X-Git-Url: http://www.git.cypherpunks.ru/?a=blobdiff_plain;f=pyderasn.py;h=9c4da4ec76b94b3561581453e2b981fa18e14ad3;hb=d9166320d60c60c2f188464c83c82547e6c4d037;hp=3afaf60fde261bd6a985d06acd63ec4ca0e55fbd;hpb=88244e6e055038f5d09f9cff63dd9507c837f123;p=pyderasn.git diff --git a/pyderasn.py b/pyderasn.py index 3afaf60..9c4da4e 100755 --- a/pyderasn.py +++ b/pyderasn.py @@ -1,7 +1,7 @@ #!/usr/bin/env python # coding: utf-8 -# PyDERASN -- Python ASN.1 DER codec with abstract structures -# Copyright (C) 2017 Sergey Matveev +# PyDERASN -- Python ASN.1 DER/BER codec with abstract structures +# Copyright (C) 2017-2018 Sergey Matveev # # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU Lesser General Public License as @@ -16,10 +16,10 @@ # You should have received a copy of the GNU Lesser General Public # License along with this program. If not, see # . -"""Python ASN.1 DER codec with abstract structures +"""Python ASN.1 DER/BER codec with abstract structures -This library allows you to marshal and unmarshal various structures in -ASN.1 DER format, like this: +This library allows you to marshal various structures in ASN.1 DER +format, unmarshal them in BER/CER/DER ones. >>> i = Integer(123) >>> raw = i.encode() @@ -68,7 +68,7 @@ ____ Most types in ASN.1 has specific tag for them. ``Obj.tag_default`` is the default tag used during coding process. You can override it with either ``IMPLICIT`` (using ``impl`` keyword argument), or -``EXPLICIT`` one (using ``expl`` keyword argument). Both arguments takes +``EXPLICIT`` one (using ``expl`` keyword argument). Both arguments take raw binary string, containing that tag. You can **not** set implicit and explicit tags simultaneously. @@ -88,10 +88,10 @@ number. Pay attention that explicit tags always have *constructed* tag Implicit tag is not explicitly shown. -Two object of the same type, but with different implicit/explicit tags +Two objects of the same type, but with different implicit/explicit tags are **not** equal. -You can get objects effective tag (either default or implicited) through +You can get object's effective tag (either default or implicited) through ``tag`` property. You can decode it using :py:func:`pyderasn.tag_decode` function:: @@ -159,12 +159,12 @@ raised. Common methods ______________ -All objects have ``ready`` boolean property, that tells if it is ready -to be encoded. If that kind of action is performed on unready object, -then :py:exc:`pyderasn.ObjNotReady` exception will be raised. +All objects have ``ready`` boolean property, that tells if object is +ready to be encoded. If that kind of action is performed on unready +object, then :py:exc:`pyderasn.ObjNotReady` exception will be raised. -All objects have ``copy()`` method, returning its copy, that can be safely -mutated. +All objects have ``copy()`` method, that returns their copy, that can be +safely mutated. .. _decoding: @@ -189,11 +189,35 @@ use following properties: Pay attention that those values do **not** include anything related to explicit tag. If you want to know information about it, then use: -``expled`` (to know if explicit tag is set), ``expl_offset`` (it is -lesser than ``offset``), ``expl_tlen``, ``expl_llen``, ``expl_vlen`` -(that actually equals to ordinary ``tlvlen``). -When error occurs, then :py:exc:`pyderasn.DecodeError` is raised. +* ``expled`` -- to know if explicit tag is set +* ``expl_offset`` (it is lesser than ``offset``) +* ``expl_tlen``, +* ``expl_llen`` +* ``expl_vlen`` (that actually equals to ordinary ``tlvlen``) +* ``fulloffset`` -- it equals to ``expl_offset`` if explicit tag is set, + ``offset`` otherwise +* ``fulllen`` -- it equals to ``expl_len`` if explicit tag is set, + ``tlvlen`` otherwise + +When error occurs, :py:exc:`pyderasn.DecodeError` is raised. + +.. _ctx: + +Context +_______ + +You can specify so called context keyword argument during ``decode()`` +invocation. It is dictionary containing various options governing +decoding process. + +Currently available context options: + +* :ref:`allow_default_values ` +* :ref:`allow_expl_oob ` +* :ref:`allow_unordered_set ` +* :ref:`bered ` +* :ref:`defines_by_path ` .. _pprinting: @@ -230,15 +254,15 @@ _____________ :py:class:`pyderasn.ObjectIdentifier` field inside :py:class:`pyderasn.Sequence` can hold mapping between OIDs and -necessary for decoding structrures. For example, CMS (:rfc:`5652`) +necessary for decoding structures. For example, CMS (:rfc:`5652`) container:: class ContentInfo(Sequence): schema = ( - ("contentType", ContentType(defines=("content", { + ("contentType", ContentType(defines=((("content",), { id_digestedData: DigestedData(), id_signedData: SignedData(), - }))), + }),))), ("content", Any(expl=tag_ctxc(0))), ) @@ -248,30 +272,51 @@ decoded with ``SignedData`` specification, if ``contentType`` equals to ``contentType`` contains unknown OID, then no automatic decoding is done. +You can specify multiple fields, that will be autodecoded -- that is why +``defines`` kwarg is a sequence. You can specify defined field +relatively or absolutely to current decode path. For example ``defines`` +for AlgorithmIdentifier of X.509's +``tbsCertificate:subjectPublicKeyInfo:algorithm:algorithm``:: + + ( + (("parameters",), { + id_ecPublicKey: ECParameters(), + id_GostR3410_2001: GostR34102001PublicKeyParameters(), + }), + (("..", "subjectPublicKey"), { + id_rsaEncryption: RSAPublicKey(), + id_GostR3410_2001: OctetString(), + }), + ), + +tells that if certificate's SPKI algorithm is GOST R 34.10-2001, then +autodecode its parameters inside SPKI's algorithm and its public key +itself. + Following types can be automatically decoded (DEFINED BY): * :py:class:`pyderasn.Any` +* :py:class:`pyderasn.BitString` (that is multiple of 8 bits) * :py:class:`pyderasn.OctetString` * :py:class:`pyderasn.SequenceOf`/:py:class:`pyderasn.SetOf` - ``Any``/``OctetString``-s + ``Any``/``BitString``/``OctetString``-s When any of those fields is automatically decoded, then ``.defined`` -attribute contains ``(OID, value)`` tuple. OID tell by which OID it was -defined, ``value`` contains corresponding decoded value. For example -above, ``content_info["content"].defined == (id_signedData, -signed_data)``. +attribute contains ``(OID, value)`` tuple. ``OID`` tells by which OID it +was defined, ``value`` contains corresponding decoded value. For example +above, ``content_info["content"].defined == (id_signedData, signed_data)``. -.. _defines_by_path_kwarg: +.. _defines_by_path_ctx: -defines_by_path kwarg -_____________________ +defines_by_path context option +______________________________ Sometimes you either can not or do not want to explicitly set *defines* in the scheme. You can dynamically apply those definitions when calling ``.decode()`` method. -Decode method takes optional ``defines_by_path`` keyword argument that -must be sequence of following tuples:: +Specify ``defines_by_path`` key in the :ref:`decode context `. Its +value must be sequence of following tuples:: (decode_path, defines) @@ -288,44 +333,89 @@ of ``PKIResponse``:: content_info, tail = ContentInfo().decode(data, defines_by_path=( ( ("contentType",), - ("content", {id_signedData: SignedData()}), + ((("content",), {id_signedData: SignedData()}),), ), ( ( "content", - decode_path_defby(id_signedData), + DecodePathDefBy(id_signedData), "encapContentInfo", "eContentType", ), - ("eContent", { + ((("eContent",), { id_cct_PKIData: PKIData(), id_cct_PKIResponse: PKIResponse(), - }), + })), ), ( ( "content", - decode_path_defby(id_signedData), + DecodePathDefBy(id_signedData), "encapContentInfo", "eContent", - decode_path_defby(id_cct_PKIResponse), + DecodePathDefBy(id_cct_PKIResponse), "controlSequence", any, "attrType", ), - ("attrValues", { + ((("attrValues",), { id_cmc_recipientNonce: RecipientNonce(), id_cmc_senderNonce: SenderNonce(), id_cmc_statusInfoV2: CMCStatusInfoV2(), id_cmc_transactionId: TransactionId(), - }), + })), ), )) -Pay attention for :py:func:`pyderasn.decode_path_defby` and ``any``. +Pay attention for :py:class:`pyderasn.DecodePathDefBy` and ``any``. First function is useful for path construction when some automatic -decoding is already done. ``any`` is used for human readability and -means literally any value it meet -- useful for sequence and set of-s. +decoding is already done. ``any`` means literally any value it meet -- +useful for SEQUENCE/SET OF-s. + +.. _bered_ctx: + +BER encoding +------------ + +By default PyDERASN accepts only DER encoded data. It always encodes to +DER. But you can optionally enable BER decoding with setting ``bered`` +:ref:`context ` argument to True. Indefinite lengths and +constructed primitive types should be parsed successfully. + +* If object is encoded in BER form (not the DER one), then ``ber_encoded`` + attribute is set to True. Only ``BOOLEAN``, ``BIT STRING``, ``OCTET + STRING``, ``SEQUENCE``, ``SET``, ``SET OF`` can contain it. +* If object has an indefinite length encoding, then its ``lenindef`` + attribute is set to True. Only ``BIT STRING``, ``OCTET STRING``, + ``SEQUENCE``, ``SET``, ``SEQUENCE OF``, ``SET OF``, ``ANY`` can + contain it. +* If object has an indefinite length encoded explicit tag, then + ``expl_lenindef`` is set to True. +* If object has either any of BER-related encoding (explicit tag + indefinite length, object's indefinite length, BER-encoding) or any + underlying component has that kind of encoding, then ``bered`` + attribute is set to True. For example SignedData CMS can have + ``ContentInfo:content:signerInfos:*`` ``bered`` value set to True, but + ``ContentInfo:content:signerInfos:*:signedAttrs`` won't. + +EOC (end-of-contents) token's length is taken in advance in object's +value length. + +.. _allow_expl_oob_ctx: + +Allow explicit tag out-of-bound +------------------------------- + +Invalid BER encoding could contain ``EXPLICIT`` tag containing more than +one value, more than one object. If you set ``allow_expl_oob`` context +option to True, then no error will be raised and that invalid encoding +will be silently further processed. But pay attention that offsets and +lengths will be invalid in that case. + +.. warning:: + + This option should be used only for skipping some decode errors, just + to see the decoded structure somehow. Primitive types --------------- @@ -368,6 +458,10 @@ CommonString ____________ .. autoclass:: pyderasn.CommonString +NumericString +_____________ +.. autoclass:: pyderasn.NumericString + UTCTime _______ .. autoclass:: pyderasn.UTCTime @@ -420,6 +514,7 @@ _____ Various ------- +.. autofunction:: pyderasn.abs_decode_path .. autofunction:: pyderasn.hexenc .. autofunction:: pyderasn.hexdec .. autofunction:: pyderasn.tag_encode @@ -427,6 +522,17 @@ Various .. autofunction:: pyderasn.tag_ctxp .. autofunction:: pyderasn.tag_ctxc .. autoclass:: pyderasn.Obj +.. autoclass:: pyderasn.DecodeError + :members: __init__ +.. autoclass:: pyderasn.NotEnoughData +.. autoclass:: pyderasn.LenIndefForm +.. autoclass:: pyderasn.TagMismatch +.. autoclass:: pyderasn.InvalidLength +.. autoclass:: pyderasn.InvalidOID +.. autoclass:: pyderasn.ObjUnknown +.. autoclass:: pyderasn.ObjNotReady +.. autoclass:: pyderasn.InvalidValueType +.. autoclass:: pyderasn.BoundsError """ from codecs import getdecoder @@ -435,6 +541,9 @@ from collections import namedtuple from collections import OrderedDict from datetime import datetime from math import ceil +from os import environ +from string import ascii_letters +from string import digits from six import add_metaclass from six import binary_type @@ -449,6 +558,13 @@ from six import text_type from six.moves import xrange as six_xrange +try: + from termcolor import colored +except ImportError: # pragma: no cover + def colored(what, *args): + return what + + __all__ = ( "Any", "BitString", @@ -456,8 +572,8 @@ __all__ = ( "Boolean", "BoundsError", "Choice", - "decode_path_defby", "DecodeError", + "DecodePathDefBy", "Enumerated", "GeneralizedTime", "GeneralString", @@ -470,6 +586,7 @@ __all__ = ( "InvalidOID", "InvalidValueType", "ISO646String", + "LenIndefForm", "NotEnoughData", "Null", "NumericString", @@ -515,6 +632,10 @@ TagClassReprs = { TagClassPrivate: "PRIVATE ", TagClassUniversal: "UNIV ", } +EOC = b"\x00\x00" +EOC_LEN = len(EOC) +LENINDEF = b"\x80" # length indefinite mark +LENINDEF_PP_CHAR = "I" if PY2 else "∞" ######################################################################## @@ -544,7 +665,7 @@ class DecodeError(Exception): c for c in ( "" if self.klass is None else self.klass.__name__, ( - ("(%s)" % ".".join(self.decode_path)) + ("(%s)" % ":".join(str(dp) for dp in self.decode_path)) if len(self.decode_path) > 0 else "" ), ("(at %d)" % self.offset) if self.offset > 0 else "", @@ -560,6 +681,10 @@ class NotEnoughData(DecodeError): pass +class LenIndefForm(DecodeError): + pass + + class TagMismatch(DecodeError): pass @@ -751,6 +876,11 @@ def len_encode(l): def len_decode(data): + """Decode length + + :returns: (decoded length, length's length, remaining data) + :raises LenIndefForm: if indefinite form encoding is met + """ if len(data) == 0: raise NotEnoughData("no data at all") first_octet = byte2int(data) @@ -760,7 +890,7 @@ def len_decode(data): if octets_num + 1 > len(data): raise NotEnoughData("encoded length is longer than data") if octets_num == 0: - raise DecodeError("long form instead of short one") + raise LenIndefForm() if byte2int(data[1:]) == 0: raise DecodeError("leading zeros") l = 0 @@ -776,9 +906,9 @@ def len_decode(data): ######################################################################## class AutoAddSlots(type): - def __new__(cls, name, bases, _dict): + def __new__(mcs, name, bases, _dict): _dict["__slots__"] = _dict.get("__slots__", ()) - return type.__new__(cls, name, bases, _dict) + return type.__new__(mcs, name, bases, _dict) @add_metaclass(AutoAddSlots) @@ -797,6 +927,9 @@ class Obj(object): "offset", "llen", "vlen", + "expl_lenindef", + "lenindef", + "ber_encoded", ) def __init__( @@ -810,14 +943,15 @@ class Obj(object): self.tag = getattr(self, "impl", self.tag_default) if impl is None else impl self._expl = getattr(self, "expl", None) if expl is None else expl if self.tag != self.tag_default and self._expl is not None: - raise ValueError( - "implicit and explicit tags can not be set simultaneously" - ) + raise ValueError("implicit and explicit tags can not be set simultaneously") if default is not None: optional = True self.optional = optional self.offset, self.llen, self.vlen = _decoded self.default = None + self.expl_lenindef = False + self.lenindef = False + self.ber_encoded = False @property def ready(self): # pragma: no cover @@ -829,6 +963,12 @@ class Obj(object): if not self.ready: raise ObjNotReady(self.__class__.