bip39 refactoring

pybtc/constants.py

@@ -3,7 +3,7 @@ import random
 import os
 
 ROOT_DIR = os.path.abspath(os.path.dirname(__file__))
-BIP0039_DIR = os.path.normpath(os.path.join(ROOT_DIR, 'bip-0039'))
+BIP0039_DIR = os.path.normpath(os.path.join(ROOT_DIR, 'bip39_word_list'))
 
 MAX_AMOUNT = 2100000000000000
 SIGHASH_ALL = 0x00000001

pybtc/functions/__init__.py

@@ -4,4 +4,5 @@ from .script import *
 from .tools import *
 from .hash import *
 from .block import *
-from .encode import *
+from .encode import *
+from .bip39_mnemonic import *

pybtc/functions/bip39_mnemonic.py

@@ -0,0 +1,157 @@
+import os
+import sys
+import time
+import random
+from secp256k1 import ffi
+parentPath = os.path.abspath("../..")
+if parentPath not in sys.path:
+    sys.path.insert(0, parentPath)
+
+from pybtc.constants import *
+from .hash import *
+from hashlib import pbkdf2_hmac
+
+
+def generate_entropy(strength=256, hex=True):
+    """
+    Generate 128-256 bits entropy bytes string
+
+    :param int strength: entropy bits strength, by default is 256 bit.
+    :param boolean hex: return HEX encoded string result flag, by default True.
+    :return: HEX encoded or bytes entropy string.
+    """
+    if strength not in [128, 160, 192, 224, 256]:
+        raise ValueError('strength should be one of the following [128, 160, 192, 224, 256]')
+    a = random.SystemRandom().randint(0, MAX_INT_PRIVATE_KEY)
+    i = int((time.time() % 0.01 ) * 100000)
+    h = a.to_bytes(32, byteorder="big")
+    # more entropy from system timer and sha256 derivation
+    while i:
+        h = hashlib.sha256(h).digest()
+        i -= 1
+        if not i and int.from_bytes(h, byteorder="big") > MAX_INT_PRIVATE_KEY:
+            i += 1
+    return h[:int(strength/8)] if not hex else h[:int(strength/8)].hex()
+
+
+def load_word_list(language='english', word_list_dir=None):
+    """
+    Load the word list from local file.
+
+    :param str language: (optional) uses word list language (chinese_simplified, chinese_traditional, english, french,
+                         italian, japanese, korean, spanish), by default is english.
+    :param str word_list_dir: (optional) path to a directory containing a list of words,
+                              by default None (use BIP39 standard list)
+    :return: list of words.
+    """
+    if not word_list_dir:
+        word_list_dir = BIP0039_DIR
+    path = os.path.join(word_list_dir, '.'.join((language, 'txt')))
+    if not os.path.exists(path):
+        raise ValueError("word list not exist")
+    with open(path) as f:
+        word_list = f.read().rstrip('\n').split('\n')
+    if len(word_list) != 2048:
+        raise ValueError("word list invalid, should contain 2048 words")
+    return word_list
+
+
+def entropy_to_mnemonic(entropy, language='english', word_list_dir=None, word_list=None):
+    """
+    Convert entropy to mnemonic words string.
+
+    :param str,bytes entropy: random entropy HEX encoded or bytes string.
+    :param str language: (optional) uses word list language (chinese_simplified, chinese_traditional, english, french,
+                         italian, japanese, korean, spanish), by default is english.
+    :param str word_list_dir: (optional) path to a directory containing a list of words,
+                              by default None (use BIP39 standard list)
+    :param list word_list: (optional) already loaded word list, by default None    
+    :return: mnemonic words string.
+    """
+    if isinstance(entropy, str):
+        entropy = bytes.fromhex(entropy)
+    if not isinstance(entropy, bytes):
