BIP-85 CLI supports different BIP-39 languages, 0.6.0 (#33)

* longer timeout (for ci?) and fix message
* add whitespace per flake8 only on 3.12?
* lint on publish

.github/workflows/ci.yaml

@@ -19,8 +19,8 @@ jobs:
         python-version: ${{ matrix.python-version }}
     - name: install
       run: |
-        make install
         make install-dev
+        make install-local
     - name: check
       run: make check
     - name: test

Makefile

@@ -1,7 +1,7 @@
 .PHONY: all build check clean git-no-unsaved git-on-main got-off-main install install-dev
-.PHONY: install-go lint publish push readme-cmds test test-network uninstall-dev
+.PHONY: install-go install-local lint publish push readme-cmds test uninstall
 
-build: clean download-wordlists check test
+build: clean download-wordlists
 	python3 -m build
 
 clean:
@@ -11,39 +11,45 @@ clean:
 check:
 	black . --check
 	isort . --check
+	flake8 . --ignore=E501,W503
 
 install:
-	pip install -r requirements.txt -r tst-requirements.txt
+	pip install -U bipsea
 
-install-dev: uninstall-dev
+install-local:
 	pip install -e .
 
+install-dev:
+	pip install -r requirements.txt -r tst-requirements.txt
+
 install-go:
 	# you must have go installed https://go.dev/doc/install	
 	go install github.com/rhysd/actionlint/cmd/actionlint@latest
 	go install github.com/mrtazz/checkmake/cmd/checkmake@latest
 
-uninstall-dev:
+uninstall:
 	pip uninstall -y bipsea
 
 lint:
 	isort .
 	black .
-	flake8 . --ignore=E501,W503
 	actionlint
+	flake8 . --ignore=E501,W503
 	checkmake Makefile
 
-publish: build git-no-unsaved git-on-main
+publish: install-local lint test readme-cmds build git-no-unsaved git-on-main
 	git pull origin main
 	python3 -m twine upload dist/*
 
-push: lint check test git-off-main git-no-unsaved
+push: lint test git-off-main git-no-unsaved
 	@branch=$$(git symbolic-ref --short HEAD); \
 	git push origin $$branch
 
 test: readme-cmds
 	pytest -sx
 
+test-publish: uninstall install readme-cmds
+
 git-off-main:
 	@branch=$$(git symbolic-ref --short HEAD); \
 	if [ "$$branch" = "main" ]; then \
@@ -72,7 +78,7 @@ readme-cmds:
 	bipsea seed -t jpn -n 15
 	bipsea seed -f eng -u "airport letter idea forget broccoli prefer panda food delay struggle ridge salute above want dinner"
 	bipsea seed -f any -u "123456123456123456"
-	bipsea seed -f any -u "$$(cat README.md)"
+	bipsea seed -f any -u "$$(cat input.txt)"
 	bipsea seed | bipsea entropy
 	bipsea seed -f eng -u "load kitchen smooth mass blood happy kidney orbit used process lady sudden" | bipsea entropy -n 12
 	bipsea seed -f eng -u "load kitchen smooth mass blood happy kidney orbit used process lady sudden" | bipsea entropy -n 12 -i 1

README.md

@@ -170,7 +170,7 @@ bipsea seed -f any -u "123456123456123456"
 You can even load the input from a file.
 
 ```
-bipsea seed -f any -u "$$(cat README.md)"
+bipsea seed -f any -u "$$(cat input.txt)"
 ```
 
 If you are now thinking, _I could use any string to derive a master key_,

input.txt

@@ -0,0 +1 @@
+Satoshi Nakamoto published the Bitcoin white paper on October 31, 2008.

src/bipsea/bip39.py

@@ -91,8 +91,7 @@
 
 
 def entropy_to_words(n_words: int, user_entropy: bytes, language: str):
-    """If caller does not provide entropy use secrets.randbits
-    * Only produces seed words in English"""
+    """If caller does not provide entropy use secrets.randbits"""
     if n_words not in N_WORDS_ALLOWED:
         raise ValueError(f"n_words must be one of {N_WORDS_ALLOWED}")
 
