimetype')
+                    # copy across fields to be preserved
+                    zi.date_time = oldzi.date_time
+                    zi.comment = oldzi.comment
+                    zi.extra = oldzi.extra
+                    zi.internal_attr = oldzi.internal_attr
+                    # external attributes are dependent on the create system, so copy both.
+                    zi.external_attr = oldzi.external_attr
+                    zi.create_system = oldzi.create_system
+                except:
+                    pass
+                outf.writestr(zi, inf.read('mimetype'))
+                for path in namelist:
+                    data = inf.read(path)
+                    zi = ZipInfo(path)
+                    zi.compress_type=ZIP_DEFLATED
+                    try:
+                        # get the file info, including time-stamp
+                        oldzi = inf.getinfo(path)
+                        # copy across useful fields
+                        zi.date_time = oldzi.date_time
+                        zi.comment = oldzi.comment
+                        zi.extra = oldzi.extra
+                        zi.internal_attr = oldzi.internal_attr
+                        # external attributes are dependent on the create system, so copy both.
+                        zi.external_attr = oldzi.external_attr
+                        zi.create_system = oldzi.create_system
+                    except:
+                        pass
+                    outf.writestr(zi, decryptor.decrypt(path, data))
+        except:
+            print u"Could not decrypt {0:s} because of an exception:\n{1:s}".format(os.path.basename(inpath), traceback.format_exc())
+            return 2
+    return 0
 
 
 def cli_main():
@@ -221,19 +323,15 @@ def cli_main():
     sys.stderr=SafeUnbuffered(sys.stderr)
     argv=unicode_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 u"usage: {0} <Name> <CC#> <keyfileout.b64>".format(progname)
+        print u"usage: {0} <keyfile.b64> <inbook.epub> <outbook.epub>".format(progname)
         return 1
-    name, ccn, keypath = argv[1:]
-    userkey = generate_key(name, ccn)
-    open(keypath,'wb').write(userkey)
-    return 0
-
+    keypath, inpath, outpath = argv[1:]
+    userkey = open(keypath,'rb').read()
+    result = decryptBook(userkey, inpath, outpath)
+    if result == 0:
+        print u"Successfully decrypted {0:s} as {1:s}".format(os.path.basename(inpath),os.path.basename(outpath))
+    return result
 
 def gui_main():
     try:
@@ -247,28 +345,33 @@ def gui_main():
     class DecryptionDialog(Tkinter.Frame):
         def __init__(self, root):
             Tkinter.Frame.__init__(self, root, border=5)
-            self.status = Tkinter.Label(self, text=u"Enter parameters")
+            self.status = Tkinter.Label(self, text=u"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=u"Account Name").grid(row=0)
-            self.name = Tkinter.Entry(body, width=40)
-            self.name.grid(row=0, column=1, sticky=sticky)
-            Tkinter.Label(body, text=u"CC#").grid(row=1)
-            self.ccn = Tkinter.Entry(body, width=40)
-            self.ccn.grid(row=1, column=1, sticky=sticky)
-            Tkinter.Label(body, text=u"Output file").grid(row=2)
-            self.keypath = Tkinter.Entry(body, width=40)
-            self.keypath.grid(row=2, column=1, sticky=sticky)
-            self.keypath.insert(2, u"bnepubkey.b64")
+            Tkinter.Label(body, text=u"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(u"bnepubkey.b64"):
+                self.keypath.insert(0, u"bnepubkey.b64")
             button = Tkinter.Button(body, text=u"...", command=self.get_keypath)
+            button.grid(row=0, column=2)
+            Tkinter.Label(body, text=u"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=u"...", command=self.get_inpath)
+            button.grid(row=1, column=2)
+            Tkinter.Label(body, text=u"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=u"...", command=self.get_outpath)
             button.grid(row=2, column=2)
             buttons = Tkinter.Frame(self)
             buttons.pack()
             botton = Tkinter.Button(
-                buttons, text=u"Generate", width=10, command=self.generate)
+                buttons, text=u"Decrypt", width=10, command=self.decrypt)
             botton.pack(side=Tkconstants.LEFT)
             Tkinter.Frame(buttons, width=10).pack(side=Tkconstants.LEFT)
             button = Tkinter.Button(
@@ -276,8 +379,8 @@ def gui_main():
             button.pack(side=Tkconstants.RIGHT)
 
         def get_keypath(self):
-            keypath = tkFileDialog.asksaveasfilename(
-                parent=None, title=u"Select B&N ePub key file to produce",
+            keypath = tkFileDialog.askopenfilename(
+                parent=None, title=u"Select Barnes & Noble \'.b64\' key file",
                 defaultextension=u".b64",
                 filetypes=[('base64-encoded files', '.b64'),
                            ('All Files', '.*')])
@@ -287,37 +390,56 @@ def gui_main():
                 self.keypath.insert(0, keypath)
             return
 
-        def generate(self):
-            name = self.name.get()
-            ccn = self.ccn.get()
+        def get_inpath(self):
+            inpath = tkFileDialog.askopenfilename(
+                parent=None, title=u"Select B&N-encrypted ePub file to decrypt",
+                defaultextension=u".epub", filetypes=[('ePub files', '.epub')])
+            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=u"Select unencrypted ePub file to produce",
+                defaultextension=u".epub", filetypes=[('ePub files', '.epub')])
+            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()
-            if not name:
-                self.status['text'] = u"Name not specified"
+            inpath = self.inpath.get()
+            outpath = self.outpath.get()
+            if not keypath or not os.path.exists(keypath):
+                self.status['text'] = u"Specified key file does not exist"
                 return
-            if not ccn:
-                self.status['text'] = u"Credit card number not specified"
+            if not inpath or not os.path.exists(inpath):
+                self.status['text'] = u"Specified input file does not exist"
                 return
-            if not keypath:
-                self.status['text'] = u"Output keyfile path not specified"
+            if not outpath:
+                self.status['text'] = u"Output file not specified"
                 return
-            self.status['text'] = u"Generating..."
+            if inpath == outpath:
+                self.status['text'] = u"Must have different input and output files"
+                return
+            userkey = open(keypath,'rb').read()
+            self.status['text'] = u"Decrypting..."
             try:
-                userkey = generate_key(name, ccn)
+                decrypt_status = decryptBook(userkey, inpath, outpath)
             except Exception, e:
-                self.status['text'] = u"Error: (0}".format(e.args[0])
+                self.status['text'] = u"Error: {0}".format(e.args[0])
                 return
-            open(keypath,'wb').write(userkey)
-            self.status['text'] = u"Keyfile successfully generated"
+            if decrypt_status == 0:
+                self.status['text'] = u"File successfully decrypted"
+            else:
+                self.status['text'] = u"The was an error decrypting the file."
 
     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(u"Barnes & Noble ePub Keyfile Generator v.{0}".format(__version__))
+    root.title(u"Barnes & Noble ePub Decrypter v.{0}".format(__version__))
     root.resizable(True, False)
     root.minsize(300, 0)
     DecryptionDialog(root).pack(fill=Tkconstants.X, expand=1)

DeDRM_calibre_plugin/DeDRM_plugin/ignoblekeygen.py

@@ -3,56 +3,47 @@
 
 from __future__ import with_statement
 
-# ineptepub.pyw, version 6.1
-# Copyright © 2009-2010 by i♥cabbages
+# ignoblekeygen.pyw, version 2.5
+# Copyright © 2009-2010 i♥cabbages
 
 # Released under the terms of the GNU General Public Licence, version 3
 # <http://www.gnu.org/licenses/>
 
 # Modified 2010–2013 by some_updates, DiapDealer and Apprentice Alf
 
-# 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.
+# Windows users: Before running this program, you must first install Python.
+#   We recommend ActiveState Python 2.7.X for Windows (x86) from
+#   http://www.activestate.com/activepython/downloads.
+#   You must also install PyCrypto from
+#   http://www.voidspace.org.uk/python/modules.shtml#pycrypto
+#   (make certain to install the version for Python 2.7).
+#   Then save this script file as ignoblekeygen.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
+# Mac OS X users: Save this script file as ignoblekeygen.pyw.  You can run this
+#   program from the command line (python ignoblekeygen.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
-#   5.7 - Fix for potential problem with PyCrypto
-#   5.8 - Revised to allow use in calibre plugins to eliminate need for duplicate code
-#   5.9 - Fixed to retain zip file metadata (e.g. file modification date)
-#   6.0 - moved unicode_argv call inside main for Windows DeDRM compatibility
-#   6.1 - Work if TkInter is missing
+#   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
+#   2.4 - Improvements to UI and now works in plugins
+#   2.5 - Additional improvement for unicode and plugin support
+#   2.6 - moved unicode_argv call inside main for Windows DeDRM compatibility
+#   2.7 - Work if TkInter is missing
 
 """
-Decrypt Adobe Digital Editions encrypted ePub books.
+Generate Barnes & Noble EPUB user key from name and credit card number.
 """
 
 __license__ = 'GPL v3'
-__version__ = "6.1"
+__version__ = "2.7"
 
 import sys
 import os
-import traceback
-import zlib
-import zipfile
-from zipfile import ZipInfo, ZipFile, ZIP_STORED, ZIP_DEFLATED
-from contextlib import closing
-import xml.etree.ElementTree as etree
+import hashlib
 
 # Wrap a stream so that output gets flushed immediately
 # and also make sure that any unicode strings get
@@ -84,8 +75,8 @@ def unicode_argv():
 
         # Versions 2.x of Python don't support Unicode in sys.argv on
         # Windows, with the underlying Windows API instead replacing multi-byte
-        # characters with '?'.
-
+        # characters with '?'.  So use shell32.GetCommandLineArgvW to get sys.argv
+        # as a list of Unicode strings and encode them as utf-8
 
         from ctypes import POINTER, byref, cdll, c_int, windll
         from ctypes.wintypes import LPCWSTR, LPWSTR
@@ -106,7 +97,9 @@ def unicode_argv():
             start = argc.value - len(sys.argv)
             return [argv[i] for i in
                     xrange(start, argc.value)]
-        return [u"ineptepub.py"]
+        # if we don't have any arguments at all, just pass back script name
+        # this should never happen
+        return [u"ignoblekeygen.py"]
     else:
         argvencoding = sys.stdin.encoding
         if argvencoding == None:
@@ -114,7 +107,7 @@ def unicode_argv():
         return [arg if (type(arg) == unicode) else unicode(arg,argvencoding) for arg in sys.argv]
 
 
-class ADEPTError(Exception):
+class IGNOBLEError(Exception):
     pass
 
 def _load_crypto_libcrypto():
@@ -128,19 +121,14 @@ def _load_crypto_libcrypto():
         libcrypto = find_library('crypto')
 
     if libcrypto is None:
-        raise ADEPTError('libcrypto not found')
+        raise IGNOBLEError('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)]
@@ -152,312 +140,80 @@ def _load_crypto_libcrypto():
         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',
+    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 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):
+        def __init__(self, userkey, iv):
             self._blocksize = len(userkey)
-            if (self._blocksize != 16) and (self._blocksize != 24) and (self._blocksize != 32) :
-                raise ADEPTError('AES improper key used')
-                return
+            self._iv = iv
             key = self._key = AES_KEY()
-            rv = AES_set_decrypt_key(userkey, len(userkey) * 8, key)
+            rv = AES_set_encrypt_key(userkey, len(userkey) * 8, key)
             if rv < 0:
-                raise ADEPTError('Failed to initialize AES key')
+                raise IGNOBLEError('Failed to initialize AES Encrypt key')
 
-        def decrypt(self, data):
+        def encrypt(self, data):
             out = create_string_buffer(len(data))
-            iv = ("\x00" * self._blocksize)
-            rv = AES_cbc_encrypt(data, out, len(data), self._key, iv, 0)
+            rv = AES_cbc_encrypt(data, out, len(data), self._key, self._iv, 1)
             if rv == 0:
-                raise ADEPTError('AES decryption failed')
+                raise IGNOBLEError('AES encryption failed')
             return out.raw
 
-    return (AES, RSA)
+    return AES
 
 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, '\x00'*16)
+        def __init__(self, key, iv):
+            self._aes = _AES.new(key, _AES.MODE_CBC, iv)
 
-        def decrypt(self, data):
-            return self._aes.decrypt(data)
+        def encrypt(self, data):
+            return self._aes.encrypt(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)
+    return AES
 
 def _load_crypto():
-    AES = RSA = None
+    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, RSA = loader()
+            AES = loader()
             break
-        except (ImportError, ADEPTError):
+        except (ImportError, IGNOBLEError):
             pass
-    return (AES, RSA)
+    return AES
 
-AES, RSA = _load_crypto()
+AES = _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#'}
+def normalize_name(name):
+    return ''.join(x for x in name.lower() if x != ' ')
 
-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 generate_key(name, ccn):
+    # remove spaces and case from name and CC numbers.
+    if type(name)==unicode:
+        name = name.encode('utf-8')
+    if type(ccn)==unicode:
+        ccn = ccn.encode('utf-8')
 
-    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
+    name = normalize_name(name) + '\x00'
+    ccn = normalize_name(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()
+    return userkey.encode('base64')
 
-# check file to make check whether it's probably an Adobe Adept encrypted ePub
-def adeptBook(inpath):
-    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:
-            return False
-        try:
-            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))
-            if len(bookkey) == 172:
-                return True
-        except:
-            # if we couldn't check, assume it is
-            return True
-    return False
 
