aboutsummaryrefslogtreecommitdiffstats
path: root/testsuite/issues/issue7/vector.vhdl
diff options
context:
space:
mode:
Diffstat (limited to 'testsuite/issues/issue7/vector.vhdl')
0 files changed, 0 insertions, 0 deletions
n70' href='#n70'>70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848
# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.

from __future__ import absolute_import, division, print_function

import binascii
import collections
import math
import re
from contextlib import contextmanager

import pytest

import six

from cryptography.exceptions import UnsupportedAlgorithm

import cryptography_vectors


HashVector = collections.namedtuple("HashVector", ["message", "digest"])
KeyedHashVector = collections.namedtuple(
    "KeyedHashVector", ["message", "digest", "key"]
)


def select_backends(names, backend_list):
    if names is None:
        return backend_list
    split_names = [x.strip() for x in names.split(',')]
    selected_backends = []
    for backend in backend_list:
        if backend.name in split_names:
            selected_backends.append(backend)

    if len(selected_backends) > 0:
        return selected_backends
    else:
        raise ValueError(
            "No backend selected. Tried to select: {0}".format(split_names)
        )


def skip_if_empty(backend_list, required_interfaces):
    if not backend_list:
        pytest.skip(
            "No backends provided supply the interface: {0}".format(
                ", ".join(iface.__name__ for iface in required_interfaces)
            )
        )


def check_backend_support(item):
    supported = item.keywords.get("supported")
    if supported and "backend" in item.funcargs:
        for mark in supported:
            if not mark.kwargs["only_if"](item.funcargs["backend"]):
                pytest.skip("{0} ({1})".format(
                    mark.kwargs["skip_message"], item.funcargs["backend"]
                ))
    elif supported:
        raise ValueError("This mark is only available on methods that take a "
                         "backend")


@contextmanager
def raises_unsupported_algorithm(reason):
    with pytest.raises(UnsupportedAlgorithm) as exc_info:
        yield exc_info

    assert exc_info.value._reason is reason


def load_vectors_from_file(filename, loader, mode="r"):
    with cryptography_vectors.open_vector_file(filename, mode) as vector_file:
        return loader(vector_file)


def load_nist_vectors(vector_data):
    test_data = None
    data = []

    for line in vector_data:
        line = line.strip()

        # Blank lines, comments, and section headers are ignored
        if not line or line.startswith("#") or (line.startswith("[") and
                                                line.endswith("]")):
            continue

        if line.strip() == "FAIL":
            test_data["fail"] = True
            continue

        # Build our data using a simple Key = Value format
        name, value = [c.strip() for c in line.split("=")]

        # Some tests (PBKDF2) contain \0, which should be interpreted as a
        # null character rather than literal.
        value = value.replace("\\0", "\0")

        # COUNT is a special token that indicates a new block of data
        if name.upper() == "COUNT":
            test_data = {}
            data.append(test_data)
            continue
        # For all other tokens we simply want the name, value stored in
        # the dictionary
        else:
            test_data[name.lower()] = value.encode("ascii")

    return data


def load_cryptrec_vectors(vector_data):
    cryptrec_list = []

    for line in vector_data:
        line = line.strip()

        # Blank lines and comments are ignored
        if not line or line.startswith("#"):
            continue

        if line.startswith("K"):
            key = line.split(" : ")[1].replace(" ", "").encode("ascii")
        elif line.startswith("P"):
            pt = line.split(" : ")[1].replace(" ", "").encode("ascii")
        elif line.startswith("C"):
            ct = line.split(" : ")[1].replace(" ", "").encode("ascii")
            # after a C is found the K+P+C tuple is complete
            # there are many P+C pairs for each K
            cryptrec_list.append({
                "key": key,
                "plaintext": pt,
                "ciphertext": ct
            })
        else:
            raise ValueError("Invalid line in file '{}'".format(line))
    return cryptrec_list


