Allow 4-character abbreviations for mnemonic when using the existing-mnemonic workflow

staking_deposit/cli/existing_mnemonic.py

@@ -5,7 +5,7 @@
 
 from staking_deposit.exceptions import ValidationError
 from staking_deposit.key_handling.key_derivation.mnemonic import (
-    verify_mnemonic,
+    reconstruct_mnemonic,
 )
 from staking_deposit.utils.constants import (
     WORD_LISTS_PATH,
@@ -23,7 +23,8 @@
 
 
 def validate_mnemonic(ctx: click.Context, param: Any, mnemonic: str) -> str:
-    if verify_mnemonic(mnemonic, WORD_LISTS_PATH):
+    mnemonic = reconstruct_mnemonic(mnemonic, WORD_LISTS_PATH)
+    if mnemonic is not None:
         return mnemonic
     else:
         raise ValidationError(load_text(['err_invalid_mnemonic']))

staking_deposit/cli/new_mnemonic.py

@@ -5,6 +5,7 @@
 
 from staking_deposit.key_handling.key_derivation.mnemonic import (
     get_mnemonic,
+    reconstruct_mnemonic,
 )
 from staking_deposit.utils.click import (
     captive_prompt_callback,
@@ -47,15 +48,14 @@
 def new_mnemonic(ctx: click.Context, mnemonic_language: str, **kwargs: Any) -> None:
     mnemonic = get_mnemonic(language=mnemonic_language, words_path=WORD_LISTS_PATH)
     test_mnemonic = ''
-    while mnemonic != test_mnemonic:
+    while mnemonic != reconstruct_mnemonic(test_mnemonic, WORD_LISTS_PATH):
         click.clear()
         click.echo(load_text(['msg_mnemonic_presentation']))
         click.echo('\n\n%s\n\n' % mnemonic)
         click.pause(load_text(['msg_press_any_key']))
 
         click.clear()
         test_mnemonic = click.prompt(load_text(['msg_mnemonic_retype_prompt']) + '\n\n')
-        test_mnemonic = test_mnemonic.lower()
     click.clear()
     # Do NOT use mnemonic_password.
     ctx.obj = {'mnemonic': mnemonic, 'mnemonic_password': ''}

staking_deposit/intl/en/cli/existing_mnemonic.json

@@ -8,7 +8,7 @@
         },
         "arg_mnemonic": {
             "help": "The mnemonic that you used to generate your keys. (It is recommended not to use this argument, and wait for the CLI to ask you for your mnemonic as otherwise it will appear in your shell history.)",
-            "prompt": "Please enter your mnemonic separated by spaces (\" \")"
+            "prompt": "Please enter your mnemonic separated by spaces (\" \"). Note: you only need to enter the first 4 letters of each word if you'd prefer."
         },
         "arg_mnemonic_password": {
             "help": "This is almost certainly not the argument you are looking for: it is for mnemonic passwords, not keystore passwords. Providing a password here when you didn't use one initially, can result in lost keys (and therefore funds)! Also note that if you used this tool to generate your mnemonic initially, then you did not use a mnemonic password. However, if you are certain you used a password to \"increase\" the security of your mnemonic, this is where you enter it.",

staking_deposit/intl/en/cli/new_mnemonic.json

@@ -10,6 +10,6 @@
         },
         "msg_mnemonic_presentation": "This is your mnemonic (seed phrase). Write it down and store it safely. It is the ONLY way to retrieve your deposit.",
         "msg_press_any_key": "Press any key when you have written down your mnemonic.",
-        "msg_mnemonic_retype_prompt": "Please type your mnemonic (separated by spaces) to confirm you have written it down"
+        "msg_mnemonic_retype_prompt": "Please type your mnemonic (separated by spaces) to confirm you have written it down. Note: you only need to enter the first 4 letters of each word if you'd prefer."
     }
 }

staking_deposit/key_handling/key_derivation/mnemonic.py

@@ -2,6 +2,7 @@
 from unicodedata import normalize
 from secrets import randbits
 from typing import (
+    List,
     Optional,
     Sequence,
 )
@@ -68,9 +69,10 @@ def determine_mnemonic_language(mnemonic: str, words_path: str) -> Sequence[str]
     languages = MNEMONIC_LANG_OPTIONS.keys()
     word_language_map = {word: lang for lang in languages for word in _get_word_list(lang, words_path)}
     try:
-        mnemonic_list = mnemonic.split(' ')
-        word_languages = [word_language_map[word] for word in mnemonic_list]
-        return list(set(word_languages))
+        mnemonic_list = [normalize('NFKC', word)[:4] for word in mnemonic.lower().split(' ')]
+        word_languages = [[lang for word, lang in word_language_map.items() if normalize('NFKC', word)[:4] == abbrev]
+                          for abbrev in mnemonic_list]
+        return list(set(sum(word_languages, [])))
     except KeyError:
         raise ValueError('Word not found in mnemonic word lists for any language.')
 