@@ -110,7 +109,7 @@ def entropy_to_words(n_words: int, user_entropy: bytes, language: str):
     elif user_entropy and (difference <= -8):
         warnings.warn(
             (
-                f"Warning: {difference + n_entropy_bits} bits in, {n_entropy_bits} bits out."
+                f"{difference + n_entropy_bits} bits in, {n_entropy_bits} bits out."
                 " Input more entropy?"
             )
         )

src/bipsea/bip85.py

@@ -71,7 +71,7 @@ def apply_85(derived_key: ExtendedKey, path: str) -> Dict[str, Union[bytes, str]
 
     if app == APPLICATIONS["words"]:
         language_index, n_words = indexes[:2]
-        n_words = int(n_words[:-1])  # chop the ' from hardened derivation
+        n_words = int(n_words[:-1])  # chop ' from hardened derivation
         if n_words not in N_WORDS_META.keys():
             raise ValueError(f"Unsupported number of words: {n_words}.")
         language = INDEX_TO_LANGUAGE[language_index]

src/bipsea/bipsea.py

@@ -21,6 +21,7 @@
 from .bip85 import (
     APPLICATIONS,
     DRNG,
+    INDEX_TO_LANGUAGE,
     PURPOSE_CODES,
     RANGES,
     apply_85,
@@ -36,24 +37,26 @@
     to_hex_string,
 )
 
-CODE_TO_LANG = {v["code"]: k for k, v in LANGUAGES.items()}
+ISO_TO_LANGUAGE = {v["code"]: k for k, v in LANGUAGES.items()}
 
 
 SEED_FROM_VALUES = [
     "any",
     "rand",
-] + list(CODE_TO_LANG.keys())
+] + list(ISO_TO_LANGUAGE.keys())
 
 
 SEED_TO_VALUES = [
     "tprv",
     "xprv",
-] + [code for code in CODE_TO_LANG.keys()]
+] + list(ISO_TO_LANGUAGE.keys())
 
-TIMEOUT = 0.1
+
+ENTROPY_TO_VALUES = list(ISO_TO_LANGUAGE.keys())
 
 N_WORDS_ALLOWED_STR = [str(n) for n in N_WORDS_ALLOWED]
-N_WORDS_ALLOWED_HELP = "|".join(N_WORDS_ALLOWED_STR)
+
+TIMEOUT = 0.1
 
 
 logger = logging.getLogger(LOGGER)
@@ -70,14 +73,14 @@ def cli():
     "-f",
     "--from",
     "from_",
-    type=click.Choice(SEED_FROM_VALUES, case_sensitive=True),
+    type=click.Choice(SEED_FROM_VALUES),
     help="Input format.",
     default="rand",
 )
 @click.option(
     "-t",
     "--to",
-    type=click.Choice(SEED_TO_VALUES, case_sensitive=True),
+    type=click.Choice(SEED_TO_VALUES),
     default="xprv",
     help="Output format.",
 )
@@ -104,7 +107,7 @@ def bip39_cmd(from_, to, input, number, passphrase, pretty):
             option_name="--from",
             message="`--from words` requires `--input STRING`, `--from rand` forbids `--input`",
         )
-    language = CODE_TO_LANG.get(from_)
+    language = ISO_TO_LANGUAGE.get(from_)
     if language or from_ == "any":
         if to == "words":
             raise click.BadOptionUsage(
@@ -130,11 +133,11 @@ def bip39_cmd(from_, to, input, number, passphrase, pretty):
                 )
     else:  # from_ == rand
         entropy = None
-        words = entropy_to_words(number, entropy, CODE_TO_LANG.get(to, "english"))
+        words = entropy_to_words(number, entropy, ISO_TO_LANGUAGE.get(to, "english"))
 
-    if to in CODE_TO_LANG.keys():
+    if to in ISO_TO_LANGUAGE.keys():
         if pretty:
-            output = "\n".join(f"{i+1}) {w}" for i, w in enumerate(words))
+            output = "\n".join(f"{i + 1}) {w}" for i, w in enumerate(words))
         else:
             output = " ".join(words)
 
@@ -167,7 +170,7 @@ def bip39_cmd(from_, to, input, number, passphrase, pretty):
     "--application",
     default="words",
     required=True,
-    type=click.Choice(APPLICATIONS.keys(), case_sensitive=True),
+    type=click.Choice(APPLICATIONS.keys()),
 )
 @click.option(
     "-n",
@@ -192,9 +195,15 @@ def bip39_cmd(from_, to, input, number, passphrase, pretty):
 @click.option(
     "-u",
     "--input",
-    help="`--input xprv123...` can be used instead of an input pipe `bipsea seed | bipsea entropy`",
+    help="alternative to a unix input pipe",
 )
-def bip85_cmd(application, number, index, special, input):
+@click.option(
+    "-t",
+    "--to",
+    type=click.Choice(ENTROPY_TO_VALUES),
+    help="output language",
+)
+def bip85_cmd(application, number, index, special, input, to):
     if not input:
         stdin, _, _ = select.select([sys.stdin], [], [], TIMEOUT)
         if stdin:
@@ -222,20 +231,29 @@ def bip85_cmd(application, number, index, special, input):
             )
     else:
         number = 24
+
     if not prv[:4] in ("tprv", "xprv"):
         no_prv()
+
     master = parse_ext_key(prv)
 