def load_hash_vectors(vector_data):
    vectors = []
    key = None
    msg = None
    md = None

    for line in vector_data:
        line = line.strip()

        if not line or line.startswith("#") or line.startswith("["):
            continue

        if line.startswith("Len"):
            length = int(line.split(" = ")[1])
        elif line.startswith("Key"):
            # HMAC vectors contain a key attribute. Hash vectors do not.
            key = line.split(" = ")[1].encode("ascii")
        elif line.startswith("Msg"):
            # In the NIST vectors they have chosen to represent an empty
            # string as hex 00, which is of course not actually an empty
            # string. So we parse the provided length and catch this edge case.
            msg = line.split(" = ")[1].encode("ascii") if length > 0 else b""
        elif line.startswith("MD"):
            md = line.split(" = ")[1]
            # after MD is found the Msg+MD (+ potential key) tuple is complete
            if key is not None:
                vectors.append(KeyedHashVector(msg, md, key))
                key = None
                msg = None
                md = None
            else:
                vectors.append(HashVector(msg, md))
                msg = None
                md = None
        else:
            raise ValueError("Unknown line in hash vector")
    return vectors


def load_pkcs1_vectors(vector_data):
    """
    Loads data out of RSA PKCS #1 vector files.
    """
    private_key_vector = None
    public_key_vector = None
    attr = None
    key = None
    example_vector = None
    examples = []
    vectors = []
    for line in vector_data:
        if (
            line.startswith("# PSS Example") or
            line.startswith("# OAEP Example") or
            line.startswith("# PKCS#1 v1.5")
        ):
            if example_vector:
                for key, value in six.iteritems(example_vector):
                    hex_str = "".join(value).replace(" ", "").encode("ascii")
                    example_vector[key] = hex_str
                examples.append(example_vector)

            attr = None
            example_vector = collections.defaultdict(list)

        if line.startswith("# Message"):
            attr = "message"
            continue
        elif line.startswith("# Salt"):
            attr = "salt"
            continue
        elif line.startswith("# Seed"):
            attr = "seed"
            continue
        elif line.startswith("# Signature"):
            attr = "signature"
            continue
        elif line.startswith("# Encryption"):
            attr = "encryption"
            continue
        elif (
            example_vector and
            line.startswith("# =============================================")
        ):
            for key, value in six.iteritems(example_vector):
                hex_str = "".join(value).replace(" ", "").encode("ascii")
                example_vector[key] = hex_str
            examples.append(example_vector)
            example_vector = None
            attr = None
        elif example_vector and line.startswith("#"):
            continue
        else:
            if attr is not None and example_vector is not None:
                example_vector[attr].append(line.strip())
                continue

        if (
            line.startswith("# Example") or
            line.startswith("# =============================================")
        ):
            if key:
                assert private_key_vector
                assert public_key_vector

                for key, value in six.iteritems(public_key_vector):
                    hex_str = "".join(value).replace(" ", "")
                    public_key_vector[key] = int(hex_str, 16)

                for key, value in six.iteritems(private_key_vector):
                    hex_str = "".join(value).replace(" ", "")
                    private_key_vector[key] = int(hex_str, 16)

                private_key_vector["examples"] = examples
                examples = []

                assert (
                    private_key_vector['public_exponent'] ==
                    public_key_vector['public_exponent']
                )

                assert (
                    private_key_vector['modulus'] ==
                    public_key_vector['modulus']
                )

                vectors.append(
                    (private_key_vector, public_key_vector)
                )

            public_key_vector = collections.defaultdict(list)
            private_key_vector = collections.defaultdict(list)
            key = None
            attr = None