-def decryptBook(userkey, inpath, outpath):
-    if AES is None:
-        raise ADEPTError(u"PyCrypto or OpenSSL must be installed.")
-    rsa = RSA(userkey)
-    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:
-            print u"{0:s} is DRM-free.".format(os.path.basename(inpath))
-            return 1
-        for name in META_NAMES:
-            namelist.remove(name)
-        try:
-            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))
-            if len(bookkey) != 172:
-                print u"{0:s} is not a secure Adobe Adept ePub.".format(os.path.basename(inpath))
-                return 1
-            bookkey = rsa.decrypt(bookkey.decode('base64'))
-            # Padded as per RSAES-PKCS1-v1_5
-            if bookkey[-17] != '\x00':
-                print u"Could not decrypt {0:s}. Wrong key".format(os.path.basename(inpath))
-                return 2
-            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')
-                zi.compress_type=ZIP_STORED
-                try:
-                    # if the mimetype is present, get its info, including time-stamp
-                    oldzi = inf.getinfo('mimetype')
-                    # copy across fields to be preserved
-                    zi.date_time = oldzi.date_time
-                    zi.comment = oldzi.comment
-                    zi.extra = oldzi.extra
-                    zi.internal_attr = oldzi.internal_attr
-                    # external attributes are dependent on the create system, so copy both.
-                    zi.external_attr = oldzi.external_attr
-                    zi.create_system = oldzi.create_system
-                except:
-                    pass
-                outf.writestr(zi, inf.read('mimetype'))
-                for path in namelist:
-                    data = inf.read(path)
-                    zi = ZipInfo(path)
-                    zi.compress_type=ZIP_DEFLATED
-                    try:
-                        # get the file info, including time-stamp
-                        oldzi = inf.getinfo(path)
-                        # copy across useful fields
-                        zi.date_time = oldzi.date_time
-                        zi.comment = oldzi.comment
-                        zi.extra = oldzi.extra
-                        zi.internal_attr = oldzi.internal_attr
-                        # external attributes are dependent on the create system, so copy both.
-                        zi.external_attr = oldzi.external_attr
-                        zi.create_system = oldzi.create_system
-                    except:
-                        pass
-                    outf.writestr(zi, decryptor.decrypt(path, data))
-        except:
-            print u"Could not decrypt {0:s} because of an exception:\n{1:s}".format(os.path.basename(inpath), traceback.format_exc())
-            return 2
-    return 0
 
 
 def cli_main():
@@ -465,15 +221,19 @@ def cli_main():
     sys.stderr=SafeUnbuffered(sys.stderr)
     argv=unicode_argv()
     progname = os.path.basename(argv[0])
-    if len(argv) != 4:
-        print u"usage: {0} <keyfile.der> <inbook.epub> <outbook.epub>".format(progname)
+    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
-    keypath, inpath, outpath = argv[1:]
-    userkey = open(keypath,'rb').read()
-    result = decryptBook(userkey, inpath, outpath)
-    if result == 0:
-        print u"Successfully decrypted {0:s} as {1:s}".format(os.path.basename(inpath),os.path.basename(outpath))
-    return result
+    if len(argv) != 4:
+        print u"usage: {0} <Name> <CC#> <keyfileout.b64>".format(progname)
+        return 1
+    name, ccn, keypath = argv[1:]
+    userkey = generate_key(name, ccn)
+    open(keypath,'wb').write(userkey)
+    return 0
+
 
 def gui_main():
     try:
@@ -487,33 +247,28 @@ def gui_main():
     class DecryptionDialog(Tkinter.Frame):
         def __init__(self, root):
             Tkinter.Frame.__init__(self, root, border=5)
-            self.status = Tkinter.Label(self, text=u"Select files for decryption")
+            self.status = Tkinter.Label(self, text=u"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=u"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(u"adeptkey.der"):
-                self.keypath.insert(0, u"adeptkey.der")
-            button = Tkinter.Button(body, text=u"...", command=self.get_keypath)
-            button.grid(row=0, column=2)
-            Tkinter.Label(body, text=u"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=u"...", command=self.get_inpath)
-            button.grid(row=1, column=2)
+            Tkinter.Label(body, text=u"Account Name").grid(row=0)
+            self.name = Tkinter.Entry(body, width=40)
+            self.name.grid(row=0, column=1, sticky=sticky)
+            Tkinter.Label(body, text=u"CC#").grid(row=1)
+            self.ccn = Tkinter.Entry(body, width=40)
+            self.ccn.grid(row=1, column=1, sticky=sticky)
             Tkinter.Label(body, text=u"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=u"...", command=self.get_outpath)
+            self.keypath = Tkinter.Entry(body, width=40)
+            self.keypath.grid(row=2, column=1, sticky=sticky)
+            self.keypath.insert(2, u"bnepubkey.b64")
+            button = Tkinter.Button(body, text=u"...", command=self.get_keypath)
             button.grid(row=2, column=2)
             buttons = Tkinter.Frame(self)
             buttons.pack()
             botton = Tkinter.Button(
-                buttons, text=u"Decrypt", width=10, command=self.decrypt)
+                buttons, text=u"Generate", width=10, command=self.generate)
             botton.pack(side=Tkconstants.LEFT)
             Tkinter.Frame(buttons, width=10).pack(side=Tkconstants.LEFT)
             button = Tkinter.Button(
@@ -521,10 +276,10 @@ def gui_main():
             button.pack(side=Tkconstants.RIGHT)
 
         def get_keypath(self):
-            keypath = tkFileDialog.askopenfilename(
-                parent=None, title=u"Select Adobe Adept \'.der\' key file",
-                defaultextension=u".der",
-                filetypes=[('Adobe Adept DER-encoded files', '.der'),
+            keypath = tkFileDialog.asksaveasfilename(
+                parent=None, title=u"Select B&N ePub key file to produce",
+                defaultextension=u".b64",
+                filetypes=[('base64-encoded files', '.b64'),
                            ('All Files', '.*')])
             if keypath:
                 keypath = os.path.normpath(keypath)
@@ -532,56 +287,37 @@ def gui_main():
                 self.keypath.insert(0, keypath)
             return
 
-        def get_inpath(self):
-            inpath = tkFileDialog.askopenfilename(
-                parent=None, title=u"Select ADEPT-encrypted ePub file to decrypt",
-                defaultextension=u".epub", filetypes=[('ePub files', '.epub')])
-            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=u"Select unencrypted ePub file to produce",
-                defaultextension=u".epub", filetypes=[('ePub files', '.epub')])
-            if outpath:
-                outpath = os.path.normpath(outpath)
-                self.outpath.delete(0, Tkconstants.END)
-                self.outpath.insert(0, outpath)
-            return
-
-        def decrypt(self):
+        def generate(self):
+            name = self.name.get()
+            ccn = self.ccn.get()
             keypath = self.keypath.get()
-            inpath = self.inpath.get()
-            outpath = self.outpath.get()
-            if not keypath or not os.path.exists(keypath):
-                self.status['text'] = u"Specified key file does not exist"
+            if not name:
+                self.status['text'] = u"Name not specified"
                 return
-            if not inpath or not os.path.exists(inpath):
-                self.status['text'] = u"Specified input file does not exist"
+            if not ccn:
+                self.status['text'] = u"Credit card number not specified"
                 return
-            if not outpath:
-                self.status['text'] = u"Output file not specified"
+            if not keypath:
+                self.status['text'] = u"Output keyfile path not specified"
                 return
-            if inpath == outpath:
-                self.status['text'] = u"Must have different input and output files"
-                return
-            userkey = open(keypath,'rb').read()
-            self.status['text'] = u"Decrypting..."
+            self.status['text'] = u"Generating..."
             try:
-                decrypt_status = decryptBook(userkey, inpath, outpath)
+                userkey = generate_key(name, ccn)
             except Exception, e:
-                self.status['text'] = u"Error: {0}".format(e.args[0])
+                self.status['text'] = u"Error: (0}".format(e.args[0])
                 return
-            if decrypt_status == 0:
-                self.status['text'] = u"File successfully decrypted"
-            else:
-                self.status['text'] = u"The was an error decrypting the file."
+            open(keypath,'wb').write(userkey)
+            self.status['text'] = u"Keyfile successfully generated"
 
     root = Tkinter.Tk()
-    root.title(u"Adobe Adept ePub Decrypter v.{0}".format(__version__))
+    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(u"Barnes & Noble ePub Keyfile Generator v.{0}".format(__version__))
     root.resizable(True, False)
     root.minsize(300, 0)
     DecryptionDialog(root).pack(fill=Tkconstants.X, expand=1)

DeDRM_calibre_plugin/DeDRM_plugin/k4mobidedrm.py

@@ -2,266 +2,331 @@
 # -*- coding: utf-8 -*-
 
 from __future__ import with_statement
-import sys
-import os, csv
-import binascii
-import zlib
+
+# ignobleepub.pyw, version 3.6
+# Copyright © 2009-2012 by DiapDealer et al.
+
+# 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
+# Special thanks to The Dark Reverser for MobiDeDrm and CMBDTC for cmbdtc_dump
+# from which this script borrows most unashamedly.
+
+
+# Changelog
+#  1.0 - Name change to k4mobidedrm. Adds Mac support, Adds plugin code
+#  1.1 - Adds support for additional kindle.info files
+#  1.2 - Better error handling for older Mobipocket
+#  1.3 - Don't try to decrypt Topaz books
+#  1.7 - Add support for Topaz books and Kindle serial numbers. Split code.
+#  1.9 - Tidy up after Topaz, minor exception changes
+#  2.1 - Topaz fix and filename sanitizing
+#  2.2 - Topaz Fix and minor Mac code fix
+#  2.3 - More Topaz fixes
+#  2.4 - K4PC/Mac key generation fix
+#  2.6 - Better handling of non-K4PC/Mac ebooks
+#  2.7 - Better trailing bytes handling in mobidedrm
+#  2.8 - Moved parsing of kindle.info files to mac & pc util files.
+#  3.1 - Updated for new calibre interface. Now __init__ in plugin.
+#  3.5 - Now support Kindle for PC/Mac 1.6
+#  3.6 - Even better trailing bytes handling in mobidedrm
+#  3.7 - Add support for Amazon Print Replica ebooks.
+#  3.8 - Improved Topaz support
+#  4.1 - Improved Topaz support and faster decryption with alfcrypto
+#  4.2 - Added support for Amazon's KF8 format ebooks
+#  4.4 - Linux calls to Wine added, and improved configuration dialog
+#  4.5 - Linux works again without Wine. Some Mac key file search changes
+#  4.6 - First attempt to handle unicode properly
+#  4.7 - Added timing reports, and changed search for Mac key files
+#  4.8 - Much better unicode handling, matching the updated inept and ignoble scripts
+#      - Moved back into plugin, __init__ in plugin now only contains plugin code.
+#  4.9 - Missed some invalid characters in cleanup_name
+#  5.0 - Extraction of info from Kindle for PC/Mac moved into kindlekey.py
+#      - tweaked GetDecryptedBook interface to leave passed parameters unchanged
+#  5.1 - moved unicode_argv call inside main for Windows DeDRM compatibility
+#  5.2 - Fixed error in command line processing of unicode arguments
+
+__version__ = '5.2'
+
+
+import sys, os, re
+import csv
+import getopt
 import re
-from struct import pack, unpack, unpack_from
 import traceback
+import time
+import htmlentitydefs
+import json
 
 class DrmException(Exception):
     pass
 
-global charMap1
-global charMap3
-global charMap4
+if 'calibre' in sys.modules:
+    inCalibre = True
+else:
+    inCalibre = False
+
+if inCalibre:
+    from calibre_plugins.dedrm import mobidedrm
+    from calibre_plugins.dedrm import topazextract
+    from calibre_plugins.dedrm import kgenpids
+    from calibre_plugins.dedrm import android
+else:
+    import mobidedrm
+    import topazextract
+    import kgenpids
+    import android
+
+# Wrap a stream so that output gets flushed immediately
+# and also make sure that any unicode strings get
+# encoded using "replace" before writing them.
+class SafeUnbuffered:
+    def __init__(self, stream):
+        self.stream = stream
+        self.encoding = stream.encoding
+        if self.encoding == None:
+            self.encoding = "utf-8"
+    def write(self, data):
+        if isinstance(data,unicode):
+            data = data.encode(self.encoding,"replace")
+        self.stream.write(data)
+        self.stream.flush()
+    def __getattr__(self, attr):
+        return getattr(self.stream, attr)
+
+iswindows = sys.platform.startswith('win')
+isosx = sys.platform.startswith('darwin')
+
+def unicode_argv():
+    if iswindows:
+        # Uses shell32.GetCommandLineArgvW to get sys.argv as a list of Unicode
+        # strings.
+
+        # Versions 2.x of Python don't support Unicode in sys.argv on
+        # Windows, with the underlying Windows API instead replacing multi-byte
+        # characters with '?'.
 
 
-charMap1 = 'n5Pr6St7Uv8Wx9YzAb0Cd1Ef2Gh3Jk4M'
-charMap3 = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/'
-charMap4 = 'ABCDEFGHIJKLMNPQRSTUVWXYZ123456789'
+        from ctypes import POINTER, byref, cdll, c_int, windll
+        from ctypes.wintypes import LPCWSTR, LPWSTR
 
-# crypto digestroutines
-import hashlib
+        GetCommandLineW = cdll.kernel32.GetCommandLineW
+        GetCommandLineW.argtypes = []
+        GetCommandLineW.restype = LPCWSTR
 
-def MD5(message):
-    ctx = hashlib.md5()
-    ctx.update(message)
-    return ctx.digest()
+        CommandLineToArgvW = windll.shell32.CommandLineToArgvW
+        CommandLineToArgvW.argtypes = [LPCWSTR, POINTER(c_int)]
+        CommandLineToArgvW.restype = POINTER(LPWSTR)
 
