Minor tweaks for first attempt at KFX support - version numbers to 6.6.0, readmes, etc. Removed KFX archive.

Initial KFX support from tomthumb1997

DeDRM_Macintosh_Application/DeDRM.app/Contents/Info.plist

@@ -24,7 +24,7 @@
 	<key>CFBundleExecutable</key>
 	<string>droplet</string>
 	<key>CFBundleGetInfoString</key>
-	<string>DeDRM AppleScript 6.5.4 Written 2010–2017 by Apprentice Alf et al.</string>
+	<string>DeDRM AppleScript 6.6.0 Written 2010–2018 by Apprentice Alf et al.</string>
 	<key>CFBundleIconFile</key>
 	<string>DeDRM</string>
 	<key>CFBundleIdentifier</key>
@@ -36,13 +36,13 @@
 	<key>CFBundlePackageType</key>
 	<string>APPL</string>
 	<key>CFBundleShortVersionString</key>
-	<string>6.5.4</string>
+	<string>6.6.0</string>
 	<key>CFBundleSignature</key>
 	<string>dplt</string>
 	<key>LSRequiresCarbon</key>
 	<true/>
 	<key>NSHumanReadableCopyright</key>
-	<string>Copyright © 2010–2017 Apprentice Alf</string>
+	<string>Copyright © 2010–2018 Apprentice Alf</string>
 	<key>WindowState</key>
 	<dict>
 		<key>bundleDividerCollapsed</key>

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

@@ -2,6 +2,10 @@
 # -*- coding: utf-8 -*-
 
 from __future__ import with_statement
+
+# __init__.py for DeDRM_plugin
+# Copyright © 2008-2018 Apprentice Harper et al.
+
 __license__   = 'GPL v3'
 __docformat__ = 'restructuredtext en'
 
@@ -58,6 +62,9 @@ __docformat__ = 'restructuredtext en'
 #   6.5.2 - Another Topaz fix
 #   6.5.3 - Warn about KFX files explicitly
 #   6.5.4 - Mac App Fix, improve PDF decryption, handle latest tcl changes in ActivePython
+#   6.5.5 - Finally a fix for the Windows non-ASCII user names.
+#   6.6.0 - Add kfx and kfx-zip as supported file types (also invoke this plugin if the original
+#           imported format was azw8 since that may be converted to kfx)
 
 
 """
@@ -65,7 +72,7 @@ Decrypt DRMed ebooks.
 """
 
 PLUGIN_NAME = u"DeDRM"
-PLUGIN_VERSION_TUPLE = (6, 5, 4)
+PLUGIN_VERSION_TUPLE = (6, 6, 0)
 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'
@@ -113,7 +120,7 @@ class DeDRM(FileTypePlugin):
     author                  = u"Apprentice Alf, Aprentice Harper, The Dark Reverser and i♥cabbages"
     version                 = PLUGIN_VERSION_TUPLE
     minimum_calibre_version = (1, 0, 0)  # Compiled python libraries cannot be imported in earlier versions.
-    file_types              = set(['epub','pdf','pdb','prc','mobi','pobi','azw','azw1','azw3','azw4','tpz'])
+    file_types              = set(['epub','pdf','pdb','prc','mobi','pobi','azw','azw1','azw3','azw4','azw8','tpz','kfx','kfx-zip'])
     on_import               = True
     priority                = 600
 
@@ -289,8 +296,8 @@ class DeDRM(FileTypePlugin):
                 except Exception, e:
                     pass
 
-            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
+            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
 
         # import the Adobe Adept ePub handler
         import calibre_plugins.dedrm.ineptepub as ineptepub
@@ -380,17 +387,19 @@ class DeDRM(FileTypePlugin):
                             except:
                                 print u"{0} v{1}: Exception when saving a new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime)
                                 traceback.print_exc()
-                            print u"{0} v{1}: Decrypted with new default key after {3:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+                            print u"{0} v{1}: Decrypted with new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
                             # Return the modified PersistentTemporary file to calibre.
                             return of.name
 
                         print u"{0} v{1}: Failed to decrypt with new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
                 except Exception, e:
+                    print u"{0} v{1}: Unexpected Exception trying a new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime)
+                    traceback.print_exc()
                     pass
 
             # Something went wrong with decryption.
-            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
+            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
 
         # Not a Barnes & Noble nor an Adobe Adept
         # Import the fixed epub.
@@ -489,8 +498,8 @@ class DeDRM(FileTypePlugin):
                 pass
 
         # Something went wrong with decryption.
-        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
+        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
 
 
     def KindleMobiDecrypt(self,path_to_ebook):
@@ -557,8 +566,8 @@ class DeDRM(FileTypePlugin):
                     pass
             if not decoded:
                 #if you reached here then no luck raise and exception
-                print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-                raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
+                print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+                raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
 
         of = self.temporary_file(book.getBookExtension())
         book.getFile(of.name)
@@ -592,8 +601,8 @@ class DeDRM(FileTypePlugin):
 
             print u"{0} v{1}: Failed to decrypt with key {2:s} after {3:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,keyname_masked,time.time()-self.starttime)
 
-        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
+        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
 
 
     def run(self, path_to_ebook):
@@ -606,7 +615,7 @@ class DeDRM(FileTypePlugin):
         self.starttime = time.time()
 
         booktype = os.path.splitext(path_to_ebook)[1].lower()[1:]
-        if booktype in ['prc','mobi','pobi','azw','azw1','azw3','azw4','tpz']:
+        if booktype in ['prc','mobi','pobi','azw','azw1','azw3','azw4','tpz','kfx-zip']:
             # Kindle/Mobipocket
             decrypted_ebook = self.KindleMobiDecrypt(path_to_ebook)
         elif booktype == 'pdb':

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

@@ -0,0 +1,981 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+# Pascal implementation by lulzkabulz. Python translation by apprenticenaomi. DeDRM integration by anon.
+# BinaryIon.pas + DrmIon.pas + IonSymbols.pas
+
+from __future__ import with_statement
+
+import collections
+import hashlib
+import hmac
+import os
+import os.path
+import struct
+
+try:
+    from cStringIO import StringIO
+except ImportError:
+    from StringIO import StringIO
+
+from Crypto.Cipher import AES
+from Crypto.Util.py3compat import bchr, bord
+
+try:
+    # lzma library from calibre 2.35.0 or later
+    import lzma.lzma1 as calibre_lzma
+except:
+    calibre_lzma = None
+    try:
+        import lzma
+    except:
+        # Need pip backports.lzma on Python <3.3
+        from backports import lzma
+
+
+TID_NULL = 0
+TID_BOOLEAN = 1
+TID_POSINT = 2
+TID_NEGINT = 3
+TID_FLOAT = 4
+TID_DECIMAL = 5
+TID_TIMESTAMP = 6
+TID_SYMBOL = 7
+TID_STRING = 8
+TID_CLOB = 9
+TID_BLOB = 0xA
+TID_LIST = 0xB
+TID_SEXP = 0xC
+TID_STRUCT = 0xD
+TID_TYPEDECL = 0xE
+TID_UNUSED = 0xF
+
+
+SID_UNKNOWN = -1
+SID_ION = 1
+SID_ION_1_0 = 2
+SID_ION_SYMBOL_TABLE = 3
+SID_NAME = 4
+SID_VERSION = 5
+SID_IMPORTS = 6
+SID_SYMBOLS = 7
+SID_MAX_ID = 8
+SID_ION_SHARED_SYMBOL_TABLE = 9
+SID_ION_1_0_MAX = 10
+
+
+LEN_IS_VAR_LEN = 0xE
+LEN_IS_NULL = 0xF
+
+
+VERSION_MARKER = b"\x01\x00\xEA"
+
+
+# asserts must always raise exceptions for proper functioning
+def _assert(test, msg="Exception"):
+    if not test:
+        raise Exception(msg)
+
+
+class SystemSymbols(object):
+    ION = '$ion'
+    ION_1_0 = '$ion_1_0'
+    ION_SYMBOL_TABLE = '$ion_symbol_table'
+    NAME = 'name'
+    VERSION = 'version'
+    IMPORTS = 'imports'
+    SYMBOLS = 'symbols'
+    MAX_ID = 'max_id'
+    ION_SHARED_SYMBOL_TABLE = '$ion_shared_symbol_table'
+
+
+class IonCatalogItem(object):
+    name = ""
+    version = 0
+    symnames = []
+
+    def __init__(self, name, version, symnames):
+        self.name = name
+        self.version = version
+        self.symnames = symnames
+
+
+class SymbolToken(object):
+    text = ""
+    sid = 0
+
+    def __init__(self, text, sid):
+        if text == "" and sid == 0:
+            raise ValueError("Symbol token must have Text or SID")
+
+        self.text = text
+        self.sid = sid
+
+
+class SymbolTable(object):
+    table = None
+
+    def __init__(self):
+        self.table = [None] * SID_ION_1_0_MAX
+        self.table[SID_ION] = SystemSymbols.ION
+        self.table[SID_ION_1_0] = SystemSymbols.ION_1_0
+        self.table[SID_ION_SYMBOL_TABLE] = SystemSymbols.ION_SYMBOL_TABLE
+        self.table[SID_NAME] = SystemSymbols.NAME
+        self.table[SID_VERSION] = SystemSymbols.VERSION
+        self.table[SID_IMPORTS] = SystemSymbols.IMPORTS
+        self.table[SID_SYMBOLS] = SystemSymbols.SYMBOLS
+        self.table[SID_MAX_ID] = SystemSymbols.MAX_ID
+        self.table[SID_ION_SHARED_SYMBOL_TABLE] = SystemSymbols.ION_SHARED_SYMBOL_TABLE
+
+    def findbyid(self, sid):
+        if sid < 1:
+            raise ValueError("Invalid symbol id")
+
+        if sid < len(self.table):
+            return self.table[sid]
+        else:
+            return ""
+
+    def import_(self, table, maxid):
+        for i in range(maxid):
+            self.table.append(table.symnames[i])
+
+    def importunknown(self, name, maxid):
+        for i in range(maxid):
+            self.table.append("%s#%d" % (name, i + 1))
+
+
+class ParserState:
+    Invalid,BeforeField,BeforeTID,BeforeValue,AfterValue,EOF = 1,2,3,4,5,6
+
+ContainerRec = collections.namedtuple("ContainerRec", "nextpos, tid, remaining")
+
+
+class BinaryIonParser(object):
+    eof = False
+    state = None
+    localremaining = 0
+    needhasnext = False
+    isinstruct = False
+    valuetid = 0
+    valuefieldid = 0
+    parenttid = 0
+    valuelen = 0
+    valueisnull = False
+    valueistrue = False
+    value = None
+    didimports = False
+
+    def __init__(self, stream):
+        self.annotations = []
+        self.catalog = []
+
+        self.stream = stream
+        self.initpos = stream.tell()
+        self.reset()
+        self.symbols = SymbolTable()
+
+    def reset(self):
+        self.state = ParserState.BeforeTID
+        self.needhasnext = True
+        self.localremaining = -1
+        self.eof = False
+        self.isinstruct = False
+        self.containerstack = []
+        self.stream.seek(self.initpos)
+
+    def addtocatalog(self, name, version, symbols):
+        self.catalog.append(IonCatalogItem(name, version, symbols))
+
+    def hasnext(self):
+        while self.needhasnext and not self.eof:
+            self.hasnextraw()
+            if len(self.containerstack) == 0 and not self.valueisnull:
+                if self.valuetid == TID_SYMBOL:
+                    if self.value == SID_ION_1_0:
+                        self.needhasnext = True
+                elif self.valuetid == TID_STRUCT:
+                    for a in self.annotations:
+                        if a == SID_ION_SYMBOL_TABLE:
+                            self.parsesymboltable()
+                            self.needhasnext = True
+                            break
+        return not self.eof
+
+    def hasnextraw(self):
+        self.clearvalue()
+        while self.valuetid == -1 and not self.eof:
+            self.needhasnext = False
+            if self.state == ParserState.BeforeField:
+                _assert(self.valuefieldid == SID_UNKNOWN)
+
+                self.valuefieldid = self.readfieldid()
+                if self.valuefieldid != SID_UNKNOWN:
+                    self.state = ParserState.BeforeTID
+                else:
+                    self.eof = True
+
+            elif self.state == ParserState.BeforeTID:
+                self.state = ParserState.BeforeValue
+                self.valuetid = self.readtypeid()
+                if self.valuetid == -1:
+                    self.state = ParserState.EOF
+                    self.eof = True
+                    break
+
+                if self.valuetid == TID_TYPEDECL:
+                    if self.valuelen == 0:
+                        self.checkversionmarker()
+                    else:
+                        self.loadannotations()
+
+            elif self.state == ParserState.BeforeValue:
+                self.skip(self.valuelen)
+                self.state = ParserState.AfterValue
+
+            elif self.state == ParserState.AfterValue:
+                if self.isinstruct:
+                    self.state = ParserState.BeforeField
+                else:
+                    self.state = ParserState.BeforeTID
+
+            else:
+                _assert(self.state == ParserState.EOF)
+
+    def next(self):
+        if self.hasnext():
+            self.needhasnext = True
+            return self.valuetid
+        else:
+            return -1
+
+    def push(self, typeid, nextposition, nextremaining):
+        self.containerstack.append(ContainerRec(nextpos=nextposition, tid=typeid, remaining=nextremaining))
+
+    def stepin(self):
+        _assert(self.valuetid in [TID_STRUCT, TID_LIST, TID_SEXP] and not self.eof,
+                "valuetid=%s eof=%s" % (self.valuetid, self.eof))
+        _assert((not self.valueisnull or self.state == ParserState.AfterValue) and
+               (self.valueisnull or self.state == ParserState.BeforeValue))
+
+        nextrem = self.localremaining
+        if nextrem != -1:
+            nextrem -= self.valuelen
+            if nextrem < 0:
+                nextrem = 0
+        self.push(self.parenttid, self.stream.tell() + self.valuelen, nextrem)
+
+        self.isinstruct = (self.valuetid == TID_STRUCT)
+        if self.isinstruct:
+            self.state = ParserState.BeforeField
+        else:
+            self.state = ParserState.BeforeTID
+
+        self.localremaining = self.valuelen
+        self.parenttid = self.valuetid
+        self.clearvalue()
+        self.needhasnext = True
+
+    def stepout(self):
+        rec = self.containerstack.pop()
+
+        self.eof = False
+        self.parenttid = rec.tid
+        if self.parenttid == TID_STRUCT:
+            self.isinstruct = True
+            self.state = ParserState.BeforeField
+        else:
+            self.isinstruct = False
+            self.state = ParserState.BeforeTID
+        self.needhasnext = True
+
+        self.clearvalue()
+        curpos = self.stream.tell()
+        if rec.nextpos > curpos:
+            self.skip(rec.nextpos - curpos)
+        else:
+            _assert(rec.nextpos == curpos)
+
+        self.localremaining = rec.remaining
+
+    def read(self, count=1):
+        if self.localremaining != -1:
+            self.localremaining -= count
+            _assert(self.localremaining >= 0)
+
+        result = self.stream.read(count)
+        if len(result) == 0:
+            raise EOFError()
+        return result
+
+    def readfieldid(self):
+        if self.localremaining != -1 and self.localremaining < 1:
+            return -1
+
+        try:
+            return self.readvaruint()
+        except EOFError:
+            return -1
+
+    def readtypeid(self):
+        if self.localremaining != -1:
+            if self.localremaining < 1:
+                return -1
+            self.localremaining -= 1
+
+        b = self.stream.read(1)
+        if len(b) < 1:
+            return -1
+        b = bord(b)
+        result = b >> 4
+        ln = b & 0xF
+
+        if ln == LEN_IS_VAR_LEN:
+            ln = self.readvaruint()
+        elif ln == LEN_IS_NULL:
+            ln = 0
+            self.state = ParserState.AfterValue
+        elif result == TID_NULL:
+            # Must have LEN_IS_NULL
+            _assert(False)
+        elif result == TID_BOOLEAN:
+            _assert(ln <= 1)
+            self.valueistrue = (ln == 1)
+            ln = 0
+            self.state = ParserState.AfterValue
+        elif result == TID_STRUCT:
+            if ln == 1:
+                ln = self.readvaruint()
+
+        self.valuelen = ln
+        return result
+
+    def readvarint(self):
+        b = bord(self.read())
+        negative = ((b & 0x40) != 0)
+        result = (b & 0x3F)
+
+        i = 0
+        while (b & 0x80) == 0 and i < 4:
+            b = bord(self.read())
+            result = (result << 7) | (b & 0x7F)
+            i += 1
+
+        _assert(i < 4 or (b & 0x80) != 0, "int overflow")
+
+        if negative:
+            return -result
+        return result
+
+    def readvaruint(self):
+        b = bord(self.read())
+        result = (b & 0x7F)
+
+        i = 0
+        while (b & 0x80) == 0 and i < 4:
+            b = bord(self.read())
+            result = (result << 7) | (b & 0x7F)
+            i += 1
+
+        _assert(i < 4 or (b & 0x80) != 0, "int overflow")
+
+        return result
+
+    def readdecimal(self):
+        if self.valuelen == 0:
+            return 0.
+
+        rem = self.localremaining - self.valuelen
+        self.localremaining = self.valuelen
+        exponent = self.readvarint()
+
+        _assert(self.localremaining > 0, "Only exponent in ReadDecimal")
+        _assert(self.localremaining <= 8, "Decimal overflow")
+
+        signed = False
+        b = [bord(x) for x in self.read(self.localremaining)]
+        if (b[0] & 0x80) != 0:
+            b[0] = b[0] & 0x7F
+            signed = True
+
+        # Convert variably sized network order integer into 64-bit little endian
+        j = 0
+        vb = [0] * 8
+        for i in range(len(b), -1, -1):
+            vb[i] = b[j]
+            j += 1
+
+        v = struct.unpack("<Q", b"".join(bchr(x) for x in vb))[0]
+
+        result = v * (10 ** exponent)
+        if signed:
+            result = -result
+
+        self.localremaining = rem
+        return result
+
+    def skip(self, count):
+        if self.localremaining != -1:
+            self.localremaining -= count
+            if self.localremaining < 0:
+                raise EOFError()
+
+        self.stream.seek(count, os.SEEK_CUR)
+
+    def parsesymboltable(self):
+        self.next() # shouldn't do anything?
+
+        _assert(self.valuetid == TID_STRUCT)
+
+        if self.didimports:
+            return
+
+        self.stepin()
+
+        fieldtype = self.next()
+        while fieldtype != -1:
+            if not self.valueisnull:
+                _assert(self.valuefieldid == SID_IMPORTS, "Unsupported symbol table field id")
+
+                if fieldtype == TID_LIST:
+                    self.gatherimports()
+
+            fieldtype = self.next()
+
+        self.stepout()
+        self.didimports = True
+
+    def gatherimports(self):
+        self.stepin()
+
+        t = self.next()
+        while t != -1:
+            if not self.valueisnull and t == TID_STRUCT:
+                self.readimport()
+
+            t = self.next()
+
+        self.stepout()
+
+    def readimport(self):
+        version = -1
+        maxid = -1
+        name = ""
+
+        self.stepin()
+
+        t = self.next()
+        while t != -1:
+            if not self.valueisnull and self.valuefieldid != SID_UNKNOWN:
+                if self.valuefieldid == SID_NAME:
+                    name = self.stringvalue()
+                elif self.valuefieldid == SID_VERSION:
+                    version = self.intvalue()
+                elif self.valuefieldid == SID_MAX_ID:
+                    maxid = self.intvalue()
+
+            t = self.next()
+
+        self.stepout()
+
+        if name == "" or name == SystemSymbols.ION:
+            return
+
+        if version < 1:
+            version = 1
+
+        table = self.findcatalogitem(name)
+        if maxid < 0:
+            _assert(table is not None and version == table.version, "Import %s lacks maxid" % name)
+            maxid = len(table.symnames)
+
+        if table is not None:
+            self.symbols.import_(table, min(maxid, len(table.symnames)))
+        else:
+            self.symbols.importunknown(name, maxid)
+
+    def intvalue(self):
+        _assert(self.valuetid in [TID_POSINT, TID_NEGINT], "Not an int")
+
+        self.preparevalue()
+        return self.value
+
+    def stringvalue(self):
+        _assert(self.valuetid == TID_STRING, "Not a string")
+
+        if self.valueisnull:
+            return ""
+
+        self.preparevalue()
+        return self.value
+
+    def symbolvalue(self):
+        _assert(self.valuetid == TID_SYMBOL, "Not a symbol")
+
+        self.preparevalue()
+        result = self.symbols.findbyid(self.value)
+        if result == "":
+            result = "SYMBOL#%d" % self.value
+        return result
+
+    def lobvalue(self):
+        _assert(self.valuetid in [TID_CLOB, TID_BLOB], "Not a LOB type: %s" % self.getfieldname())
+
+        if self.valueisnull:
+            return None
+
+        result = self.read(self.valuelen)
+        self.state = ParserState.AfterValue
+        return result
+
+    def decimalvalue(self):
+        _assert(self.valuetid == TID_DECIMAL, "Not a decimal")
+
+        self.preparevalue()
+        return self.value
+
+    def preparevalue(self):
+        if self.value is None:
+            self.loadscalarvalue()
+
+    def loadscalarvalue(self):
+        if self.valuetid not in [TID_NULL, TID_BOOLEAN, TID_POSINT, TID_NEGINT,
+                                 TID_FLOAT, TID_DECIMAL, TID_TIMESTAMP,
+                                 TID_SYMBOL, TID_STRING]:
+            return
+
+        if self.valueisnull:
+            self.value = None
+            return
+
+        if self.valuetid == TID_STRING:
+            self.value = self.read(self.valuelen).decode("UTF-8")
+
+        elif self.valuetid in (TID_POSINT, TID_NEGINT, TID_SYMBOL):
+            if self.valuelen == 0:
+                self.value = 0
+            else:
+                _assert(self.valuelen <= 4, "int too long: %d" % self.valuelen)
+                v = 0
+                for i in range(self.valuelen - 1, -1, -1):
+                    v = (v | (bord(self.read()) << (i * 8)))
+
+                if self.valuetid == TID_NEGINT:
+                    self.value = -v
+                else:
+                    self.value = v
+
+        elif self.valuetid == TID_DECIMAL:
+            self.value = self.readdecimal()
+
+        #else:
+        #    _assert(False, "Unhandled scalar type %d" % self.valuetid)
+
+        self.state = ParserState.AfterValue
+
+    def clearvalue(self):
+        self.valuetid = -1
+        self.value = None
+        self.valueisnull = False
+        self.valuefieldid = SID_UNKNOWN
+        self.annotations = []
+
+    def loadannotations(self):
+        ln = self.readvaruint()
+        maxpos = self.stream.tell() + ln
+        while self.stream.tell() < maxpos:
+            self.annotations.append(self.readvaruint())
+        self.valuetid = self.readtypeid()
+
+    def checkversionmarker(self):
+        for i in VERSION_MARKER:
+            _assert(self.read() == i, "Unknown version marker")
+
+        self.valuelen = 0
+        self.valuetid = TID_SYMBOL
+        self.value = SID_ION_1_0
+        self.valueisnull = False
+        self.valuefieldid = SID_UNKNOWN
+        self.state = ParserState.AfterValue
+
+    def findcatalogitem(self, name):
+        for result in self.catalog:
+            if result.name == name:
+                return result
+
+    def forceimport(self, symbols):
+        item = IonCatalogItem("Forced", 1, symbols)
+        self.symbols.import_(item, len(symbols))
+
+    def getfieldname(self):
+        if self.valuefieldid == SID_UNKNOWN:
+            return ""
+        return self.symbols.findbyid(self.valuefieldid)
+
+    def getfieldnamesymbol(self):
+        return SymbolToken(self.getfieldname(), self.valuefieldid)
+
+    def gettypename(self):
+        if len(self.annotations) == 0:
+            return ""
+
+        return self.symbols.findbyid(self.annotations[0])
+
+    @staticmethod
+    def printlob(b):
+        if b is None:
+            return "null"
+
+        result = ""
+        for i in b:
+            result += ("%02x " % bord(i))
+
+        if len(result) > 0:
+            result = result[:-1]
+        return result
+
+    def ionwalk(self, supert, indent, lst):
+        while self.hasnext():
+            if supert == TID_STRUCT:
+                L = self.getfieldname() + ":"
+            else:
+                L = ""
+
+            t = self.next()
+            if t in [TID_STRUCT, TID_LIST]:
+                if L != "":
+                    lst.append(indent + L)
+                L = self.gettypename()
+                if L != "":
+                    lst.append(indent + L + "::")
+                if t == TID_STRUCT:
+                    lst.append(indent + "{")
+                else:
+                    lst.append(indent + "[")
+
+                self.stepin()
+                self.ionwalk(t, indent + "  ", lst)
+                self.stepout()
+
+                if t == TID_STRUCT:
+                    lst.append(indent + "}")
+                else:
+                    lst.append(indent + "]")
+
+            else:
+                if t == TID_STRING:
+                    L += ('"%s"' % self.stringvalue())
+                elif t in [TID_CLOB, TID_BLOB]:
+                    L += ("{%s}" % self.printlob(self.lobvalue()))
+                elif t == TID_POSINT:
+                    L += str(self.intvalue())
+                elif t == TID_SYMBOL:
+                    tn = self.gettypename()
+                    if tn != "":
+                        tn += "::"
+                    L += tn + self.symbolvalue()
+                elif t == TID_DECIMAL:
+                    L += str(self.decimalvalue())
+                else:
+                    L += ("TID %d" % t)
+                lst.append(indent + L)
+
+    def print_(self, lst):
+        self.reset()
+        self.ionwalk(-1, "", lst)
+
+
+SYM_NAMES = [ 'com.amazon.drm.Envelope@1.0',
+              'com.amazon.drm.EnvelopeMetadata@1.0', 'size', 'page_size',
+              'encryption_key', 'encryption_transformation',
+              'encryption_voucher', 'signing_key', 'signing_algorithm',
+              'signing_voucher', 'com.amazon.drm.EncryptedPage@1.0',
+              'cipher_text', 'cipher_iv', 'com.amazon.drm.Signature@1.0',
+              'data', 'com.amazon.drm.EnvelopeIndexTable@1.0', 'length',
+              'offset', 'algorithm', 'encoded', 'encryption_algorithm',
+              'hashing_algorithm', 'expires', 'format', 'id',
+              'lock_parameters', 'strategy', 'com.amazon.drm.Key@1.0',
+              'com.amazon.drm.KeySet@1.0', 'com.amazon.drm.PIDv3@1.0',
+              'com.amazon.drm.PlainTextPage@1.0',
+              'com.amazon.drm.PlainText@1.0', 'com.amazon.drm.PrivateKey@1.0',
+              'com.amazon.drm.PublicKey@1.0', 'com.amazon.drm.SecretKey@1.0',
+              'com.amazon.drm.Voucher@1.0', 'public_key', 'private_key',
+              'com.amazon.drm.KeyPair@1.0', 'com.amazon.drm.ProtectedData@1.0',
+              'doctype', 'com.amazon.drm.EnvelopeIndexTableOffset@1.0',
+              'enddoc', 'license_type', 'license', 'watermark', 'key', 'value',
+              'com.amazon.drm.License@1.0', 'category', 'metadata',
+              'categorized_metadata', 'com.amazon.drm.CategorizedMetadata@1.0',
+              'com.amazon.drm.VoucherEnvelope@1.0', 'mac', 'voucher',
+              'com.amazon.drm.ProtectedData@2.0',
+              'com.amazon.drm.Envelope@2.0',
+              'com.amazon.drm.EnvelopeMetadata@2.0',
+              'com.amazon.drm.EncryptedPage@2.0',
+              'com.amazon.drm.PlainText@2.0', 'compression_algorithm',
+              'com.amazon.drm.Compressed@1.0', 'priority', 'refines']
+
+def addprottable(ion):
+    ion.addtocatalog("ProtectedData", 1, SYM_NAMES)
+
+
+def pkcs7pad(msg, blocklen):
+    paddinglen = blocklen - len(msg) % blocklen
+    padding = bchr(paddinglen) * paddinglen
+    return msg + padding
+
+
+def pkcs7unpad(msg, blocklen):
+    _assert(len(msg) % blocklen == 0)
+
+    paddinglen = bord(msg[-1])
+    _assert(paddinglen > 0 and paddinglen <= blocklen, "Incorrect padding - Wrong key")
+    _assert(msg[-paddinglen:] == bchr(paddinglen) * paddinglen, "Incorrect padding - Wrong key")
+
+    return msg[:-paddinglen]
+
+
+class DrmIonVoucher(object):
+    envelope = None
+    voucher = None
+    drmkey = None
+    license_type = "Unknown"
+
+    encalgorithm = ""
+    enctransformation = ""
+    hashalgorithm = ""
+
+    lockparams = None
+
+    ciphertext = b""
+    cipheriv = b""
+    secretkey = b""
+
+    def __init__(self, voucherenv, dsn, secret):
+        self.dsn,self.secret = dsn,secret
+
+        self.lockparams = []
+
+        self.envelope = BinaryIonParser(voucherenv)
+        addprottable(self.envelope)
+
+    def decryptvoucher(self):
+        shared = "PIDv3" + self.encalgorithm + self.enctransformation + self.hashalgorithm
+
+        self.lockparams.sort()
+        for param in self.lockparams:
+            if param == "ACCOUNT_SECRET":
+                shared += param + self.secret
+            elif param == "CLIENT_ID":
+                shared += param + self.dsn
+            else:
+                _assert(False, "Unknown lock parameter: %s" % param)
+
+        sharedsecret = shared.encode("UTF-8")
+
+        key = hmac.new(sharedsecret, sharedsecret[:5], digestmod=hashlib.sha256).digest()
+        aes = AES.new(key[:32], AES.MODE_CBC, self.cipheriv[:16])
+        b = aes.decrypt(self.ciphertext)
+        b = pkcs7unpad(b, 16)
+
+        self.drmkey = BinaryIonParser(StringIO(b))
+        addprottable(self.drmkey)
+
+        _assert(self.drmkey.hasnext() and self.drmkey.next() == TID_LIST and self.drmkey.gettypename() == "com.amazon.drm.KeySet@1.0",
+                "Expected KeySet, got %s" % self.drmkey.gettypename())
+
+        self.drmkey.stepin()
+        while self.drmkey.hasnext():
+            self.drmkey.next()
+            if self.drmkey.gettypename() != "com.amazon.drm.SecretKey@1.0":
+                continue
+
+            self.drmkey.stepin()
+            while self.drmkey.hasnext():
+                self.drmkey.next()
+                if self.drmkey.getfieldname() == "algorithm":
+                    _assert(self.drmkey.stringvalue() == "AES", "Unknown cipher algorithm: %s" % self.drmkey.stringvalue())
+                elif self.drmkey.getfieldname() == "format":
+                    _assert(self.drmkey.stringvalue() == "RAW", "Unknown key format: %s" % self.drmkey.stringvalue())
+                elif self.drmkey.getfieldname() == "encoded":
+                    self.secretkey = self.drmkey.lobvalue()
+
+            self.drmkey.stepout()
+            break
+
+        self.drmkey.stepout()
+
+    def parse(self):
+        self.envelope.reset()
+        _assert(self.envelope.hasnext(), "Envelope is empty")
+        _assert(self.envelope.next() == TID_STRUCT and self.envelope.gettypename() == "com.amazon.drm.VoucherEnvelope@1.0",
+                "Unknown type encountered in envelope, expected VoucherEnvelope")
+
+        self.envelope.stepin()
+        while self.envelope.hasnext():
+            self.envelope.next()
+            field = self.envelope.getfieldname()
+            if field == "voucher":
+                self.voucher = BinaryIonParser(StringIO(self.envelope.lobvalue()))
+                addprottable(self.voucher)
+                continue
+            elif field != "strategy":
+                continue
+
+            _assert(self.envelope.gettypename() == "com.amazon.drm.PIDv3@1.0", "Unknown strategy: %s" % self.envelope.gettypename())
+
+            self.envelope.stepin()
+            while self.envelope.hasnext():
+                self.envelope.next()
+                field = self.envelope.getfieldname()
+                if field == "encryption_algorithm":
+                    self.encalgorithm = self.envelope.stringvalue()
+                elif field == "encryption_transformation":
+                    self.enctransformation = self.envelope.stringvalue()
+                elif field == "hashing_algorithm":
+                    self.hashalgorithm = self.envelope.stringvalue()
+                elif field == "lock_parameters":
+                    self.envelope.stepin()
+                    while self.envelope.hasnext():
+                        _assert(self.envelope.next() == TID_STRING, "Expected string list for lock_parameters")
+                        self.lockparams.append(self.envelope.stringvalue())
+                    self.envelope.stepout()
+
+            self.envelope.stepout()
+
+        self.parsevoucher()
+
+    def parsevoucher(self):
+        _assert(self.voucher.hasnext(), "Voucher is empty")
+        _assert(self.voucher.next() == TID_STRUCT and self.voucher.gettypename() == "com.amazon.drm.Voucher@1.0",
+                "Unknown type, expected Voucher")
+
+        self.voucher.stepin()
+        while self.voucher.hasnext():
+            self.voucher.next()
+
+            if self.voucher.getfieldname() == "cipher_iv":
+                self.cipheriv = self.voucher.lobvalue()
+            elif self.voucher.getfieldname() == "cipher_text":
+                self.ciphertext = self.voucher.lobvalue()
+            elif self.voucher.getfieldname() == "license":
+                _assert(self.voucher.gettypename() == "com.amazon.drm.License@1.0",
+                        "Unknown license: %s" % self.voucher.gettypename())
+                self.voucher.stepin()
+                while self.voucher.hasnext():
+                    self.voucher.next()
+                    if self.voucher.getfieldname() == "license_type":
+                        self.license_type = self.voucher.stringvalue()
+                self.voucher.stepout()
+
+    def printenvelope(self, lst):
+        self.envelope.print_(lst)
+
+    def printkey(self, lst):
+        if self.voucher is None:
+            self.parse()
+        if self.drmkey is None:
+            self.decryptvoucher()
+
+        self.drmkey.print_(lst)
+
+    def printvoucher(self, lst):
+        if self.voucher is None:
+            self.parse()
+
+        self.voucher.print_(lst)
+
+    def getlicensetype(self):
+        return self.license_type
+
+
+class DrmIon(object):
+    ion = None
+    voucher = None
+    vouchername = ""
+    key = b""
+    onvoucherrequired = None
+
+    def __init__(self, ionstream, onvoucherrequired):
+        self.ion = BinaryIonParser(ionstream)
+        addprottable(self.ion)
+        self.onvoucherrequired = onvoucherrequired
+
+    def parse(self, outpages):
+        self.ion.reset()
+
+        _assert(self.ion.hasnext(), "DRMION envelope is empty")
+        _assert(self.ion.next() == TID_SYMBOL and self.ion.gettypename() == "doctype", "Expected doctype symbol")
+        _assert(self.ion.next() == TID_LIST and self.ion.gettypename() in ["com.amazon.drm.Envelope@1.0", "com.amazon.drm.Envelope@2.0"],
+                "Unknown type encountered in DRMION envelope, expected Envelope, got %s" % self.ion.gettypename())
+
+        while True:
+            if self.ion.gettypename() == "enddoc":
+                break
+
+            self.ion.stepin()
+            while self.ion.hasnext():
+                self.ion.next()
+
+                if self.ion.gettypename() in ["com.amazon.drm.EnvelopeMetadata@1.0", "com.amazon.drm.EnvelopeMetadata@2.0"]:
+                    self.ion.stepin()
+                    while self.ion.hasnext():
+                        self.ion.next()
+                        if self.ion.getfieldname() != "encryption_voucher":
+                            continue
+
+                        if self.vouchername == "":
+                            self.vouchername = self.ion.stringvalue()
+                            self.voucher = self.onvoucherrequired(self.vouchername)
+                            self.key = self.voucher.secretkey
+                            _assert(self.key is not None, "Unable to obtain secret key from voucher")
+                        else:
+                            _assert(self.vouchername == self.ion.stringvalue(),
+                                    "Unexpected: Different vouchers required for same file?")
+
+                    self.ion.stepout()
+
+                elif self.ion.gettypename() in ["com.amazon.drm.EncryptedPage@1.0", "com.amazon.drm.EncryptedPage@2.0"]:
+                    decompress = False
+                    ct = None
+                    civ = None
+                    self.ion.stepin()
+                    while self.ion.hasnext():
+                        self.ion.next()
+                        if self.ion.gettypename() == "com.amazon.drm.Compressed@1.0":
+                            decompress = True
+                        if self.ion.getfieldname() == "cipher_text":
+                            ct = self.ion.lobvalue()
+                        elif self.ion.getfieldname() == "cipher_iv":
+                            civ = self.ion.lobvalue()
+
+                    if ct is not None and civ is not None:
+                        self.processpage(ct, civ, outpages, decompress)
+                    self.ion.stepout()
+
+            self.ion.stepout()
+            if not self.ion.hasnext():
+                break
+            self.ion.next()
+
+    def print_(self, lst):
+        self.ion.print_(lst)
+
+    def processpage(self, ct, civ, outpages, decompress):
+        aes = AES.new(self.key[:16], AES.MODE_CBC, civ[:16])
+        msg = pkcs7unpad(aes.decrypt(ct), 16)
+
+        if not decompress:
+            outpages.write(msg)
+            return
+
+        _assert(msg[0] == b"\x00", "LZMA UseFilter not supported")
+
+        if calibre_lzma is not None:
+            with calibre_lzma.decompress(msg[1:], bufsize=0x1000000) as f:
+                f.seek(0)
+                outpages.write(f.read())
+            return
+
+        decomp = lzma.LZMADecompressor(format=lzma.FORMAT_ALONE)
+        while not decomp.eof:
+            segment = decomp.decompress(msg[1:])
+            msg = b"" # Contents were internally buffered after the first call
+            outpages.write(segment)