@@ -90,30 +92,48 @@ def _get_checksum(entropy: bytes) -> int:
     return int.from_bytes(SHA256(entropy), 'big') >> (256 - checksum_length)
 
 
-def verify_mnemonic(mnemonic: str, words_path: str) -> bool:
+def abbreviate_words(words: Sequence[str]) -> List[str]:
     """
-    Given a mnemonic, verify it against its own checksum."
+    Given a series of word strings, return the 4-letter version of each word (which is unique according to BIP39)
+    """
+    return [normalize('NFKC', word)[:4] for word in words]
+
+
+def reconstruct_mnemonic(mnemonic: str, words_path: str) -> Optional[str]:
+    """
+    Given a mnemonic, a reconstructed the full version (incase the abbreviated words were used)
+    then verify it against its own checksum
     """
     try:
         languages = determine_mnemonic_language(mnemonic, words_path)
     except ValueError:
-        return False
+        return None
+    reconstructed_mnemonic = None
     for language in languages:
         try:
-            word_list = _get_word_list(language, words_path)
-            mnemonic_list = mnemonic.split(' ')
+            word_list = abbreviate_words(_get_word_list(language, words_path))
+            mnemonic_list = abbreviate_words(mnemonic.lower().split(' '))
             if len(mnemonic_list) not in range(12, 25, 3):
-                return False
+                return None
             word_indices = [_word_to_index(word_list, word) for word in mnemonic_list]
             mnemonic_int = _uint11_array_to_uint(word_indices)
             checksum_length = len(mnemonic_list) // 3
             checksum = mnemonic_int & 2**checksum_length - 1
             entropy = (mnemonic_int - checksum) >> checksum_length
             entropy_bits = entropy.to_bytes(checksum_length * 4, 'big')
-            return _get_checksum(entropy_bits) == checksum
+            full_word_list = _get_word_list(language, words_path)
+            if _get_checksum(entropy_bits) == checksum:
+                """
+                This check guarantees that only one language has a valid mnemonic.
+                It is needed to ensure abbrivated words aren't valid in multiple languages
+                """
+                assert reconstructed_mnemonic is None
+                reconstructed_mnemonic = ' '.join([_index_to_word(full_word_list, index) for index in word_indices])
+            else:
+                pass
         except ValueError:
             pass
-    return False
+    return reconstructed_mnemonic
 
 
 def get_mnemonic(*, language: str, words_path: str, entropy: Optional[bytes]=None) -> str:

tests/test_cli/test_existing_menmonic.py

@@ -152,3 +152,50 @@ async def test_script() -> None:
 
     # Clean up
     clean_key_folder(my_folder_path)
+
+
+@pytest.mark.asyncio
+async def test_script_abbreviated_mnemonic() -> None:
+    my_folder_path = os.path.join(os.getcwd(), 'TESTING_TEMP_FOLDER')
+    if not os.path.exists(my_folder_path):
+        os.mkdir(my_folder_path)
+
+    if os.name == 'nt':  # Windows
+        run_script_cmd = 'sh deposit.sh'
+    else:  # Mac or Linux
+        run_script_cmd = './deposit.sh'
+
+    install_cmd = run_script_cmd + ' install'
+    proc = await asyncio.create_subprocess_shell(
+        install_cmd,
+    )
+    await proc.wait()
+
+    cmd_args = [
+        run_script_cmd,
+        '--language', 'english',
+        '--non_interactive',
+        'existing-mnemonic',
+        '--num_validators', '1',
+        '--mnemonic="aban aban aban aban aban aban aban aban aban aban aban abou"',
+        '--mnemonic-password', 'TREZOR',
+        '--validator_start_index', '1',
+        '--chain', 'mainnet',
+        '--keystore_password', 'MyPassword',
+        '--folder', my_folder_path,
+    ]
+    proc = await asyncio.create_subprocess_shell(
+        ' '.join(cmd_args),
+    )
+    await proc.wait()
+    # Check files
+    validator_keys_folder_path = os.path.join(my_folder_path, DEFAULT_VALIDATOR_KEYS_FOLDER_NAME)
+    _, _, key_files = next(os.walk(validator_keys_folder_path))
+
+    # Verify file permissions
+    if os.name == 'posix':
+        for file_name in key_files:
+            assert get_permissions(validator_keys_folder_path, file_name) == '0o440'
+
+    # Clean up
+    clean_key_folder(my_folder_path)