-def SHA1(message):
-    ctx = hashlib.sha1()
-    ctx.update(message)
-    return ctx.digest()
+        cmd = GetCommandLineW()
+        argc = c_int(0)
+        argv = CommandLineToArgvW(cmd, byref(argc))
+        if argc.value > 0:
+            # Remove Python executable and commands if present
+            start = argc.value - len(sys.argv)
+            return [argv[i] for i in
+                    xrange(start, argc.value)]
+        # if we don't have any arguments at all, just pass back script name
+        # this should never happen
+        return [u"mobidedrm.py"]
+    else:
+        argvencoding = sys.stdin.encoding
+        if argvencoding == None:
+            argvencoding = "utf-8"
+        return [arg if (type(arg) == unicode) else unicode(arg,argvencoding) for arg in sys.argv]
+
+# cleanup unicode filenames
+# borrowed from calibre from calibre/src/calibre/__init__.py
+# added in removal of control (<32) chars
+# and removal of . at start and end
+# and with some (heavily edited) code from Paul Durrant's kindlenamer.py
+def cleanup_name(name):
+    # substitute filename unfriendly characters
+    name = name.replace(u"<",u"[").replace(u">",u"]").replace(u" : ",u" – ").replace(u": ",u" – ").replace(u":",u"—").replace(u"/",u"_").replace(u"\\",u"_").replace(u"|",u"_").replace(u"\"",u"\'").replace(u"*",u"_").replace(u"?",u"")
+    # delete control characters
+    name = u"".join(char for char in name if ord(char)>=32)
+    # white space to single space, delete leading and trailing while space
+    name = re.sub(ur"\s", u" ", name).strip()
+    # remove leading dots
+    while len(name)>0 and name[0] == u".":
+        name = name[1:]
+    # remove trailing dots (Windows doesn't like them)
+    if name.endswith(u'.'):
+        name = name[:-1]
+    return name
+
+# must be passed unicode
+def unescape(text):
+    def fixup(m):
+        text = m.group(0)
+        if text[:2] == u"&#":
+            # character reference
+            try:
+                if text[:3] == u"&#x":
+                    return unichr(int(text[3:-1], 16))
+                else:
+                    return unichr(int(text[2:-1]))
+            except ValueError:
+                pass
+        else:
+            # named entity
+            try:
+                text = unichr(htmlentitydefs.name2codepoint[text[1:-1]])
+            except KeyError:
+                pass
+        return text # leave as is
+    return re.sub(u"&#?\w+;", fixup, text)
+
+def GetDecryptedBook(infile, kDatabases, serials, pids, starttime = time.time()):
+    # handle the obvious cases at the beginning
+    if not os.path.isfile(infile):
+        raise DrmException(u"Input file does not exist.")
+
+    mobi = True
+    magic3 = open(infile,'rb').read(3)
+    if magic3 == 'TPZ':
+        mobi = False
+
+    if mobi:
+        mb = mobidedrm.MobiBook(infile)
+    else:
+        mb = topazextract.TopazBook(infile)
+
+    bookname = unescape(mb.getBookTitle())
+    print u"Decrypting {1} ebook: {0}".format(bookname, mb.getBookType())
+
+    # copy list of pids
+    totalpids = list(pids)
+    # extend PID list with book-specific PIDs
+    md1, md2 = mb.getPIDMetaInfo()
+    totalpids.extend(kgenpids.getPidList(md1, md2, serials, kDatabases))
+    print u"Found {1:d} keys to try after {0:.1f} seconds".format(time.time()-starttime, len(totalpids))
+
+    try:
+        mb.processBook(totalpids)
+    except:
+        mb.cleanup
+        raise
+
+    print u"Decryption succeeded after {0:.1f} seconds".format(time.time()-starttime)
+    return mb
 
 
-# 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
+# kDatabaseFiles is a list of files created by kindlekey
+def decryptBook(infile, outdir, kDatabaseFiles, serials, pids):
+    starttime = time.time()
+    kDatabases = []
+    for dbfile in kDatabaseFiles:
+        kindleDatabase = {}
+        try:
+            with open(dbfile, 'r') as keyfilein:
+                kindleDatabase = json.loads(keyfilein.read())
+            kDatabases.append([dbfile,kindleDatabase])
+        except Exception, e:
+            print u"Error getting database from file {0:s}: {1:s}".format(dbfile,e)
+            traceback.print_exc()
 
-# 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
+
+    try:
+        book = GetDecryptedBook(infile, kDatabases, serials, pids, starttime)
+    except Exception, e:
+        print u"Error decrypting book after {1:.1f} seconds: {0}".format(e.args[0],time.time()-starttime)
+        traceback.print_exc()
+        return 1
+
+    # if we're saving to the same folder as the original, use file name_
+    # if to a different folder, use book name
+    if os.path.normcase(os.path.normpath(outdir)) == os.path.normcase(os.path.normpath(os.path.dirname(infile))):
+        outfilename = os.path.splitext(os.path.basename(infile))[0]
+    else:
+        outfilename = cleanup_name(book.getBookTitle())
+
+    # avoid excessively long file names
+    if len(outfilename)>150:
+        outfilename = outfilename[:150]
+
+    outfilename = outfilename+u"_nodrm"
+    outfile = os.path.join(outdir, outfilename + book.getBookExtension())
+
+    book.getFile(outfile)
+    print u"Saved decrypted book {1:s} after {0:.1f} seconds".format(time.time()-starttime, outfilename)
+
+    if book.getBookType()==u"Topaz":
+        zipname = os.path.join(outdir, outfilename + u"_SVG.zip")
+        book.getSVGZip(zipname)
+        print u"Saved SVG ZIP Archive for {1:s} after {0:.1f} seconds".format(time.time()-starttime, outfilename)
+
+    # remove internal temporary directory of Topaz pieces
+    book.cleanup()
+    return 0
+
+
+def usage(progname):
+    print u"Removes DRM protection from Mobipocket, Amazon KF8, Amazon Print Replica and Amazon Topaz ebooks"
+    print u"Usage:"
+    print u"    {0} [-k <kindle.k4i>] [-p <comma separated PIDs>] [-s <comma separated Kindle serial numbers>] [ -a <AmazonSecureStorage.xml> ] <infile> <outdir>".format(progname)
 
 #
-# PID generation routines
+# Main
 #
+def cli_main():
+    argv=unicode_argv()
+    progname = os.path.basename(argv[0])
+    print u"K4MobiDeDrm v{0}.\nCopyright © 2008-2013 The Dark Reverser et al.".format(__version__)
 
-# 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
+    try:
+        opts, args = getopt.getopt(argv[1:], "k:p:s:a:")
+    except getopt.GetoptError, err:
+        print u"Error in options or arguments: {0}".format(err.args[0])
+        usage(progname)
+        sys.exit(2)
+    if len(args)<2:
+        usage(progname)
+        sys.exit(2)
 
-# 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 getKindlePids(rec209, token, serialnum):
-    pids=[]
-
-    if isinstance(serialnum,unicode):
-        serialnum = serialnum.encode('ascii')
-
-    # Compute book PID
-    pidHash = SHA1(serialnum+rec209+token)
-    bookPID = encodePID(pidHash)
-    bookPID = checksumPid(bookPID)
-    pids.append(bookPID)
-
-    # compute fixed pid for old pre 2.5 firmware update pid as well
-    kindlePID = pidFromSerial(serialnum, 7) + "*"
-    kindlePID = checksumPid(kindlePID)
-    pids.append(kindlePID)
-
-    return pids
-
-
-# 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(rec209, token, kindleDatabase):
-    global charMap1
+    infile = args[0]
+    outdir = args[1]
+    kDatabaseFiles = []
+    serials = []
     pids = []
 
-    try:
-        # Get the Mazama Random number
-        MazamaRandomNumber = (kindleDatabase[1])['MazamaRandomNumber'].decode('hex').encode('ascii')
+    for o, a in opts:
+        if o == "-k":
+            if a == None :
+                raise DrmException("Invalid parameter for -k")
+            kDatabaseFiles.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(',')
+        if o == '-a':
+            if a == None:
+                continue
+            serials.extend(android.get_serials(a))
+    serials.extend(android.get_serials())
 
-        # Get the kindle account token
-        kindleAccountToken = (kindleDatabase[1])['kindle.account.tokens'].decode('hex').encode('ascii')
+    # try with built in Kindle Info files if not on Linux
+    k4 = not sys.platform.startswith('linux')
 
-        # Get the IDString used to decode the Kindle Info file
-        IDString = (kindleDatabase[1])['IDString'].decode('hex').encode('ascii')
+    return decryptBook(infile, outdir, kDatabaseFiles, serials, pids)
 
-        # Get the UserName stored when the Kindle Info file was decoded
-        UserName = (kindleDatabase[1])['UserName'].decode('hex').encode('ascii')
 
-    except KeyError:
-        print u"Keys not found in the database {0}.".format(kindleDatabase[0])
-        return pids
-
-    # Get the ID string used
-    encodedIDString = encodeHash(IDString,charMap1)
-
-    # Get the current user name
-    encodedUsername = encodeHash(UserName,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)
-    pids.append(devicePID)
-
-    # Compute book PIDs
-
-    # book pid
-    pidHash = SHA1(DSN+kindleAccountToken+rec209+token)
-    bookPID = encodePID(pidHash)
-    bookPID = checksumPid(bookPID)
-    pids.append(bookPID)
-
-    # variant 1
-    pidHash = SHA1(kindleAccountToken+rec209+token)
-    bookPID = encodePID(pidHash)
-    bookPID = checksumPid(bookPID)
-    pids.append(bookPID)
-
-    # variant 2
-    pidHash = SHA1(DSN+rec209+token)
-    bookPID = encodePID(pidHash)
-    bookPID = checksumPid(bookPID)
-    pids.append(bookPID)
-
-    return pids
-
-def getPidList(md1, md2, serials=[], kDatabases=[]):
-    pidlst = []
-
-    if kDatabases is None:
-        kDatabases = []
-    if serials is None:
-        serials = []
-
-    for kDatabase in kDatabases:
-        try:
-            pidlst.extend(getK4Pids(md1, md2, kDatabase))
-        except Exception, e:
-            print u"Error getting PIDs from database {0}: {1}".format(kDatabase[0],e.args[0])
-            traceback.print_exc()
-
-    for serialnum in serials:
-        try:
-            pidlst.extend(getKindlePids(md1, md2, serialnum))
-        except Exception, e:
-            print u"Error getting PIDs from serial number {0}: {1}".format(serialnum ,e.args[0])
-            traceback.print_exc()
-
-    return pidlst
+if __name__ == '__main__':
+    sys.stdout=SafeUnbuffered(sys.stdout)
+    sys.stderr=SafeUnbuffered(sys.stderr)
+    sys.exit(cli_main())

DeDRM_calibre_plugin/DeDRM_plugin/mobidedrm.py

@@ -1,89 +1,541 @@
 #!/usr/bin/env python
-# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
+# -*- coding: utf-8 -*-
 
-# implement just enough of des from openssl to make erdr2pml.py happy
+# mobidedrm.py, version 0.38
+# Copyright © 2008 The Dark Reverser
+#
+# Modified 2008–2012 by some_updates, DiapDealer and Apprentice Alf
 
-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
+# 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
+#  0.38 - Unicode used wherever possible, cope with absent alfcrypto
+#  0.39 - Fixed problem with TEXtREAd and getBookType interface
+#  0.40 - moved unicode_argv call inside main for Windows DeDRM compatibility
+#  0.41 - Fixed potential unicode problem in command line calls
 
-    if sys.platform.startswith('win'):
-        libcrypto = find_library('libeay32')
+
+__version__ = u"0.41"
+
+import sys
+import os
+import struct
+import binascii
+try:
+    from alfcrypto import Pukall_Cipher
+except:
+    print u"AlfCrypto not found. Using python PC1 implementation."
+
+# Wrap a stream so that output gets flushed immediately
+# and also make sure that any unicode strings get
+# encoded using "replace" before writing them.
+class SafeUnbuffered:
+    def __init__(self, stream):
+        self.stream = stream
+        self.encoding = stream.encoding
+        if self.encoding == None:
+            self.encoding = "utf-8"
+    def write(self, data):
+        if isinstance(data,unicode):
+            data = data.encode(self.encoding,"replace")
+        self.stream.write(data)
+        self.stream.flush()
+    def __getattr__(self, attr):
+        return getattr(self.stream, attr)
+
+iswindows = sys.platform.startswith('win')
+isosx = sys.platform.startswith('darwin')
+
+def unicode_argv():
+    if iswindows:
+        # Uses shell32.GetCommandLineArgvW to get sys.argv as a list of Unicode
+        # strings.
+
+        # Versions 2.x of Python don't support Unicode in sys.argv on
+        # Windows, with the underlying Windows API instead replacing multi-byte
+        # characters with '?'.
+
+
+        from ctypes import POINTER, byref, cdll, c_int, windll
+        from ctypes.wintypes import LPCWSTR, LPWSTR
+
+        GetCommandLineW = cdll.kernel32.GetCommandLineW
+        GetCommandLineW.argtypes = []
+        GetCommandLineW.restype = LPCWSTR
+
+        CommandLineToArgvW = windll.shell32.CommandLineToArgvW
+        CommandLineToArgvW.argtypes = [LPCWSTR, POINTER(c_int)]
+        CommandLineToArgvW.restype = POINTER(LPWSTR)
+
+        cmd = GetCommandLineW()
+        argc = c_int(0)
+        argv = CommandLineToArgvW(cmd, byref(argc))
+        if argc.value > 0:
+            # Remove Python executable and commands if present
+            start = argc.value - len(sys.argv)
+            return [argv[i] for i in
+                    xrange(start, argc.value)]
+        # if we don't have any arguments at all, just pass back script name
+        # this should never happen
+        return [u"mobidedrm.py"]
     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])
+        argvencoding = sys.stdin.encoding
+        if argvencoding == None:
+            argvencoding = 'utf-8'
+        return [arg if (type(arg) == unicode) else unicode(arg,argvencoding) for arg in sys.argv]
 