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

@@ -3,10 +3,13 @@
 
 from __future__ import with_statement
 
-# k4mobidedrm.py, version 5.3
-# Copyright © 2009-2015 by ApprenticeHarper et al.
+# k4mobidedrm.py
+# Copyright © 2008-2017 by Apprentice Harper et al.
 
-# engine to remove drm from Kindle and Mobipocket ebooks
+__license__ = 'GPL v3'
+__version__ = '5.5'
+
+# Engine to remove drm from Kindle and Mobipocket ebooks
 # for personal use for archiving and converting your ebooks
 
 # PLEASE DO NOT PIRATE EBOOKS!
@@ -17,12 +20,11 @@ from __future__ import with_statement
 # 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
+#    unswindle, DarkReverser, ApprenticeAlf, 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
@@ -57,9 +59,8 @@ from __future__ import with_statement
 #  5.2 - Fixed error in command line processing of unicode arguments
 #  5.3 - Changed Android support to allow passing of backup .ab files
 #  5.4 - Recognise KFX files masquerading as azw, even if we can't decrypt them yet.
-
-__version__ = '5.4'
-
+#  5.5 - Added GPL v3 licence explicitly.
+#  5.x - Invoke KFXZipBook to handle zipped KFX files
 
 import sys, os, re
 import csv
@@ -83,11 +84,13 @@ if inCalibre:
     from calibre_plugins.dedrm import topazextract
     from calibre_plugins.dedrm import kgenpids
     from calibre_plugins.dedrm import androidkindlekey
+    from calibre_plugins.dedrm import kfxdedrm
 else:
     import mobidedrm
     import topazextract
     import kgenpids
     import androidkindlekey
+    import kfxdedrm
 
 # Wrap a stream so that output gets flushed immediately
 # and also make sure that any unicode strings get
@@ -197,13 +200,15 @@ def GetDecryptedBook(infile, kDatabases, androidFiles, serials, pids, starttime
     mobi = True
     magic8 = open(infile,'rb').read(8)
     if magic8 == '\xeaDRMION\xee':
-        raise DrmException(u"KFX format detected. This format cannot be decrypted yet.")
-        
+        raise DrmException(u"The .kfx DRMION file cannot be decrypted by itself. A .kfx-zip archive containing a DRM voucher is required.")
+
     magic3 = magic8[:3]
     if magic3 == 'TPZ':
         mobi = False
 
-    if mobi:
+    if magic8[:4] == 'PK\x03\x04':
+        mb = kfxdedrm.KFXZipBook(infile)
+    elif mobi:
         mb = mobidedrm.MobiBook(infile)
     else:
         mb = topazextract.TopazBook(infile)
@@ -295,7 +300,7 @@ def usage(progname):
 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__)
+    print u"K4MobiDeDrm v{0}.\nCopyright © 2008-2017 Apprentice Harper et al.".format(__version__)
 
     try:
         opts, args = getopt.getopt(argv[1:], "k:p:s:a:")

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

@@ -0,0 +1,108 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+from __future__ import with_statement
+
+# Engine to remove drm from Kindle KFX ebooks
+
+import os
+import shutil
+import zipfile
+
+try:
+    from cStringIO import StringIO
+except ImportError:
+    from StringIO import StringIO
+
+try:
+    import ion
+except:
+    from calibre_plugins.dedrm import ion
+
+
+__license__ = 'GPL v3'
+__version__ = '1.0'
+
+
+class KFXZipBook:
+    def __init__(self, infile):
+        self.infile = infile
+        self.voucher = None
+        self.decrypted = {}
+
+    def getPIDMetaInfo(self):
+        return (None, None)
+
+    def processBook(self, totalpids):
+        with zipfile.ZipFile(self.infile, 'r') as zf:
+            for filename in zf.namelist():
+                data = zf.read(filename)
+                if data.startswith('\xeaDRMION\xee'):
+                    if self.voucher is None:
+                        self.decrypt_voucher(totalpids)
+                    print u'Decrypting KFX DRMION: {0}'.format(filename)
+                    outfile = StringIO()
+                    ion.DrmIon(StringIO(data[8:-8]), lambda name: self.voucher).parse(outfile)
+                    self.decrypted[filename] = outfile.getvalue()
+
+        if not self.decrypted:
+            print(u'The .kfx-zip archive does not contain an encrypted DRMION file')
+
+    def decrypt_voucher(self, totalpids):
+        with zipfile.ZipFile(self.infile, 'r') as zf:
+            for info in zf.infolist():
+                if info.file_size < 0x10000:
+                    data = zf.read(info.filename)
+                    if data.startswith('\xe0\x01\x00\xea') and 'ProtectedData' in data:
+                        break   # found DRM voucher
+            else:
+                raise Exception(u'The .kfx-zip archive contains an encrypted DRMION file without a DRM voucher')
+
+        print u'Decrypting KFX DRM voucher: {0}'.format(info.filename)
+
+        for pid in [''] + totalpids:
+            for dsn_len,secret_len in [(0,0), (16,0), (16,40), (32,40), (40,40)]:
+                if len(pid) == dsn_len + secret_len:
+                    break       # split pid into DSN and account secret
+            else:
+                continue
+
+            try:
+                voucher = ion.DrmIonVoucher(StringIO(data), pid[:dsn_len], pid[dsn_len:])
+                voucher.parse()
+                voucher.decryptvoucher()
+                break
+            except:
+                pass
+        else:
+            raise Exception(u'Failed to decrypt KFX DRM voucher with any key')
+
+        print u'KFX DRM voucher successfully decrypted'
+
+        license_type = voucher.getlicensetype()
+        if license_type != "Purchase":
+            raise Exception((u'This book is licensed as {0}. '
+                    'These tools are intended for use on purchased books.').format(license_type))
+
+        self.voucher = voucher
+
+    def getBookTitle(self):
+        return os.path.splitext(os.path.split(self.infile)[1])[0]
+
+    def getBookExtension(self):
+        return '.kfx-zip'
+
+    def getBookType(self):
+        return 'KFX-ZIP'
+
+    def cleanup(self):
+        pass
+
+    def getFile(self, outpath):
+        if not self.decrypted:
+            shutil.copyfile(self.infile, outpath)
+        else:
+            with zipfile.ZipFile(self.infile, 'r') as zif:
+                with zipfile.ZipFile(outpath, 'w') as zof:
+                    for info in zif.infolist():
+                        zof.writestr(info, self.decrypted.get(info.filename, zif.read(info.filename)))

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

@@ -4,10 +4,15 @@
 from __future__ import with_statement
 
 # kgenpids.py
-# Copyright © 2010-2015 by some_updates, Apprentice Alf and Apprentice Harper
+# Copyright © 2008-2017 Apprentice Harper et al.
+
+__license__ = 'GPL v3'
+__version__ = '2.1'
 