__name__) + @property + def bered(self): + """Is either object or any elements inside is BER encoded? + """ + return self.expl_lenindef or self.lenindef or self.ber_encoded + @property def decoded(self): """Is object decoded? @@ -866,7 +1006,7 @@ class Obj(object): def _encode(self): # pragma: no cover raise NotImplementedError() - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): # pragma: no cover + def _decode(self, tlv, offset, decode_path, ctx, tag_only): # pragma: no cover raise NotImplementedError() def encode(self): @@ -875,24 +1015,41 @@ class Obj(object): return raw return b"".join((self._expl, len_encode(len(raw)), raw)) - def decode(self, data, offset=0, leavemm=False, decode_path=(), defines_by_path=None): + def decode( + self, + data, + offset=0, + leavemm=False, + decode_path=(), + ctx=None, + tag_only=False, + ): """Decode the data :param data: either binary or memoryview :param int offset: initial data's offset :param bool leavemm: do we need to leave memoryview of remaining data as is, or convert it to bytes otherwise - :param defines_by_path: :ref:`Read about DEFINED BY ` + :param ctx: optional :ref:`context ` governing decoding process. + :param tag_only: decode only the tag, without length and contents + (used only in Choice and Set structures, trying to + determine if tag satisfies the scheme) :returns: (Obj, remaining data) """ + if ctx is None: + ctx = {} tlv = memoryview(data) if self._expl is None: - obj, tail = self._decode( + result = self._decode( tlv, offset, decode_path=decode_path, - defines_by_path=defines_by_path, + ctx=ctx, + tag_only=tag_only, ) + if tag_only: + return + obj, tail = result else: try: t, tlen, lv = tag_strip(tlv) @@ -911,6 +1068,36 @@ class Obj(object): ) try: l, llen, v = len_decode(lv) + except LenIndefForm as err: + if not ctx.get("bered", False): + raise err.__class__( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + llen, v = 1, lv[1:] + offset += tlen + llen + result = self._decode( + v, + offset=offset, + decode_path=decode_path, + ctx=ctx, + tag_only=tag_only, + ) + if tag_only: # pragma: no cover + return + obj, tail = result + eoc_expected, tail = tail[:EOC_LEN], tail[EOC_LEN:] + if eoc_expected.tobytes() != EOC: + raise DecodeError( + "no EOC", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + obj.vlen += EOC_LEN + obj.expl_lenindef = True except DecodeError as err: raise err.__class__( msg=err.msg, @@ -918,19 +1105,31 @@ class Obj(object): decode_path=decode_path, offset=offset, ) - if l > len(v): - raise NotEnoughData( - "encoded length is longer than data", - klass=self.__class__, + else: + if l > len(v): + raise NotEnoughData( + "encoded length is longer than data", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + result = self._decode( + v, + offset=offset + tlen + llen, decode_path=decode_path, - offset=offset, + ctx=ctx, + tag_only=tag_only, ) - obj, tail = self._decode( - v, - offset=offset + tlen + llen, - decode_path=decode_path, - defines_by_path=defines_by_path, - ) + if tag_only: # pragma: no cover + return + obj, tail = result + if obj.tlvlen < l and not ctx.get("allow_expl_oob", False): + raise DecodeError( + "explicit tag out-of-bound, longer than data", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) return obj, (tail if leavemm else tail.tobytes()) @property @@ -947,6 +1146,8 @@ class Obj(object): @property def expl_llen(self): + if self.expl_lenindef: + return 1 return len(len_encode(self.tlvlen)) @property @@ -961,11 +1162,68 @@ class Obj(object): def expl_tlvlen(self): return self.expl_tlen + self.expl_llen + self.expl_vlen + @property + def fulloffset(self): + return self.expl_offset if self.expled else self.offset + + @property + def fulllen(self): + return self.expl_tlvlen if self.expled else self.tlvlen + + def pps_lenindef(self, decode_path): + if self.lenindef and not ( + getattr(self, "defined", None) is not None and + self.defined[1].lenindef + ): + yield _pp( + asn1_type_name="EOC", + obj_name="", + decode_path=decode_path, + offset=( + self.offset + self.tlvlen - + (EOC_LEN * 2 if self.expl_lenindef else EOC_LEN) + ), + tlen=1, + llen=1, + vlen=0, + ber_encoded=True, + bered=True, + ) + if self.expl_lenindef: + yield _pp( + asn1_type_name="EOC", + obj_name="EXPLICIT", + decode_path=decode_path, + offset=self.expl_offset + self.expl_tlvlen - EOC_LEN, + tlen=1, + llen=1, + vlen=0, + ber_encoded=True, + bered=True, + ) + -def decode_path_defby(defined_by): +class DecodePathDefBy(object): """DEFINED BY representation inside decode path """ - return "DEFINED BY (%s)" % defined_by + __slots__ = ("defined_by",) + + def __init__(self, defined_by): + self.defined_by = defined_by + + def __ne__(self, their): + return not(self == their) + + def __eq__(self, their): + if not isinstance(their, self.__class__): + return False + return self.defined_by == their.defined_by + + def __str__(self): + return "DEFINED BY " + str(self.defined_by) + + def __repr__(self): + return "<%s: %s>" % (self.__class__.__name__, self.defined_by) ######################################################################## @@ -990,6 +1248,10 @@ PP = namedtuple("PP", ( "expl_tlen", "expl_llen", "expl_vlen", + "expl_lenindef", + "lenindef", + "ber_encoded", + "bered", )) @@ -1011,6 +1273,10 @@ def _pp( expl_tlen=None, expl_llen=None, expl_vlen=None, + expl_lenindef=False, + lenindef=False, + ber_encoded=False, + bered=False, ): return PP( asn1_type_name, @@ -1030,59 +1296,110 @@ def _pp( expl_tlen, expl_llen, expl_vlen, + expl_lenindef, + lenindef, + ber_encoded, + bered, ) -def pp_console_row(pp, oids=None, with_offsets=False, with_blob=True): +def _colourize(what, colour, with_colours, attrs=("bold",)): + return colored(what, colour, attrs=attrs) if with_colours else what + + +def pp_console_row( + pp, + oids=None, + with_offsets=False, + with_blob=True, + with_colours=False, + with_decode_path=False, + decode_path_len_decrease=0, +): cols = [] if with_offsets: - cols.append("%5d%s [%d,%d,%4d]" % ( + col = "%5d%s%s" % ( pp.offset, ( " " if pp.expl_offset is None else ("-%d" % (pp.offset - pp.expl_offset)) ), + LENINDEF_PP_CHAR if pp.expl_lenindef else " ", + ) + col = _colourize(col, "red", with_colours, ()) + col += _colourize("B", "red", with_colours) if pp.bered else " " + cols.append(col) + col = "[%d,%d,%4d]%s" % ( pp.tlen, pp.llen, pp.vlen, - )) - if len(pp.decode_path) > 0: - cols.append(" ." * (len(pp.decode_path))) - cols.append("%s:" % pp.decode_path[-1]) + LENINDEF_PP_CHAR if pp.lenindef else " " + ) + col = _colourize(col, "green", with_colours, ()) + cols.append(col) + decode_path_len = len(pp.decode_path) - decode_path_len_decrease + if decode_path_len > 0: + cols.append(" ." * decode_path_len) + ent = pp.decode_path[-1] + if isinstance(ent, DecodePathDefBy): + cols.append(_colourize("DEFINED BY", "red", with_colours, ("reverse",))) + value = str(ent.defined_by) + if ( + oids is not None and + ent.defined_by.asn1_type_name == + ObjectIdentifier.asn1_type_name and + value in oids + ): + cols.append(_colourize("%s:" % oids[value], "green", with_colours)) + else: + cols.append(_colourize("%s:" % value, "white", with_colours, ("reverse",))) + else: + cols.append(_colourize("%s:" % ent, "yellow", with_colours, ("reverse",))) if pp.expl is not None: klass, _, num = pp.expl - cols.append("[%s%d] EXPLICIT" % (TagClassReprs[klass], num)) + col = "[%s%d] EXPLICIT" % (TagClassReprs[klass], num) + cols.append(_colourize(col, "blue", with_colours)) if pp.impl is not None: klass, _, num = pp.impl - cols.append("[%s%d]" % (TagClassReprs[klass], num)) + col = "[%s%d]" % (TagClassReprs[klass], num) + cols.append(_colourize(col, "blue", with_colours)) if pp.asn1_type_name.replace(" ", "") != pp.obj_name.upper(): - cols.append(pp.obj_name) - cols.append(pp.asn1_type_name) + cols.append(_colourize(pp.obj_name, "magenta", with_colours)) + if pp.ber_encoded: + cols.append(_colourize("BER", "red", with_colours)) + cols.append(_colourize(pp.asn1_type_name, "cyan", with_colours)) if pp.value is not None: value = pp.value + cols.append(_colourize(value, "white", with_colours, ("reverse",))) if ( oids is not None and pp.asn1_type_name == ObjectIdentifier.asn1_type_name and value in oids ): - value = "%s (%s)" % (oids[value], pp.value) - cols.append(value) + cols.append(_colourize("(%s)" % oids[value], "green", with_colours)) if with_blob: if isinstance(pp.blob, binary_type): cols.append(hexenc(pp.blob)) elif isinstance(pp.blob, tuple): cols.append(", ".join(pp.blob)) if pp.optional: - cols.append("OPTIONAL") + cols.append(_colourize("OPTIONAL", "red", with_colours)) if pp.default: - cols.append("DEFAULT") + cols.append(_colourize("DEFAULT", "red", with_colours)) + if with_decode_path: + cols.append(_colourize( + "[%s]" % ":".join(str(p) for p in pp.decode_path), + "grey", + with_colours, + )) return " ".join(cols) -def pp_console_blob(pp): - cols = [" " * len("XXXXXYY [X,X,XXXX]")] - if len(pp.decode_path) > 0: - cols.append(" ." * (len(pp.decode_path) + 1)) +def pp_console_blob(pp, decode_path_len_decrease=0): + cols = [" " * len("XXXXXYYZZ [X,X,XXXX]Z")] + decode_path_len = len(pp.decode_path) - decode_path_len_decrease + if decode_path_len > 0: + cols.append(" ." * (decode_path_len + 1)) if isinstance(pp.blob, binary_type): blob = hexenc(pp.blob).upper() for i in range(0, len(blob), 32): @@ -1094,7 +1411,14 @@ def pp_console_blob(pp): yield " ".join(cols + [", ".join(pp.blob)]) -def pprint(obj, oids=None, big_blobs=False): +def pprint( + obj, + oids=None, + big_blobs=False, + with_colours=False, + with_decode_path=False, + decode_path_only=(), +): """Pretty print object :param Obj obj: object you want to pretty print @@ -1103,21 +1427,46 @@ def pprint(obj, oids=None, big_blobs=False): :param big_blobs: if large binary objects are met (like OctetString values), do we need to print them too, on separate lines + :param with_colours: colourize output, if ``termcolor`` library + is available + :param with_decode_path: print decode path + :param decode_path_only: print only that specified decode path """ def _pprint_pps(pps): for pp in pps: if hasattr(pp, "_fields"): + if ( + decode_path_only != () and + tuple( + str(p) for p in pp.decode_path[:len(decode_path_only)] + ) != decode_path_only + ): + continue if big_blobs: yield pp_console_row( pp, oids=oids, with_offsets=True, with_blob=False, + with_colours=with_colours, + with_decode_path=with_decode_path, + decode_path_len_decrease=len(decode_path_only), ) - for row in pp_console_blob(pp): + for row in pp_console_blob( + pp, + decode_path_len_decrease=len(decode_path_only), + ): yield row else: - yield pp_console_row(pp, oids=oids, with_offsets=True) + yield pp_console_row( + pp, + oids=oids, + with_offsets=True, + with_blob=True, + with_colours=with_colours, + with_decode_path=with_decode_path, + decode_path_len_decrease=len(decode_path_only), + ) else: for row in _pprint_pps(pp): yield row @@ -1237,7 +1586,7 @@ class Boolean(Obj): (b"\xFF" if self._value else b"\x00"), )) - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): + def _decode(self, tlv, offset, decode_path, ctx, tag_only): try: t, _, lv = tag_strip(tlv) except DecodeError as err: @@ -1253,6 +1602,8 @@ class Boolean(Obj): decode_path=decode_path, offset=offset, ) + if tag_only: + return try: l, _, v = len_decode(lv) except DecodeError as err: @@ -1277,10 +1628,14 @@ class Boolean(Obj): offset=offset, ) first_octet = byte2int(v) + ber_encoded = False if first_octet == 0: value = False elif first_octet == 0xFF: value = True + elif ctx.get("bered", False): + value = True + ber_encoded = True else: raise DecodeError( "unacceptable Boolean value", @@ -1296,6 +1651,7 @@ class Boolean(Obj): optional=self.optional, _decoded=(offset, 1, 1), ) + obj.ber_encoded = ber_encoded return obj, v[1:] def __repr__(self): @@ -1319,7 +1675,12 @@ class Boolean(Obj): expl_tlen=self.expl_tlen if self.expled else None, expl_llen=self.expl_llen if self.expled else None, expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + ber_encoded=self.ber_encoded, + bered=self.bered, ) + for pp in self.pps_lenindef(decode_path): + yield pp class Integer(Obj): @@ -1530,7 +1891,7 @@ class Integer(Obj): break return b"".join((self.tag, len_encode(len(octets)), octets)) - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): + def _decode(self, tlv, offset, decode_path, ctx, tag_only): try: t, _, lv = tag_strip(tlv) except DecodeError as err: @@ -1546,6 +1907,8 @@ class Integer(Obj): decode_path=decode_path, offset=offset, ) + if tag_only: + return try: l, llen, v = len_decode(lv) except DecodeError as err: @@ -1639,7 +2002,11 @@ class Integer(Obj): expl_tlen=self.expl_tlen if self.expled else None, expl_llen=self.expl_llen if self.expled else None, expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + bered=self.bered, ) + for pp in self.pps_lenindef(decode_path): + yield pp class BitString(Obj): @@ -1654,6 +2021,8 @@ class BitString(Obj): >>> b.bit_len 88 + >>> BitString("'0A3B5F291CD'H") + BIT STRING 44 bits 0a3b5f291cd0 >>> b = BitString("'010110000000'B") BIT STRING 12 bits 5800 >>> b.bit_len @@ -1669,19 +2038,27 @@ class BitString(Obj): class KeyUsage(BitString): schema = ( - ('digitalSignature', 0), - ('nonRepudiation', 1), - ('keyEncipherment', 2), + ("digitalSignature", 0), + ("nonRepudiation", 1), + ("keyEncipherment", 2), ) - >>> b = KeyUsage(('keyEncipherment', 'nonRepudiation')) + >>> b = KeyUsage(("keyEncipherment", "nonRepudiation")) KeyUsage BIT STRING 3 bits nonRepudiation, keyEncipherment >>> b.named ['nonRepudiation', 'keyEncipherment'] >>> b.specs {'nonRepudiation': 1, 'digitalSignature': 0, 'keyEncipherment': 2} + + .. note:: + + Pay attention that BIT STRING can be encoded both in primitive + and constructed forms. Decoder always checks constructed form tag + additionally to specified primitive one. If BER decoding is + :ref:`not enabled `, then decoder will fail, because + of DER restrictions. """ - __slots__ = ("specs",) + __slots__ = ("tag_constructed", "specs", "defined") tag_default = tag_encode(3) asn1_type_name = "BIT STRING" @@ -1718,6 +2095,13 @@ class BitString(Obj): ) if value is None: self._value = default + self.defined = None + tag_klass, _, tag_num = tag_decode(self.tag) + self.tag_constructed = tag_encode( + klass=tag_klass, + form=TagFormConstructed, + num=tag_num, + ) def _bits2octets(self, bits): if len(self.specs) > 0: @@ -1735,21 +2119,23 @@ class BitString(Obj): if isinstance(value, (string_types, binary_type)): if ( isinstance(value, string_types) and - value.startswith("'") and - value.endswith("'B") + value.startswith("'") ): - value = value[1:-2] - if not set(value) <= set(("0", "1")): - raise ValueError("B's coding contains unacceptable chars") - return self._bits2octets(value) - elif isinstance(value, binary_type): + if value.endswith("'B"): + value = value[1:-2] + if not set(value) <= set(("0", "1")): + raise ValueError("B's coding contains unacceptable chars") + return self._bits2octets(value) + elif value.endswith("'H"): + value = value[1:-2] + return ( + len(value) * 4, + hexdec(value + ("" if len(value) % 2 == 0 else "0")), + ) + if isinstance(value, binary_type): return (len(value) * 8, value) else: - raise InvalidValueType(( - self.