 
-    class DES(object):
-        def __init__(self, key):
-            if len(key) != 8 :
-                raise Exception('DES improper key used')
+class DrmException(Exception):
+    pass
+
+
+#
+# MobiBook Utility Routines
+#
+
+# Implementation of Pukall Cipher 1
+def PC1(key, src, decryption=True):
+    # if we can get it from alfcrypto, use that
+    try:
+        return Pukall_Cipher().PC1(key,src,decryption)
+    except NameError:
+        pass
+    except TypeError:
+        pass
+
+    # use slow python version, since Pukall_Cipher didn't load
+    sum1 = 0;
+    sum2 = 0;
+    keyXorVal = 0;
+    if len(key)!=16:
+         DrmException (u"PC1: Bad key length")
+    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 cleanup(self):
+        # to match function in Topaz book
+        pass
+
+    def __init__(self, infile):
+        print u"MobiDeDrm v{0:s}.\nCopyright © 2008-2012 The Dark Reverser et al.".format(__version__)
+
+        try:
+            from alfcrypto import Pukall_Cipher
+        except:
+            print u"AlfCrypto not found. Using python PC1 implementation."
+
+        # 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(u"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])
+
+        # det default values before PalmDoc test
+        self.print_replica = False
+        self.extra_data_flags = 0
+        self.meta_array = {}
+        self.mobi_length = 0
+        self.mobi_codepage = 1252
+        self.mobi_version = -1
+
+        if self.magic == 'TEXtREAd':
+            print u"PalmDoc format book detected."
             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
+        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 u"MOBI header version {0:d}, header length {1:d}".format(self.mobi_version, self.mobi_length)
+        if (self.mobi_length >= 0xE4) and (self.mobi_version >= 5):
+            self.extra_data_flags, = struct.unpack('>H', self.sect[0xF2:0xF4])
+            print u"Extra Data Flags: {0:d}".format(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
+        try:
+            exth_flag, = struct.unpack('>L', self.sect[0x80:0x84])
+            exth = ''
+            if exth_flag & 0x40:
+                exth = self.sect[16 + self.mobi_length:]
+            if (len(exth) >= 12) 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:
+            pass
+
+    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)
+
+    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 getFile(self, outpath):
+        file(outpath,'wb').write(self.mobi_data)
+
+    def getBookType(self):
+        if self.print_replica:
+            return u"Print Replica"
+        if self.mobi_version >= 8:
+            return u"Kindle Format 8"
+        if self.mobi_version >= 0:
+            return u"Mobipocket {0:d}".format(self.mobi_version)
+        return u"PalmDoc"
+
+    def getBookExtension(self):
+        if self.print_replica:
+            return u".azw4"
+        if self.mobi_version >= 8:
+            return u".azw3"
+        return u".mobi"
+
+    def processBook(self, pidlist):
+        crypto_type, = struct.unpack('>H', self.sect[0xC:0xC+2])
+        print u"Crypto Type is: {0:d}".format(crypto_type)
+        self.crypto_type = crypto_type
+        if crypto_type == 0:
+            print u"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(u"Cannot decode unknown Mobipocket encryption type {0:d}".format(crypto_type))
+        if 406 in self.meta_array:
+            data406 = self.meta_array[406]
+            val406, = struct.unpack('>Q',data406)
+            if val406 != 0:
+                raise DrmException(u"Cannot decode library or rented ebooks.")
+
+        goodpids = []
+        for pid in pidlist:
+            if len(pid)==10:
+                if checksumPid(pid[0:-2]) != pid:
+                    print u"Warning: PID {0} has incorrect checksum, should have been {1}".format(pid,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(u"Encryption not initialised. 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(u"No key found in {0:d} keys tried.".format(len(goodpids)))
+            # 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 u"File has default encryption, no specific key needed."
+        else:
+            print u"File is encoded with PID {0}.".format(checksumPid(pid))
+
+        # clear the crypto type
+        self.patchSection(0, "\0" * 2, 0xC)
+
+        # decrypt sections
+        print u"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 u".",
+            # 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 u"done"
+        return
+
+def getUnencryptedBook(infile,pidlist):
+    if not os.path.isfile(infile):
+        raise DrmException(u"Input File Not Found.")
+    book = MobiBook(infile)
+    book.processBook(pidlist)
+    return book.mobi_data
+
+
+def cli_main():
+    argv=unicode_argv()
+    progname = os.path.basename(argv[0])
+    if len(argv)<3 or len(argv)>4:
+        print u"MobiDeDrm v{0}.\nCopyright © 2008-2012 The Dark Reverser et al.".format(__version__)
+        print u"Removes protection from Kindle/Mobipocket, Kindle/KF8 and Kindle/Print Replica ebooks"
+        print u"Usage:"
+        print u"    {0} <infile> <outfile> [<Comma separated list of PIDs to try>]".format(progname)
+        return 1
+    else:
+        infile = argv[1]
+        outfile = argv[2]
+        if len(argv) is 4:
+            pidlist = argv[3].split(',')
+        else:
+            pidlist = []
+        try:
+            stripped_file = getUnencryptedBook(infile, pidlist)
+            file(outfile, 'wb').write(stripped_file)
+        except DrmException, e:
+            print u"MobiDeDRM v{0} Error: {0:s}".format(__version__,e.args[0])
+            return 1
+    return 0
+
+
+if __name__ == '__main__':
+    sys.stdout=SafeUnbuffered(sys.stdout)
+    sys.stderr=SafeUnbuffered(sys.stderr)
+    sys.exit(cli_main())

DeDRM_calibre_plugin/DeDRM_plugin/openssl_des.py

@@ -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 Exception('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

DeDRM_calibre_plugin/DeDRM_plugin/plugin-import-name-dedrm.txt

@@ -1,292 +0,0 @@
-#!/usr/bin/env python
-# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
-
-from __future__ import with_statement
-__license__ = 'GPL v3'
-
-# Standard Python modules.
-import os, sys, re, hashlib
-import json
-import traceback
-
-from calibre.utils.config import dynamic, config_dir, JSONConfig
-from calibre_plugins.dedrm.__init__ import PLUGIN_NAME, PLUGIN_VERSION
-from calibre.constants import iswindows, isosx
-
-class DeDRM_Prefs():
-    def __init__(self):
-        JSON_PATH = os.path.join(u"plugins", PLUGIN_NAME.strip().lower().replace(' ', '_') + '.json')
-        self.dedrmprefs = JSONConfig(JSON_PATH)
-
-        self.dedrmprefs.defaults['configured'] = False
-        self.dedrmprefs.defaults['bandnkeys'] = {}
-        self.dedrmprefs.defaults['adeptkeys'] = {}
-        self.dedrmprefs.defaults['ereaderkeys'] = {}
-        self.dedrmprefs.defaults['kindlekeys'] = {}
-        self.dedrmprefs.defaults['pids'] = []
-        self.dedrmprefs.defaults['serials'] = []
-        self.dedrmprefs.defaults['adobewineprefix'] = ""
-        self.dedrmprefs.defaults['kindlewineprefix'] = ""
-
-        # initialise
-        # we must actually set the prefs that are dictionaries and lists
-        # to empty dictionaries and lists, otherwise we are unable to add to them
-        # as then it just adds to the (memory only) dedrmprefs.defaults versions!
-        if self.dedrmprefs['bandnkeys'] == {}:
-            self.dedrmprefs['bandnkeys'] = {}
-        if self.dedrmprefs['adeptkeys'] == {}:
-            self.dedrmprefs['adeptkeys'] = {}
-        if self.dedrmprefs['ereaderkeys'] == {}:
-            self.dedrmprefs['ereaderkeys'] = {}
-        if self.dedrmprefs['kindlekeys'] == {}:
-            self.dedrmprefs['kindlekeys'] = {}
-        if self.dedrmprefs['pids'] == []:
-            self.dedrmprefs['pids'] = []
-        if self.dedrmprefs['serials'] == []:
-            self.dedrmprefs['serials'] = []
-
-    def __getitem__(self,kind = None):
-        if kind is not None:
-            return self.dedrmprefs[kind]
-        return self.dedrmprefs
-
-    def set(self, kind, value):
-        self.dedrmprefs[kind] = value
-
-    def writeprefs(self,value = True):
-        self.dedrmprefs['configured'] = value
-
-    def addnamedvaluetoprefs(self, prefkind, keyname, keyvalue):
-        try:
-            if keyvalue not in self.dedrmprefs[prefkind].values():
-                # ensure that the keyname is unique
-                # by adding a number (starting with 2) to the name if it is not
-                namecount = 1
-                newname = keyname
-                while newname in self.dedrmprefs[prefkind]:
-                    namecount += 1
-                    newname = "{0:s}_{1:d}".format(keyname,namecount)
-                # add to the preferences
-                self.dedrmprefs[prefkind][newname] = keyvalue
-                return (True, newname)
-        except:
-            traceback.print_exc()
-            pass
-        return (False, keyname)
-
-    def addvaluetoprefs(self, prefkind, prefsvalue):
-        # ensure the keyvalue isn't already in the preferences
-        try:
-            if prefsvalue not in self.dedrmprefs[prefkind]:
-                self.dedrmprefs[prefkind].append(prefsvalue)
-                return True
-        except:
-            traceback.print_exc()
-        return False
-
-
-def convertprefs(always = False):
-
-    def parseIgnobleString(keystuff):
-        from calibre_plugins.dedrm.ignoblekeygen import generate_key
-        userkeys = []
-        ar = keystuff.split(':')
-        for keystring in ar:
-            try:
-                name, ccn = keystring.split(',')
-                # Generate Barnes & Noble EPUB user key from name and credit card number.
-                keyname = u"{0}_{1}".format(name.strip(),ccn.strip()[-4:])
-                keyvalue = generate_key(name, ccn)
-                userkeys.append([keyname,keyvalue])
-            except Exception, e:
-                traceback.print_exc()
-                print e.args[0]
-                pass
-        return userkeys
-
-    def parseeReaderString(keystuff):
-        from calibre_plugins.dedrm.erdr2pml import getuser_key
-        userkeys = []
-        ar = keystuff.split(':')
-        for keystring in ar:
-            try:
-                name, cc = keystring.split(',')
-                # Generate eReader user key from name and credit card number.
-                keyname = u"{0}_{1}".format(name.strip(),cc.strip()[-4:])
-                keyvalue = getuser_key(name,cc).encode('hex')
-                userkeys.append([keyname,keyvalue])
-            except Exception, e:
-                traceback.print_exc()
-                print e.args[0]
-                pass
-        return userkeys
-
-    def parseKindleString(keystuff):
-        pids = []
-        serials = []
-        ar = keystuff.split(',')
-        for keystring in ar:
-            keystring = str(keystring).strip().replace(" ","")
-            if len(keystring) == 10 or len(keystring) == 8 and keystring not in pids:
-                pids.append(keystring)
-            elif len(keystring) == 16 and keystring[0] == 'B' and keystring not in serials:
-                serials.append(keystring)
-        return (pids,serials)
-
-    def getConfigFiles(extension, encoding = None):
-        # get any files with extension 'extension' in the config dir
-        userkeys = []
-        files = [f for f in os.listdir(config_dir) if f.endswith(extension)]
-        for filename in files:
-            try:
-                fpath = os.path.join(config_dir, filename)
-                key = os.path.splitext(filename)[0]
-                value = open(fpath, 'rb').read()
-                if encoding is not None:
-                    value = value.encode(encoding)
-                userkeys.append([key,value])
-            except:
-                traceback.print_exc()
-                pass
-        return userkeys
-
-    dedrmprefs = DeDRM_Prefs()
-
-    if (not always) and dedrmprefs['configured']:
-        # We've already converted old preferences,
-        # and we're not being forced to do it again, so just return
-        return
-
-
-    print u"{0} v{1}: Importing configuration data from old DeDRM plugins".format(PLUGIN_NAME, PLUGIN_VERSION)
-
-    IGNOBLEPLUGINNAME = "Ignoble Epub DeDRM"
-    EREADERPLUGINNAME = "eReader PDB 2 PML"
-    OLDKINDLEPLUGINNAME = "K4PC, K4Mac, Kindle Mobi and Topaz DeDRM"
-
-    # get prefs from older tools
-    kindleprefs = JSONConfig(os.path.join(u"plugins", u"K4MobiDeDRM"))
-    ignobleprefs = JSONConfig(os.path.join(u"plugins", u"ignoble_epub_dedrm"))
-
-    # Handle the old ignoble plugin's customization string by converting the
-    # old string to stored keys... get that personal data out of plain sight.
-    from calibre.customize.ui import config
-    sc = config['plugin_customization']
-    val = sc.pop(IGNOBLEPLUGINNAME, None)
-    if val is not None:
-        print u"{0} v{1}: Converting old Ignoble plugin configuration string.".format(PLUGIN_NAME, PLUGIN_VERSION)
-        priorkeycount = len(dedrmprefs['bandnkeys'])
-        userkeys = parseIgnobleString(str(val))
-        for keypair in userkeys:
-            name = keypair[0]
-            value = keypair[1]
-            dedrmprefs.addnamedvaluetoprefs('bandnkeys', name, value)
-        addedkeycount = len(dedrmprefs['bandnkeys'])-priorkeycount
-        print u"{0} v{1}: {2:d} Barnes and Noble {3} imported from old Ignoble plugin configuration string".format(PLUGIN_NAME, PLUGIN_VERSION, addedkeycount, u"key" if addedkeycount==1 else u"keys")
-    # Make the json write all the prefs to disk
-    dedrmprefs.writeprefs(False)
-
-    # Handle the old eReader plugin's customization string by converting the
-    # old string to stored keys... get that personal data out of plain sight.
-    val = sc.pop(EREADERPLUGINNAME, None)
-    if val is not None:
-        print u"{0} v{1}: Converting old eReader plugin configuration string.".format(PLUGIN_NAME, PLUGIN_VERSION)
-        priorkeycount = len(dedrmprefs['ereaderkeys'])
-        userkeys = parseeReaderString(str(val))
-        for keypair in userkeys:
-            name = keypair[0]
-            value = keypair[1]
-            dedrmprefs.addnamedvaluetoprefs('ereaderkeys', name, value)
-        addedkeycount = len(dedrmprefs['ereaderkeys'])-priorkeycount
-        print u"{0} v{1}: {2:d} eReader {3} imported from old eReader plugin configuration string".format(PLUGIN_NAME, PLUGIN_VERSION, addedkeycount, u"key" if addedkeycount==1 else u"keys")
-    # Make the json write all the prefs to disk
-    dedrmprefs.writeprefs(False)
-
-    # get old Kindle plugin configuration string
-    val = sc.pop(OLDKINDLEPLUGINNAME, None)
-    if val is not None:
-        print u"{0} v{1}: Converting old Kindle plugin configuration string.".format(PLUGIN_NAME, PLUGIN_VERSION)
-        priorpidcount = len(dedrmprefs['pids'])
-        priorserialcount = len(dedrmprefs['serials'])
-        pids, serials = parseKindleString(val)
-        for pid in pids:
-            dedrmprefs.addvaluetoprefs('pids',pid)
-        for serial in serials:
-            dedrmprefs.addvaluetoprefs('serials',serial)
-        addedpidcount = len(dedrmprefs['pids']) - priorpidcount
-        addedserialcount = len(dedrmprefs['serials']) - priorserialcount
-        print u"{0} v{1}: {2:d} {3} and {4:d} {5} imported from old Kindle plugin configuration string.".format(PLUGIN_NAME, PLUGIN_VERSION, addedpidcount, u"PID" if addedpidcount==1 else u"PIDs", addedserialcount, u"serial number" if addedserialcount==1 else u"serial numbers")
-    # Make the json write all the prefs to disk
-    dedrmprefs.writeprefs(False)
-
-    # copy the customisations back into calibre preferences, as we've now removed the nasty plaintext
-    config['plugin_customization'] = sc
-
-    # get any .b64 files in the config dir
-    priorkeycount = len(dedrmprefs['bandnkeys'])
-    bandnfilekeys = getConfigFiles('.b64')
-    for keypair in bandnfilekeys:
-        name = keypair[0]
-        value = keypair[1]
-        dedrmprefs.addnamedvaluetoprefs('bandnkeys', name, value)
-    addedkeycount = len(dedrmprefs['bandnkeys'])-priorkeycount
-    if addedkeycount > 0:
-        print u"{0} v{1}: {2:d} Barnes and Noble {3} imported from config folder.".format(PLUGIN_NAME, PLUGIN_VERSION, addedkeycount, u"key file" if addedkeycount==1 else u"key files")
-    # Make the json write all the prefs to disk
-    dedrmprefs.writeprefs(False)
-
-    # get any .der files in the config dir
-    priorkeycount = len(dedrmprefs['adeptkeys'])
-    adeptfilekeys = getConfigFiles('.der','hex')
-    for keypair in adeptfilekeys:
-        name = keypair[0]
-        value = keypair[1]
-        dedrmprefs.addnamedvaluetoprefs('adeptkeys', name, value)
-    addedkeycount = len(dedrmprefs['adeptkeys'])-priorkeycount
-    if addedkeycount > 0:
-        print u"{0} v{1}: {2:d} Adobe Adept {3} imported from config folder.".format(PLUGIN_NAME, PLUGIN_VERSION, addedkeycount, u"keyfile" if addedkeycount==1 else u"keyfiles")
-    # Make the json write all the prefs to disk
-    dedrmprefs.writeprefs(False)
-
-    # get ignoble json prefs
-    if 'keys' in ignobleprefs:
-        priorkeycount = len(dedrmprefs['bandnkeys'])
-        for name in ignobleprefs['keys']:
-            value = ignobleprefs['keys'][name]
-            dedrmprefs.addnamedvaluetoprefs('bandnkeys', name, value)
-        addedkeycount = len(dedrmprefs['bandnkeys']) - priorkeycount
-        # no need to delete old prefs, since they contain no recoverable private data
-        if addedkeycount > 0:
-            print u"{0} v{1}: {2:d} Barnes and Noble {3} imported from Ignoble plugin preferences.".format(PLUGIN_NAME, PLUGIN_VERSION, addedkeycount, u"key" if addedkeycount==1 else u"keys")
-    # Make the json write all the prefs to disk
-    dedrmprefs.writeprefs(False)
-
-    # get kindle json prefs
-    priorpidcount = len(dedrmprefs['pids'])
-    priorserialcount = len(dedrmprefs['serials'])
-    if 'pids' in kindleprefs:
-        pids, serials = parseKindleString(kindleprefs['pids'])
-        for pid in pids:
-            dedrmprefs.addvaluetoprefs('pids',pid)
-    if 'serials' in kindleprefs:
-        pids, serials = parseKindleString(kindleprefs['serials'])
-        for serial in serials:
-            dedrmprefs.addvaluetoprefs('serials',serial)
-    addedpidcount = len(dedrmprefs['pids']) - priorpidcount
-    if addedpidcount > 0:
-        print u"{0} v{1}: {2:d} {3} imported from Kindle plugin preferences".format(PLUGIN_NAME, PLUGIN_VERSION, addedpidcount, u"PID" if addedpidcount==1 else u"PIDs")
-    addedserialcount = len(dedrmprefs['serials']) - priorserialcount
-    if addedserialcount > 0:
-        print u"{0} v{1}: {2:d} {3} imported from Kindle plugin preferences".format(PLUGIN_NAME, PLUGIN_VERSION, addedserialcount, u"serial number" if addedserialcount==1 else u"serial numbers")
-    try:
-        if 'wineprefix' in kindleprefs and kindleprefs['wineprefix'] != "":
-            dedrmprefs.set('adobewineprefix',kindleprefs['wineprefix'])
-            dedrmprefs.set('kindlewineprefix',kindleprefs['wineprefix'])
-            print u"{0} v{1}: WINEPREFIX ‘(2)’ imported from Kindle plugin preferences".format(PLUGIN_NAME, PLUGIN_VERSION, kindleprefs['wineprefix'])
-    except:
-        traceback.print_exc()
-
-
-    # Make the json write all the prefs to disk
-    dedrmprefs.writeprefs()
-    print u"{0} v{1}: Finished setting up configuration data.".format(PLUGIN_NAME, PLUGIN_VERSION)