 # Revision history:
 #  2.0   - Fix for non-ascii Windows user names
+#  2.1   - Actual fix for non-ascii WIndows user names.
+#  x.x   - Return information needed for KFX decryption
 
 import sys
 import os, csv
@@ -168,6 +173,9 @@ def pidFromSerial(s, l):
 
 # Parse the EXTH header records and use the Kindle serial number to calculate the book pid.
 def getKindlePids(rec209, token, serialnum):
+    if rec209 is None:
+        return [serialnum]
+
     pids=[]
 
     if isinstance(serialnum,unicode):
@@ -195,20 +203,6 @@ def getK4Pids(rec209, token, kindleDatabase):
     global charMap1
     pids = []
 
-    try:
-        # Get the Mazama Random number
-        MazamaRandomNumber = (kindleDatabase[1])['MazamaRandomNumber'].decode('hex')
-
-        # Get the IDString used to decode the Kindle Info file
-        IDString = (kindleDatabase[1])['IDString'].decode('hex')
-
-        # Get the UserName stored when the Kindle Info file was decoded
-        UserName = (kindleDatabase[1])['UserName'].decode('hex')
-
-    except KeyError:
-        print u"Keys not found in the database {0}.".format(kindleDatabase[0])
-        return pids
-
     try:
         # Get the kindle account token, if present
         kindleAccountToken = (kindleDatabase[1])['kindle.account.tokens'].decode('hex')
@@ -217,14 +211,51 @@ def getK4Pids(rec209, token, kindleDatabase):
         kindleAccountToken=""
         pass
 
-    # Get the ID string used
-    encodedIDString = encodeHash(IDString,charMap1)
+    try:
+        # Get the DSN token, if present
+        DSN = (kindleDatabase[1])['DSN'].decode('hex')
+        print u"Got DSN key from database {0}".format(kindleDatabase[0])
+    except KeyError:
+        # See if we have the info to generate the DSN
+        try:
+            # Get the Mazama Random number
+            MazamaRandomNumber = (kindleDatabase[1])['MazamaRandomNumber'].decode('hex')
+            #print u"Got MazamaRandomNumber from database {0}".format(kindleDatabase[0])
 
-    # Get the current user name
-    encodedUsername = encodeHash(UserName,charMap1)
+            try:
+                # Get the SerialNumber token, if present
+                IDString = (kindleDatabase[1])['SerialNumber'].decode('hex')
+                print u"Got SerialNumber from database {0}".format(kindleDatabase[0])
+            except KeyError:
+                 # Get the IDString we added
+                IDString = (kindleDatabase[1])['IDString'].decode('hex')
 
-    # concat, hash and encode to calculate the DSN
-    DSN = encode(SHA1(MazamaRandomNumber+encodedIDString+encodedUsername),charMap1)
+            try:
+                # Get the UsernameHash token, if present
+                encodedUsername = (kindleDatabase[1])['UsernameHash'].decode('hex')
+                print u"Got UsernameHash from database {0}".format(kindleDatabase[0])
+            except KeyError:
+                # Get the UserName we added
+                UserName = (kindleDatabase[1])['UserName'].decode('hex')
+                # encode it
+                encodedUsername = encodeHash(UserName,charMap1)
+                #print u"encodedUsername",encodedUsername.encode('hex')
+        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)
+        #print u"encodedIDString",encodedIDString.encode('hex')
+
+        # concat, hash and encode to calculate the DSN
+        DSN = encode(SHA1(MazamaRandomNumber+encodedIDString+encodedUsername),charMap1)
+        #print u"DSN",DSN.encode('hex')
+        pass
+
+    if rec209 is None:
+        pids.append(DSN+kindleAccountToken)
+        return pids
 
     # Compute the device PID (for which I can tell, is used for nothing).
     table =  generatePidEncryptionTable()

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

@@ -4,7 +4,10 @@
 from __future__ import with_statement
 
 # kindlekey.py
-# Copyright © 2010-2017 by some_updates, Apprentice Alf and Apprentice Harper
+# Copyright © 2008-2017 Apprentice Harper et al.
+
+__license__ = 'GPL v3'
+__version__ = '2.5'
 
 # Revision history:
 #  1.0   - Kindle info file decryption, extracted from k4mobidedrm, etc.
@@ -23,15 +26,14 @@ from __future__ import with_statement
 #  2.2   - Fixes for Macs with bonded ethernet ports
 #          Also removed old .kinfo file support (pre-2011)
 #  2.3   - Added more field names thanks to concavegit's KFX code.
+#  2.4   - Fix for complex Mac disk setups, thanks to Tibs
+#  2.5   - Final Fix for Windows user names with non-ascii characters, thanks to oneofusoneofus
 
 
 """
 Retrieve Kindle for PC/Mac user key.
 """
 
-__license__ = 'GPL v3'
-__version__ = '2.3'
-
 import sys, os, re
 from struct import pack, unpack, unpack_from
 import json
@@ -886,10 +888,18 @@ if iswindows:
                 if errcd == 234:
                     # bad wine implementation up through wine 1.3.21
                     return "AlternateUserName"
+                # double the buffer size
                 buffer = create_unicode_buffer(len(buffer) * 2)
                 size.value = len(buffer)
-            # return low byte of the unicode value of each character of the username
-            return buffer.value.encode('utf-16-le')[::2]
+            
+            # replace any non-ASCII values with 0xfffd
+            for i in xrange(0,len(buffer)):
+                if buffer[i]>u"\u007f":
+                    #print u"swapping char "+str(i)+" ("+buffer[i]+")"
+                    buffer[i] = u"\ufffd"
+            # return utf-8 encoding of modified username
+            #print u"modified username:"+buffer.value
+            return buffer.value.encode('utf-8')
         return GetUserName
     GetUserName = GetUserName()
 
@@ -1014,7 +1024,12 @@ if iswindows:
             'SerialNumber',\
             'UsernameHash',\
             'kindle.directedid.info',\
-            'DSN'
+            'DSN',\
+            'kindle.accounttype.info',\
+            'krx.flashcardsplugin.data.encryption_key',\
+            'krx.notebookexportplugin.data.encryption_key',\
+            'proxy.http.password',\
+            'proxy.http.username'
             ]
         DB = {}
         with open(kInfoFile, 'rb') as infoReader:
@@ -1267,7 +1282,7 @@ elif isosx:
         for j in xrange(cnt):
             resline = reslst[j]
             if resline.startswith('/dev'):
-                (devpart, mpath) = resline.split(' on ')
+                (devpart, mpath) = resline.split(' on ')[:2]
                 dpart = devpart[5:]
                 names.append(dpart)
         return names

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

@@ -1,10 +1,12 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
-# mobidedrm.py, version 0.38
+# mobidedrm.py
 # Copyright © 2008 The Dark Reverser
-#
-# Modified 2008–2012 by some_updates, DiapDealer and Apprentice Alf
+# Portions © 2008–2017 Apprentice Harper et al.
+
+__license__ = 'GPL v3'
+__version__ = u"0.42"
 
 # This is a python script. You need a Python interpreter to run it.
 # For example, ActiveState Python, which exists for windows.
@@ -69,9 +71,7 @@
 #  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
-
-
-__version__ = u"0.41"
+#  0.42 - Added GPL v3 licence. updated/removed some print statements
 
 import sys
 import os
@@ -244,7 +244,7 @@ class MobiBook:
         pass
 
     def __init__(self, infile):
-        print u"MobiDeDrm v{0:s}.\nCopyright © 2008-2012 The Dark Reverser et al.".format(__version__)
+        print u"MobiDeDrm v{0:s}.\nCopyright © 2008-2017 The Dark Reverser, Apprentice Harper et al.".format(__version__)
 
         try:
             from alfcrypto import Pukall_Cipher
@@ -288,10 +288,10 @@ class MobiBook:
         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)
+        #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)
+            #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.
@@ -516,7 +516,7 @@ 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"MobiDeDrm v{0:s}.\nCopyright © 2008-2017 The Dark Reverser, Apprentice Harper 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)

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/DeDRM_App.pyw

@@ -30,8 +30,10 @@
 #   6.5.2 - Fix for a new tag in Topaz ebooks
 #   6.5.3 - Explicitly warn about KFX files
 #   6.5.4 - PDF float fix.
+#   6.5.5 - Kindle for PC/Accented characters in username fix.
+#   6.6.0 - Initial KFX support from TomThumb
 
-__version__ = '6.5.4'
+__version__ = '6.6.0'
 
 import sys
 import os, os.path
@@ -382,6 +384,9 @@ class PrefsDialog(Toplevel):
                     ('Kindle','.azw3'),
                     ('Kindle','.azw4'),
                     ('Kindle','.tpz'),
+                    ('Kindle','.azw8'),
+                    ('Kindle','.kfx'),
+                    ('Kindle','.kfx-zip'),
                     ('Kindle','.mobi'),
                     ('Kindle','.prc'),
                     ('eReader','.pdb'),
@@ -597,7 +602,7 @@ class ConvDialog(Toplevel):
             self.p2 = Process(target=processPDB, args=(q, infile, outdir, rscpath))
             self.p2.start()
             return 0
-        if ext in ['.azw', '.azw1', '.azw3', '.azw4', '.prc', '.mobi', '.pobi', '.tpz']:
+        if ext in ['.azw', '.azw1', '.azw3', '.azw4', '.prc', '.mobi', '.pobi', '.tpz', '.azw8', '.kfx', '.kfx-zip']:
             self.p2 = Process(target=processK4MOBI,args=(q, infile, outdir, rscpath))
             self.p2.start()
             return 0

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/__init__.py

@@ -2,6 +2,10 @@
 # -*- coding: utf-8 -*-
 
 from __future__ import with_statement
+
+# __init__.py for DeDRM_plugin
+# Copyright © 2008-2018 Apprentice Harper et al.
+
 __license__   = 'GPL v3'
 __docformat__ = 'restructuredtext en'
 
@@ -58,6 +62,9 @@ __docformat__ = 'restructuredtext en'
 #   6.5.2 - Another Topaz fix
 #   6.5.3 - Warn about KFX files explicitly
 #   6.5.4 - Mac App Fix, improve PDF decryption, handle latest tcl changes in ActivePython
+#   6.5.5 - Finally a fix for the Windows non-ASCII user names.
+#   6.6.0 - Add kfx and kfx-zip as supported file types (also invoke this plugin if the original
+#           imported format was azw8 since that may be converted to kfx)
 
 
 """
@@ -65,7 +72,7 @@ Decrypt DRMed ebooks.
 """
 
 PLUGIN_NAME = u"DeDRM"
-PLUGIN_VERSION_TUPLE = (6, 5, 4)
+PLUGIN_VERSION_TUPLE = (6, 6, 0)
 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'
@@ -113,7 +120,7 @@ class DeDRM(FileTypePlugin):
     author                  = u"Apprentice Alf, Aprentice Harper, The Dark Reverser and i♥cabbages"
     version                 = PLUGIN_VERSION_TUPLE
     minimum_calibre_version = (1, 0, 0)  # Compiled python libraries cannot be imported in earlier versions.
-    file_types              = set(['epub','pdf','pdb','prc','mobi','pobi','azw','azw1','azw3','azw4','tpz'])
+    file_types              = set(['epub','pdf','pdb','prc','mobi','pobi','azw','azw1','azw3','azw4','azw8','tpz','kfx','kfx-zip'])
     on_import               = True
     priority                = 600
 
@@ -289,8 +296,8 @@ class DeDRM(FileTypePlugin):
                 except Exception, e:
                     pass
 
-            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
+            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
 
         # import the Adobe Adept ePub handler
         import calibre_plugins.dedrm.ineptepub as ineptepub
@@ -380,17 +387,19 @@ class DeDRM(FileTypePlugin):
                             except:
                                 print u"{0} v{1}: Exception when saving a new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime)
                                 traceback.print_exc()
-                            print u"{0} v{1}: Decrypted with new default key after {3:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+                            print u"{0} v{1}: Decrypted with new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
                             # Return the modified PersistentTemporary file to calibre.
                             return of.name
 
                         print u"{0} v{1}: Failed to decrypt with new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
                 except Exception, e:
+                    print u"{0} v{1}: Unexpected Exception trying a new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime)
+                    traceback.print_exc()
                     pass
 
             # Something went wrong with decryption.
-            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
+            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
 
         # Not a Barnes & Noble nor an Adobe Adept
         # Import the fixed epub.
@@ -489,8 +498,8 @@ class DeDRM(FileTypePlugin):
                 pass
 
         # Something went wrong with decryption.
-        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
+        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
 
 
     def KindleMobiDecrypt(self,path_to_ebook):
@@ -557,8 +566,8 @@ class DeDRM(FileTypePlugin):
                     pass
             if not decoded:
                 #if you reached here then no luck raise and exception
-                print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-                raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
+                print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+                raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
 
         of = self.temporary_file(book.getBookExtension())
         book.getFile(of.name)
@@ -592,8 +601,8 @@ class DeDRM(FileTypePlugin):
 
             print u"{0} v{1}: Failed to decrypt with key {2:s} after {3:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,keyname_masked,time.time()-self.starttime)
 
-        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
+        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
 
 
     def run(self, path_to_ebook):
@@ -606,7 +615,7 @@ class DeDRM(FileTypePlugin):
         self.starttime = time.time()
 
         booktype = os.path.splitext(path_to_ebook)[1].lower()[1:]
-        if booktype in ['prc','mobi','pobi','azw','azw1','azw3','azw4','tpz']:
+        if booktype in ['prc','mobi','pobi','azw','azw1','azw3','azw4','tpz','kfx-zip']:
             # Kindle/Mobipocket
             decrypted_ebook = self.KindleMobiDecrypt(path_to_ebook)
         elif booktype == 'pdb':

