Merge ad33aea18d into 10b6caf9f5

The Kobo Desktop program will (when running under Wine on Linux) put all
of its data in the current working directory. This means that there
may be more than one Kobo.sqlite floating around by the user, which
leads to Obok showing an outdated list of books and the user being
confused by Obok cannot find the full list of books.

Solving this completely appears to be a bit too complicated, so this
patch is a best-effort improvement for the worst cases which can be
caused by this:

 1. If the user deletes the files but Obok has already cached the path,
    previously Obok would just error out rather than trying to search
    for a new Kobo.sqlite path and updating the cache.

 2. We search $HOME before searching /, which speeds up initial usage of
    Obok in the common case (usually Kobo Desktop will be installed in
    ~/.wine/drive_c) and also ensures that we correctly preference the
    current user's Kobo Desktop installation.

Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>

.github/workflows/main.yml

@@ -20,46 +20,33 @@ jobs:
           path: |
             DeDRM_tools_*.zip
             DeDRM_tools.zip
+     
+      - name: Prepare release
+        run: cp DeDRM_tools.zip DeDRM_alpha_${{ github.sha }}.zip
 
-#      - name: Delete old release
-#        uses: cb80/delrel@latest
-#        with:
-#          tag: autorelease
-#          token: ${{ github.token }}
-#
-#      - name: Delete old tag
-#        uses: dev-drprasad/delete-tag-and-release@v1.0
-#        with:
-#          tag_name: autorelease
-#          github_token: ${{ github.token }}
-#          delete_release: true
-#      
-#      - name: Prepare release
-#        run: cp DeDRM_tools.zip DeDRM_alpha_${{ github.sha }}.zip
-#
-#      - name: Auto-release
-#        id: autorelease
-#        uses: softprops/action-gh-release@v1
-#        with: 
-#          tag_name: autorelease
-#          token: ${{ github.token }}
-#          name: Automatic alpha release with latest changes
-#          body: |
-#            This release is automatically generated by Github for each commit. 
-#
-#            This means, every time a change is made to this repo, this release will be updated to contain an untested copy of the plugin at that stage. This will contain the most up-to-date code, but it's not tested at all and may be broken.
-#
-#            Last update based on Git commit ${{ github.sha }}.
-#          prerelease: true
-#          draft: true
-#          files: DeDRM_alpha_${{ github.sha }}.zip
-#
-#      - name: Make release public
-#        uses: irongut/EditRelease@v1.2.0
-#        with:
-#          token: ${{ github.token }}
-#          id: ${{ steps.autorelease.outputs.id }}
-#          draft: false
-#          prerelease: true
-#
-#
+
+      - uses: dev-drprasad/delete-older-releases@v0.2.1
+        with:
+          repo: noDRM/DeDRM_tools_autorelease 
+          keep_latest: 0
+          delete_tags: true
+        env:
+          GITHUB_TOKEN: ${{ secrets.AUTORELEASE_KEY }}
+
+      - name: Auto-release
+        id: autorelease
+        uses: softprops/action-gh-release@v1
+        with: 
+          tag_name: autorelease_${{ github.sha }}
+          repository: noDRM/DeDRM_tools_autorelease
+          token: ${{ secrets.AUTORELEASE_KEY }}
+          name: Automatic alpha release with latest changes
+          body: |
+            This release is automatically generated by Github for each commit. 
+
+            This means, every time a change is made to the repo, a release with an untested copy of the plugin at that stage will be created. This will contain the most up-to-date code, but it's not tested at all and may be broken.
+
+            Last update based on Git commit [${{ github.sha }}](https://github.com/noDRM/DeDRM_tools/commit/${{ github.sha }}).
+          prerelease: true
+          draft: false
+          files: DeDRM_alpha_${{ github.sha }}.zip

CHANGELOG.md

@@ -101,4 +101,5 @@ This is v10.0.9, a release candidate for v10.1.0. I don't expect there to be maj
 
 - Fix a bug where decrypting a 40-bit RC4 pdf with R=2 didn't work.
 - Fix a bug where decrypting a 256-bit AES pdf with V=5 didn't work.
-- Fix bugs in kgenpids.py and kindlekey.py that caused it to fail on Python 2 (#380).
+- Fix bugs in kgenpids.py, alfcrypto.py, mobidedrm.py and kindlekey.py that caused it to fail on Python 2 (#380).
+- Fix some bugs (Python 2 and Python 3) in erdr2pml.py (untested).