DeDRM_calibre_plugin/DeDRM_plugin/prefs.py

@@ -0,0 +1,292 @@
+#!/usr/bin/env python
+# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
+
+from __future__ import with_statement
+__license__ = 'GPL v3'
+
+# Standard Python modules.
+import os, sys, re, hashlib
+import json
+import traceback
+
+from calibre.utils.config import dynamic, config_dir, JSONConfig
+from calibre_plugins.dedrm.__init__ import PLUGIN_NAME, PLUGIN_VERSION
+from calibre.constants import iswindows, isosx
+
+class DeDRM_Prefs():
+    def __init__(self):
+        JSON_PATH = os.path.join(u"plugins", PLUGIN_NAME.strip().lower().replace(' ', '_') + '.json')
+        self.dedrmprefs = JSONConfig(JSON_PATH)
+
+        self.dedrmprefs.defaults['configured'] = False
+        self.dedrmprefs.defaults['bandnkeys'] = {}
+        self.dedrmprefs.defaults['adeptkeys'] = {}
+        self.dedrmprefs.defaults['ereaderkeys'] = {}
+        self.dedrmprefs.defaults['kindlekeys'] = {}
+        self.dedrmprefs.defaults['pids'] = []
+        self.dedrmprefs.defaults['serials'] = []
+        self.dedrmprefs.defaults['adobewineprefix'] = ""
+        self.dedrmprefs.defaults['kindlewineprefix'] = ""
+
+        # initialise
+        # we must actually set the prefs that are dictionaries and lists
+        # to empty dictionaries and lists, otherwise we are unable to add to them
+        # as then it just adds to the (memory only) dedrmprefs.defaults versions!
+        if self.dedrmprefs['bandnkeys'] == {}:
+            self.dedrmprefs['bandnkeys'] = {}
+        if self.dedrmprefs['adeptkeys'] == {}:
+            self.dedrmprefs['adeptkeys'] = {}
+        if self.dedrmprefs['ereaderkeys'] == {}:
+            self.dedrmprefs['ereaderkeys'] = {}
+        if self.dedrmprefs['kindlekeys'] == {}:
+            self.dedrmprefs['kindlekeys'] = {}
+        if self.dedrmprefs['pids'] == []:
+            self.dedrmprefs['pids'] = []
+        if self.dedrmprefs['serials'] == []:
+            self.dedrmprefs['serials'] = []
+
+    def __getitem__(self,kind = None):
+        if kind is not None:
+            return self.dedrmprefs[kind]
+        return self.dedrmprefs
+
+    def set(self, kind, value):
+        self.dedrmprefs[kind] = value
+
+    def writeprefs(self,value = True):
+        self.dedrmprefs['configured'] = value
+
+    def addnamedvaluetoprefs(self, prefkind, keyname, keyvalue):
+        try:
+            if keyvalue not in self.dedrmprefs[prefkind].values():
+                # ensure that the keyname is unique
+                # by adding a number (starting with 2) to the name if it is not
+                namecount = 1
+                newname = keyname
+                while newname in self.dedrmprefs[prefkind]:
+                    namecount += 1
+                    newname = "{0:s}_{1:d}".format(keyname,namecount)
+                # add to the preferences
+                self.dedrmprefs[prefkind][newname] = keyvalue
+                return (True, newname)
+        except:
+            traceback.print_exc()
+            pass
+        return (False, keyname)
+
+    def addvaluetoprefs(self, prefkind, prefsvalue):
+        # ensure the keyvalue isn't already in the preferences
+        try:
+            if prefsvalue not in self.dedrmprefs[prefkind]:
+                self.dedrmprefs[prefkind].append(prefsvalue)
+                return True
+        except:
+            traceback.print_exc()
+        return False
+
+
+def convertprefs(always = False):
+
+    def parseIgnobleString(keystuff):
+        from calibre_plugins.dedrm.ignoblekeygen import generate_key
+        userkeys = []
+        ar = keystuff.split(':')
+        for keystring in ar:
+            try:
+                name, ccn = keystring.split(',')
+                # Generate Barnes & Noble EPUB user key from name and credit card number.
+                keyname = u"{0}_{1}".format(name.strip(),ccn.strip()[-4:])
+                keyvalue = generate_key(name, ccn)
+                userkeys.append([keyname,keyvalue])
+            except Exception, e:
+                traceback.print_exc()
+                print e.args[0]
+                pass
+        return userkeys
+
+    def parseeReaderString(keystuff):
+        from calibre_plugins.dedrm.erdr2pml import getuser_key
+        userkeys = []
+        ar = keystuff.split(':')
+        for keystring in ar:
+            try:
+                name, cc = keystring.split(',')
+                # Generate eReader user key from name and credit card number.
+                keyname = u"{0}_{1}".format(name.strip(),cc.strip()[-4:])
+                keyvalue = getuser_key(name,cc).encode('hex')
+                userkeys.append([keyname,keyvalue])
+            except Exception, e:
+                traceback.print_exc()
+                print e.args[0]
+                pass
+        return userkeys
+
+    def parseKindleString(keystuff):
+        pids = []
+        serials = []
+        ar = keystuff.split(',')
+        for keystring in ar:
+            keystring = str(keystring).strip().replace(" ","")
+            if len(keystring) == 10 or len(keystring) == 8 and keystring not in pids:
+                pids.append(keystring)
+            elif len(keystring) == 16 and keystring[0] == 'B' and keystring not in serials:
+                serials.append(keystring)
+        return (pids,serials)
+
+    def getConfigFiles(extension, encoding = None):
+        # get any files with extension 'extension' in the config dir
+        userkeys = []
+        files = [f for f in os.listdir(config_dir) if f.endswith(extension)]
+        for filename in files:
+            try:
+                fpath = os.path.join(config_dir, filename)
+                key = os.path.splitext(filename)[0]
+                value = open(fpath, 'rb').read()
+                if encoding is not None:
+                    value = value.encode(encoding)
+                userkeys.append([key,value])
+            except:
+                traceback.print_exc()
+                pass
+        return userkeys
+
+    dedrmprefs = DeDRM_Prefs()
+
+    if (not always) and dedrmprefs['configured']:
+        # We've already converted old preferences,
+        # and we're not being forced to do it again, so just return
+        return
+
+
+    print u"{0} v{1}: Importing configuration data from old DeDRM plugins".format(PLUGIN_NAME, PLUGIN_VERSION)
+
+    IGNOBLEPLUGINNAME = "Ignoble Epub DeDRM"
+    EREADERPLUGINNAME = "eReader PDB 2 PML"
+    OLDKINDLEPLUGINNAME = "K4PC, K4Mac, Kindle Mobi and Topaz DeDRM"
+
+    # get prefs from older tools
+    kindleprefs = JSONConfig(os.path.join(u"plugins", u"K4MobiDeDRM"))
+    ignobleprefs = JSONConfig(os.path.join(u"plugins", u"ignoble_epub_dedrm"))
+
+    # Handle the old ignoble plugin's customization string by converting the
+    # old string to stored keys... get that personal data out of plain sight.
+    from calibre.customize.ui import config
+    sc = config['plugin_customization']
+    val = sc.pop(IGNOBLEPLUGINNAME, None)
+    if val is not None:
+        print u"{0} v{1}: Converting old Ignoble plugin configuration string.".format(PLUGIN_NAME, PLUGIN_VERSION)
+        priorkeycount = len(dedrmprefs['bandnkeys'])
+        userkeys = parseIgnobleString(str(val))
+        for keypair in userkeys:
+            name = keypair[0]
+            value = keypair[1]
+            dedrmprefs.addnamedvaluetoprefs('bandnkeys', name, value)
+        addedkeycount = len(dedrmprefs['bandnkeys'])-priorkeycount
+        print u"{0} v{1}: {2:d} Barnes and Noble {3} imported from old Ignoble plugin configuration string".format(PLUGIN_NAME, PLUGIN_VERSION, addedkeycount, u"key" if addedkeycount==1 else u"keys")
+    # Make the json write all the prefs to disk
+    dedrmprefs.writeprefs(False)
+
+    # Handle the old eReader plugin's customization string by converting the
+    # old string to stored keys... get that personal data out of plain sight.
+    val = sc.pop(EREADERPLUGINNAME, None)
+    if val is not None:
+        print u"{0} v{1}: Converting old eReader plugin configuration string.".format(PLUGIN_NAME, PLUGIN_VERSION)
+        priorkeycount = len(dedrmprefs['ereaderkeys'])
+        userkeys = parseeReaderString(str(val))
+        for keypair in userkeys:
+            name = keypair[0]
+            value = keypair[1]
+            dedrmprefs.addnamedvaluetoprefs('ereaderkeys', name, value)
+        addedkeycount = len(dedrmprefs['ereaderkeys'])-priorkeycount
+        print u"{0} v{1}: {2:d} eReader {3} imported from old eReader plugin configuration string".format(PLUGIN_NAME, PLUGIN_VERSION, addedkeycount, u"key" if addedkeycount==1 else u"keys")
+    # Make the json write all the prefs to disk
+    dedrmprefs.writeprefs(False)
+
+    # get old Kindle plugin configuration string
+    val = sc.pop(OLDKINDLEPLUGINNAME, None)
+    if val is not None:
+        print u"{0} v{1}: Converting old Kindle plugin configuration string.".format(PLUGIN_NAME, PLUGIN_VERSION)
+        priorpidcount = len(dedrmprefs['pids'])
+        priorserialcount = len(dedrmprefs['serials'])
+        pids, serials = parseKindleString(val)
+        for pid in pids:
+            dedrmprefs.addvaluetoprefs('pids',pid)
+        for serial in serials:
+            dedrmprefs.addvaluetoprefs('serials',serial)
+        addedpidcount = len(dedrmprefs['pids']) - priorpidcount
+        addedserialcount = len(dedrmprefs['serials']) - priorserialcount
+        print u"{0} v{1}: {2:d} {3} and {4:d} {5} imported from old Kindle plugin configuration string.".format(PLUGIN_NAME, PLUGIN_VERSION, addedpidcount, u"PID" if addedpidcount==1 else u"PIDs", addedserialcount, u"serial number" if addedserialcount==1 else u"serial numbers")
+    # Make the json write all the prefs to disk
+    dedrmprefs.writeprefs(False)
+
+    # copy the customisations back into calibre preferences, as we've now removed the nasty plaintext
+    config['plugin_customization'] = sc
+
+    # get any .b64 files in the config dir
+    priorkeycount = len(dedrmprefs['bandnkeys'])
+    bandnfilekeys = getConfigFiles('.b64')
+    for keypair in bandnfilekeys:
+        name = keypair[0]
+        value = keypair[1]
+        dedrmprefs.addnamedvaluetoprefs('bandnkeys', name, value)
+    addedkeycount = len(dedrmprefs['bandnkeys'])-priorkeycount
+    if addedkeycount > 0:
+        print u"{0} v{1}: {2:d} Barnes and Noble {3} imported from config folder.".format(PLUGIN_NAME, PLUGIN_VERSION, addedkeycount, u"key file" if addedkeycount==1 else u"key files")
+    # Make the json write all the prefs to disk
+    dedrmprefs.writeprefs(False)
+
+    # get any .der files in the config dir
+    priorkeycount = len(dedrmprefs['adeptkeys'])
+    adeptfilekeys = getConfigFiles('.der','hex')
+    for keypair in adeptfilekeys:
+        name = keypair[0]
+        value = keypair[1]
+        dedrmprefs.addnamedvaluetoprefs('adeptkeys', name, value)
+    addedkeycount = len(dedrmprefs['adeptkeys'])-priorkeycount
+    if addedkeycount > 0:
+        print u"{0} v{1}: {2:d} Adobe Adept {3} imported from config folder.".format(PLUGIN_NAME, PLUGIN_VERSION, addedkeycount, u"keyfile" if addedkeycount==1 else u"keyfiles")
+    # Make the json write all the prefs to disk
+    dedrmprefs.writeprefs(False)
+
+    # get ignoble json prefs
+    if 'keys' in ignobleprefs:
+        priorkeycount = len(dedrmprefs['bandnkeys'])
+        for name in ignobleprefs['keys']:
+            value = ignobleprefs['keys'][name]
+            dedrmprefs.addnamedvaluetoprefs('bandnkeys', name, value)
+        addedkeycount = len(dedrmprefs['bandnkeys']) - priorkeycount
+        # no need to delete old prefs, since they contain no recoverable private data
+        if addedkeycount > 0:
+            print u"{0} v{1}: {2:d} Barnes and Noble {3} imported from Ignoble plugin preferences.".format(PLUGIN_NAME, PLUGIN_VERSION, addedkeycount, u"key" if addedkeycount==1 else u"keys")
+    # Make the json write all the prefs to disk
+    dedrmprefs.writeprefs(False)
+
+    # get kindle json prefs
+    priorpidcount = len(dedrmprefs['pids'])
+    priorserialcount = len(dedrmprefs['serials'])
+    if 'pids' in kindleprefs:
+        pids, serials = parseKindleString(kindleprefs['pids'])
+        for pid in pids:
+            dedrmprefs.addvaluetoprefs('pids',pid)
+    if 'serials' in kindleprefs:
+        pids, serials = parseKindleString(kindleprefs['serials'])
+        for serial in serials:
+            dedrmprefs.addvaluetoprefs('serials',serial)
+    addedpidcount = len(dedrmprefs['pids']) - priorpidcount
+    if addedpidcount > 0:
+        print u"{0} v{1}: {2:d} {3} imported from Kindle plugin preferences".format(PLUGIN_NAME, PLUGIN_VERSION, addedpidcount, u"PID" if addedpidcount==1 else u"PIDs")
+    addedserialcount = len(dedrmprefs['serials']) - priorserialcount
+    if addedserialcount > 0:
+        print u"{0} v{1}: {2:d} {3} imported from Kindle plugin preferences".format(PLUGIN_NAME, PLUGIN_VERSION, addedserialcount, u"serial number" if addedserialcount==1 else u"serial numbers")
+    try:
+        if 'wineprefix' in kindleprefs and kindleprefs['wineprefix'] != "":
+            dedrmprefs.set('adobewineprefix',kindleprefs['wineprefix'])
+            dedrmprefs.set('kindlewineprefix',kindleprefs['wineprefix'])
+            print u"{0} v{1}: WINEPREFIX ‘(2)’ imported from Kindle plugin preferences".format(PLUGIN_NAME, PLUGIN_VERSION, kindleprefs['wineprefix'])
+    except:
+        traceback.print_exc()
+
+
+    # Make the json write all the prefs to disk
+    dedrmprefs.writeprefs()
+    print u"{0} v{1}: Finished setting up configuration data.".format(PLUGIN_NAME, PLUGIN_VERSION)