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/ion.py

@@ -0,0 +1,981 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+# Pascal implementation by lulzkabulz. Python translation by apprenticenaomi. DeDRM integration by anon.
+# BinaryIon.pas + DrmIon.pas + IonSymbols.pas
+
+from __future__ import with_statement
+
+import collections
+import hashlib
+import hmac
+import os
+import os.path
+import struct
+
+try:
+    from cStringIO import StringIO
+except ImportError:
+    from StringIO import StringIO
+
+from Crypto.Cipher import AES
+from Crypto.Util.py3compat import bchr, bord
+
+try:
+    # lzma library from calibre 2.35.0 or later
+    import lzma.lzma1 as calibre_lzma
+except:
+    calibre_lzma = None
+    try:
+        import lzma
+    except:
+        # Need pip backports.lzma on Python <3.3
+        from backports import lzma
+
+
+TID_NULL = 0
+TID_BOOLEAN = 1
+TID_POSINT = 2
+TID_NEGINT = 3
+TID_FLOAT = 4
+TID_DECIMAL = 5
+TID_TIMESTAMP = 6
+TID_SYMBOL = 7
+TID_STRING = 8
+TID_CLOB = 9
+TID_BLOB = 0xA
+TID_LIST = 0xB
+TID_SEXP = 0xC
+TID_STRUCT = 0xD
+TID_TYPEDECL = 0xE
+TID_UNUSED = 0xF
+
+
+SID_UNKNOWN = -1
+SID_ION = 1
+SID_ION_1_0 = 2
+SID_ION_SYMBOL_TABLE = 3
+SID_NAME = 4
+SID_VERSION = 5
+SID_IMPORTS = 6
+SID_SYMBOLS = 7
+SID_MAX_ID = 8
+SID_ION_SHARED_SYMBOL_TABLE = 9
+SID_ION_1_0_MAX = 10
+
+
+LEN_IS_VAR_LEN = 0xE
+LEN_IS_NULL = 0xF
+
+
+VERSION_MARKER = b"\x01\x00\xEA"
+
+
+# asserts must always raise exceptions for proper functioning
+def _assert(test, msg="Exception"):
+    if not test:
+        raise Exception(msg)
+
+
+class SystemSymbols(object):
+    ION = '$ion'
+    ION_1_0 = '$ion_1_0'
+    ION_SYMBOL_TABLE = '$ion_symbol_table'
+    NAME = 'name'
+    VERSION = 'version'
+    IMPORTS = 'imports'
+    SYMBOLS = 'symbols'
+    MAX_ID = 'max_id'
+    ION_SHARED_SYMBOL_TABLE = '$ion_shared_symbol_table'
+
+
+class IonCatalogItem(object):
+    name = ""
+    version = 0
+    symnames = []
+
+    def __init__(self, name, version, symnames):
+        self.name = name
+        self.version = version
+        self.symnames = symnames
+
+
+class SymbolToken(object):
+    text = ""
+    sid = 0
+
+    def __init__(self, text, sid):
+        if text == "" and sid == 0:
+            raise ValueError("Symbol token must have Text or SID")
+
+        self.text = text
+        self.sid = sid
+
+
+class SymbolTable(object):
+    table = None
+
+    def __init__(self):
+        self.table = [None] * SID_ION_1_0_MAX
+        self.table[SID_ION] = SystemSymbols.ION
+        self.table[SID_ION_1_0] = SystemSymbols.ION_1_0
+        self.table[SID_ION_SYMBOL_TABLE] = SystemSymbols.ION_SYMBOL_TABLE
+        self.table[SID_NAME] = SystemSymbols.NAME
+        self.table[SID_VERSION] = SystemSymbols.VERSION
+        self.table[SID_IMPORTS] = SystemSymbols.IMPORTS
+        self.table[SID_SYMBOLS] = SystemSymbols.SYMBOLS
+        self.table[SID_MAX_ID] = SystemSymbols.MAX_ID
+        self.table[SID_ION_SHARED_SYMBOL_TABLE] = SystemSymbols.ION_SHARED_SYMBOL_TABLE
+
+    def findbyid(self, sid):
+        if sid < 1:
+            raise ValueError("Invalid symbol id")
+
+        if sid < len(self.table):
+            return self.table[sid]
+        else:
+            return ""
+
+    def import_(self, table, maxid):
+        for i in range(maxid):
+            self.table.append(table.symnames[i])
+
+    def importunknown(self, name, maxid):
+        for i in range(maxid):
+            self.table.append("%s#%d" % (name, i + 1))
+
+
+class ParserState:
+    Invalid,BeforeField,BeforeTID,BeforeValue,AfterValue,EOF = 1,2,3,4,5,6
+
+ContainerRec = collections.namedtuple("ContainerRec", "nextpos, tid, remaining")
+
+
+class BinaryIonParser(object):
+    eof = False
+    state = None
+    localremaining = 0
+    needhasnext = False
+    isinstruct = False
+    valuetid = 0
+    valuefieldid = 0
+    parenttid = 0
+    valuelen = 0
+    valueisnull = False
+    valueistrue = False
+    value = None
+    didimports = False
+
+    def __init__(self, stream):
+        self.annotations = []
+        self.catalog = []
+
+        self.stream = stream
+        self.initpos = stream.tell()
+        self.reset()
+        self.symbols = SymbolTable()
+
+    def reset(self):
+        self.state = ParserState.BeforeTID
+        self.needhasnext = True
+        self.localremaining = -1
+        self.eof = False
+        self.isinstruct = False
+        self.containerstack = []
+        self.stream.seek(self.initpos)
+
+    def addtocatalog(self, name, version, symbols):
+        self.catalog.append(IonCatalogItem(name, version, symbols))
+
+    def hasnext(self):
+        while self.needhasnext and not self.eof:
+            self.hasnextraw()
+            if len(self.containerstack) == 0 and not self.valueisnull:
+                if self.valuetid == TID_SYMBOL:
+                    if self.value == SID_ION_1_0:
+                        self.needhasnext = True
+                elif self.valuetid == TID_STRUCT:
+                    for a in self.annotations:
+                        if a == SID_ION_SYMBOL_TABLE:
+                            self.parsesymboltable()
+                            self.needhasnext = True
+                            break
+        return not self.eof
+
+    def hasnextraw(self):
+        self.clearvalue()
+        while self.valuetid == -1 and not self.eof:
+            self.needhasnext = False
+            if self.state == ParserState.BeforeField:
+                _assert(self.valuefieldid == SID_UNKNOWN)
+
+                self.valuefieldid = self.readfieldid()
+                if self.valuefieldid != SID_UNKNOWN:
+                    self.state = ParserState.BeforeTID
+                else:
+                    self.eof = True
+
+            elif self.state == ParserState.BeforeTID:
+                self.state = ParserState.BeforeValue
+                self.valuetid = self.readtypeid()
+                if self.valuetid == -1:
+                    self.state = ParserState.EOF
+                    self.eof = True
+                    break
+
+                if self.valuetid == TID_TYPEDECL:
+                    if self.valuelen == 0:
+                        self.checkversionmarker()
+                    else:
+                        self.loadannotations()
+
+            elif self.state == ParserState.BeforeValue:
+                self.skip(self.valuelen)
+                self.state = ParserState.AfterValue
+
+            elif self.state == ParserState.AfterValue:
+                if self.isinstruct:
+                    self.state = ParserState.BeforeField
+                else:
+                    self.state = ParserState.BeforeTID
+
+            else:
+                _assert(self.state == ParserState.EOF)
+
+    def next(self):
+        if self.hasnext():
+            self.needhasnext = True
+            return self.valuetid
+        else:
+            return -1
+
+    def push(self, typeid, nextposition, nextremaining):
+        self.containerstack.append(ContainerRec(nextpos=nextposition, tid=typeid, remaining=nextremaining))
+
+    def stepin(self):
+        _assert(self.valuetid in [TID_STRUCT, TID_LIST, TID_SEXP] and not self.eof,
+                "valuetid=%s eof=%s" % (self.valuetid, self.eof))
+        _assert((not self.valueisnull or self.state == ParserState.AfterValue) and
+               (self.valueisnull or self.state == ParserState.BeforeValue))
+
+        nextrem = self.localremaining
+        if nextrem != -1:
+            nextrem -= self.valuelen
+            if nextrem < 0:
+                nextrem = 0
+        self.push(self.parenttid, self.stream.tell() + self.valuelen, nextrem)
+
+        self.isinstruct = (self.valuetid == TID_STRUCT)
+        if self.isinstruct:
+            self.state = ParserState.BeforeField
+        else:
+            self.state = ParserState.BeforeTID
+
+        self.localremaining = self.valuelen
+        self.parenttid = self.valuetid
+        self.clearvalue()
+        self.needhasnext = True
+
+    def stepout(self):
+        rec = self.containerstack.pop()
+
+        self.eof = False
+        self.parenttid = rec.tid
+        if self.parenttid == TID_STRUCT:
+            self.isinstruct = True
+            self.state = ParserState.BeforeField
+        else:
+            self.isinstruct = False
+            self.state = ParserState.BeforeTID
+        self.needhasnext = True
+
+        self.clearvalue()
+        curpos = self.stream.tell()
+        if rec.nextpos > curpos:
+            self.skip(rec.nextpos - curpos)
+        else:
+            _assert(rec.nextpos == curpos)
+
+        self.localremaining = rec.remaining
+
+    def read(self, count=1):
+        if self.localremaining != -1:
+            self.localremaining -= count
+            _assert(self.localremaining >= 0)
+
+        result = self.stream.read(count)
+        if len(result) == 0:
+            raise EOFError()
+        return result
+
+    def readfieldid(self):
+        if self.localremaining != -1 and self.localremaining < 1:
+            return -1
+
+        try:
+            return self.readvaruint()
+        except EOFError:
+            return -1
+
+    def readtypeid(self):
+        if self.localremaining != -1:
+            if self.localremaining < 1:
+                return -1
+            self.localremaining -= 1
+
+        b = self.stream.read(1)
+        if len(b) < 1:
+            return -1
+        b = bord(b)
+        result = b >> 4
+        ln = b & 0xF
+
+        if ln == LEN_IS_VAR_LEN:
+            ln = self.readvaruint()
+        elif ln == LEN_IS_NULL:
+            ln = 0
+            self.state = ParserState.AfterValue
+        elif result == TID_NULL:
+            # Must have LEN_IS_NULL
+            _assert(False)
+        elif result == TID_BOOLEAN:
+            _assert(ln <= 1)
+            self.valueistrue = (ln == 1)
+            ln = 0
+            self.state = ParserState.AfterValue
+        elif result == TID_STRUCT:
+            if ln == 1:
+                ln = self.readvaruint()
+
+        self.valuelen = ln
+        return result
+
+    def readvarint(self):
+        b = bord(self.read())
+        negative = ((b & 0x40) != 0)
+        result = (b & 0x3F)
+
+        i = 0
+        while (b & 0x80) == 0 and i < 4:
+            b = bord(self.read())
+            result = (result << 7) | (b & 0x7F)
+            i += 1
+
+        _assert(i < 4 or (b & 0x80) != 0, "int overflow")
+
+        if negative:
+            return -result
+        return result
+
+    def readvaruint(self):
+        b = bord(self.read())
+        result = (b & 0x7F)
+
+        i = 0
+        while (b & 0x80) == 0 and i < 4:
+            b = bord(self.read())
+            result = (result << 7) | (b & 0x7F)
+            i += 1
+
+        _assert(i < 4 or (b & 0x80) != 0, "int overflow")
+
+        return result
+
+    def readdecimal(self):
+        if self.valuelen == 0:
+            return 0.
+
+        rem = self.localremaining - self.valuelen
+        self.localremaining = self.valuelen
+        exponent = self.readvarint()
+
+        _assert(self.localremaining > 0, "Only exponent in ReadDecimal")
+        _assert(self.localremaining <= 8, "Decimal overflow")
+
+        signed = False
+        b = [bord(x) for x in self.read(self.localremaining)]
+        if (b[0] & 0x80) != 0:
+            b[0] = b[0] & 0x7F
+            signed = True
+
+        # Convert variably sized network order integer into 64-bit little endian
+        j = 0
+        vb = [0] * 8
+        for i in range(len(b), -1, -1):
+            vb[i] = b[j]
+            j += 1
+
+        v = struct.unpack("<Q", b"".join(bchr(x) for x in vb))[0]
+
+        result = v * (10 ** exponent)
+        if signed:
+            result = -result
+
+        self.localremaining = rem
+        return result
+
+    def skip(self, count):
+        if self.localremaining != -1:
+            self.localremaining -= count
+            if self.localremaining < 0:
+                raise EOFError()
+
+        self.stream.seek(count, os.SEEK_CUR)
+
+    def parsesymboltable(self):
+        self.next() # shouldn't do anything?
+
+        _assert(self.valuetid == TID_STRUCT)
+
+        if self.didimports:
+            return
+
+        self.stepin()
+
+        fieldtype = self.next()
+        while fieldtype != -1:
+            if not self.valueisnull:
+                _assert(self.valuefieldid == SID_IMPORTS, "Unsupported symbol table field id")
+
+                if fieldtype == TID_LIST:
+                    self.gatherimports()
+
+            fieldtype = self.next()
+
+        self.stepout()
+        self.didimports = True
+
+    def gatherimports(self):
+        self.stepin()
+
+        t = self.next()
+        while t != -1:
+            if not self.valueisnull and t == TID_STRUCT:
+                self.readimport()
+
+            t = self.next()
+
+        self.stepout()
+
+    def readimport(self):
+        version = -1
+        maxid = -1
+        name = ""
+
+        self.stepin()
+
+        t = self.next()
+        while t != -1:
+            if not self.valueisnull and self.valuefieldid != SID_UNKNOWN:
+                if self.valuefieldid == SID_NAME:
+                    name = self.stringvalue()
+                elif self.valuefieldid == SID_VERSION:
+                    version = self.intvalue()
+                elif self.valuefieldid == SID_MAX_ID:
+                    maxid = self.intvalue()
+
+            t = self.next()
+
+        self.stepout()
+
+        if name == "" or name == SystemSymbols.ION:
+            return
+
+        if version < 1:
+            version = 1
+
+        table = self.findcatalogitem(name)
+        if maxid < 0:
+            _assert(table is not None and version == table.version, "Import %s lacks maxid" % name)
+            maxid = len(table.symnames)
+
+        if table is not None:
+            self.symbols.import_(table, min(maxid, len(table.symnames)))
+        else:
+            self.symbols.importunknown(name, maxid)
+
+    def intvalue(self):
+        _assert(self.valuetid in [TID_POSINT, TID_NEGINT], "Not an int")
+
+        self.preparevalue()
+        return self.value
+
+    def stringvalue(self):
+        _assert(self.valuetid == TID_STRING, "Not a string")
+
+        if self.valueisnull:
+            return ""
+
+        self.preparevalue()
+        return self.value
+
+    def symbolvalue(self):
+        _assert(self.valuetid == TID_SYMBOL, "Not a symbol")
+
+        self.preparevalue()
+        result = self.symbols.findbyid(self.value)
+        if result == "":
+            result = "SYMBOL#%d" % self.value
+        return result
+
+    def lobvalue(self):
+        _assert(self.valuetid in [TID_CLOB, TID_BLOB], "Not a LOB type: %s" % self.getfieldname())
+
+        if self.valueisnull:
+            return None
+
+        result = self.read(self.valuelen)
+        self.state = ParserState.AfterValue
+        return result
+
+    def decimalvalue(self):
+        _assert(self.valuetid == TID_DECIMAL, "Not a decimal")
+
+        self.preparevalue()
+        return self.value
+
+    def preparevalue(self):
+        if self.value is None:
+            self.loadscalarvalue()
+
+    def loadscalarvalue(self):
+        if self.valuetid not in [TID_NULL, TID_BOOLEAN, TID_POSINT, TID_NEGINT,
+                                 TID_FLOAT, TID_DECIMAL, TID_TIMESTAMP,
+                                 TID_SYMBOL, TID_STRING]:
+            return
+
+        if self.valueisnull:
+            self.value = None
+            return
+
+        if self.valuetid == TID_STRING:
+            self.value = self.read(self.valuelen).decode("UTF-8")
+
+        elif self.valuetid in (TID_POSINT, TID_NEGINT, TID_SYMBOL):
+            if self.valuelen == 0:
+                self.value = 0
+            else:
+                _assert(self.valuelen <= 4, "int too long: %d" % self.valuelen)
+                v = 0
+                for i in range(self.valuelen - 1, -1, -1):
+                    v = (v | (bord(self.read()) << (i * 8)))
+
+                if self.valuetid == TID_NEGINT:
+                    self.value = -v
+                else:
+                    self.value = v
+
+        elif self.valuetid == TID_DECIMAL:
+            self.value = self.readdecimal()
+
+        #else:
+        #    _assert(False, "Unhandled scalar type %d" % self.valuetid)
+
+        self.state = ParserState.AfterValue
+
+    def clearvalue(self):
+        self.valuetid = -1
+        self.value = None
+        self.valueisnull = False
+        self.valuefieldid = SID_UNKNOWN
+        self.annotations = []
+
+    def loadannotations(self):
+        ln = self.readvaruint()
+        maxpos = self.stream.tell() + ln
+        while self.stream.tell() < maxpos:
+            self.annotations.append(self.readvaruint())
+        self.valuetid = self.readtypeid()
+
+    def checkversionmarker(self):
+        for i in VERSION_MARKER:
+            _assert(self.read() == i, "Unknown version marker")
+
+        self.valuelen = 0
+        self.valuetid = TID_SYMBOL
+        self.value = SID_ION_1_0
+        self.valueisnull = False
+        self.valuefieldid = SID_UNKNOWN
+        self.state = ParserState.AfterValue
+
+    def findcatalogitem(self, name):
+        for result in self.catalog:
+            if result.name == name:
+                return result
+
+    def forceimport(self, symbols):
+        item = IonCatalogItem("Forced", 1, symbols)
+        self.symbols.import_(item, len(symbols))
+
+    def getfieldname(self):
+        if self.valuefieldid == SID_UNKNOWN:
+            return ""
+        return self.symbols.findbyid(self.valuefieldid)
+
+    def getfieldnamesymbol(self):
+        return SymbolToken(self.getfieldname(), self.valuefieldid)
+
+    def gettypename(self):
+        if len(self.annotations) == 0:
+            return ""
+
+        return self.symbols.findbyid(self.annotations[0])
+
+    @staticmethod
+    def printlob(b):
+        if b is None:
+            return "null"
+
+        result = ""
+        for i in b:
+            result += ("%02x " % bord(i))
+
+        if len(result) > 0:
+            result = result[:-1]
+        return result
+
+    def ionwalk(self, supert, indent, lst):
+        while self.hasnext():
+            if supert == TID_STRUCT:
+                L = self.getfieldname() + ":"
+            else:
+                L = ""
+
+            t = self.next()
+            if t in [TID_STRUCT, TID_LIST]:
+                if L != "":
+                    lst.append(indent + L)
+                L = self.gettypename()
+                if L != "":
+                    lst.append(indent + L + "::")
+                if t == TID_STRUCT:
+                    lst.append(indent + "{")
+                else:
+                    lst.append(indent + "[")
+
+                self.stepin()
+                self.ionwalk(t, indent + "  ", lst)
+                self.stepout()
+
+                if t == TID_STRUCT:
+                    lst.append(indent + "}")
+                else:
+                    lst.append(indent + "]")
+
+            else:
+                if t == TID_STRING:
+                    L += ('"%s"' % self.stringvalue())
+                elif t in [TID_CLOB, TID_BLOB]:
+                    L += ("{%s}" % self.printlob(self.lobvalue()))
+                elif t == TID_POSINT:
+                    L += str(self.intvalue())
+                elif t == TID_SYMBOL:
+                    tn = self.gettypename()
+                    if tn != "":
+                        tn += "::"
+                    L += tn + self.symbolvalue()
+                elif t == TID_DECIMAL:
+                    L += str(self.decimalvalue())
+                else:
+                    L += ("TID %d" % t)
+                lst.append(indent + L)
+
+    def print_(self, lst):
+        self.reset()
+        self.ionwalk(-1, "", lst)
+
+
+SYM_NAMES = [ 'com.amazon.drm.Envelope@1.0',
+              'com.amazon.drm.EnvelopeMetadata@1.0', 'size', 'page_size',
+              'encryption_key', 'encryption_transformation',
+              'encryption_voucher', 'signing_key', 'signing_algorithm',
+              'signing_voucher', 'com.amazon.drm.EncryptedPage@1.0',
+              'cipher_text', 'cipher_iv', 'com.amazon.drm.Signature@1.0',
+              'data', 'com.amazon.drm.EnvelopeIndexTable@1.0', 'length',
+              'offset', 'algorithm', 'encoded', 'encryption_algorithm',
+              'hashing_algorithm', 'expires', 'format', 'id',
+              'lock_parameters', 'strategy', 'com.amazon.drm.Key@1.0',
+              'com.amazon.drm.KeySet@1.0', 'com.amazon.drm.PIDv3@1.0',
+              'com.amazon.drm.PlainTextPage@1.0',
+              'com.amazon.drm.PlainText@1.0', 'com.amazon.drm.PrivateKey@1.0',
+              'com.amazon.drm.PublicKey@1.0', 'com.amazon.drm.SecretKey@1.0',
+              'com.amazon.drm.Voucher@1.0', 'public_key', 'private_key',
+              'com.amazon.drm.KeyPair@1.0', 'com.amazon.drm.ProtectedData@1.0',
+              'doctype', 'com.amazon.drm.EnvelopeIndexTableOffset@1.0',
+              'enddoc', 'license_type', 'license', 'watermark', 'key', 'value',
+              'com.amazon.drm.License@1.0', 'category', 'metadata',
+              'categorized_metadata', 'com.amazon.drm.CategorizedMetadata@1.0',
+              'com.amazon.drm.VoucherEnvelope@1.0', 'mac', 'voucher',
+              'com.amazon.drm.ProtectedData@2.0',
+              'com.amazon.drm.Envelope@2.0',
+              'com.amazon.drm.EnvelopeMetadata@2.0',
+              'com.amazon.drm.EncryptedPage@2.0',
+              'com.amazon.drm.PlainText@2.0', 'compression_algorithm',
+              'com.amazon.drm.Compressed@1.0', 'priority', 'refines']
+
+def addprottable(ion):
+    ion.addtocatalog("ProtectedData", 1, SYM_NAMES)
+
+
+def pkcs7pad(msg, blocklen):
+    paddinglen = blocklen - len(msg) % blocklen
+    padding = bchr(paddinglen) * paddinglen
+    return msg + padding
+
+
+def pkcs7unpad(msg, blocklen):
+    _assert(len(msg) % blocklen == 0)
+
+    paddinglen = bord(msg[-1])
+    _assert(paddinglen > 0 and paddinglen <= blocklen, "Incorrect padding - Wrong key")
+    _assert(msg[-paddinglen:] == bchr(paddinglen) * paddinglen, "Incorrect padding - Wrong key")
+
+    return msg[:-paddinglen]
+
+
+class DrmIonVoucher(object):
+    envelope = None
+    voucher = None
+    drmkey = None
+    license_type = "Unknown"
+
+    encalgorithm = ""
+    enctransformation = ""
+    hashalgorithm = ""
+
+    lockparams = None
+
+    ciphertext = b""
+    cipheriv = b""
+    secretkey = b""
+
+    def __init__(self, voucherenv, dsn, secret):
+        self.dsn,self.secret = dsn,secret
+
+        self.lockparams = []
+
+        self.envelope = BinaryIonParser(voucherenv)
+        addprottable(self.envelope)
+
+    def decryptvoucher(self):
+        shared = "PIDv3" + self.encalgorithm + self.enctransformation + self.hashalgorithm
+
+        self.lockparams.sort()
+        for param in self.lockparams:
+            if param == "ACCOUNT_SECRET":
+                shared += param + self.secret
+            elif param == "CLIENT_ID":
+                shared += param + self.dsn
+            else:
+                _assert(False, "Unknown lock parameter: %s" % param)
+
+        sharedsecret = shared.encode("UTF-8")
+
+        key = hmac.new(sharedsecret, sharedsecret[:5], digestmod=hashlib.sha256).digest()
+        aes = AES.new(key[:32], AES.MODE_CBC, self.cipheriv[:16])
+        b = aes.decrypt(self.ciphertext)
+        b = pkcs7unpad(b, 16)
+
+        self.drmkey = BinaryIonParser(StringIO(b))
+        addprottable(self.drmkey)
+
+        _assert(self.drmkey.hasnext() and self.drmkey.next() == TID_LIST and self.drmkey.gettypename() == "com.amazon.drm.KeySet@1.0",
+                "Expected KeySet, got %s" % self.drmkey.gettypename())
+
+        self.drmkey.stepin()
+        while self.drmkey.hasnext():
+            self.drmkey.next()
+            if self.drmkey.gettypename() != "com.amazon.drm.SecretKey@1.0":
+                continue
+
+            self.drmkey.stepin()
+            while self.drmkey.hasnext():
+                self.drmkey.next()
+                if self.drmkey.getfieldname() == "algorithm":
+                    _assert(self.drmkey.stringvalue() == "AES", "Unknown cipher algorithm: %s" % self.drmkey.stringvalue())
+                elif self.drmkey.getfieldname() == "format":
+                    _assert(self.drmkey.stringvalue() == "RAW", "Unknown key format: %s" % self.drmkey.stringvalue())
+                elif self.drmkey.getfieldname() == "encoded":
+                    self.secretkey = self.drmkey.lobvalue()
+
+            self.drmkey.stepout()
+            break
+
+        self.drmkey.stepout()
+
+    def parse(self):
+        self.envelope.reset()
+        _assert(self.envelope.hasnext(), "Envelope is empty")
+        _assert(self.envelope.next() == TID_STRUCT and self.envelope.gettypename() == "com.amazon.drm.VoucherEnvelope@1.0",
+                "Unknown type encountered in envelope, expected VoucherEnvelope")
+
+        self.envelope.stepin()
+        while self.envelope.hasnext():
+            self.envelope.next()
+            field = self.envelope.getfieldname()
+            if field == "voucher":
+                self.voucher = BinaryIonParser(StringIO(self.envelope.lobvalue()))
+                addprottable(self.voucher)
+                continue
+            elif field != "strategy":
+                continue
+
+            _assert(self.envelope.gettypename() == "com.amazon.drm.PIDv3@1.0", "Unknown strategy: %s" % self.envelope.gettypename())
+
+            self.envelope.stepin()
+            while self.envelope.hasnext():
+                self.envelope.next()
+                field = self.envelope.getfieldname()
+                if field == "encryption_algorithm":
+                    self.encalgorithm = self.envelope.stringvalue()
+                elif field == "encryption_transformation":
+                    self.enctransformation = self.envelope.stringvalue()
+                elif field == "hashing_algorithm":
+                    self.hashalgorithm = self.envelope.stringvalue()
+                elif field == "lock_parameters":
+                    self.envelope.stepin()
+                    while self.envelope.hasnext():
+                        _assert(self.envelope.next() == TID_STRING, "Expected string list for lock_parameters")
+                        self.lockparams.append(self.envelope.stringvalue())
+                    self.envelope.stepout()
+
+            self.envelope.stepout()
+
+        self.parsevoucher()
+
+    def parsevoucher(self):
+        _assert(self.voucher.hasnext(), "Voucher is empty")
+        _assert(self.voucher.next() == TID_STRUCT and self.voucher.gettypename() == "com.amazon.drm.Voucher@1.0",
+                "Unknown type, expected Voucher")
+
+        self.voucher.stepin()
+        while self.voucher.hasnext():
+            self.voucher.next()
+
+            if self.voucher.getfieldname() == "cipher_iv":
+                self.cipheriv = self.voucher.lobvalue()
+            elif self.voucher.getfieldname() == "cipher_text":
+                self.ciphertext = self.voucher.lobvalue()
+            elif self.voucher.getfieldname() == "license":
+                _assert(self.voucher.gettypename() == "com.amazon.drm.License@1.0",
+                        "Unknown license: %s" % self.voucher.gettypename())
+                self.voucher.stepin()
+                while self.voucher.hasnext():
+                    self.voucher.next()
+                    if self.voucher.getfieldname() == "license_type":
+                        self.license_type = self.voucher.stringvalue()
+                self.voucher.stepout()
+
+    def printenvelope(self, lst):
+        self.envelope.print_(lst)
+
+    def printkey(self, lst):
+        if self.voucher is None:
+            self.parse()
+        if self.drmkey is None:
+            self.decryptvoucher()
+
+        self.drmkey.print_(lst)
+
+    def printvoucher(self, lst):
+        if self.voucher is None:
+            self.parse()
+
+        self.voucher.print_(lst)
+
+    def getlicensetype(self):
+        return self.license_type
+
+
+class DrmIon(object):
+    ion = None
+    voucher = None
+    vouchername = ""
+    key = b""
+    onvoucherrequired = None
+
+    def __init__(self, ionstream, onvoucherrequired):
+        self.ion = BinaryIonParser(ionstream)
+        addprottable(self.ion)
+        self.onvoucherrequired = onvoucherrequired
+
+    def parse(self, outpages):
+        self.ion.reset()
+
+        _assert(self.ion.hasnext(), "DRMION envelope is empty")
+        _assert(self.ion.next() == TID_SYMBOL and self.ion.gettypename() == "doctype", "Expected doctype symbol")
+        _assert(self.ion.next() == TID_LIST and self.ion.gettypename() in ["com.amazon.drm.Envelope@1.0", "com.amazon.drm.Envelope@2.0"],
+                "Unknown type encountered in DRMION envelope, expected Envelope, got %s" % self.ion.gettypename())
+
+        while True:
+            if self.ion.gettypename() == "enddoc":
+                break
+
+            self.ion.stepin()
+            while self.ion.hasnext():
+                self.ion.next()
+
+                if self.ion.gettypename() in ["com.amazon.drm.EnvelopeMetadata@1.0", "com.amazon.drm.EnvelopeMetadata@2.0"]:
+                    self.ion.stepin()
+                    while self.ion.hasnext():
+                        self.ion.next()
+                        if self.ion.getfieldname() != "encryption_voucher":
+                            continue
+
+                        if self.vouchername == "":
+                            self.vouchername = self.ion.stringvalue()
+                            self.voucher = self.onvoucherrequired(self.vouchername)
+                            self.key = self.voucher.secretkey
+                            _assert(self.key is not None, "Unable to obtain secret key from voucher")
+                        else:
+                            _assert(self.vouchername == self.ion.stringvalue(),
+                                    "Unexpected: Different vouchers required for same file?")
+
+                    self.ion.stepout()
+
+                elif self.ion.gettypename() in ["com.amazon.drm.EncryptedPage@1.0", "com.amazon.drm.EncryptedPage@2.0"]:
+                    decompress = False
+                    ct = None
+                    civ = None
+                    self.ion.stepin()
+                    while self.ion.hasnext():
+                        self.ion.next()
+                        if self.ion.gettypename() == "com.amazon.drm.Compressed@1.0":
+                            decompress = True
+                        if self.ion.getfieldname() == "cipher_text":
+                            ct = self.ion.lobvalue()
+                        elif self.ion.getfieldname() == "cipher_iv":
+                            civ = self.ion.lobvalue()
+
+                    if ct is not None and civ is not None:
+                        self.processpage(ct, civ, outpages, decompress)
+                    self.ion.stepout()
+
+            self.ion.stepout()
+            if not self.ion.hasnext():
+                break
+            self.ion.next()
+
+    def print_(self, lst):
+        self.ion.print_(lst)
+
+    def processpage(self, ct, civ, outpages, decompress):
+        aes = AES.new(self.key[:16], AES.MODE_CBC, civ[:16])
+        msg = pkcs7unpad(aes.decrypt(ct), 16)
+
+        if not decompress:
+            outpages.write(msg)
+            return
+
+        _assert(msg[0] == b"\x00", "LZMA UseFilter not supported")
+
+        if calibre_lzma is not None:
+            with calibre_lzma.decompress(msg[1:], bufsize=0x1000000) as f:
+                f.seek(0)
+                outpages.write(f.read())
+            return
+
+        decomp = lzma.LZMADecompressor(format=lzma.FORMAT_ALONE)
+        while not decomp.eof:
+            segment = decomp.decompress(msg[1:])
+            msg = b"" # Contents were internally buffered after the first call
+            outpages.write(segment)

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/k4mobidedrm.py

@@ -3,10 +3,13 @@
 
 from __future__ import with_statement
 
-# k4mobidedrm.py, version 5.5
-# Copyright © 2009-2017 by ApprenticeHarper et al.
+# k4mobidedrm.py
+# Copyright © 2008-2017 by Apprentice Harper et al.
 