__class__, - string_types, - binary_type, - )) + raise InvalidValueType((self.__class__, string_types, binary_type)) if isinstance(value, tuple): if ( len(value) == 2 and @@ -1863,22 +2249,7 @@ class BitString(Obj): octets, )) - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): - try: - t, _, lv = tag_strip(tlv) - except DecodeError as err: - raise err.__class__( - msg=err.msg, - klass=self.__class__, - decode_path=decode_path, - offset=offset, - ) - if t != self.tag: - raise TagMismatch( - klass=self.__class__, - decode_path=decode_path, - offset=offset, - ) + def _decode_chunk(self, lv, offset, decode_path, ctx): try: l, llen, v = len_decode(lv) except DecodeError as err: @@ -1917,7 +2288,7 @@ class BitString(Obj): decode_path=decode_path, offset=offset, ) - if byte2int(v[-1:]) & ((1 << pad_size) - 1) != 0: + if byte2int(v[l - 1:l]) & ((1 << pad_size) - 1) != 0: raise DecodeError( "invalid pad", klass=self.__class__, @@ -1936,6 +2307,134 @@ class BitString(Obj): ) return obj, tail + def _decode(self, tlv, offset, decode_path, ctx, tag_only): + try: + t, tlen, lv = tag_strip(tlv) + except DecodeError as err: + raise err.__class__( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + if t == self.tag: + if tag_only: # pragma: no cover + return + return self._decode_chunk(lv, offset, decode_path, ctx) + if t == self.tag_constructed: + if not ctx.get("bered", False): + raise DecodeError( + "unallowed BER constructed encoding", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + if tag_only: # pragma: no cover + return + lenindef = False + try: + l, llen, v = len_decode(lv) + except LenIndefForm: + llen, l, v = 1, 0, lv[1:] + lenindef = True + except DecodeError as err: + raise err.__class__( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + if l > len(v): + raise NotEnoughData( + "encoded length is longer than data", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + if not lenindef and l == 0: + raise NotEnoughData( + "zero length", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + chunks = [] + sub_offset = offset + tlen + llen + vlen = 0 + while True: + if lenindef: + if v[:EOC_LEN].tobytes() == EOC: + break + else: + if vlen == l: + break + if vlen > l: + raise DecodeError( + "chunk out of bounds", + klass=self.__class__, + decode_path=decode_path + (str(len(chunks) - 1),), + offset=chunks[-1].offset, + ) + sub_decode_path = decode_path + (str(len(chunks)),) + try: + chunk, v_tail = BitString().decode( + v, + offset=sub_offset, + decode_path=sub_decode_path, + leavemm=True, + ctx=ctx, + ) + except TagMismatch: + raise DecodeError( + "expected BitString encoded chunk", + klass=self.__class__, + decode_path=sub_decode_path, + offset=sub_offset, + ) + chunks.append(chunk) + sub_offset += chunk.tlvlen + vlen += chunk.tlvlen + v = v_tail + if len(chunks) == 0: + raise DecodeError( + "no chunks", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + values = [] + bit_len = 0 + for chunk_i, chunk in enumerate(chunks[:-1]): + if chunk.bit_len % 8 != 0: + raise DecodeError( + "BitString chunk is not multiple of 8 bits", + klass=self.__class__, + decode_path=decode_path + (str(chunk_i),), + offset=chunk.offset, + ) + values.append(bytes(chunk)) + bit_len += chunk.bit_len + chunk_last = chunks[-1] + values.append(bytes(chunk_last)) + bit_len += chunk_last.bit_len + obj = self.__class__( + value=(bit_len, b"".join(values)), + impl=self.tag, + expl=self._expl, + default=self.default, + optional=self.optional, + _specs=self.specs, + _decoded=(offset, llen, vlen + (EOC_LEN if lenindef else 0)), + ) + obj.lenindef = lenindef + obj.ber_encoded = True + return obj, (v[EOC_LEN:] if lenindef else v) + raise TagMismatch( + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + def __repr__(self): return pp_console_row(next(self.pps())) @@ -1965,7 +2464,18 @@ class BitString(Obj): expl_tlen=self.expl_tlen if self.expled else None, expl_llen=self.expl_llen if self.expled else None, expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + lenindef=self.lenindef, + ber_encoded=self.ber_encoded, + bered=self.bered, ) + defined_by, defined = self.defined or (None, None) + if defined_by is not None: + yield defined.pps( + decode_path=decode_path + (DecodePathDefBy(defined_by),) + ) + for pp in self.pps_lenindef(decode_path): + yield pp class OctetString(Obj): @@ -1983,8 +2493,16 @@ class OctetString(Obj): pyderasn.BoundsError: unsatisfied bounds: 4 <= 5 <= 4 >>> OctetString(b"hell", bounds=(4, 4)) OCTET STRING 4 bytes 68656c6c + + .. note:: + + Pay attention that OCTET STRING can be encoded both in primitive + and constructed forms. Decoder always checks constructed form tag + additionally to specified primitive one. If BER decoding is + :ref:`not enabled `, then decoder will fail, because + of DER restrictions. """ - __slots__ = ("_bound_min", "_bound_max", "defined") + __slots__ = ("tag_constructed", "_bound_min", "_bound_max", "defined") tag_default = tag_encode(4) asn1_type_name = "OCTET STRING" @@ -2033,6 +2551,12 @@ class OctetString(Obj): if self._value is None: self._value = default self.defined = None + tag_klass, _, tag_num = tag_decode(self.tag) + self.tag_constructed = tag_encode( + klass=tag_klass, + form=TagFormConstructed, + num=tag_num, + ) def _value_sanitize(self, value): if issubclass(value.__class__, OctetString): @@ -2110,22 +2634,7 @@ class OctetString(Obj): self._value, )) - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): - try: - t, _, lv = tag_strip(tlv) - except DecodeError as err: - raise err.__class__( - msg=err.msg, - klass=self.__class__, - decode_path=decode_path, - offset=offset, - ) - if t != self.tag: - raise TagMismatch( - klass=self.__class__, - decode_path=decode_path, - offset=offset, - ) + def _decode_chunk(self, lv, offset, decode_path, ctx): try: l, llen, v = len_decode(lv) except DecodeError as err: @@ -2153,6 +2662,13 @@ class OctetString(Obj): optional=self.optional, _decoded=(offset, llen, l), ) + except DecodeError as err: + raise DecodeError( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) except BoundsError as err: raise DecodeError( msg=str(err), @@ -2162,6 +2678,120 @@ class OctetString(Obj): ) return obj, tail + def _decode(self, tlv, offset, decode_path, ctx, tag_only): + try: + t, tlen, lv = tag_strip(tlv) + except DecodeError as err: + raise err.__class__( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + if t == self.tag: + if tag_only: + return + return self._decode_chunk(lv, offset, decode_path, ctx) + if t == self.tag_constructed: + if not ctx.get("bered", False): + raise DecodeError( + "unallowed BER constructed encoding", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + if tag_only: + return + lenindef = False + try: + l, llen, v = len_decode(lv) + except LenIndefForm: + llen, l, v = 1, 0, lv[1:] + lenindef = True + except DecodeError as err: + raise err.__class__( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + if l > len(v): + raise NotEnoughData( + "encoded length is longer than data", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + chunks = [] + sub_offset = offset + tlen + llen + vlen = 0 + while True: + if lenindef: + if v[:EOC_LEN].tobytes() == EOC: + break + else: + if vlen == l: + break + if vlen > l: + raise DecodeError( + "chunk out of bounds", + klass=self.