DeDRM_calibre_plugin/DeDRM_plugin/stylexml2css.py

@@ -178,7 +178,12 @@ class DocParser(object):
                             if val == "":
                                 val = 0
 
-                            if not ((attr == 'hang') and (int(val) == 0)) :
+                            if not ((attr == 'hang') and (int(val) == 0)):
+                                try:
+                                    f = float(val)
+                                except:
+                                    print "Warning: unrecognised val, ignoring"
+                                    val = 0
                                 pv = float(val)/scale
                                 cssargs[attr] = (self.attr_val_map[attr], pv)
                                 keep = True

DeDRM_calibre_plugin/DeDRM_plugin/subasyncio.py

@@ -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()

DeDRM_calibre_plugin/DeDRM_plugin/topazextract.py

@@ -1,148 +1,538 @@
 #!/usr/bin/env python
-# vim:ts=4:sw=4:softtabstop=4:smarttab:expandtab
+# -*- coding: utf-8 -*-
 
-import os, sys
-import signal
-import threading
-import subprocess
-from subprocess import Popen, PIPE, STDOUT
+# topazextract.py
+# Mostly written by some_updates based on code from many others
 
-# **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
+# Changelog
+#  4.9  - moved unicode_argv call inside main for Windows DeDRM compatibility
+#  5.0  - Fixed potential unicode problem with command line interface
 
-# create an asynchronous subprocess whose output can be collected in
-# a non-blocking manner
+__version__ = '5.0'
 
-# What a mess!  Have to use threads just to get non-blocking io
-# in a cross-platform manner
+import sys
+import os, csv, getopt
+import zlib, zipfile, tempfile, shutil
+import traceback
+from struct import pack
+from struct import unpack
+from alfcrypto import Topaz_Cipher
 
-# 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
+class SafeUnbuffered:
+    def __init__(self, stream):
+        self.stream = stream
+        self.encoding = stream.encoding
+        if self.encoding == None:
+            self.encoding = "utf-8"
     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()
+        if isinstance(data,unicode):
+            data = data.encode(self.encoding,"replace")
+        self.stream.write(data)
+        self.stream.flush()
+    def __getattr__(self, attr):
+        return getattr(self.stream, attr)
 
-    # close stdinput of subprocess
-    def closeinput(self):
-        self.__lock.acquire()
-        self.__quit = True
-        self.__inputsem.release()
-        self.__lock.release()
+iswindows = sys.platform.startswith('win')
+isosx = sys.platform.startswith('darwin')
+
+def unicode_argv():
+    if iswindows:
+        # Uses shell32.GetCommandLineArgvW to get sys.argv as a list of Unicode
+        # strings.
+
+        # Versions 2.x of Python don't support Unicode in sys.argv on
+        # Windows, with the underlying Windows API instead replacing multi-byte
+        # characters with '?'.
+
+
+        from ctypes import POINTER, byref, cdll, c_int, windll
+        from ctypes.wintypes import LPCWSTR, LPWSTR
+
+        GetCommandLineW = cdll.kernel32.GetCommandLineW
+        GetCommandLineW.argtypes = []
+        GetCommandLineW.restype = LPCWSTR
+
+        CommandLineToArgvW = windll.shell32.CommandLineToArgvW
+        CommandLineToArgvW.argtypes = [LPCWSTR, POINTER(c_int)]
+        CommandLineToArgvW.restype = POINTER(LPWSTR)
+
+        cmd = GetCommandLineW()
+        argc = c_int(0)
+        argv = CommandLineToArgvW(cmd, byref(argc))
+        if argc.value > 0:
+            # Remove Python executable and commands if present
+            start = argc.value - len(sys.argv)
+            return [argv[i] for i in
+                    xrange(start, argc.value)]
+        # if we don't have any arguments at all, just pass back script name
+        # this should never happen
+        return [u"mobidedrm.py"]
+    else:
+        argvencoding = sys.stdin.encoding
+        if argvencoding == None:
+            argvencoding = 'utf-8'
+        return [arg if (type(arg) == unicode) else unicode(arg,argvencoding) for arg in sys.argv]
+
+#global switch
+debug = False
+
+if 'calibre' in sys.modules:
+    inCalibre = True
+    from calibre_plugins.dedrm import kgenpids
+else:
+    inCalibre = False
+    import kgenpids
+
+
+class DrmException(Exception):
+    pass
+
+
+# recursive zip creation support routine
+def zipUpDir(myzip, tdir, localname):
+    currentdir = tdir
+    if localname != u"":
+        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 DrmException(u"Didn't find PID magic numbers in record")
+    elif fields[1] != 8 or fields[3] != 8 :
+        raise DrmException(u"Record didn't contain correct length fields")
+    elif fields[2] != PID :
+        raise DrmException(u"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 DrmException:
+            pass
+        data = data[1+length:]
+    if len(records) == 0:
+        raise DrmException(u"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 DrmException(u"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)
+            if debug: print "%d records in header " % nbValues,
+            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 DrmException(u"Parse Error : Invalid Header")
+            tag = bookReadString(self.fo)
+            record = bookReadHeaderRecordData()
+            return [tag,record]
+        nbRecords = bookReadEncodedNumber(self.fo)
+        if debug: print "Headers: %d" % nbRecords
+        for i in range (0,nbRecords):
+            result = parseTopazHeaderRecord()
+            if debug: print result[0], ": ", result[1]
+            self.bookHeaderRecords[result[0]] = result[1]
+        if ord(self.fo.read(1))  != 0x64 :
+            raise DrmException(u"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 DrmException(u"Parse Error : Record Names Don't Match")
+        flags = ord(self.fo.read(1))
+        nbRecords = ord(self.fo.read(1))
+        if debug: print "Metadata Records: %d" % nbRecords
+        for i in range (0,nbRecords) :
+            keyval = bookReadString(self.fo)
+            content = bookReadString(self.fo)
+            if debug: print keyval
+            if debug: 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.decode('utf-8')
+
+    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 DrmException("Parse Error : Invalid Record, record not found")
+
+        self.fo.seek(self.bookPayloadOffset + recordOffset)
+
+        tag = bookReadString(self.fo)
+        if tag != name :
+            raise DrmException("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 DrmException("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 DrmException("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 DrmException, e:
+            print u"no dkey record found, book may not be encrypted"
+            print u"attempting to extrct files without a book key"
+            self.createBookDirectory()
+            self.extractFiles()
+            print u"Successfully Extracted Topaz contents"
+            if inCalibre:
+                from calibre_plugins.dedrm import genbook
+            else:
+                import genbook
+
+            rv = genbook.generateBook(self.outdir, raw, fixedimage)
+            if rv == 0:
+                print u"Book 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 u"Trying: {0}".format(pid)
+            bookKeys = []
+            data = keydata
+            try:
+                bookKeys+=decryptDkeyRecords(data,pid)
+            except DrmException, e:
+                pass
+            else:
+                bookKey = bookKeys[0]
+                print u"Book Key Found! ({0})".format(bookKey.encode('hex'))
+                break
+
+        if not bookKey:
+            raise DrmException(u"No key found in {0:d} keys tried. Read the FAQs at Alf's blog: http://apprenticealf.wordpress.com/".format(len(pidlst)))
+
+        self.setBookKey(bookKey)
+        self.createBookDirectory()
+        self.extractFiles() 
+        print u"Successfully Extracted Topaz contents"
+        if inCalibre:
+            from calibre_plugins.dedrm import genbook
+        else:
+            import genbook
+
+        rv = genbook.generateBook(self.outdir, raw, fixedimage)
+        if rv == 0:
+            print u"Book 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,u"img")
+        if not os.path.exists(destdir):
+            os.makedirs(destdir)
+        destdir =  os.path.join(outdir,u"color_img")
+        if not os.path.exists(destdir):
+            os.makedirs(destdir)
+        destdir =  os.path.join(outdir,u"page")
+        if not os.path.exists(destdir):
+            os.makedirs(destdir)
+        destdir =  os.path.join(outdir,u"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 = u".dat"
+                if name == 'img': ext = u".jpg"
+                if name == 'color' : ext = u".jpg"
+                print u"Processing Section: {0}\n. . .".format(name),
+                for index in range (0,len(self.bookHeaderRecords[name])) :
+                    fname = u"{0}{1:04d}{2}".format(name,index,ext)
+                    destdir = outdir
+                    if name == 'img':
+                        destdir =  os.path.join(outdir,u"img")
+                    if name == 'color':
+                        destdir =  os.path.join(outdir,u"color_img")
+                    if name == 'page':
+                        destdir =  os.path.join(outdir,u"page")
+                    if name == 'glyphs':
+                        destdir =  os.path.join(outdir,u"glyphs")
+                    outputFile = os.path.join(destdir,fname)
+                    print u".",
+                    record = self.getBookPayloadRecord(name,index)
+                    if record != '':
+                        file(outputFile, 'wb').write(record)
+                print u" "
+
+    def getFile(self, zipname):
+        htmlzip = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
+        htmlzip.write(os.path.join(self.outdir,u"book.html"),u"book.html")
+        htmlzip.write(os.path.join(self.outdir,u"book.opf"),u"book.opf")
+        if os.path.isfile(os.path.join(self.outdir,u"cover.jpg")):
+            htmlzip.write(os.path.join(self.outdir,u"cover.jpg"),u"cover.jpg")
+        htmlzip.write(os.path.join(self.outdir,u"style.css"),u"style.css")
+        zipUpDir(htmlzip, self.outdir, u"img")
+        htmlzip.close()
+
+    def getBookType(self):
+        return u"Topaz"
+
+    def getBookExtension(self):
+        return u".htmlz"
+
+    def getSVGZip(self, zipname):
+        svgzip = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
+        svgzip.write(os.path.join(self.outdir,u"index_svg.xhtml"),u"index_svg.xhtml")
+        zipUpDir(svgzip, self.outdir, u"svg")
+        zipUpDir(svgzip, self.outdir, u"img")
+        svgzip.close()
+
+    def cleanup(self):
+        if os.path.isdir(self.outdir):
+            shutil.rmtree(self.outdir, True)
+
+def usage(progname):
+    print u"Removes DRM protection from Topaz ebooks and extracts the contents"
+    print u"Usage:"
+    print u"    {0} [-k <kindle.k4i>] [-p <comma separated PIDs>] [-s <comma separated Kindle serial numbers>] <infile> <outdir>".format(progname)
+
+# Main
+def cli_main():
+    argv=unicode_argv()
+    progname = os.path.basename(argv[0])
+    print u"TopazExtract v{0}.".format(__version__)
+
+    try:
+        opts, args = getopt.getopt(argv[1:], "k:p:s:x")
+    except getopt.GetoptError, err:
+        print u"Error in options or arguments: {0}".format(err.args[0])
+        usage(progname)
+        return 1
+    if len(args)<2:
+        usage(progname)
+        return 1
+
+    infile = args[0]
+    outdir = args[1]
+    if not os.path.isfile(infile):
+        print u"Input File {0} Does Not Exist.".format(infile)
+        return 1
+
+    if not os.path.exists(outdir):
+        print u"Output Directory {0} Does Not Exist.".format(outdir)
+        return 1
+
+    kDatabaseFiles = []
+    serials = []
+    pids = []
+
+    for o, a in opts:
+        if o == '-k':
+            if a == None :
+                raise DrmException("Invalid parameter for -k")
+            kDatabaseFiles.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 = [serial.replace(" ","") for serial in a.split(',')]
+
+    bookname = os.path.splitext(os.path.basename(infile))[0]
+
+    tb = TopazBook(infile)
+    title = tb.getBookTitle()
+    print u"Processing Book: {0}".format(title)
+    md1, md2 = tb.getPIDMetaInfo()
+    pids.extend(kgenpids.getPidList(md1, md2, serials, kDatabaseFiles))
+
+    try:
+        print u"Decrypting Book"
+        tb.processBook(pids)
+
+        print u"   Creating HTML ZIP Archive"
+        zipname = os.path.join(outdir, bookname + u"_nodrm.htmlz")
+        tb.getFile(zipname)
+
+        print u"   Creating SVG ZIP Archive"
+        zipname = os.path.join(outdir, bookname + u"_SVG.zip")
+        tb.getSVGZip(zipname)
+
+        # removing internal temporary directory of pieces
+        tb.cleanup()
+
+    except DrmException, e:
+        print u"Decryption failed\n{0}".format(traceback.format_exc())
+
+        try:
+            tb.cleanup()
+        except:
+            pass
+        return 1
+
+    except Exception, e:
+        print u"Decryption failed\m{0}".format(traceback.format_exc())
+        try:
+            tb.cleanup()
+        except:
+            pass
+        return 1
+
+    return 0
+
+
+if __name__ == '__main__':
+    sys.stdout=SafeUnbuffered(sys.stdout)
+    sys.stderr=SafeUnbuffered(sys.stderr)
+    sys.exit(cli_main())