tests/test_cli/test_new_mnemonic.py

@@ -9,6 +9,7 @@
 
 from staking_deposit.cli import new_mnemonic
 from staking_deposit.deposit import cli
+from staking_deposit.key_handling.key_derivation.mnemonic import abbreviate_words
 from staking_deposit.utils.constants import DEFAULT_VALIDATOR_KEYS_FOLDER_NAME, ETH1_ADDRESS_WITHDRAWAL_PREFIX
 from staking_deposit.utils.intl import load_text
 from .helpers import clean_key_folder, get_permissions, get_uuid
@@ -17,7 +18,7 @@
 def test_new_mnemonic_bls_withdrawal(monkeypatch) -> None:
     # monkeypatch get_mnemonic
     def mock_get_mnemonic(language, words_path, entropy=None) -> str:
-        return "fakephrase"
+        return "abandon abandon abandon abandon abandon abandon abandon abandon abandon abandon abandon about"
 
     monkeypatch.setattr(new_mnemonic, "get_mnemonic", mock_get_mnemonic)
 
@@ -28,7 +29,8 @@ def mock_get_mnemonic(language, words_path, entropy=None) -> str:
         os.mkdir(my_folder_path)
 
     runner = CliRunner()
-    inputs = ['english', 'english', '1', 'mainnet', 'MyPassword', 'MyPassword', 'fakephrase']
+    inputs = ['english', 'english', '1', 'mainnet', 'MyPassword', 'MyPassword',
+              'abandon abandon abandon abandon abandon abandon abandon abandon abandon abandon abandon about']
     data = '\n'.join(inputs)
     result = runner.invoke(cli, ['new-mnemonic', '--folder', my_folder_path], input=data)
     assert result.exit_code == 0
@@ -56,7 +58,7 @@ def mock_get_mnemonic(language, words_path, entropy=None) -> str:
 def test_new_mnemonic_eth1_address_withdrawal(monkeypatch) -> None:
     # monkeypatch get_mnemonic
     def mock_get_mnemonic(language, words_path, entropy=None) -> str:
-        return "fakephrase"
+        return "abandon abandon abandon abandon abandon abandon abandon abandon abandon abandon abandon about"
 
     monkeypatch.setattr(new_mnemonic, "get_mnemonic", mock_get_mnemonic)
 
@@ -67,7 +69,8 @@ def mock_get_mnemonic(language, words_path, entropy=None) -> str:
         os.mkdir(my_folder_path)
 
     runner = CliRunner()
-    inputs = ['english', '1', 'mainnet', 'MyPassword', 'MyPassword', 'fakephrase']
+    inputs = ['english', '1', 'mainnet', 'MyPassword', 'MyPassword',
+              'abandon abandon abandon abandon abandon abandon abandon abandon abandon abandon abandon about']
     data = '\n'.join(inputs)
     eth1_withdrawal_address = '0x00000000219ab540356cbb839cbe05303d7705fa'
     arguments = [
@@ -178,3 +181,76 @@ async def test_script() -> None:
 