        if private_key_vector is None or public_key_vector is None:
            continue

        if line.startswith("# Private key"):
            key = private_key_vector
        elif line.startswith("# Public key"):
            key = public_key_vector
        elif line.startswith("# Modulus:"):
            attr = "modulus"
        elif line.startswith("# Public exponent:"):
            attr = "public_exponent"
        elif line.startswith("# Exponent:"):
            if key is public_key_vector:
                attr = "public_exponent"
            else:
                assert key is private_key_vector
                attr = "private_exponent"
        elif line.startswith("# Prime 1:"):
            attr = "p"
        elif line.startswith("# Prime 2:"):
            attr = "q"
        elif line.startswith("# Prime exponent 1:"):
            attr = "dmp1"
        elif line.startswith("# Prime exponent 2:"):
            attr = "dmq1"
        elif line.startswith("# Coefficient:"):
            attr = "iqmp"
        elif line.startswith("#"):
            attr = None
        else:
            if key is not None and attr is not None:
                key[attr].append(line.strip())
    return vectors


def load_rsa_nist_vectors(vector_data):
    test_data = None
    p = None
    salt_length = None
    data = []

    for line in vector_data:
        line = line.strip()

        # Blank lines and section headers are ignored
        if not line or line.startswith("["):
            continue

        if line.startswith("# Salt len:"):
            salt_length = int(line.split(":")[1].strip())
            continue
        elif line.startswith("#"):
            continue

        # Build our data using a simple Key = Value format
        name, value = [c.strip() for c in line.split("=")]

        if name == "n":
            n = int(value, 16)
        elif name == "e" and p is None:
            e = int(value, 16)
        elif name == "p":
            p = int(value, 16)
        elif name == "q":
            q = int(value, 16)
        elif name == "SHAAlg":
            if p is None:
                test_data = {
                    "modulus": n,
                    "public_exponent": e,
                    "salt_length": salt_length,
                    "algorithm": value,
                    "fail": False
                }
            else:
                test_data = {
                    "modulus": n,
                    "p": p,
                    "q": q,
                    "algorithm": value
                }
                if salt_length is not None:
                    test_data["salt_length"] = salt_length
            data.append(test_data)
        elif name == "e" and p is not None:
            test_data["public_exponent"] = int(value, 16)
        elif name == "d":
            test_data["private_exponent"] = int(value, 16)
        elif name == "Result":
            test_data["fail"] = value.startswith("F")
        # For all other tokens we simply want the name, value stored in
        # the dictionary
        else:
            test_data[name.lower()] = value.encode("ascii")

    return data


def load_fips_dsa_key_pair_vectors(vector_data):
    """
    Loads data out of the FIPS DSA KeyPair vector files.
    """
    vectors = []
    # When reading_key_data is set to True it tells the loader to continue
    # constructing dictionaries. We set reading_key_data to False during the
    # blocks of the vectors of N=224 because we don't support it.
    reading_key_data = True
    for line in vector_data:
        line = line.strip()

        if not line or line.startswith("#"):
            continue
        elif line.startswith("[mod = L=1024"):
            continue
        elif line.startswith("[mod = L=2048, N=224"):
            reading_key_data = False
            continue
        elif line.startswith("[mod = L=2048, N=256"):
            reading_key_data = True
            continue
        elif line.startswith("[mod = L=3072"):
            continue

        if reading_key_data:
            if line.startswith("P"):
                vectors.append({'p': int(line.split("=")[1], 16)})
            elif line.startswith("Q"):
                vectors[-1]['q'] = int(line.split("=")[1], 16)
            elif line.startswith("G"):
                vectors[-1]['g'] = int(line.split("=")[1], 16)
            elif line.startswith("X") and 'x' not in vectors[-1]:
                vectors[-1]['x'] = int(line.split("=")[1], 16)
            elif line.startswith("X") and 'x' in vectors[-1]:
                vectors.append({'p': vectors[-1]['p'],
                                'q': vectors[-1]['q'],
                                'g': vectors[-1]['g'],
                                'x': int(line.split("=")[1], 16)
                                })
            elif line.startswith("Y"):
                vectors[-1]['y'] = int(line.split("=")[1], 16)