__class__, + decode_path=decode_path + (str(len(chunks) - 1),), + offset=chunks[-1].offset, + ) + sub_decode_path = decode_path + (str(len(chunks)),) + try: + chunk, v_tail = OctetString().decode( + v, + offset=sub_offset, + decode_path=sub_decode_path, + leavemm=True, + ctx=ctx, + ) + except TagMismatch: + raise DecodeError( + "expected OctetString encoded chunk", + klass=self.__class__, + decode_path=sub_decode_path, + offset=sub_offset, + ) + chunks.append(chunk) + sub_offset += chunk.tlvlen + vlen += chunk.tlvlen + v = v_tail + try: + obj = self.__class__( + value=b"".join(bytes(chunk) for chunk in chunks), + bounds=(self._bound_min, self._bound_max), + impl=self.tag, + expl=self._expl, + default=self.default, + optional=self.optional, + _decoded=(offset, llen, vlen + (EOC_LEN if lenindef else 0)), + ) + except DecodeError as err: + raise DecodeError( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + except BoundsError as err: + raise DecodeError( + msg=str(err), + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + obj.lenindef = lenindef + obj.ber_encoded = True + return obj, (v[EOC_LEN:] if lenindef else v) + raise TagMismatch( + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + def __repr__(self): return pp_console_row(next(self.pps())) @@ -2184,12 +2814,18 @@ class OctetString(Obj): expl_tlen=self.expl_tlen if self.expled else None, expl_llen=self.expl_llen if self.expled else None, expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + lenindef=self.lenindef, + ber_encoded=self.ber_encoded, + bered=self.bered, ) defined_by, defined = self.defined or (None, None) if defined_by is not None: yield defined.pps( - decode_path=decode_path + (decode_path_defby(defined_by),) + decode_path=decode_path + (DecodePathDefBy(defined_by),) ) + for pp in self.pps_lenindef(decode_path): + yield pp class Null(Obj): @@ -2259,7 +2895,7 @@ class Null(Obj): def _encode(self): return self.tag + len_encode(0) - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): + def _decode(self, tlv, offset, decode_path, ctx, tag_only): try: t, _, lv = tag_strip(tlv) except DecodeError as err: @@ -2275,6 +2911,8 @@ class Null(Obj): decode_path=decode_path, offset=offset, ) + if tag_only: # pragma: no cover + return try: l, _, v = len_decode(lv) except DecodeError as err: @@ -2318,7 +2956,11 @@ class Null(Obj): expl_tlen=self.expl_tlen if self.expled else None, expl_llen=self.expl_llen if self.expled else None, expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + bered=self.bered, ) + for pp in self.pps_lenindef(decode_path): + yield pp class ObjectIdentifier(Obj): @@ -2346,7 +2988,7 @@ class ObjectIdentifier(Obj): def __init__( self, value=None, - defines=None, + defines=(), impl=None, expl=None, default=None, @@ -2357,12 +2999,14 @@ class ObjectIdentifier(Obj): :param value: set the value. Either tuples of integers, string of "."-concatenated integers, or :py:class:`pyderasn.ObjectIdentifier` object - :param defines: tuple of two elements. First one is a name of - field inside :py:class:`pyderasn.Sequence`, - defining with that OID. Second element is a - ``{OID: pyderasn.Obj()}`` dictionary, mapping - between current OID value and structure applied - to defined field. + :param defines: sequence of tuples. Each tuple has two elements. + First one is relative to current one decode + path, aiming to the field defined by that OID. + Read about relative path in + :py:func:`pyderasn.abs_decode_path`. Second + tuple element is ``{OID: pyderasn.Obj()}`` + dictionary, mapping between current OID value + and structure applied to defined field. :ref:`Read about DEFINED BY ` :param bytes impl: override default tag with ``IMPLICIT`` one :param bytes expl: override default tag with ``EXPLICIT`` one @@ -2502,7 +3146,7 @@ class ObjectIdentifier(Obj): v = b"".join(octets) return b"".join((self.tag, len_encode(len(v)), v)) - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): + def _decode(self, tlv, offset, decode_path, ctx, tag_only): try: t, _, lv = tag_strip(tlv) except DecodeError as err: @@ -2518,6 +3162,8 @@ class ObjectIdentifier(Obj): decode_path=decode_path, offset=offset, ) + if tag_only: # pragma: no cover + return try: l, llen, v = len_decode(lv) except DecodeError as err: @@ -2602,7 +3248,11 @@ class ObjectIdentifier(Obj): expl_tlen=self.expl_tlen if self.expled else None, expl_llen=self.expl_llen if self.expled else None, expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + bered=self.bered, ) + for pp in self.pps_lenindef(decode_path): + yield pp class Enumerated(Integer): @@ -2707,7 +3357,7 @@ class CommonString(OctetString): >>> PrintableString("привет мир") Traceback (most recent call last): - UnicodeEncodeError: 'ascii' codec can't encode characters in position 0-5: ordinal not in range(128) + pyderasn.DecodeError: 'ascii' codec can't encode characters in position 0-5: ordinal not in range(128) >>> BMPString("ада", bounds=(2, 2)) Traceback (most recent call last): @@ -2763,14 +3413,17 @@ class CommonString(OctetString): value_raw = value else: raise InvalidValueType((self.__class__, text_type, binary_type)) - value_raw = ( - value_decoded.encode(self.encoding) - if value_raw is None else value_raw - ) - value_decoded = ( - value_raw.decode(self.encoding) - if value_decoded is None else value_decoded - ) + try: + value_raw = ( + value_decoded.encode(self.encoding) + if value_raw is None else value_raw + ) + value_decoded = ( + value_raw.decode(self.encoding) + if value_decoded is None else value_decoded + ) + except (UnicodeEncodeError, UnicodeDecodeError) as err: + raise DecodeError(str(err)) if not self._bound_min <= len(value_decoded) <= self._bound_max: raise BoundsError( self._bound_min, @@ -2817,7 +3470,16 @@ class CommonString(OctetString): tlen=self.tlen, llen=self.llen, vlen=self.vlen, + expl_offset=self.expl_offset if self.expled else None, + expl_tlen=self.expl_tlen if self.expled else None, + expl_llen=self.expl_llen if self.expled else None, + expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + ber_encoded=self.ber_encoded, + bered=self.bered, ) + for pp in self.pps_lenindef(decode_path): + yield pp class UTF8String(CommonString): @@ -2828,17 +3490,40 @@ class UTF8String(CommonString): class NumericString(CommonString): + """Numeric string + + Its value is properly sanitized: only ASCII digits with spaces can + be stored. + """ __slots__ = () tag_default = tag_encode(18) encoding = "ascii" asn1_type_name = "NumericString" + allowable_chars = set(digits.encode("ascii") + b" ") + + def _value_sanitize(self, value): + value = super(NumericString, self)._value_sanitize(value) + if not set(value) <= self.allowable_chars: + raise DecodeError("non-numeric value") + return value class PrintableString(CommonString): + """Printable string + + Its value is properly sanitized: see X.680 41.4 table 10. + """ __slots__ = () tag_default = tag_encode(19) encoding = "ascii" asn1_type_name = "PrintableString" + allowable_chars = set((ascii_letters + digits + " '()+,-./:=?").encode("ascii")) + + def _value_sanitize(self, value): + value = super(PrintableString, self)._value_sanitize(value) + if not set(value) <= self.allowable_chars: + raise DecodeError("non-printable value") + return