new seed_version: remove whitespaces only between CJK characters

gui/qt/installwizard.py

@@ -16,7 +16,7 @@ from amountedit import AmountEdit
 import sys
 import threading
 from electrum.plugins import run_hook
-
+from electrum.mnemonic import prepare_seed
 
 MSG_ENTER_ANYTHING    = _("Please enter a wallet seed, a master public key, a list of Bitcoin addresses, or a list of private keys")
 MSG_SHOW_MPK          = _("This is your master public key")
@@ -114,8 +114,7 @@ class InstallWizard(QDialog):
         r = self.enter_seed_dialog(MSG_VERIFY_SEED, sid, func)
         if not r:
             return
-
-        if r != seed:
+        if prepare_seed(r) != prepare_seed(seed):
             QMessageBox.warning(None, _('Error'), _('Incorrect seed'), _('OK'))
             return False
         else:

lib/bitcoin.py

@@ -150,7 +150,7 @@ hmac_sha_512 = lambda x,y: hmac.new(x, y, hashlib.sha512).digest()
 
 def is_new_seed(x, prefix=version.SEED_BIP44):
     import mnemonic
-    x = mnemonic.Mnemonic.prepare_seed(x)
+    x = mnemonic.prepare_seed(x)
     s = hmac_sha_512("Seed version", x.encode('utf8')).encode('hex')
     return s.startswith(prefix)
 

lib/mnemonic.py

@@ -32,6 +32,60 @@ from bitcoin import is_old_seed, is_new_seed
 import version
 
 
+# http://www.asahi-net.or.jp/~ax2s-kmtn/ref/unicode/e_asia.html
+CJK_INTERVALS = [
+    (0x4E00, 0x9FFF, 'CJK Unified Ideographs'),
+    (0x3400, 0x4DBF, 'CJK Unified Ideographs Extension A'),
+    (0x20000, 0x2A6DF, 'CJK Unified Ideographs Extension B'),
+    (0x2A700, 0x2B73F, 'CJK Unified Ideographs Extension C'),
+    (0x2B740, 0x2B81F, 'CJK Unified Ideographs Extension D'),
+    (0xF900, 0xFAFF, 'CJK Compatibility Ideographs'),
+    (0x2F800, 0x2FA1D, 'CJK Compatibility Ideographs Supplement'),
+    (0x3190, 0x319F , 'Kanbun'),
+    (0x2E80, 0x2EFF, 'CJK Radicals Supplement'),
+    (0x2F00, 0x2FDF, 'CJK Radicals'),
+    (0x31C0, 0x31EF, 'CJK Strokes'),
+    (0x2FF0, 0x2FFF, 'Ideographic Description Characters'),
+    (0xE0100, 0xE01EF, 'Variation Selectors Supplement'),
+    (0x3100, 0x312F, 'Bopomofo'),
+    (0x31A0, 0x31BF, 'Bopomofo Extended'),
+    (0xFF00, 0xFFEF, 'Halfwidth and Fullwidth Forms'),
+    (0x3040, 0x309F, 'Hiragana'),
+    (0x30A0, 0x30FF, 'Katakana'),
+    (0x31F0, 0x31FF, 'Katakana Phonetic Extensions'),
+    (0x1B000, 0x1B0FF, 'Kana Supplement'),
+    (0xAC00, 0xD7AF, 'Hangul Syllables'),
+    (0x1100, 0x11FF, 'Hangul Jamo'),
+    (0xA960, 0xA97F, 'Hangul Jamo Extended A'),
+    (0xD7B0, 0xD7FF, 'Hangul Jamo Extended B'),
+    (0x3130, 0x318F, 'Hangul Compatibility Jamo'),
+    (0xA4D0, 0xA4FF, 'Lisu'),
+    (0x16F00, 0x16F9F, 'Miao'),
+    (0xA000, 0xA48F, 'Yi Syllables'),
+    (0xA490, 0xA4CF, 'Yi Radicals'),
+]
+
+def is_CJK(c):
+    n = ord(c)
+    for imin,imax,name in CJK_INTERVALS:
+        if n>=imin and n<=imax: return True
+    return False
+
+
+def prepare_seed(seed):
+    # normalize
+    seed = unicodedata.normalize('NFKD', unicode(seed))
+    # lower
+    seed = seed.lower()
+    # remove accents
+    seed = u''.join([c for c in seed if not unicodedata.combining(c)])
+    # normalize whitespaces
+    seed = u' '.join(seed.split())
+    # remove whitespaces between CJK
+    seed = u''.join([seed[i] for i in range(len(seed)) if not (seed[i] in string.whitespace and is_CJK(seed[i-1]) and is_CJK(seed[i+1]))])
+    return seed
+
+
 filenames = {
     'en':'english.txt',
     'es':'spanish.txt',
@@ -63,19 +117,9 @@ class Mnemonic(object):
     @classmethod
     def mnemonic_to_seed(self, mnemonic, passphrase):
         PBKDF2_ROUNDS = 2048
-        mnemonic = self.prepare_seed(mnemonic)
+        mnemonic = prepare_seed(mnemonic)
         return pbkdf2.PBKDF2(mnemonic, 'mnemonic' + passphrase, iterations = PBKDF2_ROUNDS, macmodule = hmac, digestmodule = hashlib.sha512).read(64)
 
-    @classmethod
-    def prepare_seed(self, seed):
-        # normalize
-        seed = unicodedata.normalize('NFKD', unicode(seed))
-        # lower
-        seed = seed.lower()
-        # remove accents and whitespaces
-        seed = u''.join([c for c in seed if not unicodedata.combining(c) and not c in string.whitespace])
-        return seed
-
     def mnemonic_encode(self, i):
         n = len(self.wordlist)
         words = []
@@ -120,4 +164,3 @@ class Mnemonic(object):
                 break
         print_error('%d words'%len(seed.split()))
         return seed
-

lib/version.py

@@ -1,6 +1,6 @@
 ELECTRUM_VERSION = "2.0"    # version of the client package
 PROTOCOL_VERSION = '0.9'    # protocol version requested
-NEW_SEED_VERSION = 9        # electrum versions >= 2.0
+NEW_SEED_VERSION = 10       # electrum versions >= 2.0
 OLD_SEED_VERSION = 4        # electrum versions < 2.0