DeDRM_plugin/__calibre_compat_code.py

@@ -15,6 +15,6 @@ if "calibre" in sys.modules and sys.version_info[0] == 2:
         sys.path.insert(0, os.path.join(config_dir, "plugins", "DeDRM.zip"))
 
 # Explicitly set the package identifier so we are allowed to import stuff ...
-#__package__ = "DeDRM_plugin"
+__package__ = "calibre_plugins.dedrm"
 
 #@@CALIBRE_COMPAT_CODE_END@@

DeDRM_plugin/__init__.py

@@ -96,7 +96,10 @@ import traceback
 #@@CALIBRE_COMPAT_CODE@@
 
 try: 
-    import __version
+    try: 
+        from . import __version
+    except:
+        import __version
 except: 
     print("#############################")
     print("Failed to load the DeDRM plugin")
@@ -134,8 +137,10 @@ try:
 except:
     config_dir = ""
 
-
-import utilities
+try: 
+    from . import utilities
+except: 
+    import utilities
 
 
 PLUGIN_NAME = __version.PLUGIN_NAME
@@ -915,6 +920,9 @@ class DeDRM(FileTypePlugin):
             # perhaps we need to get a new default Kindle for Mac/PC key
             defaultkeys = []
             print("{0} v{1}: Failed to decrypt with error: {2}".format(PLUGIN_NAME, PLUGIN_VERSION,e.args[0]))
+
+            traceback.print_exc()
+
             print("{0} v{1}: Looking for new default Kindle Key after {2:.1f} seconds".format(PLUGIN_NAME, PLUGIN_VERSION, time.time()-self.starttime))
 
             try:

DeDRM_plugin/adobekey.py

@@ -44,10 +44,11 @@ __version__ = '7.4'
 import sys, os, struct, getopt
 from base64 import b64decode
 
+#@@CALIBRE_COMPAT_CODE@@
 
 
-from utilities import SafeUnbuffered
-from argv_utils import unicode_argv
+from .utilities import SafeUnbuffered
+from .argv_utils import unicode_argv
 
 
 try:

DeDRM_plugin/alfcrypto.py

@@ -8,6 +8,7 @@
 # pbkdf2.py Copyright © 2009 Daniel Holth <dholth@fastmail.fm>
 # pbkdf2.py This code may be freely used and modified for any purpose.
 
+import sys
 import hmac
 from struct import pack
 import hashlib
@@ -25,7 +26,10 @@ class Pukall_Cipher(object):
             raise Exception("PC1: Bad key length")
         wkey = []
         for i in range(8):
-            wkey.append(key[i*2]<<8 | key[i*2+1])
+            if sys.version_info[0] == 2:
+                wkey.append(ord(key[i*2])<<8 | ord(key[i*2+1]))
+            else: 
+                wkey.append(key[i*2]<<8 | key[i*2+1])
         dst = bytearray(len(src))
         for i in range(len(src)):
             temp1 = 0;
@@ -37,7 +41,12 @@ class Pukall_Cipher(object):
                 sum2  = (sum2+sum1)&0xFFFF
                 temp1 = (temp1*20021+1)&0xFFFF
                 byteXorVal ^= temp1 ^ sum2
-            curByte = src[i]
+
+            if sys.version_info[0] == 2:
+                curByte = ord(src[i])
+            else:
+                curByte = src[i]
+
             if not decryption:
                 keyXorVal = curByte * 257;
             curByte = ((curByte ^ (byteXorVal >> 8)) ^ byteXorVal) & 0xFF
@@ -45,7 +54,12 @@ class Pukall_Cipher(object):
                 keyXorVal = curByte * 257;
             for j in range(8):
                 wkey[j] ^= keyXorVal;
-            dst[i] = curByte
+            
+            if sys.version_info[0] == 2:
+                dst[i] = chr(curByte)
+            else: 
+                dst[i] = curByte
+                
         return bytes(dst)
 
