Update app.py

app.py

@@ -1,9 +1,217 @@
-# test_index.py
+from itertools import permutations, combinations
+import hashlib, hmac, os, time, math, gc
+from multiprocessing import Pool
 from eth_account import Account
+import coincurve
+from eth_utils import keccak, to_checksum_address
 
-Account.enable_unaudited_hdwallet_features()
-phrase = "lesson cushion crane affair federal decline limb spend reunion chief cart zebra"
+# ============================================================
+words = [
+    "lesson", "cushion", "crane", "affair", "federal",
+    "decline", "limb", "spend", "reunion", "chief", "cart", "zebra",
+]
 
-for idx in range(10):
-    acct = Account.from_mnemonic(phrase, account_path=f"m/44'/60'/0'/0/{idx}")
-    print(f"Index {idx}: {acct.address}")
+PHRASE_MODE = "replace"
+USE_BIP39   = True
+BIP39_PASSPHRASE = ""
+HD_DERIVATION = True
+ADDRESS_INDEX = 0
+CHECK_BIP39_CHECKSUM = True
+ADDRESSES_FILE = "12730802_addresses.csv"
+MATCHED_FILE   = f"matched_keys_{PHRASE_MODE}.txt"
+BATCH_SIZE  = 5000
+NUM_WORKERS = 96
+# ============================================================
+
+
+# ---------- secp256k1 ----------
+_SECP256K1_N = 0xFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFEBAAEDCE6AF48A03BBFD25E8CD0364141
+_HARDENED    = 0x80000000
+
+
+def _bip32_derive_child_priv(parent_priv, parent_chain, index):
+    if index >= _HARDENED:
+        data = b'\x00' + parent_priv + index.to_bytes(4, 'big')
+    else:
+        pub = coincurve.PrivateKey(parent_priv).public_key.format(compressed=True)
+        data = pub + index.to_bytes(4, 'big')
+    I = hmac.new(parent_chain, data, hashlib.sha512).digest()
+    child_int = (int.from_bytes(I[:32], 'big') + int.from_bytes(parent_priv, 'big')) % _SECP256K1_N
+    if child_int == 0:
+        raise ValueError("Invalid child key")
+    return child_int.to_bytes(32, 'big'), I[32:]
+
+
+def seed_to_privkey_hd(seed, address_index=0):
+    I = hmac.new(b"Bitcoin seed", seed, hashlib.sha512).digest()
+    key, chain = I[:32], I[32:]
+    path = [44 + _HARDENED, 60 + _HARDENED, _HARDENED, 0, address_index]
+    for idx in path:
+        key, chain = _bip32_derive_child_priv(key, chain, idx)
+    return key
+
+
+# ---------- BIP39 ----------
+try:
+    import bip39 as _bip39_rust
+    _USE_RUST_BIP39 = True
+except ImportError:
+    _bip39_rust = None
+    _USE_RUST_BIP39 = False
+
+_MNEMONIC = None
+
+def _get_mnemonic():
+    global _MNEMONIC
+    if _MNEMONIC is None:
+        from mnemonic import Mnemonic
+        _MNEMONIC = Mnemonic("english")
+    return _MNEMONIC
+
+def is_valid_bip39_fast(phrase):
+    if _USE_RUST_BIP39:
+        try:
+            _bip39_rust.bip39_mnemonic_to_entropy(phrase)
+            return True
+        except Exception:
+            return False
+    return _get_mnemonic().check(phrase)
+
+def phrase_to_seed(phrase):
+    if _USE_RUST_BIP39:
+        return _bip39_rust.bip39_mnemonic_to_seed(phrase, BIP39_PASSPHRASE)
+    return _get_mnemonic().to_seed(phrase, passphrase=BIP39_PASSPHRASE)
+
+
+# ---------- Address file (lowercase) ----------
+def load_addresses(path):
+    s = set()
+    with open(path, "r", encoding="utf-8") as f:
+        for line in f:
+            a = line.strip().strip('"').strip("'").lstrip("\ufeff").lower()
+            if a.startswith("0x"):
+                a = a[2:]
+            if a and len(a) == 40:
+                s.add(a)
+    return frozenset(s)
+
+def normalize(addr):
+    a = addr.strip().lower()
+    if a.startswith("0x"):
+        a = a[2:]
+    return a
+
+