     # Clean up
     clean_key_folder(my_folder_path)
+
+
+@pytest.mark.asyncio
+async def test_script_abbreviated_mnemonic() -> None:
+    my_folder_path = os.path.join(os.getcwd(), 'TESTING_TEMP_FOLDER')
+    if not os.path.exists(my_folder_path):
+        os.mkdir(my_folder_path)
+
+    if os.name == 'nt':  # Windows
+        run_script_cmd = 'sh deposit.sh'
+    else:  # Mac or Linux
+        run_script_cmd = './deposit.sh'
+
+    install_cmd = run_script_cmd + ' install'
+    proc = await asyncio.create_subprocess_shell(
+        install_cmd,
+    )
+    await proc.wait()
+
+    cmd_args = [
+        run_script_cmd,
+        '--language', 'english',
+        '--non_interactive',
+        'new-mnemonic',
+        '--num_validators', '5',
+        '--mnemonic_language', 'english',
+        '--chain', 'mainnet',
+        '--keystore_password', 'MyPassword',
+        '--folder', my_folder_path,
+    ]
+    proc = await asyncio.create_subprocess_shell(
+        ' '.join(cmd_args),
+        stdin=asyncio.subprocess.PIPE,
+        stdout=asyncio.subprocess.PIPE,
+    )
+
+    seed_phrase = ''
+    parsing = False
+    mnemonic_json_file = os.path.join(os.getcwd(), 'staking_deposit/../staking_deposit/cli/', 'new_mnemonic.json')
+    async for out in proc.stdout:
+        output = out.decode('utf-8').rstrip()
+        if output.startswith(load_text(['msg_mnemonic_presentation'], mnemonic_json_file, 'new_mnemonic')):
+            parsing = True
+        elif output.startswith(load_text(['msg_mnemonic_retype_prompt'], mnemonic_json_file, 'new_mnemonic')):
+            parsing = False
+        elif parsing:
+            seed_phrase += output
+            if len(seed_phrase) > 0:
+                abbreviated_mnemonic = ' '.join(abbreviate_words(seed_phrase.split(' ')))
+                encoded_phrase = abbreviated_mnemonic.encode()
+                proc.stdin.write(encoded_phrase)
+                proc.stdin.write(b'\n')
+
+    assert len(seed_phrase) > 0
+
+    # Check files
+    validator_keys_folder_path = os.path.join(my_folder_path, DEFAULT_VALIDATOR_KEYS_FOLDER_NAME)
+    _, _, key_files = next(os.walk(validator_keys_folder_path))
+
+    all_uuid = [
+        get_uuid(validator_keys_folder_path + '/' + key_file)
+        for key_file in key_files
+        if key_file.startswith('keystore')
+    ]
+    assert len(set(all_uuid)) == 5
+
+    # Verify file permissions
+    if os.name == 'posix':
+        for file_name in key_files:
+            assert get_permissions(validator_keys_folder_path, file_name) == '0o440'
+
+    # Clean up
+    clean_key_folder(my_folder_path)

tests/test_key_handling/test_key_derivation/test_mnemonic.py

@@ -4,6 +4,7 @@
 from typing import (
     Sequence,
 )
+from unicodedata import normalize
 
 from staking_deposit.utils.constants import (
     MNEMONIC_LANG_OPTIONS,
@@ -13,7 +14,7 @@
     _get_word_list,
     get_seed,
     get_mnemonic,
-    verify_mnemonic,
+    reconstruct_mnemonic,
 )
 
 
@@ -40,13 +41,29 @@ def test_bip39(language: str, test: Sequence[str]) -> None:
 
 
 @pytest.mark.parametrize(
-    'test_mnemonic,is_valid',
-    [(test_mnemonic[1], True)
+    'test_mnemonic',
+    [(test_mnemonic[1])
      for _, language_test_vectors in test_vectors.items()
      for test_mnemonic in language_test_vectors]
 )
-def test_verify_mnemonic(test_mnemonic: str, is_valid: bool) -> None:
-    assert verify_mnemonic(test_mnemonic, WORD_LISTS_PATH) == is_valid
+def test_reconstruct_mnemonic(test_mnemonic: str) -> None:
+    assert reconstruct_mnemonic(test_mnemonic, WORD_LISTS_PATH) is not None
+
+
+def abbreviate_mnemonic(mnemonic: str) -> str:
+    words = str.split(mnemonic)
+    words = [normalize('NFKC', word) for word in words]
+    return str.join(' ', words)
+
+
+@pytest.mark.parametrize(
+    'test_mnemonic',
+    [abbreviate_mnemonic(test_mnemonic[1])
+     for _, language_test_vectors in test_vectors.items()
+     for test_mnemonic in language_test_vectors]
+)
+def test_reconstruct_abbreviated_mnemonic(test_mnemonic: str) -> None:
+    assert reconstruct_mnemonic(test_mnemonic, WORD_LISTS_PATH) is not None
 
 
 @pytest.mark.parametrize(