DeDRM_calibre_plugin/DeDRM_plugin/utilities.py

@@ -1,538 +1,39 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
-# topazextract.py
-# Mostly written by some_updates based on code from many others
+from __future__ import with_statement
 
-# Changelog
-#  4.9  - moved unicode_argv call inside main for Windows DeDRM compatibility
-#  5.0  - Fixed potential unicode problem with command line interface
+__license__ = 'GPL v3'
 
-__version__ = '5.0'
+DETAILED_MESSAGE = \
+'You have personal information stored in this plugin\'s customization '+ \
+'string from a previous version of this plugin.\n\n'+ \
+'This new version of the plugin can convert that info '+ \
+'into key data that the new plugin can then use (which doesn\'t '+ \
+'require personal information to be stored/displayed in an insecure '+ \
+'manner like the old plugin did).\n\nIf you choose NOT to migrate this data at this time '+ \
+'you will be prompted to save that personal data to a file elsewhere; and you\'ll have '+ \
+'to manually re-configure this plugin with your information.\n\nEither way... ' + \
+'this new version of the plugin will not be responsible for storing that personal '+ \
+'info in plain sight any longer.'
 
-import sys
-import os, csv, getopt
-import zlib, zipfile, tempfile, shutil
-import traceback
-from struct import pack
-from struct import unpack
-from alfcrypto import Topaz_Cipher
-
-class SafeUnbuffered:
-    def __init__(self, stream):
-        self.stream = stream
-        self.encoding = stream.encoding
-        if self.encoding == None:
-            self.encoding = "utf-8"
-    def write(self, data):
-        if isinstance(data,unicode):
-            data = data.encode(self.encoding,"replace")
-        self.stream.write(data)
-        self.stream.flush()
-    def __getattr__(self, attr):
-        return getattr(self.stream, attr)
-
-iswindows = sys.platform.startswith('win')
-isosx = sys.platform.startswith('darwin')
-
-def unicode_argv():
-    if iswindows:
-        # Uses shell32.GetCommandLineArgvW to get sys.argv as a list of Unicode
-        # strings.
-
-        # Versions 2.x of Python don't support Unicode in sys.argv on
-        # Windows, with the underlying Windows API instead replacing multi-byte
-        # characters with '?'.
-
-
-        from ctypes import POINTER, byref, cdll, c_int, windll
-        from ctypes.wintypes import LPCWSTR, LPWSTR
-
-        GetCommandLineW = cdll.kernel32.GetCommandLineW
-        GetCommandLineW.argtypes = []
-        GetCommandLineW.restype = LPCWSTR
-
-        CommandLineToArgvW = windll.shell32.CommandLineToArgvW
-        CommandLineToArgvW.argtypes = [LPCWSTR, POINTER(c_int)]
-        CommandLineToArgvW.restype = POINTER(LPWSTR)
-
-        cmd = GetCommandLineW()
-        argc = c_int(0)
-        argv = CommandLineToArgvW(cmd, byref(argc))
-        if argc.value > 0:
-            # Remove Python executable and commands if present
-            start = argc.value - len(sys.argv)
-            return [argv[i] for i in
-                    xrange(start, argc.value)]
-        # if we don't have any arguments at all, just pass back script name
-        # this should never happen
-        return [u"mobidedrm.py"]
+def uStrCmp (s1, s2, caseless=False):
+    import unicodedata as ud
+    str1 = s1 if isinstance(s1, unicode) else unicode(s1)
+    str2 = s2 if isinstance(s2, unicode) else unicode(s2)
+    if caseless:
+        return ud.normalize('NFC', str1.lower()) == ud.normalize('NFC', str2.lower())
     else:
-        argvencoding = sys.stdin.encoding
-        if argvencoding == None:
-            argvencoding = 'utf-8'
-        return [arg if (type(arg) == unicode) else unicode(arg,argvencoding) for arg in sys.argv]
+        return ud.normalize('NFC', str1) == ud.normalize('NFC', str2)
 
-#global switch
-debug = False
-
-if 'calibre' in sys.modules:
-    inCalibre = True
-    from calibre_plugins.dedrm import kgenpids
-else:
-    inCalibre = False
-    import kgenpids
-
-
-class DrmException(Exception):
-    pass
-
-
-# recursive zip creation support routine
-def zipUpDir(myzip, tdir, localname):
-    currentdir = tdir
-    if localname != u"":
-        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 DrmException(u"Didn't find PID magic numbers in record")
-    elif fields[1] != 8 or fields[3] != 8 :
-        raise DrmException(u"Record didn't contain correct length fields")
-    elif fields[2] != PID :
-        raise DrmException(u"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])
+def parseCustString(keystuff):
+    userkeys = []
+    ar = keystuff.split(':')
+    for i in ar:
         try:
-            key = decryptDkeyRecord(data[1:length+1],PID)
-            records.append(key)
-        except DrmException:
-            pass
-        data = data[1+length:]
-    if len(records) == 0:
-        raise DrmException(u"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 DrmException(u"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)
-            if debug: print "%d records in header " % nbValues,
-            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 DrmException(u"Parse Error : Invalid Header")
-            tag = bookReadString(self.fo)
-            record = bookReadHeaderRecordData()
-            return [tag,record]
-        nbRecords = bookReadEncodedNumber(self.fo)
-        if debug: print "Headers: %d" % nbRecords
-        for i in range (0,nbRecords):
-            result = parseTopazHeaderRecord()
-            if debug: print result[0], ": ", result[1]
-            self.bookHeaderRecords[result[0]] = result[1]
-        if ord(self.fo.read(1))  != 0x64 :
-            raise DrmException(u"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 DrmException(u"Parse Error : Record Names Don't Match")
-        flags = ord(self.fo.read(1))
-        nbRecords = ord(self.fo.read(1))
-        if debug: print "Metadata Records: %d" % nbRecords
-        for i in range (0,nbRecords) :
-            keyval = bookReadString(self.fo)
-            content = bookReadString(self.fo)
-            if debug: print keyval
-            if debug: 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.decode('utf-8')
-
-    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 DrmException("Parse Error : Invalid Record, record not found")
-
-        self.fo.seek(self.bookPayloadOffset + recordOffset)
-
-        tag = bookReadString(self.fo)
-        if tag != name :
-            raise DrmException("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 DrmException("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 DrmException("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 DrmException, e:
-            print u"no dkey record found, book may not be encrypted"
-            print u"attempting to extrct files without a book key"
-            self.createBookDirectory()
-            self.extractFiles()
-            print u"Successfully Extracted Topaz contents"
-            if inCalibre:
-                from calibre_plugins.dedrm import genbook
-            else:
-                import genbook
-
-            rv = genbook.generateBook(self.outdir, raw, fixedimage)
-            if rv == 0:
-                print u"Book 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 u"Trying: {0}".format(pid)
-            bookKeys = []
-            data = keydata
-            try:
-                bookKeys+=decryptDkeyRecords(data,pid)
-            except DrmException, e:
-                pass
-            else:
-                bookKey = bookKeys[0]
-                print u"Book Key Found! ({0})".format(bookKey.encode('hex'))
-                break
-
-        if not bookKey:
-            raise DrmException(u"No key found in {0:d} keys tried. Read the FAQs at Alf's blog: http://apprenticealf.wordpress.com/".format(len(pidlst)))
-
-        self.setBookKey(bookKey)
-        self.createBookDirectory()
-        self.extractFiles()
-        print u"Successfully Extracted Topaz contents"
-        if inCalibre:
-            from calibre_plugins.dedrm import genbook
-        else:
-            import genbook
-
-        rv = genbook.generateBook(self.outdir, raw, fixedimage)
-        if rv == 0:
-            print u"Book 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,u"img")
-        if not os.path.exists(destdir):
-            os.makedirs(destdir)
-        destdir =  os.path.join(outdir,u"color_img")
-        if not os.path.exists(destdir):
-            os.makedirs(destdir)
-        destdir =  os.path.join(outdir,u"page")
-        if not os.path.exists(destdir):
-            os.makedirs(destdir)
-        destdir =  os.path.join(outdir,u"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 = u".dat"
-                if name == 'img': ext = u".jpg"
-                if name == 'color' : ext = u".jpg"
-                print u"Processing Section: {0}\n. . .".format(name),
-                for index in range (0,len(self.bookHeaderRecords[name])) :
-                    fname = u"{0}{1:04d}{2}".format(name,index,ext)
-                    destdir = outdir
-                    if name == 'img':
-                        destdir =  os.path.join(outdir,u"img")
-                    if name == 'color':
-                        destdir =  os.path.join(outdir,u"color_img")
-                    if name == 'page':
-                        destdir =  os.path.join(outdir,u"page")
-                    if name == 'glyphs':
-                        destdir =  os.path.join(outdir,u"glyphs")
-                    outputFile = os.path.join(destdir,fname)
-                    print u".",
-                    record = self.getBookPayloadRecord(name,index)
-                    if record != '':
-                        file(outputFile, 'wb').write(record)
-                print u" "
-
-    def getFile(self, zipname):
-        htmlzip = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
-        htmlzip.write(os.path.join(self.outdir,u"book.html"),u"book.html")
-        htmlzip.write(os.path.join(self.outdir,u"book.opf"),u"book.opf")
-        if os.path.isfile(os.path.join(self.outdir,u"cover.jpg")):
-            htmlzip.write(os.path.join(self.outdir,u"cover.jpg"),u"cover.jpg")
-        htmlzip.write(os.path.join(self.outdir,u"style.css"),u"style.css")
-        zipUpDir(htmlzip, self.outdir, u"img")
-        htmlzip.close()
-
-    def getBookType(self):
-        return u"Topaz"
-
-    def getBookExtension(self):
-        return u".htmlz"
-
-    def getSVGZip(self, zipname):
-        svgzip = zipfile.ZipFile(zipname,'w',zipfile.ZIP_DEFLATED, False)
-        svgzip.write(os.path.join(self.outdir,u"index_svg.xhtml"),u"index_svg.xhtml")
-        zipUpDir(svgzip, self.outdir, u"svg")
-        zipUpDir(svgzip, self.outdir, u"img")
-        svgzip.close()
-
-    def cleanup(self):
-        if os.path.isdir(self.outdir):
-            shutil.rmtree(self.outdir, True)
-
-def usage(progname):
-    print u"Removes DRM protection from Topaz ebooks and extracts the contents"
-    print u"Usage:"
-    print u"    {0} [-k <kindle.k4i>] [-p <comma separated PIDs>] [-s <comma separated Kindle serial numbers>] <infile> <outdir>".format(progname)
-
-# Main
-def cli_main():
-    argv=unicode_argv()
-    progname = os.path.basename(argv[0])
-    print u"TopazExtract v{0}.".format(__version__)
-
-    try:
-        opts, args = getopt.getopt(argv[1:], "k:p:s:x")
-    except getopt.GetoptError, err:
-        print u"Error in options or arguments: {0}".format(err.args[0])
-        usage(progname)
-        return 1
-    if len(args)<2:
-        usage(progname)
-        return 1
-
-    infile = args[0]
-    outdir = args[1]
-    if not os.path.isfile(infile):
-        print u"Input File {0} Does Not Exist.".format(infile)
-        return 1
-
-    if not os.path.exists(outdir):
-        print u"Output Directory {0} Does Not Exist.".format(outdir)
-        return 1
-
-    kDatabaseFiles = []
-    serials = []
-    pids = []
-
-    for o, a in opts:
-        if o == '-k':
-            if a == None :
-                raise DrmException("Invalid parameter for -k")
-            kDatabaseFiles.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 = [serial.replace(" ","") for serial in a.split(',')]
-
-    bookname = os.path.splitext(os.path.basename(infile))[0]
-
-    tb = TopazBook(infile)
-    title = tb.getBookTitle()
-    print u"Processing Book: {0}".format(title)
-    md1, md2 = tb.getPIDMetaInfo()
-    pids.extend(kgenpids.getPidList(md1, md2, serials, kDatabaseFiles))
-
-    try:
-        print u"Decrypting Book"
-        tb.processBook(pids)
-
-        print u"   Creating HTML ZIP Archive"
-        zipname = os.path.join(outdir, bookname + u"_nodrm.htmlz")
-        tb.getFile(zipname)
-
-        print u"   Creating SVG ZIP Archive"
-        zipname = os.path.join(outdir, bookname + u"_SVG.zip")
-        tb.getSVGZip(zipname)
-
-        # removing internal temporary directory of pieces
-        tb.cleanup()
-
-    except DrmException, e:
-        print u"Decryption failed\n{0}".format(traceback.format_exc())
-
-        try:
-            tb.cleanup()
+            name, ccn = i.split(',')
+            # Generate Barnes & Noble EPUB user key from name and credit card number.
+            userkeys.append(generate_key(name, ccn))
         except:
             pass
-        return 1
-
-    except Exception, e:
-        print u"Decryption failed\m{0}".format(traceback.format_exc())
-        try:
-            tb.cleanup()
-        except:
-            pass
-        return 1
-
-    return 0
-
-
-if __name__ == '__main__':
-    sys.stdout=SafeUnbuffered(sys.stdout)
-    sys.stderr=SafeUnbuffered(sys.stderr)
-    sys.exit(cli_main())
+    return userkeys