 class Topaz_Cipher(object):
@@ -103,7 +117,7 @@ class KeyIVGen(object):
         def xorbytes( a, b ):
             if len(a) != len(b):
                 raise Exception("xorbytes(): lengths differ")
-            return bytes([x ^ y for x, y in zip(a, b)])
+            return bytes(bytearray([x ^ y for x, y in zip(a, b)]))
 
         def prf( h, data ):
             hm = h.copy()

DeDRM_plugin/config.py

@@ -29,7 +29,7 @@ from calibre.constants import iswindows, isosx
 
 from __init__ import PLUGIN_NAME, PLUGIN_VERSION
 from __version import RESOURCE_NAME as help_file_name
-from utilities import uStrCmp
+from .utilities import uStrCmp
 
 import prefs
 import androidkindlekey

DeDRM_plugin/epubtest.py

@@ -53,12 +53,12 @@ import sys, struct, os, traceback
 import zlib
 import zipfile
 import xml.etree.ElementTree as etree
-from argv_utils import unicode_argv
+from .argv_utils import unicode_argv
 
 NSMAP = {'adept': 'http://ns.adobe.com/adept',
          'enc': 'http://www.w3.org/2001/04/xmlenc#'}
 
-from utilities import SafeUnbuffered
+from .utilities import SafeUnbuffered
 
 
 _FILENAME_LEN_OFFSET = 26

DeDRM_plugin/erdr2pml.py

@@ -79,12 +79,13 @@ except ImportError:
 
 #@@CALIBRE_COMPAT_CODE@@
 
-from utilities import SafeUnbuffered
+from .utilities import SafeUnbuffered
+from .argv_utils import unicode_argv
 
 iswindows = sys.platform.startswith('win')
 isosx = sys.platform.startswith('darwin')
 
-from argv_utils import unicode_argv
+
 
 import cgi
 import logging
@@ -141,14 +142,20 @@ def sanitizeFileName(name):
 
 def fixKey(key):
     def fixByte(b):
+        if sys.version_info[0] == 2:
+            b = ord(b)
+
         return b ^ ((b ^ (b<<1) ^ (b<<2) ^ (b<<3) ^ (b<<4) ^ (b<<5) ^ (b<<6) ^ (b<<7) ^ 0x80) & 0x80)
-    return bytes([fixByte(a) for a in key])
+    return bytes(bytearray([fixByte(a) for a in key]))
 
 def deXOR(text, sp, table):
-    r=''
+    r=b''
     j = sp
     for i in range(len(text)):
-        r += chr(ord(table[j]) ^ ord(text[i]))
+        if sys.version_info[0] == 2:
+            r += chr(ord(table[j]) ^ ord(text[i]))
+        else: 
+            r += bytes(bytearray([table[j] ^ text[i]]))
         j = j + 1
         if j == len(table):
             j = 0

DeDRM_plugin/ignoblekeyGenPassHash.py

@@ -50,9 +50,9 @@ try:
 except ImportError:
     from Crypto.Cipher import AES
 
-from utilities import SafeUnbuffered
+from .utilities import SafeUnbuffered
 
-from argv_utils import unicode_argv
+from .argv_utils import unicode_argv
 
 class IGNOBLEError(Exception):
     pass

DeDRM_plugin/ignoblekeyNookStudy.py

@@ -27,14 +27,14 @@ import hashlib
 import getopt
 import re
 
-from utilities import SafeUnbuffered
+from .utilities import SafeUnbuffered
 
 try:
     from calibre.constants import iswindows
 except:
     iswindows = sys.platform.startswith('win')
 