    return vectors


def load_fips_dsa_sig_vectors(vector_data):
    """
    Loads data out of the FIPS DSA SigVer vector files.
    """
    vectors = []
    sha_regex = re.compile(
        r"\[mod = L=...., N=..., SHA-(?P<sha>1|224|256|384|512)\]"
    )
    # When reading_key_data is set to True it tells the loader to continue
    # constructing dictionaries. We set reading_key_data to False during the
    # blocks of the vectors of N=224 because we don't support it.
    reading_key_data = True

    for line in vector_data:
        line = line.strip()

        if not line or line.startswith("#"):
            continue

        sha_match = sha_regex.match(line)
        if sha_match:
            digest_algorithm = "SHA-{0}".format(sha_match.group("sha"))

        if line.startswith("[mod = L=2048, N=224"):
            reading_key_data = False
            continue
        elif line.startswith("[mod = L=2048, N=256"):
            reading_key_data = True
            continue

        if not reading_key_data or line.startswith("[mod"):
            continue

        name, value = [c.strip() for c in line.split("=")]

        if name == "P":
            vectors.append({'p': int(value, 16),
                            'digest_algorithm': digest_algorithm})
        elif name == "Q":
            vectors[-1]['q'] = int(value, 16)
        elif name == "G":
            vectors[-1]['g'] = int(value, 16)
        elif name == "Msg" and 'msg' not in vectors[-1]:
            hexmsg = value.strip().encode("ascii")
            vectors[-1]['msg'] = binascii.unhexlify(hexmsg)
        elif name == "Msg" and 'msg' in vectors[-1]:
            hexmsg = value.strip().encode("ascii")
            vectors.append({'p': vectors[-1]['p'],
                            'q': vectors[-1]['q'],
                            'g': vectors[-1]['g'],
                            'digest_algorithm':
                            vectors[-1]['digest_algorithm'],
                            'msg': binascii.unhexlify(hexmsg)})
        elif name == "X":
            vectors[-1]['x'] = int(value, 16)
        elif name == "Y":
            vectors[-1]['y'] = int(value, 16)
        elif name == "R":
            vectors[-1]['r'] = int(value, 16)
        elif name == "S":
            vectors[-1]['s'] = int(value, 16)
        elif name == "Result":
            vectors[-1]['result'] = value.split("(")[0].strip()

    return vectors


# http://tools.ietf.org/html/rfc4492#appendix-A
_ECDSA_CURVE_NAMES = {
    "P-192": "secp192r1",
    "P-224": "secp224r1",
    "P-256": "secp256r1",
    "P-384": "secp384r1",
    "P-521": "secp521r1",

    "K-163": "sect163k1",
    "K-233": "sect233k1",
    "K-256": "secp256k1",
    "K-283": "sect283k1",
    "K-409": "sect409k1",
    "K-571": "sect571k1",

    "B-163": "sect163r2",
    "B-233": "sect233r1",
    "B-283": "sect283r1",
    "B-409": "sect409r1",
    "B-571": "sect571r1",
}


def load_fips_ecdsa_key_pair_vectors(vector_data):
    """
    Loads data out of the FIPS ECDSA KeyPair vector files.
    """
    vectors = []
    key_data = None
    for line in vector_data:
        line = line.strip()

        if not line or line.startswith("#"):
            continue

        if line[1:-1] in _ECDSA_CURVE_NAMES:
            curve_name = _ECDSA_CURVE_NAMES[line[1:-1]]

        elif line.startswith("d = "):
            if key_data is not None:
                vectors.append(key_data)

            key_data = {
                "curve": curve_name,
                "d": int(line.split("=")[1], 16)
            }

        elif key_data is not None:
            if line.startswith("Qx = "):
                key_data["x"] = int(line.split("=")[1], 16)
            elif line.startswith("Qy = "):
                key_data["y"] = int(line.split("=")[1], 16)

    assert key_data is not None
    vectors.append(key_data)