+# ---------- Generator ----------
+def generate_phrases():
+    n = len(words)
+    if PHRASE_MODE == "order":
+        for perm in permutations(words):
+            yield " ".join(perm)
+    elif PHRASE_MODE == "replace":
+        for i in range(n):
+            for w in words:
+                new = words.copy()
+                new[i] = w
+                yield " ".join(new)
+    elif PHRASE_MODE == "subset":
+        for r in range(1, n + 1):
+            for combo in combinations(words, r):
+                yield " ".join(combo)
+
+def count_total():
+    n = len(words)
+    if PHRASE_MODE == "order":   return math.factorial(n)
+    if PHRASE_MODE == "replace": return n * n
+    if PHRASE_MODE == "subset":  return 2**n - 1
+
+
+# ---------- Worker ----------
+_KNOWN = None
+
+def init_worker(known):
+    global _KNOWN
+    _KNOWN = known
+    _get_mnemonic()
+
+def process_batch(batch):
+    hits = []
+    known = _KNOWN
+    m = _get_mnemonic()
+    for phrase in batch:
+        if CHECK_BIP39_CHECKSUM and not is_valid_bip39_fast(phrase):
+            continue
+        try:
+            seed = phrase_to_seed(phrase)
+            if HD_DERIVATION:
+                priv = seed_to_privkey_hd(seed, ADDRESS_INDEX)
+            else:
+                priv = seed[:32]
+            pub = coincurve.PrivateKey(priv).public_key.format(compressed=False)[1:]
+            addr = to_checksum_address(keccak(pub)[-20:])
+        except Exception:
+            continue
+        if normalize(addr) in known:
+            hits.append((phrase, priv.hex(), addr))
+    return hits
+
+def batch_generator():
+    batch = []
+    for phrase in generate_phrases():
+        batch.append(phrase)
+        if len(batch) >= BATCH_SIZE:
+            yield batch
+            batch = []
+    if batch:
+        yield batch
+
+
+# ---------- Main ----------
+def main():
+    if not os.path.exists(ADDRESSES_FILE):
+        raise SystemExit(f"File nahi mili: {ADDRESSES_FILE}")
+
+    total = count_total()
+    print(f"\nMode: {PHRASE_MODE} | Total: {total:,} | Workers: {NUM_WORKERS}")
+    print(f"HD: {HD_DERIVATION} (index {ADDRESS_INDEX})\n")
+
+    known = load_addresses(ADDRESSES_FILE)
+    print(f"Loaded {len(known):,} lowercase addresses\n")
+
+    matches = []
+    done = 0
+    start = time.time()
+    last_log = start
+
+    if os.path.exists(MATCHED_FILE):
+        os.remove(MATCHED_FILE)
+
+    gc.collect()
+    gc.freeze()
+
+    with Pool(processes=NUM_WORKERS,
+              initializer=init_worker,
+              initargs=(known,),
+              maxtasksperchild=1000) as pool:
+        for hits in pool.imap_unordered(process_batch, batch_generator(), chunksize=1):
+            done += BATCH_SIZE
+            for phrase, key_hex, addr in hits:
+                print(f"\n✅ MATCH: {phrase}\n   priv: {key_hex}\n   addr: {addr}\n")
+                matches.append((phrase, key_hex, addr))
+                with open(MATCHED_FILE, "a", encoding="utf-8") as f:
+                    f.write(f"{phrase} | {key_hex} | {addr}\n")
+
+            now = time.time()
+            if now - last_log >= 5:
+                elapsed = now - start
+                rate = done / elapsed if elapsed else 0
+                eta = (total - done) / rate if rate else 0
+                print(f"  ...{done:,}/{total:,} ({100*done/total:.2f}%) ETA {eta/60:.1f}min")
+                last_log = now
+
+    elapsed = time.time() - start
+    print(f"\nDone {elapsed:.1f}s. {len(matches)} match(es).")
+
+
+if __name__ == "__main__":
+    main()