-from argv_utils import unicode_argv
+from .argv_utils import unicode_argv
 
 class DrmException(Exception):
     pass

DeDRM_plugin/ineptepub.py

@@ -70,9 +70,8 @@ def unpad(data, padding=16):
 
     return data[:-pad_len]
 
-from utilities import SafeUnbuffered
-
-from argv_utils import unicode_argv
+from .utilities import SafeUnbuffered
+from .argv_utils import unicode_argv
 
 
 class ADEPTError(Exception):

DeDRM_plugin/ineptpdf.py

@@ -93,12 +93,12 @@ def unpad(data, padding=16):
     return data[:-pad_len]
 
 
-from utilities import SafeUnbuffered
+from .utilities import SafeUnbuffered
 
 iswindows = sys.platform.startswith('win')
 isosx = sys.platform.startswith('darwin')
 
-from argv_utils import unicode_argv
+from .argv_utils import unicode_argv
 
 class ADEPTError(Exception):
     pass

DeDRM_plugin/k4mobidedrm.py

@@ -88,9 +88,9 @@ import kgenpids
 import androidkindlekey
 import kfxdedrm
 
-from utilities import SafeUnbuffered
+from .utilities import SafeUnbuffered
 
-from argv_utils import unicode_argv
+from .argv_utils import unicode_argv
 
 
 # cleanup unicode filenames

DeDRM_plugin/kgenpids.py

@@ -190,6 +190,10 @@ def getKindlePids(rec209, token, serialnum):
     if isinstance(serialnum,str):
         serialnum = serialnum.encode('utf-8')
 
+    if sys.version_info[0] == 2:
+        if isinstance(serialnum,unicode):
+            serialnum = serialnum.encode('utf-8')
+
     if rec209 is None:
         return [serialnum]
 

DeDRM_plugin/kindlekey.py

@@ -62,7 +62,9 @@ except NameError:
 
 # Routines common to Mac and PC
 
-from utilities import SafeUnbuffered
+from .utilities import SafeUnbuffered
+from .argv_utils import unicode_argv
+    
 
 try:
     from calibre.constants import iswindows, isosx
@@ -70,7 +72,7 @@ except:
     iswindows = sys.platform.startswith('win')
     isosx = sys.platform.startswith('darwin')
 
-from argv_utils import unicode_argv
+
 
 class DrmException(Exception):
     pass
@@ -238,9 +240,14 @@ if iswindows:
 
             # replace any non-ASCII values with 0xfffd
             for i in range(0,len(buffer)):
-                if buffer[i]>"\u007f":
-                    #print "swapping char "+str(i)+" ("+buffer[i]+")"
-                    buffer[i] = "\ufffd"
+                if sys.version_info[0] == 2:
+                    if buffer[i]>u"\u007f":
+                        #print "swapping char "+str(i)+" ("+buffer[i]+")"
+                        buffer[i] = u"\ufffd"
+                else: 
+                    if buffer[i]>"\u007f":
+                        #print "swapping char "+str(i)+" ("+buffer[i]+")"
+                        buffer[i] = "\ufffd"
             # return utf-8 encoding of modified username
             #print "modified username:"+buffer.value
             return buffer.value.encode('utf-8')

DeDRM_plugin/kindlepid.py

@@ -16,24 +16,24 @@
 import sys
 import binascii
 
-from utilities import SafeUnbuffered
+from .utilities import SafeUnbuffered
 
-from argv_utils import unicode_argv
+from .argv_utils import unicode_argv
 
-letters = 'ABCDEFGHIJKLMNPQRSTUVWXYZ123456789'
+letters = b'ABCDEFGHIJKLMNPQRSTUVWXYZ123456789'
 