+        raise TypeError("entropy should be bytes or hex encoded string")
+    if len(entropy) not in [16, 20, 24, 28, 32]:
+        raise ValueError(
+            'entropy length should be one of the following: [16, 20, 24, 28, 32]')
+
+    if word_list is None:
+        word_list = load_word_list(language, word_list_dir)
+    elif not isinstance(word_list, list) or len(word_list) != 2048:
+        raise TypeError("invalid wordl ist type")
+
+    # checksum
+    mask = 0b10000000
+    data_int = int.from_bytes(entropy, byteorder="big")
+    data_bit_len = len(entropy) * 8 // 32
+    fbyte_hash = sha256(entropy)[0]
+
+    while data_bit_len:
+        data_bit_len -= 1
+        data_int = (data_int << 1) | 1 if fbyte_hash & mask else data_int << 1
+        mask = mask >> 1
+
+    mnemonic = []
+    while data_int:
+        mnemonic.append(word_list[data_int & 0b11111111111])
+        data_int = data_int >> 11
+
+    return " ".join(mnemonic[::-1])
+
+
+def mnemonic_to_entropy(mnemonic, language='english', word_list_dir=None,
+                        word_list=None, hex=True):
+    """
+    Converting mnemonic words to entropy.
+    
+    :param str mnemonic: mnemonic words string (space separated)
+    :param str language: (optional) uses word list language (chinese_simplified, chinese_traditional, english, french,
+                         italian, japanese, korean, spanish), by default is english.
+    :param str word_list_dir: (optional) path to a directory containing a list of words,
+                              by default None (use BIP39 standard list)
+    :param list word_list: (optional) already loaded word list, by default None    
+    :param boolean hex: return HEX encoded string result flag, by default True.
+    :return: bytes string.
+    """
+    if word_list is None:
+        word_list = load_word_list(language, word_list_dir)
+    elif not isinstance(word_list, list) or len(word_list) != 2048:
+        raise TypeError("invalid word list type")
+
+    mnemonic = mnemonic.split()
+    word_count = len(mnemonic)
+    if word_count not in [12, 15, 18, 21, 24]:
+        raise ValueError('Number of words must be one of the following: [12, 15, 18, 21, 24]')
+
+    codes = {w: c for c, w in enumerate(word_list)}
+    entropy_int = 0
+    bit_size = word_count * 11
+    chk_sum_bit_len = word_count * 11 % 32
+    for w in mnemonic:
+        entropy_int = (entropy_int << 11) | codes[w]
+    chk_sum = entropy_int & (2 ** chk_sum_bit_len - 1)
+    entropy_int = entropy_int >> chk_sum_bit_len
+    entropy = entropy_int.to_bytes((bit_size - chk_sum_bit_len) // 8, byteorder="big")
+    fb = sha256(entropy)[0]
+    assert (fb >> (8 - chk_sum_bit_len)) == chk_sum
+    return entropy if not hex else entropy.hex()
+
+
+def mnemonic_to_seed(mnemonic, passphrase="", hex=True):
+    """
+    Converting mnemonic words string to seed for uses in key derivation (BIP-0032).
+
+    :param str mnemonic: mnemonic words string (space separated)
+    :param str passphrase: (optional) passphrase to get ability use 2FA approach for 
+                          creating seed, by default empty string.
+    :param boolean hex: return HEX encoded string result flag, by default True.
+    :return: HEX encoded or bytes string.
+    """
+    if not isinstance(mnemonic, str):
+        raise TypeError("mnemonic should be string")
+    if not isinstance(passphrase, str):
+        raise TypeError("mnemonic should be string")
+
+    seed = pbkdf2_hmac('sha512', mnemonic.encode(), ("mnemonic"+passphrase).encode(), 2048)
+    return seed if not hex else seed.hex()