-# engine to remove drm from Kindle and Mobipocket ebooks
+__license__ = 'GPL v3'
+__version__ = '5.5'
+
+# Engine to remove drm from Kindle and Mobipocket ebooks
 # for personal use for archiving and converting your ebooks
 
 # PLEASE DO NOT PIRATE EBOOKS!
@@ -17,12 +20,11 @@ from __future__ import with_statement
 # 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
+#    unswindle, DarkReverser, ApprenticeAlf, 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
@@ -57,15 +59,12 @@ from __future__ import with_statement
 #  5.2 - Fixed error in command line processing of unicode arguments
 #  5.3 - Changed Android support to allow passing of backup .ab files
 #  5.4 - Recognise KFX files masquerading as azw, even if we can't decrypt them yet.
-#  5.5 - Support multiple input files
-
-__version__ = '5.5'
-
+#  5.5 - Added GPL v3 licence explicitly.
+#  5.x - Invoke KFXZipBook to handle zipped KFX files
 
 import sys, os, re
 import csv
 import getopt
-import glob
 import re
 import traceback
 import time
@@ -85,11 +84,13 @@ if inCalibre:
     from calibre_plugins.dedrm import topazextract
     from calibre_plugins.dedrm import kgenpids
     from calibre_plugins.dedrm import androidkindlekey
+    from calibre_plugins.dedrm import kfxdedrm
 else:
     import mobidedrm
     import topazextract
     import kgenpids
     import androidkindlekey
+    import kfxdedrm
 
 # Wrap a stream so that output gets flushed immediately
 # and also make sure that any unicode strings get
@@ -199,13 +200,15 @@ def GetDecryptedBook(infile, kDatabases, androidFiles, serials, pids, starttime
     mobi = True
     magic8 = open(infile,'rb').read(8)
     if magic8 == '\xeaDRMION\xee':
-        raise DrmException(u"KFX format detected. This format cannot be decrypted yet.")
-        
+        raise DrmException(u"The .kfx DRMION file cannot be decrypted by itself. A .kfx-zip archive containing a DRM voucher is required.")
+
     magic3 = magic8[:3]
     if magic3 == 'TPZ':
         mobi = False
 
-    if mobi:
+    if magic8[:4] == 'PK\x03\x04':
+        mb = kfxdedrm.KFXZipBook(infile)
+    elif mobi:
         mb = mobidedrm.MobiBook(infile)
     else:
         mb = topazextract.TopazBook(infile)
@@ -289,7 +292,7 @@ def decryptBook(infile, outdir, kDatabaseFiles, androidFiles, 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>] [ -a <AmazonSecureStorage.xml|backup.ab> ] <infile, infiles, or indir> <outdir>".format(progname)
+    print u"    {0} [-k <kindle.k4i>] [-p <comma separated PIDs>] [-s <comma separated Kindle serial numbers>] [ -a <AmazonSecureStorage.xml|backup.ab> ] <infile> <outdir>".format(progname)
 
 #
 # Main
@@ -297,7 +300,7 @@ def usage(progname):
 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__)
+    print u"K4MobiDeDrm v{0}.\nCopyright © 2008-2017 Apprentice Harper et al.".format(__version__)
 
     try:
         opts, args = getopt.getopt(argv[1:], "k:p:s:a:")
@@ -309,7 +312,8 @@ def cli_main():
         usage(progname)
         sys.exit(2)
 
-    outdir = args.pop()
+    infile = args[0]
+    outdir = args[1]
     kDatabaseFiles = []
     androidFiles = []
     serials = []
@@ -336,21 +340,7 @@ def cli_main():
     # try with built in Kindle Info files if not on Linux
     k4 = not sys.platform.startswith('linux')
 
-    filenames = []
-    for filename in args:
-        if os.path.isdir(filename):
-            for file_extension in ['.azw', '.azw1', '.azw3', '.azw4', '.prc', '.mobi', '.pobi']:
-                filenames += glob.glob(os.path.join(filename, '*%s' % file_extension))
-        else:
-            # Assume a filename
-            filenames.append(filename)
-
-    for infile in filenames:
-        result = decryptBook(infile, outdir, kDatabaseFiles, androidFiles, serials, pids)
-        if result != 0:
-            print u'Error with %r' % infile
-    # return last result only
-    return result
+    return decryptBook(infile, outdir, kDatabaseFiles, androidFiles, serials, pids)
 
 
 if __name__ == '__main__':

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/kfxdedrm.py

@@ -0,0 +1,108 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+from __future__ import with_statement
+
+# Engine to remove drm from Kindle KFX ebooks
+
+import os
+import shutil
+import zipfile
+
+try:
+    from cStringIO import StringIO
+except ImportError:
+    from StringIO import StringIO
+
+try:
+    import ion
+except:
+    from calibre_plugins.dedrm import ion
+
+
+__license__ = 'GPL v3'
+__version__ = '1.0'
+
+
+class KFXZipBook:
+    def __init__(self, infile):
+        self.infile = infile
+        self.voucher = None
+        self.decrypted = {}
+
+    def getPIDMetaInfo(self):
+        return (None, None)
+
+    def processBook(self, totalpids):
+        with zipfile.ZipFile(self.infile, 'r') as zf:
+            for filename in zf.namelist():
+                data = zf.read(filename)
+                if data.startswith('\xeaDRMION\xee'):
+                    if self.voucher is None:
+                        self.decrypt_voucher(totalpids)
+                    print u'Decrypting KFX DRMION: {0}'.format(filename)
+                    outfile = StringIO()
+                    ion.DrmIon(StringIO(data[8:-8]), lambda name: self.voucher).parse(outfile)
+                    self.decrypted[filename] = outfile.getvalue()
+
+        if not self.decrypted:
+            print(u'The .kfx-zip archive does not contain an encrypted DRMION file')
+
+    def decrypt_voucher(self, totalpids):
+        with zipfile.ZipFile(self.infile, 'r') as zf:
+            for info in zf.infolist():
+                if info.file_size < 0x10000:
+                    data = zf.read(info.filename)
+                    if data.startswith('\xe0\x01\x00\xea') and 'ProtectedData' in data:
+                        break   # found DRM voucher
+            else:
+                raise Exception(u'The .kfx-zip archive contains an encrypted DRMION file without a DRM voucher')
+
+        print u'Decrypting KFX DRM voucher: {0}'.format(info.filename)
+
+        for pid in [''] + totalpids:
+            for dsn_len,secret_len in [(0,0), (16,0), (16,40), (32,40), (40,40)]:
+                if len(pid) == dsn_len + secret_len:
+                    break       # split pid into DSN and account secret
+            else:
+                continue
+
+            try:
+                voucher = ion.DrmIonVoucher(StringIO(data), pid[:dsn_len], pid[dsn_len:])
+                voucher.parse()
+                voucher.decryptvoucher()
+                break
+            except:
+                pass
+        else:
+            raise Exception(u'Failed to decrypt KFX DRM voucher with any key')
+
+        print u'KFX DRM voucher successfully decrypted'
+
+        license_type = voucher.getlicensetype()
+        if license_type != "Purchase":
+            raise Exception((u'This book is licensed as {0}. '
+                    'These tools are intended for use on purchased books.').format(license_type))
+
+        self.voucher = voucher
+
+    def getBookTitle(self):
+        return os.path.splitext(os.path.split(self.infile)[1])[0]
+
+    def getBookExtension(self):
+        return '.kfx-zip'
+
+    def getBookType(self):
+        return 'KFX-ZIP'
+
+    def cleanup(self):
+        pass
+
+    def getFile(self, outpath):
+        if not self.decrypted:
+            shutil.copyfile(self.infile, outpath)
+        else:
+            with zipfile.ZipFile(self.infile, 'r') as zif:
+                with zipfile.ZipFile(outpath, 'w') as zof:
+                    for info in zif.infolist():
+                        zof.writestr(info, self.decrypted.get(info.filename, zif.read(info.filename)))

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/kgenpids.py

@@ -4,10 +4,15 @@
 from __future__ import with_statement
 
 # kgenpids.py
-# Copyright © 2010-2015 by some_updates, Apprentice Alf and Apprentice Harper
+# Copyright © 2008-2017 Apprentice Harper et al.
+
+__license__ = 'GPL v3'
+__version__ = '2.1'
 
 # Revision history:
 #  2.0   - Fix for non-ascii Windows user names
+#  2.1   - Actual fix for non-ascii WIndows user names.
+#  x.x   - Return information needed for KFX decryption
 
 import sys
 import os, csv
@@ -168,6 +173,9 @@ def pidFromSerial(s, l):
 
 # Parse the EXTH header records and use the Kindle serial number to calculate the book pid.
 def getKindlePids(rec209, token, serialnum):
+    if rec209 is None:
+        return [serialnum]
+
     pids=[]
 
     if isinstance(serialnum,unicode):
@@ -195,20 +203,6 @@ def getK4Pids(rec209, token, kindleDatabase):
     global charMap1
     pids = []
 
-    try:
-        # Get the Mazama Random number
-        MazamaRandomNumber = (kindleDatabase[1])['MazamaRandomNumber'].decode('hex')
-
-        # Get the IDString used to decode the Kindle Info file
-        IDString = (kindleDatabase[1])['IDString'].decode('hex')
-
-        # Get the UserName stored when the Kindle Info file was decoded
-        UserName = (kindleDatabase[1])['UserName'].decode('hex')
-
-    except KeyError:
-        print u"Keys not found in the database {0}.".format(kindleDatabase[0])
-        return pids
-
     try:
         # Get the kindle account token, if present
         kindleAccountToken = (kindleDatabase[1])['kindle.account.tokens'].decode('hex')
@@ -217,14 +211,51 @@ def getK4Pids(rec209, token, kindleDatabase):
         kindleAccountToken=""
         pass
 
-    # Get the ID string used
-    encodedIDString = encodeHash(IDString,charMap1)
+    try:
+        # Get the DSN token, if present
+        DSN = (kindleDatabase[1])['DSN'].decode('hex')
+        print u"Got DSN key from database {0}".format(kindleDatabase[0])
+    except KeyError:
+        # See if we have the info to generate the DSN
+        try:
+            # Get the Mazama Random number
+            MazamaRandomNumber = (kindleDatabase[1])['MazamaRandomNumber'].decode('hex')
+            #print u"Got MazamaRandomNumber from database {0}".format(kindleDatabase[0])
 
-    # Get the current user name
-    encodedUsername = encodeHash(UserName,charMap1)
+            try:
+                # Get the SerialNumber token, if present
+                IDString = (kindleDatabase[1])['SerialNumber'].decode('hex')
+                print u"Got SerialNumber from database {0}".format(kindleDatabase[0])
+            except KeyError:
+                 # Get the IDString we added
+                IDString = (kindleDatabase[1])['IDString'].decode('hex')
 
-    # concat, hash and encode to calculate the DSN
-    DSN = encode(SHA1(MazamaRandomNumber+encodedIDString+encodedUsername),charMap1)
+            try:
+                # Get the UsernameHash token, if present
+                encodedUsername = (kindleDatabase[1])['UsernameHash'].decode('hex')
+                print u"Got UsernameHash from database {0}".format(kindleDatabase[0])
+            except KeyError:
+                # Get the UserName we added
+                UserName = (kindleDatabase[1])['UserName'].decode('hex')
+                # encode it
+                encodedUsername = encodeHash(UserName,charMap1)
+                #print u"encodedUsername",encodedUsername.encode('hex')
+        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)
+        #print u"encodedIDString",encodedIDString.encode('hex')
+
+        # concat, hash and encode to calculate the DSN
+        DSN = encode(SHA1(MazamaRandomNumber+encodedIDString+encodedUsername),charMap1)
+        #print u"DSN",DSN.encode('hex')
+        pass
+
+    if rec209 is None:
+        pids.append(DSN+kindleAccountToken)
+        return pids
 
     # Compute the device PID (for which I can tell, is used for nothing).
     table =  generatePidEncryptionTable()

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/kindlekey.py

@@ -4,7 +4,10 @@
 from __future__ import with_statement
 
 # kindlekey.py
-# Copyright © 2010-2017 by some_updates, Apprentice Alf and Apprentice Harper
+# Copyright © 2008-2017 Apprentice Harper et al.
+
+__license__ = 'GPL v3'
+__version__ = '2.5'
 
 # Revision history:
 #  1.0   - Kindle info file decryption, extracted from k4mobidedrm, etc.
@@ -23,15 +26,14 @@ from __future__ import with_statement
 #  2.2   - Fixes for Macs with bonded ethernet ports
 #          Also removed old .kinfo file support (pre-2011)
 #  2.3   - Added more field names thanks to concavegit's KFX code.
+#  2.4   - Fix for complex Mac disk setups, thanks to Tibs
+#  2.5   - Final Fix for Windows user names with non-ascii characters, thanks to oneofusoneofus
 
 
 """
 Retrieve Kindle for PC/Mac user key.
 """
 
-__license__ = 'GPL v3'
-__version__ = '2.3'
-
 import sys, os, re
 from struct import pack, unpack, unpack_from
 import json
@@ -886,10 +888,18 @@ if iswindows:
                 if errcd == 234:
                     # bad wine implementation up through wine 1.3.21
                     return "AlternateUserName"
+                # double the buffer size
                 buffer = create_unicode_buffer(len(buffer) * 2)
                 size.value = len(buffer)
-            # return low byte of the unicode value of each character of the username
-            return buffer.value.encode('utf-16-le')[::2]
+            
+            # replace any non-ASCII values with 0xfffd
+            for i in xrange(0,len(buffer)):
+                if buffer[i]>u"\u007f":
+                    #print u"swapping char "+str(i)+" ("+buffer[i]+")"
+                    buffer[i] = u"\ufffd"
+            # return utf-8 encoding of modified username
+            #print u"modified username:"+buffer.value
+            return buffer.value.encode('utf-8')
         return GetUserName
     GetUserName = GetUserName()
 
@@ -1014,7 +1024,12 @@ if iswindows:
             'SerialNumber',\
             'UsernameHash',\
             'kindle.directedid.info',\
-            'DSN'
+            'DSN',\
+            'kindle.accounttype.info',\
+            'krx.flashcardsplugin.data.encryption_key',\
+            'krx.notebookexportplugin.data.encryption_key',\
+            'proxy.http.password',\
+            'proxy.http.username'
             ]
         DB = {}
         with open(kInfoFile, 'rb') as infoReader:
@@ -1267,7 +1282,7 @@ elif isosx:
         for j in xrange(cnt):
             resline = reslst[j]
             if resline.startswith('/dev'):
-                (devpart, mpath) = resline.split(' on ')
+                (devpart, mpath) = resline.split(' on ')[:2]
                 dpart = devpart[5:]
                 names.append(dpart)
         return names

DeDRM_Windows_Application/DeDRM_App/DeDRM_lib/lib/mobidedrm.py

@@ -1,10 +1,12 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
-# mobidedrm.py, version 0.38
+# mobidedrm.py
 # Copyright © 2008 The Dark Reverser
-#
-# Modified 2008–2012 by some_updates, DiapDealer and Apprentice Alf
+# Portions © 2008–2017 Apprentice Harper et al.
+
+__license__ = 'GPL v3'
+__version__ = u"0.42"
 
 # This is a python script. You need a Python interpreter to run it.
 # For example, ActiveState Python, which exists for windows.
@@ -69,9 +71,7 @@
 #  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
-
-
-__version__ = u"0.41"
+#  0.42 - Added GPL v3 licence. updated/removed some print statements
 
 import sys
 import os
@@ -244,7 +244,7 @@ class MobiBook:
         pass
 
     def __init__(self, infile):
-        print u"MobiDeDrm v{0:s}.\nCopyright © 2008-2012 The Dark Reverser et al.".format(__version__)
+        print u"MobiDeDrm v{0:s}.\nCopyright © 2008-2017 The Dark Reverser, Apprentice Harper et al.".format(__version__)
 
         try:
             from alfcrypto import Pukall_Cipher
@@ -288,10 +288,10 @@ class MobiBook:
         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)
+        #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)
+            #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.
@@ -516,7 +516,7 @@ 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"MobiDeDrm v{0:s}.\nCopyright © 2008-2017 The Dark Reverser, Apprentice Harper 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)

DeDRM_calibre_plugin/DeDRM_plugin/__init__.py

@@ -2,6 +2,10 @@
 # -*- coding: utf-8 -*-
 
 from __future__ import with_statement
+
+# __init__.py for DeDRM_plugin
+# Copyright © 2008-2018 Apprentice Harper et al.
+
 __license__   = 'GPL v3'
 __docformat__ = 'restructuredtext en'
 
@@ -58,6 +62,9 @@ __docformat__ = 'restructuredtext en'
 #   6.5.2 - Another Topaz fix
 #   6.5.3 - Warn about KFX files explicitly
 #   6.5.4 - Mac App Fix, improve PDF decryption, handle latest tcl changes in ActivePython
+#   6.5.5 - Finally a fix for the Windows non-ASCII user names.
+#   6.6.0 - Add kfx and kfx-zip as supported file types (also invoke this plugin if the original
+#           imported format was azw8 since that may be converted to kfx)
 
 
 """
@@ -65,7 +72,7 @@ Decrypt DRMed ebooks.
 """
 
 PLUGIN_NAME = u"DeDRM"
-PLUGIN_VERSION_TUPLE = (6, 5, 4)
+PLUGIN_VERSION_TUPLE = (6, 6, 0)
 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'
@@ -113,7 +120,7 @@ class DeDRM(FileTypePlugin):
     author                  = u"Apprentice Alf, Aprentice Harper, The Dark Reverser and i♥cabbages"
     version                 = PLUGIN_VERSION_TUPLE
     minimum_calibre_version = (1, 0, 0)  # Compiled python libraries cannot be imported in earlier versions.
-    file_types              = set(['epub','pdf','pdb','prc','mobi','pobi','azw','azw1','azw3','azw4','tpz'])
+    file_types              = set(['epub','pdf','pdb','prc','mobi','pobi','azw','azw1','azw3','azw4','azw8','tpz','kfx','kfx-zip'])
     on_import               = True
     priority                = 600
 
@@ -289,8 +296,8 @@ class DeDRM(FileTypePlugin):
                 except Exception, e:
                     pass
 
-            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
+            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
 
         # import the Adobe Adept ePub handler
         import calibre_plugins.dedrm.ineptepub as ineptepub
@@ -380,17 +387,19 @@ class DeDRM(FileTypePlugin):
                             except:
                                 print u"{0} v{1}: Exception when saving a new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime)
                                 traceback.print_exc()
-                            print u"{0} v{1}: Decrypted with new default key after {3:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+                            print u"{0} v{1}: Decrypted with new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
                             # Return the modified PersistentTemporary file to calibre.
                             return of.name
 
                         print u"{0} v{1}: Failed to decrypt with new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
                 except Exception, e:
+                    print u"{0} v{1}: Unexpected Exception trying a new default key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime)
+                    traceback.print_exc()
                     pass
 
             # Something went wrong with decryption.
-            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
+            print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+            raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
 
         # Not a Barnes & Noble nor an Adobe Adept
         # Import the fixed epub.
@@ -489,8 +498,8 @@ class DeDRM(FileTypePlugin):
                 pass
 
         # Something went wrong with decryption.
-        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
+        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
 
 
     def KindleMobiDecrypt(self,path_to_ebook):
@@ -557,8 +566,8 @@ class DeDRM(FileTypePlugin):
                     pass
             if not decoded:
                 #if you reached here then no luck raise and exception
-                print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-                raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
+                print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+                raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime))
 
         of = self.temporary_file(book.getBookExtension())
         book.getFile(of.name)
@@ -592,8 +601,8 @@ class DeDRM(FileTypePlugin):
 
             print u"{0} v{1}: Failed to decrypt with key {2:s} after {3:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION,keyname_masked,time.time()-self.starttime)
 
-        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.\nRead the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
-        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds.".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
+        print u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION,time.time()-self.starttime)
+        raise DeDRMError(u"{0} v{1}: Ultimately failed to decrypt after {2:.1f} seconds. Read the FAQs at Harper's repository: https://github.com/apprenticeharper/DeDRM_tools/blob/master/FAQs.md".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
 
 
     def run(self, path_to_ebook):
@@ -606,7 +615,7 @@ class DeDRM(FileTypePlugin):
         self.starttime = time.time()
 
         booktype = os.path.splitext(path_to_ebook)[1].lower()[1:]
-        if booktype in ['prc','mobi','pobi','azw','azw1','azw3','azw4','tpz']:
+        if booktype in ['prc','mobi','pobi','azw','azw1','azw3','azw4','tpz','kfx-zip']:
             # Kindle/Mobipocket
             decrypted_ebook = self.KindleMobiDecrypt(path_to_ebook)
         elif booktype == 'pdb':