 def crc32(s):
     return (~binascii.crc32(s,-1))&0xFFFFFFFF
 
 def checksumPid(s):
-    crc = crc32(s.encode('ascii'))
+    crc = crc32(s)
     crc = crc ^ (crc >> 16)
     res = s
     l = len(letters)
     for i in (0,1):
         b = crc & 0xff
         pos = (b // l) ^ (b % l)
-        res += letters[pos%l]
+        res += bytes(bytearray([letters[pos%l]]))
         crc >>= 8
 
     return res
@@ -43,16 +43,19 @@ def pidFromSerial(s, l):
 
     arr1 = [0]*l
     for i in range(len(s)):
-        arr1[i%l] ^= s[i]
+        if sys.version_info[0] == 2:
+            arr1[i%l] ^= ord(s[i])
+        else: 
+            arr1[i%l] ^= s[i]
 
     crc_bytes = [crc >> 24 & 0xff, crc >> 16 & 0xff, crc >> 8 & 0xff, crc & 0xff]
     for i in range(l):
         arr1[i] ^= crc_bytes[i&3]
 
-    pid = ''
+    pid = b""
     for i in range(l):
         b = arr1[i] & 0xff
-        pid+=letters[(b >> 7) + ((b >> 5 & 3) ^ (b & 0x1f))]
+        pid+=bytes(bytearray([letters[(b >> 7) + ((b >> 5 & 3) ^ (b & 0x1f))]]))
 
     return pid
 

DeDRM_plugin/mobidedrm.py

@@ -80,11 +80,14 @@ import sys
 import os
 import struct
 import binascii
-from alfcrypto import Pukall_Cipher
 
-from utilities import SafeUnbuffered
 
-from argv_utils import unicode_argv
+#@@CALIBRE_COMPAT_CODE@@
+
+
+from .alfcrypto import Pukall_Cipher
+from .utilities import SafeUnbuffered
+from .argv_utils import unicode_argv
 
 
 class DrmException(Exception):
@@ -103,19 +106,26 @@ def PC1(key, src, decryption=True):
     except: 
         raise
 
-# accepts unicode returns unicode
+letters = b'ABCDEFGHIJKLMNPQRSTUVWXYZ123456789'
+
+def crc32(s):
+    return (~binascii.crc32(s,-1))&0xFFFFFFFF
+
 def checksumPid(s):
-    letters = 'ABCDEFGHIJKLMNPQRSTUVWXYZ123456789'
-    crc = (~binascii.crc32(s.encode('utf-8'),-1))&0xFFFFFFFF
+
+    s = s.encode()
+
+
+    crc = crc32(s)
     crc = crc ^ (crc >> 16)
     res = s
     l = len(letters)
     for i in (0,1):
         b = crc & 0xff
         pos = (b // l) ^ (b % l)
-        res += letters[pos%l]
+        res += bytes(bytearray([letters[pos%l]]))
         crc >>= 8
-    return res
+    return res.decode()
 
 # expects bytearray
 def getSizeOfTrailingDataEntries(ptr, size, flags):
@@ -124,7 +134,11 @@ def getSizeOfTrailingDataEntries(ptr, size, flags):
         if size <= 0:
             return result
         while True:
-            v = ptr[size-1]
+            if sys.version_info[0] == 2:
+                v = ord(ptr[size-1])
+            else:
+                v = ptr[size-1]
+
             result |= (v & 0x7F) << bitpos
             bitpos += 7
             size -= 1
@@ -140,7 +154,10 @@ def getSizeOfTrailingDataEntries(ptr, size, flags):
     # if multibyte data is included in the encryped data, we'll
     # have already cleared this flag.
     if flags & 1:
-        num += (ptr[size - num - 1] & 0x3) + 1
+        if sys.version_info[0] == 2:
+            num += (ord(ptr[size - num - 1]) & 0x3) + 1
+        else: 
+            num += (ptr[size - num - 1] & 0x3) + 1
     return num
 