pybtc/functions/key.py

@@ -11,6 +11,7 @@ from pybtc.constants import *
 from .hash import *
 from .encode import *
 from .hash import *
+from .bip39_mnemonic import generate_entropy
 
 
 def create_private_key(compressed=True, testnet=False, wif=True, hex=False):
@@ -27,20 +28,11 @@ def create_private_key(compressed=True, testnet=False, wif=True, hex=False):
              raw bytes string in case wif and hex flags set to False.
 
     """
-    a = random.SystemRandom().randint(0, MAX_INT_PRIVATE_KEY)
-    i = int((time.time() % 0.01 ) * 100000)
-    h = a.to_bytes(32, byteorder="big")
-    # more entropy from system timer and sha256 derivation
-    while i:
-        h = hashlib.sha256(h).digest()
-        i -= 1
-        if not i and int.from_bytes(h, byteorder="big") > MAX_INT_PRIVATE_KEY:
-            i += 1
     if wif:
-        return private_key_to_wif(h, compressed=compressed, testnet=testnet)
+        return private_key_to_wif(generate_entropy(hex=False), compressed=compressed, testnet=testnet)
     elif hex:
-        return h.hex()
-    return h
+        return generate_entropy()
+    return generate_entropy(hex=False)
 
 
 def private_key_to_wif(h, compressed=True, testnet=False):

pybtc/functions/tools.py

@@ -146,8 +146,7 @@ def read_var_int(stream):
     :return: bytes.
     """
     l = stream.read(1)
-    bytes_length = get_var_int_len(l)
-    return b"%s%s" % (l, stream.read(bytes_length - 1))
+    return b"".join((l, stream.read(get_var_int_len(l) - 1)))
 
 
 def read_var_list(stream, data_type):

pybtc/hdwallet.py

@@ -402,122 +402,3 @@ def deserialize_xkey(encode_key):
     return None
 
 
-
-# Mnemonic code for generating deterministic keys
-# BIP-0039
-
-def create_passphrase(bits=256, language='english'):
-    """
-    Creating the passphrase.
-
-    :param int bits: size of entropy is 128-256 bits, by default is 256.
-    :param str language: uses wordlist language (chinese_simplified, chinese_traditional, english, french, italian, japanese, korean, spanish), by default is english.
-    :return: string is passphrase.
-    """
-    if bits in [128, 160, 192, 224, 256]:
-        entropy = os.urandom(bits // 8)
-        mnemonic = create_mnemonic(entropy, language)
-        return ' '.join(mnemonic)
-    else:
-        raise ValueError('Strength should be one of the following [128, 160, 192, 224, 256], but it is not (%d).' % bits)
-
-
-def create_mnemonic(entropy, language='english'):
-    """
-    Generating the mnemonic.
-
-    :param bytes entropy: random entropy bytes.
-    :param str language: uses wordlist language (chinese_simplified, chinese_traditional, english, french, italian, japanese, korean, spanish), by default is english.
-    :return: list of words.
-    """
-    mnemonic = []
-    wordlist = create_wordlist(language)
-    entropy_int = int.from_bytes(entropy, byteorder="big")
-    entropy_bit_len = len(entropy) * 8
-    chk_sum_bit_len = entropy_bit_len // 32
-    fbyte_hash = sha256(entropy)[0]
-    entropy_int = add_checksum_ent(entropy)
-    while entropy_int:
-        mnemonic.append(wordlist[entropy_int & 0b11111111111])
-        entropy_int = entropy_int >> 11
-    return mnemonic[::-1]
-
-
-def create_wordlist(language='english', wordlist_dir=None):
-    """
-    Creating the wordlist.
-
-    :param str language: uses wordlist language (chinese_simplified, chinese_traditional, english, french, italian, japanese, korean, spanish), by default is english.
-    :param str wordlist_dir: path to a file containing a list of words.
-    :return: list of words.
-    """
-    if not wordlist_dir:
-        wordlist_dir = BIP0039_DIR
-    f = None
-    path = os.path.join(wordlist_dir, '.'.join((language, 'txt')))
-    assert os.path.exists(path)
-    f = open(path)
-    content = f.read().rstrip('\n')
-    assert content
-    f.close()
-    return content.split('\n')
-
-
-def add_checksum_ent(data):
-    """
-    Adding a checksum of a entropy to a entropy.
-
-    :param bytes data: random entropy bytes.
-    :return: bytes string.
-    """
-    mask = 0b10000000
-    data_int = int.from_bytes(data, byteorder="big")
-    data_bit_len = len(data) * 8 // 32
-    fbyte_hash = sha256(data)[0]
-    while data_bit_len:
-        data_bit_len -= 1
-        data_int = (data_int << 1) | 1 if fbyte_hash & mask else data_int << 1
-        mask = mask >> 1
-    return data_int
-
-
-def mnemonic_to_entropy(passphrase, language):
-    """
-    Converting passphrase to entropy.
-
-    :param str passphrase: key passphrase.
-    :param str language: uses wordlist language.
-    :return: bytes string.
-    """
-    mnemonic = passphrase.split()
-    if len(mnemonic) in [12, 15, 18, 21, 24]:
-        wordlist = create_wordlist(language)
-        codes = dict()
-        for code, word in enumerate(wordlist):
-            codes[word] = code
-        word_count = len(mnemonic)
-        entropy_int = None
-        bit_size = word_count * 11
-        chk_sum_bit_len = word_count * 11 % 32
-        for word in mnemonic:
-            entropy_int = (entropy_int << 11) | codes[word] if entropy_int else codes[word]
-        chk_sum = entropy_int & (2 ** chk_sum_bit_len - 1)
-        entropy_int = entropy_int >> chk_sum_bit_len
-        entropy = entropy_int.to_bytes((bit_size - chk_sum_bit_len) // 8, byteorder="big")
-        fb = sha256(entropy)[0]
-        assert (fb >> (8 - chk_sum_bit_len)) & chk_sum
-        return entropy
-    else:
-        raise ValueError('Number of words must be one of the following: [12, 15, 18, 21, 24], but it is not (%d).' % len(mnemonic))
-
-
-def mnemonic_to_seed(passphrase, password):
-    """
-    Converting passphrase to seed for uses in key derivation (BIP-0032).
-
-    :param str passphrase: key passphrase.
-    :param str password: password for key passphrase.
-    :return: bytes string.
-    """
-    return pbkdf2_hmac('sha512', passphrase.encode(), (passphrase + password).encode(), 2048)
-