DeDRM_calibre_plugin/DeDRM_plugin/ion.py

@@ -0,0 +1,981 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+# Pascal implementation by lulzkabulz. Python translation by apprenticenaomi. DeDRM integration by anon.
+# BinaryIon.pas + DrmIon.pas + IonSymbols.pas
+
+from __future__ import with_statement
+
+import collections
+import hashlib
+import hmac
+import os
+import os.path
+import struct
+
+try:
+    from cStringIO import StringIO
+except ImportError:
+    from StringIO import StringIO
+
+from Crypto.Cipher import AES
+from Crypto.Util.py3compat import bchr, bord
+
+try:
+    # lzma library from calibre 2.35.0 or later
+    import lzma.lzma1 as calibre_lzma
+except:
+    calibre_lzma = None
+    try:
+        import lzma
+    except:
+        # Need pip backports.lzma on Python <3.3
+        from backports import lzma
+
+
+TID_NULL = 0
+TID_BOOLEAN = 1
+TID_POSINT = 2
+TID_NEGINT = 3
+TID_FLOAT = 4
+TID_DECIMAL = 5
+TID_TIMESTAMP = 6
+TID_SYMBOL = 7
+TID_STRING = 8
+TID_CLOB = 9
+TID_BLOB = 0xA
+TID_LIST = 0xB
+TID_SEXP = 0xC
+TID_STRUCT = 0xD
+TID_TYPEDECL = 0xE
+TID_UNUSED = 0xF
+
+
+SID_UNKNOWN = -1
+SID_ION = 1
+SID_ION_1_0 = 2
+SID_ION_SYMBOL_TABLE = 3
+SID_NAME = 4
+SID_VERSION = 5
+SID_IMPORTS = 6
+SID_SYMBOLS = 7
+SID_MAX_ID = 8
+SID_ION_SHARED_SYMBOL_TABLE = 9
+SID_ION_1_0_MAX = 10
+
+
+LEN_IS_VAR_LEN = 0xE
+LEN_IS_NULL = 0xF
+
+
+VERSION_MARKER = b"\x01\x00\xEA"
+
+
+# asserts must always raise exceptions for proper functioning
+def _assert(test, msg="Exception"):
+    if not test:
+        raise Exception(msg)
+
+
+class SystemSymbols(object):
+    ION = '$ion'
+    ION_1_0 = '$ion_1_0'
+    ION_SYMBOL_TABLE = '$ion_symbol_table'
+    NAME = 'name'
+    VERSION = 'version'
+    IMPORTS = 'imports'
+    SYMBOLS = 'symbols'
+    MAX_ID = 'max_id'
+    ION_SHARED_SYMBOL_TABLE = '$ion_shared_symbol_table'
+
+
+class IonCatalogItem(object):
+    name = ""
+    version = 0
+    symnames = []
+
+    def __init__(self, name, version, symnames):
+        self.name = name
+        self.version = version
+        self.symnames = symnames
+
+
+class SymbolToken(object):
+    text = ""
+    sid = 0
+
+    def __init__(self, text, sid):
+        if text == "" and sid == 0:
+            raise ValueError("Symbol token must have Text or SID")
+
+        self.text = text
+        self.sid = sid
+
+
+class SymbolTable(object):
+    table = None
+
+    def __init__(self):
+        self.table = [None] * SID_ION_1_0_MAX
+        self.table[SID_ION] = SystemSymbols.ION
+        self.table[SID_ION_1_0] = SystemSymbols.ION_1_0
+        self.table[SID_ION_SYMBOL_TABLE] = SystemSymbols.ION_SYMBOL_TABLE
+        self.table[SID_NAME] = SystemSymbols.NAME
+        self.table[SID_VERSION] = SystemSymbols.VERSION
+        self.table[SID_IMPORTS] = SystemSymbols.IMPORTS
+        self.table[SID_SYMBOLS] = SystemSymbols.SYMBOLS
+        self.table[SID_MAX_ID] = SystemSymbols.MAX_ID
+        self.table[SID_ION_SHARED_SYMBOL_TABLE] = SystemSymbols.ION_SHARED_SYMBOL_TABLE
+
+    def findbyid(self, sid):
+        if sid < 1:
+            raise ValueError("Invalid symbol id")
+
+        if sid < len(self.table):
+            return self.table[sid]
+        else:
+            return ""
+
+    def import_(self, table, maxid):
+        for i in range(maxid):
+            self.table.append(table.symnames[i])
+
+    def importunknown(self, name, maxid):
+        for i in range(maxid):
+            self.table.append("%s#%d" % (name, i + 1))
+
+
+class ParserState:
+    Invalid,BeforeField,BeforeTID,BeforeValue,AfterValue,EOF = 1,2,3,4,5,6
+
+ContainerRec = collections.namedtuple("ContainerRec", "nextpos, tid, remaining")
+
+
+class BinaryIonParser(object):
+    eof = False
+    state = None
+    localremaining = 0
+    needhasnext = False
+    isinstruct = False
+    valuetid = 0
+    valuefieldid = 0
+    parenttid = 0
+    valuelen = 0
+    valueisnull = False
+    valueistrue = False
+    value = None
+    didimports = False
+
+    def __init__(self, stream):
+        self.annotations = []
+        self.catalog = []
+
+        self.stream = stream
+        self.initpos = stream.tell()
+        self.reset()
+        self.symbols = SymbolTable()
+
+    def reset(self):
+        self.state = ParserState.BeforeTID
+        self.needhasnext = True
+        self.localremaining = -1
+        self.eof = False
+        self.isinstruct = False
+        self.containerstack = []
+        self.stream.seek(self.initpos)
+
+    def addtocatalog(self, name, version, symbols):
+        self.catalog.append(IonCatalogItem(name, version, symbols))
+
+    def hasnext(self):
+        while self.needhasnext and not self.eof:
+            self.hasnextraw()
+            if len(self.containerstack) == 0 and not self.valueisnull:
+                if self.valuetid == TID_SYMBOL:
+                    if self.value == SID_ION_1_0:
+                        self.needhasnext = True
+                elif self.valuetid == TID_STRUCT:
+                    for a in self.annotations:
+                        if a == SID_ION_SYMBOL_TABLE:
+                            self.parsesymboltable()
+                            self.needhasnext = True
+                            break
+        return not self.eof
+
+    def hasnextraw(self):
+        self.clearvalue()
+        while self.valuetid == -1 and not self.eof:
+            self.needhasnext = False
+            if self.state == ParserState.BeforeField:
+                _assert(self.valuefieldid == SID_UNKNOWN)
+
+                self.valuefieldid = self.readfieldid()
+                if self.valuefieldid != SID_UNKNOWN:
+                    self.state = ParserState.BeforeTID
+                else:
+                    self.eof = True
+
+            elif self.state == ParserState.BeforeTID:
+                self.state = ParserState.BeforeValue
+                self.valuetid = self.readtypeid()
+                if self.valuetid == -1:
+                    self.state = ParserState.EOF
+                    self.eof = True
+                    break
+
+                if self.valuetid == TID_TYPEDECL:
+                    if self.valuelen == 0:
+                        self.checkversionmarker()
+                    else:
+                        self.loadannotations()
+
+            elif self.state == ParserState.BeforeValue:
+                self.skip(self.valuelen)
+                self.state = ParserState.AfterValue
+
+            elif self.state == ParserState.AfterValue:
+                if self.isinstruct:
+                    self.state = ParserState.BeforeField
+                else:
+                    self.state = ParserState.BeforeTID
+
+            else:
+                _assert(self.state == ParserState.EOF)
+
+    def next(self):
+        if self.hasnext():
+            self.needhasnext = True
+            return self.valuetid
+        else:
+            return -1
+
+    def push(self, typeid, nextposition, nextremaining):
+        self.containerstack.append(ContainerRec(nextpos=nextposition, tid=typeid, remaining=nextremaining))
+
+    def stepin(self):
+        _assert(self.valuetid in [TID_STRUCT, TID_LIST, TID_SEXP] and not self.eof,
+                "valuetid=%s eof=%s" % (self.valuetid, self.eof))
+        _assert((not self.valueisnull or self.state == ParserState.AfterValue) and
+               (self.valueisnull or self.state == ParserState.BeforeValue))
+
+        nextrem = self.localremaining
+        if nextrem != -1:
+            nextrem -= self.valuelen
+            if nextrem < 0:
+                nextrem = 0
+        self.push(self.parenttid, self.stream.tell() + self.valuelen, nextrem)
+
+        self.isinstruct = (self.valuetid == TID_STRUCT)
+        if self.isinstruct:
+            self.state = ParserState.BeforeField
+        else:
+            self.state = ParserState.BeforeTID
+
+        self.localremaining = self.valuelen
+        self.parenttid = self.valuetid
+        self.clearvalue()
+        self.needhasnext = True
+
+    def stepout(self):
+        rec = self.containerstack.pop()
+
+        self.eof = False
+        self.parenttid = rec.tid
+        if self.parenttid == TID_STRUCT:
+            self.isinstruct = True
+            self.state = ParserState.BeforeField
+        else:
+            self.isinstruct = False
+            self.state = ParserState.BeforeTID
+        self.needhasnext = True
+
+        self.clearvalue()
+        curpos = self.stream.tell()
+        if rec.nextpos > curpos:
+            self.skip(rec.nextpos - curpos)
+        else:
+            _assert(rec.nextpos == curpos)
+
+        self.localremaining = rec.remaining
+
+    def read(self, count=1):
+        if self.localremaining != -1:
+            self.localremaining -= count
+            _assert(self.localremaining >= 0)
+
+        result = self.stream.read(count)
+        if len(result) == 0:
+            raise EOFError()
+        return result
+
+    def readfieldid(self):
+        if self.localremaining != -1 and self.localremaining < 1:
+            return -1
+
+        try:
+            return self.readvaruint()
+        except EOFError:
+            return -1
+
+    def readtypeid(self):
+        if self.localremaining != -1:
+            if self.localremaining < 1:
+                return -1
+            self.localremaining -= 1
+
+        b = self.stream.read(1)
+        if len(b) < 1:
+            return -1
+        b = bord(b)
+        result = b >> 4
+        ln = b & 0xF
+
+        if ln == LEN_IS_VAR_LEN:
+            ln = self.readvaruint()
+        elif ln == LEN_IS_NULL:
+            ln = 0
+            self.state = ParserState.AfterValue
+        elif result == TID_NULL:
+            # Must have LEN_IS_NULL
+            _assert(False)
+        elif result == TID_BOOLEAN:
+            _assert(ln <= 1)
+            self.valueistrue = (ln == 1)
+            ln = 0
+            self.state = ParserState.AfterValue
+        elif result == TID_STRUCT:
+            if ln == 1:
+                ln = self.readvaruint()
+
+        self.valuelen = ln
+        return result
+
+    def readvarint(self):
+        b = bord(self.read())
+        negative = ((b & 0x40) != 0)
+        result = (b & 0x3F)
+
+        i = 0
+        while (b & 0x80) == 0 and i < 4:
+            b = bord(self.read())
+            result = (result << 7) | (b & 0x7F)
+            i += 1
+
+        _assert(i < 4 or (b & 0x80) != 0, "int overflow")
+
+        if negative:
+            return -result
+        return result
+
+    def readvaruint(self):
+        b = bord(self.read())
+        result = (b & 0x7F)
+
+        i = 0
+        while (b & 0x80) == 0 and i < 4:
+            b = bord(self.read())
+            result = (result << 7) | (b & 0x7F)
+            i += 1
+
+        _assert(i < 4 or (b & 0x80) != 0, "int overflow")
+
+        return result
+
+    def readdecimal(self):
+        if self.valuelen == 0:
+            return 0.
+
+        rem = self.localremaining - self.valuelen
+        self.localremaining = self.valuelen
+        exponent = self.readvarint()
+
+        _assert(self.localremaining > 0, "Only exponent in ReadDecimal")
+        _assert(self.localremaining <= 8, "Decimal overflow")
+
+        signed = False
+        b = [bord(x) for x in self.read(self.localremaining)]
+        if (b[0] & 0x80) != 0:
+            b[0] = b[0] & 0x7F
+            signed = True
+
+        # Convert variably sized network order integer into 64-bit little endian
+        j = 0
+        vb = [0] * 8
+        for i in range(len(b), -1, -1):
+            vb[i] = b[j]
+            j += 1
+
+        v = struct.unpack("<Q", b"".join(bchr(x) for x in vb))[0]
+
+        result = v * (10 ** exponent)
+        if signed:
+            result = -result
+
+        self.localremaining = rem
+        return result
+
+    def skip(self, count):
+        if self.localremaining != -1:
+            self.localremaining -= count
+            if self.localremaining < 0:
+                raise EOFError()
+
+        self.stream.seek(count, os.SEEK_CUR)
+
+    def parsesymboltable(self):
+        self.next() # shouldn't do anything?
+
+        _assert(self.valuetid == TID_STRUCT)
+
+        if self.didimports:
+            return
+
+        self.stepin()
+
+        fieldtype = self.next()
+        while fieldtype != -1:
+            if not self.valueisnull:
+                _assert(self.valuefieldid == SID_IMPORTS, "Unsupported symbol table field id")
+
+                if fieldtype == TID_LIST:
+                    self.gatherimports()
+
+            fieldtype = self.next()
+
+        self.stepout()
+        self.didimports = True
+
+    def gatherimports(self):
+        self.stepin()
+
+        t = self.next()
+        while t != -1:
+            if not self.valueisnull and t == TID_STRUCT:
+                self.readimport()
+
+            t = self.next()
+
+        self.stepout()
+
+    def readimport(self):
+        version = -1
+        maxid = -1
+        name = ""
+
+        self.stepin()
+
+        t = self.next()
+        while t != -1:
+            if not self.valueisnull and self.valuefieldid != SID_UNKNOWN:
+                if self.valuefieldid == SID_NAME:
+                    name = self.stringvalue()
+                elif self.valuefieldid == SID_VERSION:
+                    version = self.intvalue()
+                elif self.valuefieldid == SID_MAX_ID:
+                    maxid = self.intvalue()
+
+            t = self.next()
+
+        self.stepout()
+
+        if name == "" or name == SystemSymbols.ION:
+            return
+
+        if version < 1:
+            version = 1
+
+        table = self.findcatalogitem(name)
+        if maxid < 0:
+            _assert(table is not None and version == table.version, "Import %s lacks maxid" % name)
+            maxid = len(table.symnames)
+
+        if table is not None:
+            self.symbols.import_(table, min(maxid, len(table.symnames)))
+        else:
+            self.symbols.importunknown(name, maxid)
+
+    def intvalue(self):
+        _assert(self.valuetid in [TID_POSINT, TID_NEGINT], "Not an int")
+
+        self.preparevalue()
+        return self.value
+
+    def stringvalue(self):
+        _assert(self.valuetid == TID_STRING, "Not a string")
+
+        if self.valueisnull:
+            return ""
+
+        self.preparevalue()
+        return self.value
+
+    def symbolvalue(self):
+        _assert(self.valuetid == TID_SYMBOL, "Not a symbol")
+
+        self.preparevalue()
+        result = self.symbols.findbyid(self.value)
+        if result == "":
+            result = "SYMBOL#%d" % self.value
+        return result
+
+    def lobvalue(self):
+        _assert(self.valuetid in [TID_CLOB, TID_BLOB], "Not a LOB type: %s" % self.getfieldname())
+
+        if self.valueisnull:
+            return None
+
+        result = self.read(self.valuelen)
+        self.state = ParserState.AfterValue
+        return result
+
+    def decimalvalue(self):
+        _assert(self.valuetid == TID_DECIMAL, "Not a decimal")
+
+        self.preparevalue()
+        return self.value
+
+    def preparevalue(self):
+        if self.value is None:
+            self.loadscalarvalue()
+
+    def loadscalarvalue(self):
+        if self.valuetid not in [TID_NULL, TID_BOOLEAN, TID_POSINT, TID_NEGINT,
+                                 TID_FLOAT, TID_DECIMAL, TID_TIMESTAMP,
+                                 TID_SYMBOL, TID_STRING]:
+            return
+
+        if self.valueisnull:
+            self.value = None
+            return
+
+        if self.valuetid == TID_STRING:
+            self.value = self.read(self.valuelen).decode("UTF-8")
+
+        elif self.valuetid in (TID_POSINT, TID_NEGINT, TID_SYMBOL):
+            if self.valuelen == 0:
+                self.value = 0
+            else:
+                _assert(self.valuelen <= 4, "int too long: %d" % self.valuelen)
+                v = 0
+                for i in range(self.valuelen - 1, -1, -1):
+                    v = (v | (bord(self.read()) << (i * 8)))
+
+                if self.valuetid == TID_NEGINT:
+                    self.value = -v
+                else:
+                    self.value = v
+
+        elif self.valuetid == TID_DECIMAL:
+            self.value = self.readdecimal()
+
+        #else:
+        #    _assert(False, "Unhandled scalar type %d" % self.valuetid)
+
+        self.state = ParserState.AfterValue
+
+    def clearvalue(self):
+        self.valuetid = -1
+        self.value = None
+        self.valueisnull = False
+        self.valuefieldid = SID_UNKNOWN
+        self.annotations = []
+
+    def loadannotations(self):
+        ln = self.readvaruint()
+        maxpos = self.stream.tell() + ln
+        while self.stream.tell() < maxpos:
+            self.annotations.append(self.readvaruint())
+        self.valuetid = self.readtypeid()
+
+    def checkversionmarker(self):
+        for i in VERSION_MARKER:
+            _assert(self.read() == i, "Unknown version marker")
+
+        self.valuelen = 0
+        self.valuetid = TID_SYMBOL
+        self.value = SID_ION_1_0
+        self.valueisnull = False
+        self.valuefieldid = SID_UNKNOWN
+        self.state = ParserState.AfterValue
+
+    def findcatalogitem(self, name):
+        for result in self.catalog:
+            if result.name == name:
+                return result
+
+    def forceimport(self, symbols):
+        item = IonCatalogItem("Forced", 1, symbols)
+        self.symbols.import_(item, len(symbols))
+
+    def getfieldname(self):
+        if self.valuefieldid == SID_UNKNOWN:
+            return ""
+        return self.symbols.findbyid(self.valuefieldid)
+
+    def getfieldnamesymbol(self):
+        return SymbolToken(self.getfieldname(), self.valuefieldid)
+
+    def gettypename(self):
+        if len(self.annotations) == 0:
+            return ""
+
+        return self.symbols.findbyid(self.annotations[0])
+
+    @staticmethod
+    def printlob(b):
+        if b is None:
+            return "null"
+
+        result = ""
+        for i in b:
+            result += ("%02x " % bord(i))
+
+        if len(result) > 0:
+            result = result[:-1]
+        return result
+
+    def ionwalk(self, supert, indent, lst):
+        while self.hasnext():
+            if supert == TID_STRUCT:
+                L = self.getfieldname() + ":"
+            else:
+                L = ""
+
+            t = self.next()
+            if t in [TID_STRUCT, TID_LIST]:
+                if L != "":
+                    lst.append(indent + L)
+                L = self.gettypename()
+                if L != "":
+                    lst.append(indent + L + "::")
+                if t == TID_STRUCT:
+                    lst.append(indent + "{")
+                else:
+                    lst.append(indent + "[")
+
+                self.stepin()
+                self.ionwalk(t, indent + "  ", lst)
+                self.stepout()
+
+                if t == TID_STRUCT:
+                    lst.append(indent + "}")
+                else:
+                    lst.append(indent + "]")
+
+            else:
+                if t == TID_STRING:
+                    L += ('"%s"' % self.stringvalue())
+                elif t in [TID_CLOB, TID_BLOB]:
+                    L += ("{%s}" % self.printlob(self.lobvalue()))
+                elif t == TID_POSINT:
+                    L += str(self.intvalue())
+                elif t == TID_SYMBOL:
+                    tn = self.gettypename()
+                    if tn != "":
+                        tn += "::"
+                    L += tn + self.symbolvalue()
+                elif t == TID_DECIMAL:
+                    L += str(self.decimalvalue())
+                else:
+                    L += ("TID %d" % t)
+                lst.append(indent + L)
+
+    def print_(self, lst):
+        self.reset()
+        self.ionwalk(-1, "", lst)
+
+
+SYM_NAMES = [ 'com.amazon.drm.Envelope@1.0',
+              'com.amazon.drm.EnvelopeMetadata@1.0', 'size', 'page_size',
+              'encryption_key', 'encryption_transformation',
+              'encryption_voucher', 'signing_key', 'signing_algorithm',
+              'signing_voucher', 'com.amazon.drm.EncryptedPage@1.0',
+              'cipher_text', 'cipher_iv', 'com.amazon.drm.Signature@1.0',
+              'data', 'com.amazon.drm.EnvelopeIndexTable@1.0', 'length',
+              'offset', 'algorithm', 'encoded', 'encryption_algorithm',
+              'hashing_algorithm', 'expires', 'format', 'id',
+              'lock_parameters', 'strategy', 'com.amazon.drm.Key@1.0',
+              'com.amazon.drm.KeySet@1.0', 'com.amazon.drm.PIDv3@1.0',
+              'com.amazon.drm.PlainTextPage@1.0',
+              'com.amazon.drm.PlainText@1.0', 'com.amazon.drm.PrivateKey@1.0',
+              'com.amazon.drm.PublicKey@1.0', 'com.amazon.drm.SecretKey@1.0',
+              'com.amazon.drm.Voucher@1.0', 'public_key', 'private_key',
+              'com.amazon.drm.KeyPair@1.0', 'com.amazon.drm.ProtectedData@1.0',
+              'doctype', 'com.amazon.drm.EnvelopeIndexTableOffset@1.0',
+              'enddoc', 'license_type', 'license', 'watermark', 'key', 'value',
+              'com.amazon.drm.License@1.0', 'category', 'metadata',
+              'categorized_metadata', 'com.amazon.drm.CategorizedMetadata@1.0',
+              'com.amazon.drm.VoucherEnvelope@1.0', 'mac', 'voucher',
+              'com.amazon.drm.ProtectedData@2.0',
+              'com.amazon.drm.Envelope@2.0',
+              'com.amazon.drm.EnvelopeMetadata@2.0',
+              'com.amazon.drm.EncryptedPage@2.0',
+              'com.amazon.drm.PlainText@2.0', 'compression_algorithm',
+              'com.amazon.drm.Compressed@1.0', 'priority', 'refines']
+
+def addprottable(ion):
+    ion.addtocatalog("ProtectedData", 1, SYM_NAMES)
+
+
+def pkcs7pad(msg, blocklen):
+    paddinglen = blocklen - len(msg) % blocklen
+    padding = bchr(paddinglen) * paddinglen
+    return msg + padding
+
+
+def pkcs7unpad(msg, blocklen):
+    _assert(len(msg) % blocklen == 0)
+
+    paddinglen = bord(msg[-1])
+    _assert(paddinglen > 0 and paddinglen <= blocklen, "Incorrect padding - Wrong key")
+    _assert(msg[-paddinglen:] == bchr(paddinglen) * paddinglen, "Incorrect padding - Wrong key")
+
+    return msg[:-paddinglen]
+
+
+class DrmIonVoucher(object):
+    envelope = None
+    voucher = None
+    drmkey = None
+    license_type = "Unknown"
+
+    encalgorithm = ""
+    enctransformation = ""
+    hashalgorithm = ""
+
+    lockparams = None
+
+    ciphertext = b""
+    cipheriv = b""
+    secretkey = b""
+
+    def __init__(self, voucherenv, dsn, secret):
+        self.dsn,self.secret = dsn,secret
+
+        self.lockparams = []
+
+        self.envelope = BinaryIonParser(voucherenv)
+        addprottable(self.envelope)
+
+    def decryptvoucher(self):
+        shared = "PIDv3" + self.encalgorithm + self.enctransformation + self.hashalgorithm
+
+        self.lockparams.sort()
+        for param in self.lockparams:
+            if param == "ACCOUNT_SECRET":
+                shared += param + self.secret
+            elif param == "CLIENT_ID":
+                shared += param + self.dsn
+            else:
+                _assert(False, "Unknown lock parameter: %s" % param)
+
+        sharedsecret = shared.encode("UTF-8")
+
+        key = hmac.new(sharedsecret, sharedsecret[:5], digestmod=hashlib.sha256).digest()
+        aes = AES.new(key[:32], AES.MODE_CBC, self.cipheriv[:16])
+        b = aes.decrypt(self.ciphertext)
+        b = pkcs7unpad(b, 16)
+
+        self.drmkey = BinaryIonParser(StringIO(b))
+        addprottable(self.drmkey)
+
+        _assert(self.drmkey.hasnext() and self.drmkey.next() == TID_LIST and self.drmkey.gettypename() == "com.amazon.drm.KeySet@1.0",
+                "Expected KeySet, got %s" % self.drmkey.gettypename())
+
+        self.drmkey.stepin()
+        while self.drmkey.hasnext():
+            self.drmkey.next()
+            if self.drmkey.gettypename() != "com.amazon.drm.SecretKey@1.0":
+                continue
+
+            self.drmkey.stepin()
+            while self.drmkey.hasnext():
+                self.drmkey.next()
+                if self.drmkey.getfieldname() == "algorithm":
+                    _assert(self.drmkey.stringvalue() == "AES", "Unknown cipher algorithm: %s" % self.drmkey.stringvalue())
+                elif self.drmkey.getfieldname() == "format":
+                    _assert(self.drmkey.stringvalue() == "RAW", "Unknown key format: %s" % self.drmkey.stringvalue())
+                elif self.drmkey.getfieldname() == "encoded":
+                    self.secretkey = self.drmkey.lobvalue()
+
+            self.drmkey.stepout()
+            break
+
+        self.drmkey.stepout()
+
+    def parse(self):
+        self.envelope.reset()
+        _assert(self.envelope.hasnext(), "Envelope is empty")
+        _assert(self.envelope.next() == TID_STRUCT and self.envelope.gettypename() == "com.amazon.drm.VoucherEnvelope@1.0",
+                "Unknown type encountered in envelope, expected VoucherEnvelope")
+
+        self.envelope.stepin()
+        while self.envelope.hasnext():
+            self.envelope.next()
+            field = self.envelope.getfieldname()
+            if field == "voucher":
+                self.voucher = BinaryIonParser(StringIO(self.envelope.lobvalue()))
+                addprottable(self.voucher)
+                continue
+            elif field != "strategy":
+                continue
+
+            _assert(self.envelope.gettypename() == "com.amazon.drm.PIDv3@1.0", "Unknown strategy: %s" % self.envelope.gettypename())
+
+            self.envelope.stepin()
+            while self.envelope.hasnext():
+                self.envelope.next()
+                field = self.envelope.getfieldname()
+                if field == "encryption_algorithm":
+                    self.encalgorithm = self.envelope.stringvalue()
+                elif field == "encryption_transformation":
+                    self.enctransformation = self.envelope.stringvalue()
+                elif field == "hashing_algorithm":
+                    self.hashalgorithm = self.envelope.stringvalue()
+                elif field == "lock_parameters":
+                    self.envelope.stepin()
+                    while self.envelope.hasnext():
+                        _assert(self.envelope.next() == TID_STRING, "Expected string list for lock_parameters")
+                        self.lockparams.append(self.envelope.stringvalue())
+                    self.envelope.stepout()
+
+            self.envelope.stepout()
+
+        self.parsevoucher()
+
+    def parsevoucher(self):
+        _assert(self.voucher.hasnext(), "Voucher is empty")
+        _assert(self.voucher.next() == TID_STRUCT and self.voucher.gettypename() == "com.amazon.drm.Voucher@1.0",
+                "Unknown type, expected Voucher")
+
+        self.voucher.stepin()
+        while self.voucher.hasnext():
+            self.voucher.next()
+
+            if self.voucher.getfieldname() == "cipher_iv":
+                self.cipheriv = self.voucher.lobvalue()
+            elif self.voucher.getfieldname() == "cipher_text":
+                self.ciphertext = self.voucher.lobvalue()
+            elif self.voucher.getfieldname() == "license":
+                _assert(self.voucher.gettypename() == "com.amazon.drm.License@1.0",
+                        "Unknown license: %s" % self.voucher.gettypename())
+                self.voucher.stepin()
+                while self.voucher.hasnext():
+                    self.voucher.next()
+                    if self.voucher.getfieldname() == "license_type":
+                        self.license_type = self.voucher.stringvalue()
+                self.voucher.stepout()
+
+    def printenvelope(self, lst):
+        self.envelope.print_(lst)
+
+    def printkey(self, lst):
+        if self.voucher is None:
+            self.parse()
+        if self.drmkey is None:
+            self.decryptvoucher()
+
+        self.drmkey.print_(lst)
+
+    def printvoucher(self, lst):
+        if self.voucher is None:
+            self.parse()
+
+        self.voucher.print_(lst)
+
+    def getlicensetype(self):
+        return self.license_type
+
+
+class DrmIon(object):
+    ion = None
+    voucher = None
+    vouchername = ""
+    key = b""
+    onvoucherrequired = None
+
+    def __init__(self, ionstream, onvoucherrequired):
+        self.ion = BinaryIonParser(ionstream)
+        addprottable(self.ion)
+        self.onvoucherrequired = onvoucherrequired
+
+    def parse(self, outpages):
+        self.ion.reset()
+
+        _assert(self.ion.hasnext(), "DRMION envelope is empty")
+        _assert(self.ion.next() == TID_SYMBOL and self.ion.gettypename() == "doctype", "Expected doctype symbol")
+        _assert(self.ion.next() == TID_LIST and self.ion.gettypename() in ["com.amazon.drm.Envelope@1.0", "com.amazon.drm.Envelope@2.0"],
+                "Unknown type encountered in DRMION envelope, expected Envelope, got %s" % self.ion.gettypename())
+
+        while True:
+            if self.ion.gettypename() == "enddoc":
+                break
+
+            self.ion.stepin()
+            while self.ion.hasnext():
+                self.ion.next()
+
+                if self.ion.gettypename() in ["com.amazon.drm.EnvelopeMetadata@1.0", "com.amazon.drm.EnvelopeMetadata@2.0"]:
+                    self.ion.stepin()
+                    while self.ion.hasnext():
+                        self.ion.next()
+                        if self.ion.getfieldname() != "encryption_voucher":
+                            continue
+
+                        if self.vouchername == "":
+                            self.vouchername = self.ion.stringvalue()
+                            self.voucher = self.onvoucherrequired(self.vouchername)
+                            self.key = self.voucher.secretkey
+                            _assert(self.key is not None, "Unable to obtain secret key from voucher")
+                        else:
+                            _assert(self.vouchername == self.ion.stringvalue(),
+                                    "Unexpected: Different vouchers required for same file?")
+
+                    self.ion.stepout()
+
+                elif self.ion.gettypename() in ["com.amazon.drm.EncryptedPage@1.0", "com.amazon.drm.EncryptedPage@2.0"]:
+                    decompress = False
+                    ct = None
+                    civ = None
+                    self.ion.stepin()
+                    while self.ion.hasnext():
+                        self.ion.next()
+                        if self.ion.gettypename() == "com.amazon.drm.Compressed@1.0":
+                            decompress = True
+                        if self.ion.getfieldname() == "cipher_text":
+                            ct = self.ion.lobvalue()
+                        elif self.ion.getfieldname() == "cipher_iv":
+                            civ = self.ion.lobvalue()
+
+                    if ct is not None and civ is not None:
+                        self.processpage(ct, civ, outpages, decompress)
+                    self.ion.stepout()
+
+            self.ion.stepout()
+            if not self.ion.hasnext():
+                break
+            self.ion.next()
+
+    def print_(self, lst):
+        self.ion.print_(lst)
+
+    def processpage(self, ct, civ, outpages, decompress):
+        aes = AES.new(self.key[:16], AES.MODE_CBC, civ[:16])
+        msg = pkcs7unpad(aes.decrypt(ct), 16)
+
+        if not decompress:
+            outpages.write(msg)
+            return
+
+        _assert(msg[0] == b"\x00", "LZMA UseFilter not supported")
+
+        if calibre_lzma is not None:
+            with calibre_lzma.decompress(msg[1:], bufsize=0x1000000) as f:
+                f.seek(0)
+                outpages.write(f.read())
+            return
+
+        decomp = lzma.LZMADecompressor(format=lzma.FORMAT_ALONE)
+        while not decomp.eof:
+            segment = decomp.decompress(msg[1:])
+            msg = b"" # Contents were internally buffered after the first call
+            outpages.write(segment)