    return vectors


def load_fips_ecdsa_signing_vectors(vector_data):
    """
    Loads data out of the FIPS ECDSA SigGen vector files.
    """
    vectors = []

    curve_rx = re.compile(
        r"\[(?P<curve>[PKB]-[0-9]{3}),SHA-(?P<sha>1|224|256|384|512)\]"
    )

    data = None
    for line in vector_data:
        line = line.strip()

        curve_match = curve_rx.match(line)
        if curve_match:
            curve_name = _ECDSA_CURVE_NAMES[curve_match.group("curve")]
            digest_name = "SHA-{0}".format(curve_match.group("sha"))

        elif line.startswith("Msg = "):
            if data is not None:
                vectors.append(data)

            hexmsg = line.split("=")[1].strip().encode("ascii")

            data = {
                "curve": curve_name,
                "digest_algorithm": digest_name,
                "message": binascii.unhexlify(hexmsg)
            }

        elif data is not None:
            if line.startswith("Qx = "):
                data["x"] = int(line.split("=")[1], 16)
            elif line.startswith("Qy = "):
                data["y"] = int(line.split("=")[1], 16)
            elif line.startswith("R = "):
                data["r"] = int(line.split("=")[1], 16)
            elif line.startswith("S = "):
                data["s"] = int(line.split("=")[1], 16)
            elif line.startswith("d = "):
                data["d"] = int(line.split("=")[1], 16)
            elif line.startswith("Result = "):
                data["fail"] = line.split("=")[1].strip()[0] == "F"

    assert data is not None
    vectors.append(data)
    return vectors


def load_kasvs_dh_vectors(vector_data):
    """
    Loads data out of the KASVS key exchange vector data
    """

    result_rx = re.compile(r"([FP]) \(([0-9]+) -")

    vectors = []
    data = {
        "fail_z": False,
        "fail_agree": False
    }

    for line in vector_data:
        line = line.strip()

        if not line or line.startswith("#"):
            continue

        if line.startswith("P = "):
            data["p"] = int(line.split("=")[1], 16)
        elif line.startswith("Q = "):
            data["q"] = int(line.split("=")[1], 16)
        elif line.startswith("G = "):
            data["g"] = int(line.split("=")[1], 16)
        elif line.startswith("Z = "):
            z_hex = line.split("=")[1].strip().encode("ascii")
            data["z"] = binascii.unhexlify(z_hex)
        elif line.startswith("XstatCAVS = "):
            data["x1"] = int(line.split("=")[1], 16)
        elif line.startswith("YstatCAVS = "):
            data["y1"] = int(line.split("=")[1], 16)
        elif line.startswith("XstatIUT = "):
            data["x2"] = int(line.split("=")[1], 16)
        elif line.startswith("YstatIUT = "):
            data["y2"] = int(line.split("=")[1], 16)
        elif line.startswith("Result = "):
            result_str = line.split("=")[1].strip()
            match = result_rx.match(result_str)

            if match.group(1) == "F":
                if int(match.group(2)) in (5, 10):
                    data["fail_z"] = True
                else:
                    data["fail_agree"] = True

            vectors.append(data)

            data = {
                "p": data["p"],
                "q": data["q"],
                "g": data["g"],
                "fail_z": False,
                "fail_agree": False
            }

    return vectors


def load_kasvs_ecdh_vectors(vector_data):
    """
    Loads data out of the KASVS key exchange vector data
    """

    curve_name_map = {
        "P-192": "secp192r1",
        "P-224": "secp224r1",
        "P-256": "secp256r1",
        "P-384": "secp384r1",
        "P-521": "secp521r1",
    }

    result_rx = re.compile(r"([FP]) \(([0-9]+) -")

    tags = []
    sets = {}
    vectors = []

    # find info in header
    for line in vector_data:
        line = line.strip()

        if line.startswith("#"):
            parm = line.split("Parameter set(s) supported:")
            if len(parm) == 2:
                names = parm[1].strip().split()