     path = f"m/{PURPOSE_CODES['BIP-85']}"
     app_code = APPLICATIONS[application]
     path += f"/{app_code}"
-    if application == "words":
-        if number not in N_WORDS_ALLOWED:
+
+    if to:
+        if application != "words":
             raise click.BadOptionUsage(
-                option_name="--number",
-                message=f"`--application wif` requires `--number NUMBER` in {N_WORDS_ALLOWED_HELP}",
+                option_name="--to",
+                message="--to requires `--application words`",
             )
-        path += f"/0'/{number}'/{index}'"
+    else:
+        to = "eng"
+
+    if application == "words":
+        language = ISO_TO_LANGUAGE[to]
+        code_85 = next(i for i, l in INDEX_TO_LANGUAGE.items() if l == language)
+        path += f"/{code_85}/{number}'/{index}'"
     elif application in ("wif", "xprv"):
         path += f"/{index}'"
     elif application in ("base64", "base85", "hex"):
@@ -275,8 +293,8 @@ def check_range(number: int, application: str):
 
 def no_prv():
     raise click.BadOptionUsage(
-        option_name="[incoming pipe]",
-        message="Bad input. Need xprv or tprv. Try `bipsea seed` | bipsea entropy`",
+        option_name="--input",
+        message="Missing xprv or tprv from pipe or --input. Try `bipsea seed | bipsea entropy`",
     )
 
 

src/bipsea/util.py

@@ -9,7 +9,7 @@
 from collections import Counter
 from typing import List, Sequence
 
-__version__ = "0.5.0"
+__version__ = "0.6.0"
 __app_name__ = "bipsea"
 
 LOGGER = __app_name__
@@ -40,18 +40,17 @@ def shannon_entropy(input: List[str]) -> float:
 
 def relative_entropy(input: Sequence, universe: set = ASCII_INPUTS) -> float:
     input_set = set(list(input))
-    warn = False
-    if not input_set <= universe:
-        warnings.warn(
-            f"Unexpected input characters {input_set - universe}, expect bad entropy"
-        )
-        warn = True
+    overage = input_set - universe
 
     ideal = math.log(len(universe), 2)
     actual = shannon_entropy(input)
     ratio = actual / ideal
 
-    if not warn:
+    if overage:
+        warnings.warn(
+            f"Some inputs outside (ASCII) universe: {overage}, can't estimate entropy"
+        )
+    else:
         assert 0 < ratio <= 1.001
 
     return ratio

tests/test_cli.py

@@ -7,7 +7,7 @@
 from data.bip85_vectors import BIP_39, HEX, PWD_BASE85, WIF
 
 from bipsea.bip39 import LANGUAGES, verify_seed_words
-from bipsea.bipsea import CODE_TO_LANG, N_WORDS_ALLOWED, cli
+from bipsea.bipsea import ISO_TO_LANGUAGE, N_WORDS_ALLOWED, cli
 from bipsea.util import ASCII_INPUTS, LOGGER
 
 logger = logging.getLogger(LOGGER)
@@ -92,7 +92,7 @@ def test_seed_command_from_rand(runner, n, code):
         assert len(words) == int(n)
         assert result.exit_code == 0
         if style != "--pretty":
-            assert verify_seed_words(words, CODE_TO_LANG[code])
+            assert verify_seed_words(words, ISO_TO_LANGUAGE[code])
 
 
 def test_seed_command_from_custom_words(runner):
@@ -125,7 +125,7 @@ def test_seed_from_and_to_words(runner):
     assert "--from rand" in result.output
 
 
-@pytest.mark.parametrize("n", [-1, 11, 13, 0])
+@pytest.mark.parametrize("n", [-1, 11, 13, 0, 25])
 def test_seed_bad_n(runner, n):
     result = runner.invoke(cli, ["seed", "--from", "eng", "-n", n])
     assert result.exit_code != 0