 
@@ -299,7 +316,10 @@ class MobiBook:
         for pid in pidlist:
             bigpid = pid.encode('utf-8').ljust(16,b'\0')
             temp_key = PC1(keyvec1, bigpid, False)
-            temp_key_sum = sum(temp_key) & 0xff
+            if sys.version_info[0] == 2:
+                temp_key_sum = sum(map(ord,temp_key)) & 0xff
+            else:
+                temp_key_sum = sum(temp_key) & 0xff
             found_key = None
             for i in range(count):
                 verification, size, type, cksum, cookie = struct.unpack('>LLLBxxx32s', data[i*0x30:i*0x30+0x30])
@@ -315,7 +335,11 @@ class MobiBook:
             # Then try the default encoding that doesn't require a PID
             pid = '00000000'
             temp_key = keyvec1
-            temp_key_sum = sum(temp_key) & 0xff
+            if sys.version_info[0] == 2:
+                temp_key_sum = sum(map(ord,temp_key)) & 0xff
+            else:
+                temp_key_sum = sum(temp_key) & 0xff
+            
             for i in range(count):
                 verification, size, type, cksum, cookie = struct.unpack('>LLLBxxx32s', data[i*0x30:i*0x30+0x30])
                 if cksum == temp_key_sum:

DeDRM_plugin/topazextract.py

@@ -24,10 +24,10 @@ import traceback
 from struct import pack
 from struct import unpack
 
-from alfcrypto import Topaz_Cipher
-from utilities import SafeUnbuffered
+from .alfcrypto import Topaz_Cipher
+from .utilities import SafeUnbuffered
 
-from argv_utils import unicode_argv
+from .argv_utils import unicode_argv
 
 
 #global switch

Obok_plugin/obok/obok.py

@@ -327,36 +327,50 @@ class KoboLibrary(object):
                 elif sys.platform.startswith('darwin'):
                     self.kobodir = os.path.join(os.environ['HOME'], "Library", "Application Support", "Kobo", "Kobo Desktop Edition")
                 elif sys.platform.startswith('linux'):
+                    # Since on Linux, you have to run Kobo Desktop within Wine,
+                    # there is no guarantee where Kobo.sqlite (and the rest of
+                    # the kobo directory) might be.
+                    #
+                    # It should also be noted that Kobo Desktop will store all
+                    # of it files in the current directory where you run it,
+                    # meaning that a user might accidentally create several
+                    # Kobo.sqlite files which are all separate and then be
+                    # confused why Obok can't find any of the new books. Sadly
+                    # there isn't a trivial way to deal with this.
 
-                    #sets ~/.config/calibre as the location to store the kobodir location info file and creates this directory if necessary
-                    kobodir_cache_dir = os.path.join(os.environ['HOME'], ".config", "calibre")
+                    # We cache the kobodir we find in ~/.config/calibre.
+                    kobodir_cache_dir = os.path.join(os.environ['HOME'], ".config", "calibre", "plugins", "obok")
                     if not os.path.isdir(kobodir_cache_dir):
                         os.mkdir(kobodir_cache_dir)
-                    
-                    #appends the name of the file we're storing the kobodir location info to the above path
-                    kobodir_cache_file = str(kobodir_cache_dir) + "/" + "kobo location"
-                    
-                    """if the above file does not exist, recursively searches from the root
-                    of the filesystem until kobodir is found and stores the location of kobodir
-                    in that file so this loop can be skipped in the future"""
-                    original_stdout = sys.stdout
-                    if not os.path.isfile(kobodir_cache_file):
-                        for root, dirs, files in os.walk('/'):
-                            for file in files:
-                                if file == 'Kobo.sqlite':
-                                    kobo_linux_path = str(root)
-                                    with open(kobodir_cache_file, 'w') as f:
-                                        sys.stdout = f
-                                        print(kobo_linux_path, end='')
-                                        sys.stdout = original_stdout
+                    kobodir_cache_file = os.path.join(kobodir_cache_dir, "kobo-location")
 