DeDRM_calibre_plugin/DeDRM_plugin/wineutils.py

@@ -0,0 +1,60 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+from __future__ import with_statement
+
+__license__ = 'GPL v3'
+
+# Standard Python modules.
+import os, sys, re, hashlib
+from calibre_plugins.dedrm.__init__ import PLUGIN_NAME, PLUGIN_VERSION
+
+def WineGetKeys(scriptpath, extension, wineprefix=""):
+    import subprocess
+    from subprocess import Popen, PIPE, STDOUT
+
+    import subasyncio
+    from subasyncio import Process
+
+    if extension == u".k4i":
+        import json
+
+    basepath, script = os.path.split(scriptpath)
+    print u"{0} v{1}: Running {2} under Wine".format(PLUGIN_NAME, PLUGIN_VERSION, script)
+
+    outdirpath = os.path.join(basepath, u"winekeysdir")
+    if not os.path.exists(outdirpath):
+        os.makedirs(outdirpath)
+
+    if wineprefix != "" and os.path.exists(wineprefix):
+         cmdline = u"WINEPREFIX=\"{2}\" wine python.exe \"{0}\" \"{1}\"".format(scriptpath,outdirpath,wineprefix)
+    else:
+        cmdline = u"wine python.exe \"{0}\" \"{1}\"".format(scriptpath,outdirpath)
+    print u"{0} v{1}: Command line: “{2}”".format(PLUGIN_NAME, PLUGIN_VERSION, cmdline)
+
+    try:
+        cmdline = cmdline.encode(sys.getfilesystemencoding())
+        p2 = Process(cmdline, shell=True, bufsize=1, stdin=None, stdout=sys.stdout, stderr=STDOUT, close_fds=False)
+        result = p2.wait("wait")
+    except Exception, e:
+        print u"{0} v{1}: Wine subprocess call error: {2}".format(PLUGIN_NAME, PLUGIN_VERSION, e.args[0])
+        return []
+
+    winekeys = []
+    # get any files with extension in the output dir
+    files = [f for f in os.listdir(outdirpath) if f.endswith(extension)]
+    for filename in files:
+        try:
+            fpath = os.path.join(outdirpath, filename)
+            with open(fpath, 'rb') as keyfile:
+                if extension == u".k4i":
+                    new_key_value = json.loads(keyfile.read())
+                else:
+                    new_key_value = keyfile.read()
+            winekeys.append(new_key_value)
+        except:
+            print u"{0} v{1}: Error loading file {2}".format(PLUGIN_NAME, PLUGIN_VERSION, filename)
+            traceback.print_exc()
+        os.remove(fpath)
+    print u"{0} v{1}: Found and decrypted {2} {3}".format(PLUGIN_NAME, PLUGIN_VERSION, len(winekeys), u"key file" if len(winekeys) == 1 else u"key files")
+    return winekeys

DeDRM_calibre_plugin/DeDRM_plugin/zipfix.py

@@ -1,60 +1,188 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
-from __future__ import with_statement
+# zipfix.py, version 1.1
+# Copyright © 2010-2013 by some_updates, DiapDealer and Apprentice Alf
+
+# Released under the terms of the GNU General Public Licence, version 3
+# <http://www.gnu.org/licenses/>
+
+# Revision history:
+#   1.0 - Initial release
+#   1.1 - Updated to handle zip file metadata correctly
+
+"""
+Re-write zip (or ePub) fixing problems with file names (and mimetype entry).
+"""
 
 __license__ = 'GPL v3'
+__version__ = "1.1"
 
-# Standard Python modules.
-import os, sys, re, hashlib
-from calibre_plugins.dedrm.__init__ import PLUGIN_NAME, PLUGIN_VERSION
+import sys
+import zlib
+import zipfilerugged
+import os
+import os.path
+import getopt
+from struct import unpack
 
-def WineGetKeys(scriptpath, extension, wineprefix=""):
-    import subprocess
-    from subprocess import Popen, PIPE, STDOUT
 
-    import subasyncio
-    from subasyncio import Process
+_FILENAME_LEN_OFFSET = 26
+_EXTRA_LEN_OFFSET = 28
+_FILENAME_OFFSET = 30
+_MAX_SIZE = 64 * 1024
+_MIMETYPE = 'application/epub+zip'
 
-    if extension == u".k4i":
-        import json
+class ZipInfo(zipfilerugged.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
 
-    basepath, script = os.path.split(scriptpath)
-    print u"{0} v{1}: Running {2} under Wine".format(PLUGIN_NAME, PLUGIN_VERSION, script)
+class fixZip:
+    def __init__(self, zinput, zoutput):
+        self.ztype = 'zip'
+        if zinput.lower().find('.epub') >= 0 :
+            self.ztype = 'epub'
+        self.inzip = zipfilerugged.ZipFile(zinput,'r')
+        self.outzip = zipfilerugged.ZipFile(zoutput,'w')
+        # open the input zip for reading only as a raw file
+        self.bzf = file(zinput,'rb')
 
-    outdirpath = os.path.join(basepath, u"winekeysdir")
-    if not os.path.exists(outdirpath):
-        os.makedirs(outdirpath)
+    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
 
-    if wineprefix != "" and os.path.exists(wineprefix):
-         cmdline = u"WINEPREFIX=\"{2}\" wine python.exe \"{0}\" \"{1}\"".format(scriptpath,outdirpath,wineprefix)
+    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:
-        cmdline = u"wine python.exe \"{0}\" \"{1}\"".format(scriptpath,outdirpath)
-    print u"{0} v{1}: Command line: “{2}”".format(PLUGIN_NAME, PLUGIN_VERSION, cmdline)
+                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
 
-    try:
-        cmdline = cmdline.encode(sys.getfilesystemencoding())
-        p2 = Process(cmdline, shell=True, bufsize=1, stdin=None, stdout=sys.stdout, stderr=STDOUT, close_fds=False)
-        result = p2.wait("wait")
-    except Exception, e:
-        print u"{0} v{1}: Wine subprocess call error: {2}".format(PLUGIN_NAME, PLUGIN_VERSION, e.args[0])
-        return []
+    def getfiledata(self, zi):
+        # get file name length and exta data length to find start of file data
+        local_header_offset = zi.header_offset
 
-    winekeys = []
-    # get any files with extension in the output dir
-    files = [f for f in os.listdir(outdirpath) if f.endswith(extension)]
-    for filename in files:
+        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 == zipfilerugged.ZIP_STORED:
+            data = self.bzf.read(zi.file_size)
+
+        # if compressed we must decompress it using zlib
+        if zi.compress_type == zipfilerugged.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':
+            # first get a ZipInfo with current time and no compression
+            mimeinfo = ZipInfo('mimetype',compress_type=zipfilerugged.ZIP_STORED)
+            mimeinfo.internal_attr = 1 # text file
             try:
-            fpath = os.path.join(outdirpath, filename)
-            with open(fpath, 'rb') as keyfile:
-                if extension == u".k4i":
-                    new_key_value = json.loads(keyfile.read())
-                else:
-                    new_key_value = keyfile.read()
-            winekeys.append(new_key_value)
+                # if the mimetype is present, get its info, including time-stamp
+                oldmimeinfo = self.inzip.getinfo('mimetype')
+                # copy across useful fields
+                mimeinfo.date_time = oldmimeinfo.date_time
+                mimeinfo.comment = oldmimeinfo.comment
+                mimeinfo.extra = oldmimeinfo.extra
+                mimeinfo.internal_attr = oldmimeinfo.internal_attr
+                mimeinfo.external_attr = oldmimeinfo.external_attr
+                mimeinfo.create_system = oldmimeinfo.create_system
             except:
-            print u"{0} v{1}: Error loading file {2}".format(PLUGIN_NAME, PLUGIN_VERSION, filename)
-            traceback.print_exc()
-        os.remove(fpath)
-    print u"{0} v{1}: Found and decrypted {2} {3}".format(PLUGIN_NAME, PLUGIN_VERSION, len(winekeys), u"key file" if len(winekeys) == 1 else u"key files")
-    return winekeys
+                pass
+            self.outzip.writestr(mimeinfo, _MIMETYPE)
+
+        # write the rest of the files
+        for zinfo in self.inzip.infolist():
+            if zinfo.filename != "mimetype" or self.ztype != 'epub':
+                data = None
+                try:
+                    data = self.inzip.read(zinfo.filename)
+                except zipfilerugged.BadZipfile or zipfilerugged.error:
+                    local_name = self.getlocalname(zinfo)
+                    data = self.getfiledata(zinfo)
+                    zinfo.filename = local_name
+
+                # create new ZipInfo with only the useful attributes from the old info
+                nzinfo = ZipInfo(zinfo.filename, zinfo.date_time, compress_type=zinfo.compress_type)
+                nzinfo.comment=zinfo.comment
+                nzinfo.extra=zinfo.extra
+                nzinfo.internal_attr=zinfo.internal_attr
+                nzinfo.external_attr=zinfo.external_attr
+                nzinfo.create_system=zinfo.create_system
+                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())

Other_Tools/DRM_Key_Scripts/Kindle_for_Mac_and_PC/kindlekey.pyw

@@ -19,6 +19,7 @@ from __future__ import with_statement
 #  1.6   - Fixed a problem getting the disk serial numbers
 #  1.7   - Work if TkInter is missing
 #  1.8   - Fixes for Kindle for Mac, and non-ascii in Windows user names
+#  1.9   - Fixes for Unicode in Windows user names
 
 
 """
@@ -26,7 +27,7 @@ Retrieve Kindle for PC/Mac user key.
 """
 
 __license__ = 'GPL v3'
-__version__ = '1.8'
+__version__ = '1.9'
 
 import sys, os, re
 from struct import pack, unpack, unpack_from
@@ -907,18 +908,34 @@ if iswindows:
         return CryptUnprotectData
     CryptUnprotectData = CryptUnprotectData()
 
+    # Returns Environmental Variables that contain unicode
+    def getEnvironmentVariable(name):
+        import ctypes
+        name = unicode(name) # make sure string argument is unicode
+        n = ctypes.windll.kernel32.GetEnvironmentVariableW(name, None, 0)
+        if n == 0:
+            return None
+        buf = ctypes.create_unicode_buffer(u'\0'*n)
+        ctypes.windll.kernel32.GetEnvironmentVariableW(name, buf, n)
+        return buf.value
 
     # Locate all of the kindle-info style files and return as list
     def getKindleInfoFiles():
         kInfoFiles = []
         # 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
+        # or the python interface to the 32 vs 64 bit registry is broken
         path = ""
         if 'LOCALAPPDATA' in os.environ.keys():
-            path = os.environ['LOCALAPPDATA']
+			# Python 2.x does not return unicode env. Use Python 3.x
+            path = winreg.ExpandEnvironmentStrings(u"%LOCALAPPDATA%")
+            # this is just another alternative.
+            # path = getEnvironmentVariable('LOCALAPPDATA')
+            if not os.path.isdir(path):
+                path = ""
         else:
             # User Shell Folders show take precedent over Shell Folders if present
             try:
+                # this will still break
                 regkey = winreg.OpenKey(winreg.HKEY_CURRENT_USER, "Software\\Microsoft\\Windows\\CurrentVersion\\Explorer\\User Shell Folders\\")
                 path = winreg.QueryValueEx(regkey, 'Local AppData')[0]
                 if not os.path.isdir(path):
@@ -937,13 +954,14 @@ if iswindows:
         if path == "":
             print ('Could not find the folder in which to look for kinfoFiles.')
         else:
-            print('searching for kinfoFiles in ' + path)
+            # Probably not the best. To Fix (shouldn't ignore in encoding) or use utf-8
+            print(u'searching for kinfoFiles in ' + path.encode('ascii', 'ignore'))
 
             # look for (K4PC 1.9.0 and later) .kinf2011 file
             kinfopath = path +'\\Amazon\\Kindle\\storage\\.kinf2011'
             if os.path.isfile(kinfopath):
                 found = True
-                print('Found K4PC 1.9+ kinf2011 file: ' + kinfopath)
+                print('Found K4PC 1.9+ kinf2011 file: ' + kinfopath.encode('ascii','ignore'))
                 kInfoFiles.append(kinfopath)
 
             # look for (K4PC 1.6.0 and later) rainier.2.1.1.kinf file
@@ -1142,7 +1160,7 @@ if iswindows:
                 cleartext = CryptUnprotectData(encryptedValue, entropy, 1)
                 DB[keyname] = cleartext
 
-        if 'MazamaRandomNumber' in DB and 'kindle.account.tokens' in DB:
+        if 'kindle.account.tokens' in DB:
             print u"Decrypted key file using IDString '{0:s}' and UserName '{1:s}'".format(GetIDString(), GetUserName().decode("latin-1"))
             # store values used in decryption
             DB['IDString'] = GetIDString()
@@ -1758,7 +1776,7 @@ elif isosx:
                         break
             except:
                 pass
-        if 'MazamaRandomNumber' in DB and 'kindle.account.tokens' in DB:
+        if 'kindle.account.tokens' in DB:
             # store values used in decryption
             print u"Decrypted key file using IDString '{0:s}' and UserName '{1:s}'".format(IDString, GetUserName())
             DB['IDString'] = IDString