setup.py

@@ -16,6 +16,6 @@ setup(name='pybtc',
       install_requires=[ 'secp256k1'],
       include_package_data=True,
       package_data={
-          'pybtc': ['bip-0039/*.txt'],
+          'pybtc': ['bip39_word_list/*.txt'],
       },
       zip_safe=False)

tests/test.py

@@ -5,5 +5,5 @@ import test
 testLoad = unittest.TestLoader()
 suites = testLoad.loadTestsFromModule(test)
 
-runner = unittest.TextTestRunner(verbosity=2)
+runner = unittest.TextTestRunner(verbosity=3)
 runner.run(suites)

tests/test/__init__.py

@@ -1,12 +1,13 @@
-from .hash_functions import *
-from .integer import *
-from .address_functions import *
-from .address_class import *
-from .ecdsa import *
-from .transaction_deserialize import *
-from .transaction_constructor import *
-from .sighash import *
-from .block import *
+# from .hash_functions import *
+# from .integer import *
+# from .address_functions import *
+# from .address_class import *
+# from .ecdsa import *
+# from .transaction_deserialize import *
+# from .transaction_constructor import *
+# from .sighash import *
+# from .block import *
+from .mnemonic import *
 
 # from .script_deserialize import *
 # from .create_transaction import *

tests/test/block.py

@@ -609,6 +609,10 @@ class BlockDeserializeTests(unittest.TestCase):
                      "f = open('./test/raw_block.txt');"
                      "fc = f.readline();"
                      "pybtc.Block(fc[:-1], format='decoded')")
+        cProfile.run("import pybtc;"
+                     "f = open('./test/raw_block.txt');"
+                     "fc = f.readline();"
+                     "pybtc.Block(fc[:-1], format='raw')")
         # print(">>>",block.bits)
         # print(">>>",block.hash)
         # print(">>>",block.timestamp)

tests/test/mnemonic.py

@@ -0,0 +1,27 @@
+import unittest
+import os, sys
+parentPath = os.path.abspath("..")
+if parentPath not in sys.path:
+    sys.path.insert(0, parentPath)
+
+from pybtc import *
+
+
+class BlockDeserializeTests(unittest.TestCase):
+    @classmethod
+    def setUpClass(cls):
+        print("\nTesting Block class deserialization:\n")
+
+    def test_mnemonic_functions(self):
+        mnemonic = 'young crime force door joy subject situate hen pen sweet brisk snake nephew sauce ' \
+                   'point skate life truly hockey scout assault lab impulse boss'
+        entropy = "ff46716c20b789aff26b59a27b74716699457f29d650815d2db1e0a0d8f81c88"
+        seed = "a870edd6272a4f0962a7595612d96645f683a3378fd9b067340eb11ebef45cb3d28fb64678cadc43969846" \
+               "0a3d48bd57b2ae562b6d2b3c9fb5462d21e474191c"
+        self.assertEqual(entropy_to_mnemonic(entropy), mnemonic)
+
+        self.assertEqual(mnemonic_to_entropy(mnemonic), entropy)
+        self.assertEqual(mnemonic_to_seed(mnemonic), seed)
+
+        print(generate_entropy())
+        print(generate_entropy(128))