-                    f = open(kobodir_cache_file, 'r' )
-                    self.kobodir = f.read()
+                    try:
+                        # If the cached version exists and the path does really
+                        # contain Kobo.sqlite, use that.
+                        with open(kobodir_cache_file, "r") as f:
+                            cached_kobodir = f.read().strip()
+                            assert os.path.isfile(os.path.join(cached_kobodir, "Kobo.sqlite"))
+                            self.kobodir = cached_kobodir
+                    except (AssertionError, FileNotFoundError):
+                        # If there was no cached version, search the entire
+                        # filesystem tree for a directory containing
+                        # "Kobo.sqlite".
+                        #
+                        # We first search $HOME to avoid picking another user's
+                        # Kobo.sqlite file, but then fallback to / if there was
+                        # nothing in $HOME.
+                        for candidate_root in (os.environ["HOME"], "/"):
+                            for root, _, files in os.walk(candidate_root):
+                                if "Kobo.sqlite" in files:
+                                    with open(kobodir_cache_file, "w") as f:
+                                        f.write("%s/\n" % (root,))
+                                    self.kobodir = root
+                                    break
 
-            # desktop versions use Kobo.sqlite
+            # Desktop versions use Kobo.sqlite.
             kobodb = os.path.join(self.kobodir, "Kobo.sqlite")
             # check for existence of file
-            if (not(os.path.isfile(kobodb))):
+            if not os.path.isfile(kobodb):
                 # give up here, we haven't found anything useful
                 self.kobodir = u""
                 kobodb  = u""
@@ -430,7 +444,7 @@ class KoboLibrary(object):
         macaddrs = []
         if sys.platform.startswith('win'):
             c = re.compile('\s?(' + '[0-9a-f]{2}[:\-]' * 5 + '[0-9a-f]{2})(\s|$)', re.IGNORECASE)
-            try: 
+            try:
                 output = subprocess.Popen('ipconfig /all', shell=True, stdout=subprocess.PIPE, text=True).stdout
                 for line in output:
                     m = c.search(line)

README.md

@@ -3,11 +3,13 @@ DeDRM tools for ebooks
 
 This is a fork of Apprentice Harper's version of the DeDRM tools. Apprentice Harper said that the original version of the plugin [is no longer maintained](https://github.com/apprenticeharper/DeDRM_tools#no-longer-maintained), so I've taken over, merged a bunch of open PRs, and added a ton more features and bugfixes. 
 
-The latest stable (released) version is v10.0.3 which [can be downloaded here](https://github.com/noDRM/DeDRM_tools/releases/tag/v10.0.3).
+The latest stable (released) version is v10.0.3 which [can be downloaded here](https://github.com/noDRM/DeDRM_tools/releases/tag/v10.0.3). The latest beta is v10.0.9, as a release candidate for v10.1.0. It [can be downloaded here](https://github.com/noDRM/DeDRM_tools/releases/tag/v10.0.9). 
 
-Take a look at [the CHANGELOG](https://github.com/noDRM/DeDRM_tools/blob/master/CHANGELOG.md) to see a list of changes since the last version by Apprentice Harper (v7.2.1). This plugin will start with version v10.0.0.
+The latest alpha version is available [at this link](https://github.com/noDRM/DeDRM_tools_autorelease/releases). This version is completely untested and will contain the latest code changes in this repository. With each commit in this repository, a new automatic alpha version will be uploaded there. If you want the most up-to-date code to test things and are okay with the plugin occasionally breaking, you can download this version.
 
-The v10.0.0 versions of this plugin should both work with Calibre 5.x (Python 3) as well as Calibre 4.x and lower (Python 2). If you encounter issues with this plugin in Calibre 4.x or lower, please open a bug report.
+Take a look at [the CHANGELOG](https://github.com/noDRM/DeDRM_tools/blob/master/CHANGELOG.md) to see a list of changes since the last version by Apprentice Harper (v7.2.1). 
+
+My version of the plugin should both work with Calibre 5.x/6.x (Python 3) as well as Calibre 4.x and lower (Python 2). If you encounter issues with this plugin in Calibre 4.x or lower, please open a bug report. 
 
 # Original README from Apprentice Harper