DeDRM_calibre_plugin/DeDRM_plugin/k4mobidedrm.py

@@ -3,10 +3,13 @@
 
 from __future__ import with_statement
 
-# k4mobidedrm.py, version 5.3
-# Copyright © 2009-2015 by ApprenticeHarper et al.
+# k4mobidedrm.py
+# Copyright © 2008-2017 by Apprentice Harper et al.
 
-# engine to remove drm from Kindle and Mobipocket ebooks
+__license__ = 'GPL v3'
+__version__ = '5.5'
+
+# Engine to remove drm from Kindle and Mobipocket ebooks
 # for personal use for archiving and converting your ebooks
 
 # PLEASE DO NOT PIRATE EBOOKS!
@@ -17,12 +20,11 @@ from __future__ import with_statement
 # 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
+#    unswindle, DarkReverser, ApprenticeAlf, 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
@@ -57,9 +59,8 @@ from __future__ import with_statement
 #  5.2 - Fixed error in command line processing of unicode arguments
 #  5.3 - Changed Android support to allow passing of backup .ab files
 #  5.4 - Recognise KFX files masquerading as azw, even if we can't decrypt them yet.
-
-__version__ = '5.4'
-
+#  5.5 - Added GPL v3 licence explicitly.
+#  5.x - Invoke KFXZipBook to handle zipped KFX files
 
 import sys, os, re
 import csv
@@ -83,11 +84,13 @@ if inCalibre:
     from calibre_plugins.dedrm import topazextract
     from calibre_plugins.dedrm import kgenpids
     from calibre_plugins.dedrm import androidkindlekey
+    from calibre_plugins.dedrm import kfxdedrm
 else:
     import mobidedrm
     import topazextract
     import kgenpids
     import androidkindlekey
+    import kfxdedrm
 
 # Wrap a stream so that output gets flushed immediately
 # and also make sure that any unicode strings get
@@ -197,13 +200,15 @@ def GetDecryptedBook(infile, kDatabases, androidFiles, serials, pids, starttime
     mobi = True
     magic8 = open(infile,'rb').read(8)
     if magic8 == '\xeaDRMION\xee':
-        raise DrmException(u"KFX format detected. This format cannot be decrypted yet.")
-        
+        raise DrmException(u"The .kfx DRMION file cannot be decrypted by itself. A .kfx-zip archive containing a DRM voucher is required.")
+
     magic3 = magic8[:3]
     if magic3 == 'TPZ':
         mobi = False
 
-    if mobi:
+    if magic8[:4] == 'PK\x03\x04':
+        mb = kfxdedrm.KFXZipBook(infile)
+    elif mobi:
         mb = mobidedrm.MobiBook(infile)
     else:
         mb = topazextract.TopazBook(infile)
@@ -295,7 +300,7 @@ def usage(progname):
 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__)
+    print u"K4MobiDeDrm v{0}.\nCopyright © 2008-2017 Apprentice Harper et al.".format(__version__)
 
     try:
         opts, args = getopt.getopt(argv[1:], "k:p:s:a:")

DeDRM_calibre_plugin/DeDRM_plugin/kfxdedrm.py

@@ -0,0 +1,108 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+from __future__ import with_statement
+
+# Engine to remove drm from Kindle KFX ebooks
+
+import os
+import shutil
+import zipfile
+
+try:
+    from cStringIO import StringIO
+except ImportError:
+    from StringIO import StringIO
+
+try:
+    import ion
+except:
+    from calibre_plugins.dedrm import ion
+
+
+__license__ = 'GPL v3'
+__version__ = '1.0'
+
+
+class KFXZipBook:
+    def __init__(self, infile):
+        self.infile = infile
+        self.voucher = None
+        self.decrypted = {}
+
+    def getPIDMetaInfo(self):
+        return (None, None)
+
+    def processBook(self, totalpids):
+        with zipfile.ZipFile(self.infile, 'r') as zf:
+            for filename in zf.namelist():
+                data = zf.read(filename)
+                if data.startswith('\xeaDRMION\xee'):
+                    if self.voucher is None:
+                        self.decrypt_voucher(totalpids)
+                    print u'Decrypting KFX DRMION: {0}'.format(filename)
+                    outfile = StringIO()
+                    ion.DrmIon(StringIO(data[8:-8]), lambda name: self.voucher).parse(outfile)
+                    self.decrypted[filename] = outfile.getvalue()
+
+        if not self.decrypted:
+            print(u'The .kfx-zip archive does not contain an encrypted DRMION file')
+
+    def decrypt_voucher(self, totalpids):
+        with zipfile.ZipFile(self.infile, 'r') as zf:
+            for info in zf.infolist():
+                if info.file_size < 0x10000:
+                    data = zf.read(info.filename)
+                    if data.startswith('\xe0\x01\x00\xea') and 'ProtectedData' in data:
+                        break   # found DRM voucher
+            else:
+                raise Exception(u'The .kfx-zip archive contains an encrypted DRMION file without a DRM voucher')
+
+        print u'Decrypting KFX DRM voucher: {0}'.format(info.filename)
+
+        for pid in [''] + totalpids:
+            for dsn_len,secret_len in [(0,0), (16,0), (16,40), (32,40), (40,40)]:
+                if len(pid) == dsn_len + secret_len:
+                    break       # split pid into DSN and account secret
+            else:
+                continue
+
+            try:
+                voucher = ion.DrmIonVoucher(StringIO(data), pid[:dsn_len], pid[dsn_len:])
+                voucher.parse()
+                voucher.decryptvoucher()
+                break
+            except:
+                pass
+        else:
+            raise Exception(u'Failed to decrypt KFX DRM voucher with any key')
+
+        print u'KFX DRM voucher successfully decrypted'
+
+        license_type = voucher.getlicensetype()
+        if license_type != "Purchase":
+            raise Exception((u'This book is licensed as {0}. '
+                    'These tools are intended for use on purchased books.').format(license_type))
+
+        self.voucher = voucher
+
+    def getBookTitle(self):
+        return os.path.splitext(os.path.split(self.infile)[1])[0]
+
+    def getBookExtension(self):
+        return '.kfx-zip'
+
+    def getBookType(self):
+        return 'KFX-ZIP'
+
+    def cleanup(self):
+        pass
+
+    def getFile(self, outpath):
+        if not self.decrypted:
+            shutil.copyfile(self.infile, outpath)
+        else:
+            with zipfile.ZipFile(self.infile, 'r') as zif:
+                with zipfile.ZipFile(outpath, 'w') as zof:
+                    for info in zif.infolist():
+                        zof.writestr(info, self.decrypted.get(info.filename, zif.read(info.filename)))

DeDRM_calibre_plugin/DeDRM_plugin/kgenpids.py

@@ -4,10 +4,15 @@
 from __future__ import with_statement
 
 # kgenpids.py
-# Copyright © 2010-2015 by some_updates, Apprentice Alf and Apprentice Harper
+# Copyright © 2008-2017 Apprentice Harper et al.
+
+__license__ = 'GPL v3'
+__version__ = '2.1'
 
 # Revision history:
 #  2.0   - Fix for non-ascii Windows user names
+#  2.1   - Actual fix for non-ascii WIndows user names.
+#  x.x   - Return information needed for KFX decryption
 
 import sys
 import os, csv
@@ -168,6 +173,9 @@ def pidFromSerial(s, l):
 
 # Parse the EXTH header records and use the Kindle serial number to calculate the book pid.
 def getKindlePids(rec209, token, serialnum):
+    if rec209 is None:
+        return [serialnum]
+
     pids=[]
 
     if isinstance(serialnum,unicode):
@@ -195,20 +203,6 @@ def getK4Pids(rec209, token, kindleDatabase):
     global charMap1
     pids = []
 
-    try:
-        # Get the Mazama Random number
-        MazamaRandomNumber = (kindleDatabase[1])['MazamaRandomNumber'].decode('hex')
-
-        # Get the IDString used to decode the Kindle Info file
-        IDString = (kindleDatabase[1])['IDString'].decode('hex')
-
-        # Get the UserName stored when the Kindle Info file was decoded
-        UserName = (kindleDatabase[1])['UserName'].decode('hex')
-
-    except KeyError:
-        print u"Keys not found in the database {0}.".format(kindleDatabase[0])
-        return pids
-
     try:
         # Get the kindle account token, if present
         kindleAccountToken = (kindleDatabase[1])['kindle.account.tokens'].decode('hex')
@@ -217,14 +211,51 @@ def getK4Pids(rec209, token, kindleDatabase):
         kindleAccountToken=""
         pass
 
-    # Get the ID string used
-    encodedIDString = encodeHash(IDString,charMap1)
+    try:
+        # Get the DSN token, if present
+        DSN = (kindleDatabase[1])['DSN'].decode('hex')
+        print u"Got DSN key from database {0}".format(kindleDatabase[0])
+    except KeyError:
+        # See if we have the info to generate the DSN
+        try:
+            # Get the Mazama Random number
+            MazamaRandomNumber = (kindleDatabase[1])['MazamaRandomNumber'].decode('hex')
+            #print u"Got MazamaRandomNumber from database {0}".format(kindleDatabase[0])
 
-    # Get the current user name
-    encodedUsername = encodeHash(UserName,charMap1)
+            try:
+                # Get the SerialNumber token, if present
+                IDString = (kindleDatabase[1])['SerialNumber'].decode('hex')
+                print u"Got SerialNumber from database {0}".format(kindleDatabase[0])
+            except KeyError:
+                 # Get the IDString we added
+                IDString = (kindleDatabase[1])['IDString'].decode('hex')
 
-    # concat, hash and encode to calculate the DSN
-    DSN = encode(SHA1(MazamaRandomNumber+encodedIDString+encodedUsername),charMap1)
+            try:
+                # Get the UsernameHash token, if present
+                encodedUsername = (kindleDatabase[1])['UsernameHash'].decode('hex')
+                print u"Got UsernameHash from database {0}".format(kindleDatabase[0])
+            except KeyError:
+                # Get the UserName we added
+                UserName = (kindleDatabase[1])['UserName'].decode('hex')
+                # encode it
+                encodedUsername = encodeHash(UserName,charMap1)
+                #print u"encodedUsername",encodedUsername.encode('hex')
+        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)
+        #print u"encodedIDString",encodedIDString.encode('hex')
+
+        # concat, hash and encode to calculate the DSN
+        DSN = encode(SHA1(MazamaRandomNumber+encodedIDString+encodedUsername),charMap1)
+        #print u"DSN",DSN.encode('hex')
+        pass
+
+    if rec209 is None:
+        pids.append(DSN+kindleAccountToken)
+        return pids
 
     # Compute the device PID (for which I can tell, is used for nothing).
     table =  generatePidEncryptionTable()

DeDRM_calibre_plugin/DeDRM_plugin/kindlekey.py

@@ -4,7 +4,10 @@
 from __future__ import with_statement
 
 # kindlekey.py
-# Copyright © 2010-2017 by some_updates, Apprentice Alf and Apprentice Harper
+# Copyright © 2008-2017 Apprentice Harper et al.
+
+__license__ = 'GPL v3'
+__version__ = '2.5'
 
 # Revision history:
 #  1.0   - Kindle info file decryption, extracted from k4mobidedrm, etc.
@@ -23,15 +26,14 @@ from __future__ import with_statement
 #  2.2   - Fixes for Macs with bonded ethernet ports
 #          Also removed old .kinfo file support (pre-2011)
 #  2.3   - Added more field names thanks to concavegit's KFX code.
+#  2.4   - Fix for complex Mac disk setups, thanks to Tibs
+#  2.5   - Final Fix for Windows user names with non-ascii characters, thanks to oneofusoneofus
 
 
 """
 Retrieve Kindle for PC/Mac user key.
 """
 
-__license__ = 'GPL v3'
-__version__ = '2.3'
-
 import sys, os, re
 from struct import pack, unpack, unpack_from
 import json
@@ -886,10 +888,18 @@ if iswindows:
                 if errcd == 234:
                     # bad wine implementation up through wine 1.3.21
                     return "AlternateUserName"
+                # double the buffer size
                 buffer = create_unicode_buffer(len(buffer) * 2)
                 size.value = len(buffer)
-            # return low byte of the unicode value of each character of the username
-            return buffer.value.encode('utf-16-le')[::2]
+            
+            # replace any non-ASCII values with 0xfffd
+            for i in xrange(0,len(buffer)):
+                if buffer[i]>u"\u007f":
+                    #print u"swapping char "+str(i)+" ("+buffer[i]+")"
+                    buffer[i] = u"\ufffd"
+            # return utf-8 encoding of modified username
+            #print u"modified username:"+buffer.value
+            return buffer.value.encode('utf-8')
         return GetUserName
     GetUserName = GetUserName()
 
@@ -1014,7 +1024,12 @@ if iswindows:
             'SerialNumber',\
             'UsernameHash',\
             'kindle.directedid.info',\
-            'DSN'
+            'DSN',\
+            'kindle.accounttype.info',\
+            'krx.flashcardsplugin.data.encryption_key',\
+            'krx.notebookexportplugin.data.encryption_key',\
+            'proxy.http.password',\
+            'proxy.http.username'
             ]
         DB = {}
         with open(kInfoFile, 'rb') as infoReader:
@@ -1267,7 +1282,7 @@ elif isosx:
         for j in xrange(cnt):
             resline = reslst[j]
             if resline.startswith('/dev'):
-                (devpart, mpath) = resline.split(' on ')
+                (devpart, mpath) = resline.split(' on ')[:2]
                 dpart = devpart[5:]
                 names.append(dpart)
         return names

DeDRM_calibre_plugin/DeDRM_plugin/mobidedrm.py

@@ -1,10 +1,12 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
-# mobidedrm.py, version 0.38
+# mobidedrm.py
 # Copyright © 2008 The Dark Reverser
-#
-# Modified 2008–2012 by some_updates, DiapDealer and Apprentice Alf
+# Portions © 2008–2017 Apprentice Harper et al.
+
+__license__ = 'GPL v3'
+__version__ = u"0.42"
 
 # This is a python script. You need a Python interpreter to run it.
 # For example, ActiveState Python, which exists for windows.
@@ -69,9 +71,7 @@
 #  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
-
-
-__version__ = u"0.41"
+#  0.42 - Added GPL v3 licence. updated/removed some print statements
 
 import sys
 import os
@@ -244,7 +244,7 @@ class MobiBook:
         pass
 
     def __init__(self, infile):
-        print u"MobiDeDrm v{0:s}.\nCopyright © 2008-2012 The Dark Reverser et al.".format(__version__)
+        print u"MobiDeDrm v{0:s}.\nCopyright © 2008-2017 The Dark Reverser, Apprentice Harper et al.".format(__version__)
 
         try:
             from alfcrypto import Pukall_Cipher
@@ -288,10 +288,10 @@ class MobiBook:
         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)
+        #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)
+            #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.
@@ -516,7 +516,7 @@ 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"MobiDeDrm v{0:s}.\nCopyright © 2008-2017 The Dark Reverser, Apprentice Harper 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)

DeDRM_calibre_plugin/DeDRM_plugin_ReadMe.txt

@@ -4,7 +4,7 @@ DeDRM_plugin.zip
 This calibre plugin replaces many previously separate DRM removal plugins. Before you install this plugin, you should uninstall any older individual DRM removal plugins, e.g. K4MobiDeDRM. The exception is the obok plugin, which should not be removed.
 
 This plugin will remove the DRM from
- - Kindle ebooks (files from Kindle for Mac/PC (v1.17*) and eInk Kindles**).
+ - Kindle ebooks (files from Kindle for Mac/PC* and eInk Kindles**).
  - Barnes and Noble ePubs
  - Adobe Digital Editions (v2.0.1***) ePubs (including Kobo and Google ePubs downloaded to ADE)
  - Adobe Digital Editions (v2.0.1) PDFs
@@ -13,11 +13,11 @@ This plugin will remove the DRM from
  
 These tools do NOT work with kepubs downloaded using Kobo's desktop app (see the separate obok plugin) nor Apple's iBooks FairPlay DRM (see details about Requiem at the end of this file.)
 
-* With Kindle for PC/Mac 1.19 and later, Amazon included support for their new KFX format which uses a new DRM scheme that these tools cannot remove. Using 1.17 or earlier prevents downloads of the new format.
+* With Kindle for PC/Mac 1.19 and later, Amazon included support for their new KFX format. While the tools now include a first attempt at supporting drm removal for KFX format, we recommend using Kindle for PC/Mac 1.17 or earlier which prevents downloads of the new format, as conversions from the olde KF8 format are likely to be more successful.
 
-** Some later Kindles support Amazon's new KFX format which uses a new DRM scheme that these tools cannot remove. To avoid this problem, instead of using files downloaded directly to your Kindle, download from Amazon's web site 'for transfer via USB'. This will give you an older format file that the tools can decrypt.
+** Some later Kindles support Amazon's new KFX format. And some books download in a split azw3/azw6 format. For best results, instead of using files downloaded directly to your Kindle, download from Amazon's web site 'for transfer via USB'. This will give you an single file to import. See also the FAQ entry about this.
 
-*** With Adobe Digital Editions 3.0 and later, Adobe have introduced a new, optional, DRM scheme. To avoid this new scheme, you should use Adobe Digital Editions 2.0.1. Some books are required to use the new DRM scheme and so will not download with ADE 2.0.1. If you still want such a book, you will need to use ADE 3.0 or later to download it, but you should remember that no tools to remove Adobe's new DRM scheme exist as of April 2017.
+*** With Adobe Digital Editions 3.0 and later, Adobe have introduced a new, optional, DRM scheme. To avoid this new scheme, you should use Adobe Digital Editions 2.0.1. Some books are required to use the new DRM scheme and so will not download with ADE 2.0.1. If you still want such a book, you will need to use ADE 3.0 or later to download it, but you should remember that no tools to remove Adobe's new DRM scheme exist as of October 2017.
 
 
 Installation
@@ -67,6 +67,7 @@ The original mobidedrm and erdr2pml scripts were by The Dark Reverser
 The original topaz DRM removal script was by CMBDTC
 The original topaz format conversion scripts were by some_updates, clarknova and Bart Simpson
 The original obok script was by Physisticated
+The original KFX format decryption was by lulzkabulz, converted to python by Apprentice Naomi and integrated into the tools by tomthumb1997
 
 The alfcrypto library is by some_updates
 The ePub encryption detection script is by Apprentice Alf, adapted from a script by Paul Durrant
@@ -82,7 +83,7 @@ Linux Systems Only
 Instructions for installing Wine, Kindle for PC, Adobe Digital Editions, Python and PyCrypto
 --------------------------------------------------------------------------------------------
 
-These instructions have been tested with Wine 1.4 on Ubuntu. 
+These instructions have been tested with Wine 1.4 on Ubuntu but are now very out of date. 
 
  1. First download the software you're going to to have to install.
     a. Adobe Digital Editions 1.7.x from http://helpx.adobe.com/digital-editions/kb/cant-install-digital-editions.html
@@ -98,7 +99,7 @@ These instructions have been tested with Wine 1.4 on Ubuntu.
  6. Install Python 2.7.x using `msiexec /i python-2.7.8.msi`. Accept all defaults.
  7. Install PyCrypto 2.1. Accept all defaults.
  8. Unzip DeDRM_plugin.zip and move kindlekey.py to somewhere in drive_c, such as ~/.wine/drive_c/DeDRM/libraryfiles/kindlekey.py.
- 9. Run `wine 'C:\Python27/python.exe' 'C:DeDRM/libraryfiles/kindlekey.py'`, or wherever you copied kindlekey.py to.
+ 9. Run `wine 'C:\Python27/python.exe' 'C:\DeDRM/libraryfiles/kindlekey.py'`, or wherever you copied kindlekey.py to.
  10. Import the resulting key file to the Calibre plugin through the Kindle for Mac/PC ebooks option.
 
 

FAQs.md

@@ -19,20 +19,41 @@ Just download and use these tools, that's all! Uh, almost. There are a few, uh,
 But otherwise, if your ebook is from Amazon, Kobo, Barnes & Noble or any of the ebook stores selling ebooks compatible with Adobe Digital Editions 2.0.1, you should be able to remove the DRM that's been applied to your ebooks.
 
 ### A Recent Change to Kindle for PC/Kindle for Mac
-Starting with version 1.19, Kindle for PC/Mac uses Amazon's new KFX format which these tools can't handle. Stick with version 1.17 or earlier. Kindle for PC 1.17 can be downloaded from https://s3.amazonaws.com/kindleforpc/44183/KindleForPC-installer-1.17.44183.exe and Kindle for Mac 1.17 can be downloaded from https://s3.amazonaws.com/kindleformac/44182/KindleForMac-44182.dmg
+Starting with version 1.19, Kindle for PC/Mac uses Amazon's new KFX format which these tools can't handle. There are two options to get the older formats that the tools can decrypt. Either stick with version 1.17 or earlier, or modify the executable by changing a file name.
 
-If you have already installed or have been updated to 1.19, just go to the preferences and uncheck the auto update checkbox. The download and install 1.17 over the top of the 1.19 installation. You'll also need to delete the KFX folders from your My Kindle Content folder.
+Version 1.17 of Kindle is are no longer available directly from Amazon, so you will need to search for the proper file name and find it on a third party site. The name is "KindleForPC-installer-1.17.44170.exe" for PC and "KindleForMac-44182.dmg" for Mac.
+Verify the one of the following cryptographic hash values, using software of your choice, before installing the downloaded file in order to avoid viruses. If the hash does not match, delete the downloaded file and try again from another site.
+Kindle for PC:
+MD-5: 53F793B562F4823721AA47D7DE099869
+SHA-1: 73C404D719F0DD8D4AE1C2C96612B095D6C86255
+SHA-256: 14E0F0053F1276C0C7C446892DC170344F707FBFE99B695176 2C120144163200
+Kindle for Mac:
+MD-5: E7E36D5369E1F3CF1D28E5D9115DF15F
+SHA-1: 7AB9A86B954CB23D622BD79E3257F8E2182D791C
+SHA-256: 28DC21246A9C7CDEDD2D6F0F4082E6BF7EF9DB9CE9D485548E 8A9E1D19EAE2AC. 
+
+You will need to go to the preferences and uncheck the auto update checkbox. Then download and install 1.17 over the top of the 1.19 installation. You'll also need to delete the KFX folders from your My Kindle Content folder.
 
 A second possible solution is to use 1.19 or later, but disable KFX by renaming or disabling a necessary component of the application. This may or may not work on versions after 1.20. In a command window, enter the following commands when Kindle for PC/Mac is not running:
 
 #### Windows
 ren %localappdata%\Amazon\Kindle\application\renderer-test.exe renderer-test.xxx
 
+PC Note: The renderer-test program may be in a different location in some Kindle for PC installations. If the rename command fails look in other folders, such as C:\Program Files\Amazon\Kindle.
+
 #### Macintosh
 chmod -x /Applications/Kindle.app/Contents/MacOS/renderer-test
 
+Mac Note: If the chmod command fails with a permission error try again using sudo before chmod - sudo chmod [...]
+
 After restarting the Kindle program any books previously downloaded in KFX format will no longer open. You will need to remove them from your device and re-download them. All future downloads will use the older Kindle formats instead of KFX although they will continue to be placed in one individual subdirectory per book.
 
+#### Another Note on KFX
+It now possible, but not easy, to convert books from KFX to other formats in calibre by installing the optional KFX Input plugin. The lack of automatic DRM removal makes this process difficult so it is not recommended unless there is no other alternative, such as for Indic language books only available in KFX. There is a windows-only KFX DRM rmeoval program in the repository, but not yet integrated into the tools.
+
+#### Thanks
+Thanks to jhowell for his investigations into KFX format and workarounds. Some of these instructions are from [his thread on the subject](https://www.mobileread.com/forums/showthread.php?t=283371) at MobileRead.
+
 ## Where can I get the latest version of these free DRM removal tools?
 Right here at github. Just go to the [releases page](https://github.com/apprenticeharper/DeDRM_tools/releases) and download the latest zip archive of the tools, named DeDRM\_tools\_X.X.X.zip, where X.X.X is the version number. You do not need to download the source code archive.
 
@@ -65,6 +86,9 @@ Try right-clicking and select open. That might give you the option to open it an
 ### I can't open the Macintosh Application at all. I get 'The aplication "DeDRM" can't be opened'
 Some unzip applications do not respect the execution bit setting. Try unzipping the main tools archive using the built-in Mac unzip utility.
 
+### I can't open the Macintosh Application at all. I get 'spawn_via_launchd() failed, errno=111'
+There seems to be a bug in Apple's launch services. Try using the free [Maintenance utility](https://www.titanium-software.fr/en/maintenance.html) from Titanium Software to clear the launch cache and database.
+
 # Using the Tools
 ## I can’t get the tools to work on my rented or library ebooks.
 The tools are not designed to remove DRM from rented or library ebooks.
@@ -100,8 +124,6 @@ If you cannot read the ebook on your current device or installed software, the t
 ## I have installed the calibre plugin, and the book is not already in calibre, but the DRM does not get removed. It is a Kindle book.
 If you are on Windows 8 and using the Windows 8 AppStore Kindle app, you must download and install the Kindle for PC application directly from the Amazon website. The tools do not work with the Windows 8 AppStore Kindle app.
 
-If you are on Windows, using the Kindle for PC application, and your windows user name has accented or other non-ASCII characters in it, this will cause the plugin to fail. Create a new user account with an ASCII-only username. Install and register Kindle for PC there, and see if things work running calibre and the plugin in that user account.
-
 If this book is from an eInk Kindle (e.g. Paperwhite), you must enter the serial number into the configuration dialog. The serial number is sixteen characters long, and is case-sensitive.
 
 If this book is from Kindle for Mac or Kindle for PC, you must have the Kindle Software installed on the same computer and user account as your copy of calibre.

Other_Tools/DRM_Key_Scripts/Kindle_for_Mac_and_PC/kindlekey.pyw

@@ -4,7 +4,10 @@
 from __future__ import with_statement
 
 # kindlekey.py
-# Copyright © 2010-2017 by some_updates, Apprentice Alf and Apprentice Harper
+# Copyright © 2008-2017 Apprentice Harper et al.
+
+__license__ = 'GPL v3'
+__version__ = '2.5'
 
 # Revision history:
 #  1.0   - Kindle info file decryption, extracted from k4mobidedrm, etc.
@@ -23,15 +26,14 @@ from __future__ import with_statement
 #  2.2   - Fixes for Macs with bonded ethernet ports
 #          Also removed old .kinfo file support (pre-2011)
 #  2.3   - Added more field names thanks to concavegit's KFX code.
+#  2.4   - Fix for complex Mac disk setups, thanks to Tibs
+#  2.5   - Final Fix for Windows user names with non-ascii characters, thanks to oneofusoneofus
 
 
 """
 Retrieve Kindle for PC/Mac user key.
 """
 
-__license__ = 'GPL v3'
-__version__ = '2.3'
-
 import sys, os, re
 from struct import pack, unpack, unpack_from
 import json
@@ -886,10 +888,18 @@ if iswindows:
                 if errcd == 234:
                     # bad wine implementation up through wine 1.3.21
                     return "AlternateUserName"
+                # double the buffer size
                 buffer = create_unicode_buffer(len(buffer) * 2)
                 size.value = len(buffer)
-            # return low byte of the unicode value of each character of the username
-            return buffer.value.encode('utf-16-le')[::2]
+            
+            # replace any non-ASCII values with 0xfffd
+            for i in xrange(0,len(buffer)):
+                if buffer[i]>u"\u007f":
+                    #print u"swapping char "+str(i)+" ("+buffer[i]+")"
+                    buffer[i] = u"\ufffd"
+            # return utf-8 encoding of modified username
+            #print u"modified username:"+buffer.value
+            return buffer.value.encode('utf-8')
         return GetUserName
     GetUserName = GetUserName()
 
@@ -1014,7 +1024,12 @@ if iswindows:
             'SerialNumber',\
             'UsernameHash',\
             'kindle.directedid.info',\
-            'DSN'
+            'DSN',\
+            'kindle.accounttype.info',\
+            'krx.flashcardsplugin.data.encryption_key',\
+            'krx.notebookexportplugin.data.encryption_key',\
+            'proxy.http.password',\
+            'proxy.http.username'
             ]
         DB = {}
         with open(kInfoFile, 'rb') as infoReader:
@@ -1267,7 +1282,7 @@ elif isosx:
         for j in xrange(cnt):
             resline = reslst[j]
             if resline.startswith('/dev'):
-                (devpart, mpath) = resline.split(' on ')
+                (devpart, mpath) = resline.split(' on ')[:2]
                 dpart = devpart[5:]
                 names.append(dpart)
         return names

ReadMe_First.txt

@@ -1,11 +1,11 @@
 Welcome to the tools!
 =====================
 
-This ReadMe_First.txt is meant to give users a quick overview of what is available and how to get started. This document is part of the Tools v6.5.4 archive from Apprentice Harper's github repository: https://github.com/apprenticeharper/DeDRM_tools/
+This ReadMe_First.txt is meant to give users a quick overview of what is available and how to get started. This document is part of the Tools v6.6.0 archive from Apprentice Harper's github repository: https://github.com/apprenticeharper/DeDRM_tools/
 
 The is archive includes tools to remove DRM from:
 
- - Kindle ebooks (files from Kindle for Mac/PC (v1.17*) and eInk Kindles**).
+ - Kindle ebooks (files from Kindle for Mac/PC* and eInk Kindles**).
  - Adobe Digital Editions (v2.0.1***) ePubs (including Kobo and Google ePubs downloaded to ADE)
  - Kobo kePubs from the Kobo Desktop application
  - Barnes and Noble ePubs
@@ -17,15 +17,15 @@ The is archive includes tools to remove DRM from:
 
 These tools do NOT work with Apple's iBooks FairPlay DRM (see end of this file.)
 
-* With Kindle for PC/Mac 1.19 and later, Amazon included support for their new KFX format which uses a new DRM scheme that these tools cannot remove. Using 1.17 or earlier prevents downloads of the new format.
+* With Kindle for PC/Mac 1.19 and later, Amazon included support for their new KFX format. While the tools now include a first attempt at supporting drm removal for KFX format, we recommend using Kindle for PC/Mac 1.17 or earlier which prevents downloads of the new format, as conversions from the olde KF8 format are likely to be more successful.
 
-** Some later Kindles support Amazon's new KFX format which uses a new DRM scheme that these tools cannot remove. To avoid this problem, instead of using files downloaded directly to your Kindle, download from Amazon's web site 'for transfer via USB'. This will give you an older format file that the tools can decrypt. See also the FAQ entry about this.
+** Some later Kindles support Amazon's new KFX format. And some books download in a split azw3/azw6 format. For best results, instead of using files downloaded directly to your Kindle, download from Amazon's web site 'for transfer via USB'. This will give you an single file to import. See also the FAQ entry about this.
 
 *** With Adobe Digital Editions 3.0 and later, Adobe have introduced a new, optional, DRM scheme. To avoid this new scheme, you should use Adobe Digital Editions 2.0.1. Some books are required to use the new DRM scheme and so will not download with ADE 2.0.1. If you still want such a book, you will need to use ADE 3.0 or later to download it, but you should remember that no tools to remove Adobe's new DRM scheme exist as of June 2017.
 
 About the tools
 ---------------
-These tools are updated and maintained by Apprentice Alf and Apprentice Harper. You can find the latest updates at Apprentice Harper's github repository https://github.com/apprenticeharper/DeDRM_tools/ and get support by creating an issue at the repository (github account required) or by posting a comment at Apprentice Alf's blog: http://www.apprenticealf.wordpress.com/
+These tools are updated and maintained by Apprentice Harper and many others. You can find the latest updates at Apprentice Harper's github repository https://github.com/apprenticeharper/DeDRM_tools/ and get support by creating an issue at the repository (github account required) or by posting a comment at Apprentice Alf's blog: http://www.apprenticealf.wordpress.com/
 
 If you re-post these tools, a link to the repository and/or the blog would be appreciated.
 
@@ -48,7 +48,7 @@ For instructions, see the obok_plugin_ReadMe.txt file in the Obok_calibre_plugin
 
 DeDRM application for Mac OS X users: (Mac OS X 10.4 and above)
 ---------------------------------------------------------------
-This application is a stand-alone DRM removal application for Mac OS X users. It is only needed for people who cannot or will not use the calibre plugin.
+This application is a stand-alone DRM removal application for Mac OS X users. It is only needed for people who cannot or will not use the calibre plugin. KFX support has not been tested yet.
 
 For instructions, see the "DeDRM ReadMe.rtf" file in the DeDRM_Macintosh_Application folder.
 
@@ -60,7 +60,7 @@ DeDRM application for Windows users: (Windows XP through Windows 10)
 ***This program requires that Python and PyCrypto be properly installed.***
 ***See below for details on recommended versions and how to install them.***
 
-This application is a stand-alone application for Windows users. It is only needed for people who cannot or will not use the calibre plugin.
+This application is a stand-alone application for Windows users. It is only needed for people who cannot or will not use the calibre plugin. KFX support has not been tested yet.
 
 For instructions, see the DeDRM_App_ReadMe.txt file in the DeDRM_Windows_Applications folder.
 
@@ -90,12 +90,13 @@ A link to the tool for removing DRM from ScuolaBooks PDFs, created by "Hex".
 
 Windows and Python
 ------------------
-We **strongly** recommend ActiveState's Active Python 2.7 Community Edition for Windows. This can be downloaded for free from:
+We **strongly** recommend using calibre and the plugin.
+
+If you really want to use the WIndows app or the individual scripts, you'll need to install python.
+ActiveState's Active Python 2.7 Community Edition for Windowscan be downloaded for free from:
 
 http://www.activestate.com/activepython/downloads
 
-We do **NOT** recommend the version of Python from python.org as it is missing various Windows specific libraries, does not install the Tk Widget kit (for graphical user interfaces) by default, and does not properly update the system PATH environment variable. Therefore using the default python.org build on Windows is simply an exercise in frustration for most Windows users.
-
 In addition, Windows Users need PyCrypto:
 
     There are many places to get PyCrypto installers for Windows. One such place is:
@@ -137,6 +138,8 @@ The original inept and ignoble scripts were by i♥cabbages
 The original mobidedrm and erdr2pml scripts were by The Dark Reverser
 The original topaz DRM removal script was by CMBDTC
 The original topaz format conversion scripts were by some_updates, clarknova and Bart Simpson
+The original KFX format decryption was by lulzkabulz, converted to python by Apprentice Naomi and integrated into the tools by tomthumb1997
+
 The original obok script was by Physisticated
 
 The alfcrypto library is by some_updates