value class TeletexString(CommonString): @@ -2937,11 +3622,14 @@ class UTCTime(CommonString): if isinstance(value, datetime): return value.strftime(self.fmt).encode("ascii") if isinstance(value, binary_type): - value_decoded = value.decode("ascii") + try: + value_decoded = value.decode("ascii") + except (UnicodeEncodeError, UnicodeDecodeError) as err: + raise DecodeError("invalid UTCTime encoding") if len(value_decoded) == LEN_YYMMDDHHMMSSZ: try: datetime.strptime(value_decoded, self.fmt) - except ValueError: + except (TypeError, ValueError): raise DecodeError("invalid UTCTime format") return value else: @@ -2998,7 +3686,16 @@ class UTCTime(CommonString): tlen=self.tlen, llen=self.llen, vlen=self.vlen, + expl_offset=self.expl_offset if self.expled else None, + expl_tlen=self.expl_tlen if self.expled else None, + expl_llen=self.expl_llen if self.expled else None, + expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + ber_encoded=self.ber_encoded, + bered=self.bered, ) + for pp in self.pps_lenindef(decode_path): + yield pp class GeneralizedTime(UTCTime): @@ -3028,11 +3725,14 @@ class GeneralizedTime(UTCTime): self.fmt_ms if value.microsecond > 0 else self.fmt ).encode("ascii") if isinstance(value, binary_type): - value_decoded = value.decode("ascii") + try: + value_decoded = value.decode("ascii") + except (UnicodeEncodeError, UnicodeDecodeError) as err: + raise DecodeError("invalid GeneralizedTime encoding") if len(value_decoded) == LEN_YYYYMMDDHHMMSSZ: try: datetime.strptime(value_decoded, self.fmt) - except ValueError: + except (TypeError, ValueError): raise DecodeError( "invalid GeneralizedTime (without ms) format", ) @@ -3040,7 +3740,7 @@ class GeneralizedTime(UTCTime): elif len(value_decoded) >= LEN_YYYYMMDDHHMMSSDMZ: try: datetime.strptime(value_decoded, self.fmt_ms) - except ValueError: + except (TypeError, ValueError): raise DecodeError( "invalid GeneralizedTime (with ms) format", ) @@ -3106,8 +3806,8 @@ class Choice(Obj): class GeneralName(Choice): schema = ( - ('rfc822Name', IA5String(impl=tag_ctxp(1))), - ('dNSName', IA5String(impl=tag_ctxp(2))), + ("rfc822Name", IA5String(impl=tag_ctxp(1))), + ("dNSName", IA5String(impl=tag_ctxp(2))), ) >>> gn = GeneralName() @@ -3191,6 +3891,13 @@ class Choice(Obj): def ready(self): return self._value is not None and self._value[1].ready + @property + def bered(self): + return self.expl_lenindef or ( + (self._value is not None) and + self._value[1].bered + ) + def copy(self): obj = self.__class__(schema=self.specs) obj._expl = self._expl @@ -3269,32 +3976,45 @@ class Choice(Obj): self._assert_ready() return self._value[1].encode() - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): + def _decode(self, tlv, offset, decode_path, ctx, tag_only): for choice, spec in self.specs.items(): + sub_decode_path = decode_path + (choice,) try: - value, tail = spec.decode( + spec.decode( tlv, offset=offset, leavemm=True, - decode_path=decode_path + (choice,), - defines_by_path=defines_by_path, + decode_path=sub_decode_path, + ctx=ctx, + tag_only=True, ) except TagMismatch: continue - obj = self.__class__( - schema=self.specs, - expl=self._expl, - default=self.default, - optional=self.optional, - _decoded=(offset, 0, value.tlvlen), + break + else: + raise TagMismatch( + klass=self.__class__, + decode_path=decode_path, + offset=offset, ) - obj._value = (choice, value) - return obj, tail - raise TagMismatch( - klass=self.__class__, - decode_path=decode_path, + if tag_only: # pragma: no cover + return + value, tail = spec.decode( + tlv, offset=offset, + leavemm=True, + decode_path=sub_decode_path, + ctx=ctx, ) + obj = self.__class__( + schema=self.specs, + expl=self._expl, + default=self.default, + optional=self.optional, + _decoded=(offset, 0, value.fulllen), + ) + obj._value = (choice, value) + return obj, tail def __repr__(self): value = pp_console_row(next(self.pps())) @@ -3316,9 +4036,13 @@ class Choice(Obj): tlen=self.tlen, llen=self.llen, vlen=self.vlen, + expl_lenindef=self.expl_lenindef, + bered=self.bered, ) if self.ready: yield self.value.pps(decode_path=decode_path + (self.choice,)) + for pp in self.pps_lenindef(decode_path): + yield pp class PrimitiveTypes(Choice): @@ -3402,6 +4126,14 @@ class Any(Obj): def ready(self): return self._value is not None + @property + def bered(self): + if self.expl_lenindef or self.lenindef: + return True + if self.defined is None: + return False + return self.defined[1].bered + def copy(self): obj = self.__class__() obj._value = self._value @@ -3444,10 +4176,50 @@ class Any(Obj): self._assert_ready() return self._value - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): + def _decode(self, tlv, offset, decode_path, ctx, tag_only): try: t, tlen, lv = tag_strip(tlv) + except DecodeError as err: + raise err.__class__( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + try: l, llen, v = len_decode(lv) + except LenIndefForm as err: + if not ctx.get("bered", False): + raise err.__class__( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + llen, vlen, v = 1, 0, lv[1:] + sub_offset = offset + tlen + llen + chunk_i = 0 + while v[:EOC_LEN].tobytes() != EOC: + chunk, v = Any().decode( + v, + offset=sub_offset, + decode_path=decode_path + (str(chunk_i),), + leavemm=True, + ctx=ctx, + ) + vlen += chunk.tlvlen + sub_offset += chunk.tlvlen + chunk_i += 1 + tlvlen = tlen + llen + vlen + EOC_LEN + obj = self.__class__( + value=tlv[:tlvlen].tobytes(), + expl=self._expl, + optional=self.optional, + _decoded=(offset, 0, tlvlen), + ) + obj.lenindef = True + obj.tag = t + return obj, v[EOC_LEN:] except DecodeError as err: raise err.__class__( msg=err.msg, @@ -3494,12 +4266,17 @@ class Any(Obj): expl_tlen=self.expl_tlen if self.expled else None, expl_llen=self.expl_llen if self.expled else None, expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + lenindef=self.lenindef, + bered=self.bered, ) defined_by, defined = self.defined or (None, None) if defined_by is not None: yield defined.pps( - decode_path=decode_path + (decode_path_defby(defined_by),) + decode_path=decode_path + (DecodePathDefBy(defined_by),) ) + for pp in self.pps_lenindef(decode_path): + yield pp ######################################################################## @@ -3519,6 +4296,32 @@ def get_def_by_path(defines_by_path, sub_decode_path): return define +def abs_decode_path(decode_path, rel_path): + """Create an absolute decode path from current and relative ones + + :param decode_path: current decode path, starting point. + Tuple of strings + :param rel_path: relative path to ``decode_path``. Tuple of strings. + If first tuple's element is "/", then treat it as + an absolute path, ignoring ``decode_path`` as + starting point. Also this tuple can contain ".." + elements, stripping the leading element from + ``decode_path`` + + >>> abs_decode_path(("foo", "bar"), ("baz", "whatever")) + ("foo", "bar", "baz", "whatever") + >>> abs_decode_path(("foo", "bar", "baz"), ("..", "..", "whatever")) + ("foo", "whatever") + >>> abs_decode_path(("foo", "bar"), ("/", "baz", "whatever")) + ("baz", "whatever") + """ + if rel_path[0] == "/": + return rel_path[1:] + if rel_path[0] == "..": + return abs_decode_path(decode_path[:-1], rel_path[1:]) + return decode_path + rel_path + + class Sequence(Obj): """``SEQUENCE`` structure type @@ -3545,7 +4348,7 @@ class Sequence(Obj): pyderasn.InvalidValueType: invalid value type, expected: >>> ext["extnID"] = ObjectIdentifier("1.2.3") - You can know if sequence is ready to be encoded: + You can determine if sequence is ready to be encoded: >>> ext.ready False @@ -3569,7 +4372,17 @@ class Sequence(Obj): >>> tbs = TBSCertificate() >>> tbs["version"] = Version("v2") # no need to explicitly add ``expl`` - You can know if value exists/set in the sequence and take its value: + Assign ``None`` to remove value from sequence. + + You can set values in Sequence during its initialization: + + >>> AlgorithmIdentifier(( + ("algorithm", ObjectIdentifier("1.2.3")), + ("parameters", Any(Null())) + )) + AlgorithmIdentifier SEQUENCE[algorithm: OBJECT IDENTIFIER 1.2.3; parameters: ANY 0500 OPTIONAL] + + You can determine if value exists/set in the sequence and take its value: >>> "extnID" in ext, "extnValue" in ext, "critical" in ext (True, True, False) @@ -3588,13 +4401,14 @@ class Sequence(Obj): All defaulted values are always optional. - .. warning:: + .. _allow_default_values_ctx: - When decoded DER contains defaulted value inside, then - technically this is not valid DER encoding. But we allow - and pass it. Of course reencoding of that kind of DER will - result in different binary representation (validly without - defaulted value inside). + DER prohibits default value encoding and will raise an error if + default value is unexpectedly met during decode. + If :ref:`bered ` context option is set, then no error + will be raised, but ``bered`` attribute set. You can disable strict + defaulted values existence validation by setting + ``"allow_default_values": True`` :ref:`context ` option. Two sequences are equal if they have equal specification (schema), implicit/explicit tagging and the same values. @@ -3621,9 +4435,17 @@ class Sequence(Obj): ) self._value = {} if value is not None: - self._value = self._value_sanitize(value) + if issubclass(value.__class__, Sequence): + self._value = value._value + elif hasattr(value, "__iter__"): + for seq_key, seq_value in value: + self[seq_key] = seq_value + else: + raise InvalidValueType((Sequence,)) if default is not None: - default_value = self._value_sanitize(default) + if not issubclass(default.__class__, Sequence): + raise InvalidValueType((Sequence,)) + default_value = default._value default_obj = self.__class__(impl=self.tag, expl=self._expl) default_obj.specs = self.specs default_obj._value = default_value @@ -3631,11 +4453,6 @@ class Sequence(Obj): if value is None: self._value = default_obj.copy()._value - def _value_sanitize(self, value): - if not issubclass(value.__class__, Sequence): - raise InvalidValueType((Sequence,)) - return value._value - @property def ready(self): for name, spec in self.specs.items(): @@ -3649,6 +4466,12 @@ class Sequence(Obj): return False return True + @property + def bered(self): + if self.expl_lenindef or self.lenindef or self.ber_encoded: + return True + return any(value.bered for value in self._value.values()) + def copy(self): obj = self.__class__(schema=self.specs) obj.tag = self.tag @@ -3732,7 +4555,7 @@ class Sequence(Obj): v = b"".join(self._encoded_values()) return b"".join((self.tag, len_encode(len(v)), v)) - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): + def _decode(self, tlv, offset, decode_path, ctx, tag_only): try: t, tlen, lv = tag_strip(tlv) except DecodeError as err: @@ -3748,8 +4571,22 @@ class Sequence(Obj): decode_path=decode_path, offset=offset, ) + if tag_only: # pragma: no cover + return + lenindef = False + ctx_bered = ctx.get("bered", False) try: l, llen, v = len_decode(lv) + except LenIndefForm as err: + if not ctx_bered: + raise err.__class__( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + l, llen, v = 0, 1, lv[1:] + lenindef = True except DecodeError as err: raise err.__class__( msg=err.msg, @@ -3764,12 +4601,18 @@ class Sequence(Obj): decode_path=decode_path, offset=offset, ) - v, tail = v[:l], v[l:] + if not lenindef: + v, tail = v[:l], v[l:] + vlen = 0 sub_offset = offset + tlen + llen values = {} - defines = {} + ber_encoded = False + ctx_allow_default_values = ctx.get("allow_default_values", False) for name, spec in self.specs.items(): - if len(v) == 0 and spec.optional: + if spec.optional and ( + (lenindef and v[:EOC_LEN].tobytes() == EOC) or + len(v) == 0 + ): continue sub_decode_path = decode_path + (name,) try: @@ -3778,28 +4621,31 @@ class Sequence(Obj): sub_offset, leavemm=True, decode_path=sub_decode_path, - defines_by_path=defines_by_path, + ctx=ctx, ) except TagMismatch: if spec.optional: continue raise - defined = defines.pop(name, None) + defined = get_def_by_path(ctx.get("_defines", ()), sub_decode_path) if defined is not None: defined_by, defined_spec = defined if issubclass(value.__class__, SequenceOf): for i, _value in enumerate(value): sub_sub_decode_path = sub_decode_path + ( str(i), - decode_path_defby(defined_by), + DecodePathDefBy(defined_by), ) defined_value, defined_tail = defined_spec.decode( memoryview(bytes(_value)), - sub_offset + value.tlen + value.llen, + sub_offset + ( + (value.tlen + value.llen + value.expl_tlen + value.expl_llen) + if value.expled else (value.tlen + value.llen) + ), leavemm=True, decode_path=sub_sub_decode_path, - defines_by_path=defines_by_path, + ctx=ctx, ) if len(defined_tail) > 0: raise DecodeError( @@ -3812,37 +4658,63 @@ class Sequence(Obj): else: defined_value, defined_tail = defined_spec.decode( memoryview(bytes(value)), - sub_offset + value.tlen + value.llen, + sub_offset + ( + (value.tlen + value.llen + value.expl_tlen + value.expl_llen) + if value.expled else (value.tlen + value.llen) + ), leavemm=True, - decode_path=sub_decode_path + (decode_path_defby(defined_by),), - defines_by_path=defines_by_path, + decode_path=sub_decode_path + (DecodePathDefBy(defined_by),), + ctx=ctx, ) if len(defined_tail) > 0: raise DecodeError( "remaining data", klass=self.__class__, - decode_path=sub_decode_path + (decode_path_defby(defined_by),), + decode_path=sub_decode_path + (DecodePathDefBy(defined_by),), offset=offset, ) value.defined = (defined_by, defined_value) - sub_offset += (value.expl_tlvlen if value.expled else value.tlvlen) + value_len = value.fulllen + vlen += value_len + sub_offset += value_len v = v_tail if spec.default is not None and value == spec.default: - # Encoded default values are not valid in DER, - # but we allow that anyway - continue + if ctx_bered or ctx_allow_default_values: + ber_encoded = True + else: + raise DecodeError( + "DEFAULT value met", + klass=self.__class__, + decode_path=sub_decode_path, + offset=sub_offset, + ) values[name] = value - spec_defines = getattr(spec, "defines", None) - if defines_by_path is not None and spec_defines is None: - spec_defines = get_def_by_path(defines_by_path, sub_decode_path) - if spec_defines is not None: - what, schema = spec_defines - defined = schema.get(value, None) - if defined is not None: - defines[what] = (value, defined) - if len(v) > 0: + spec_defines = getattr(spec, "defines", ()) + if len(spec_defines) == 0: + defines_by_path = ctx.get("defines_by_path", ()) + if len(defines_by_path) > 0: + spec_defines = get_def_by_path(defines_by_path, sub_decode_path) + if spec_defines is not None and len(spec_defines) > 0: + for rel_path, schema in spec_defines: + defined = schema.get(value, None) + if defined is not None: + ctx.setdefault("_defines", []).append(( + abs_decode_path(sub_decode_path[:-1], rel_path), + (value, defined), + )) + if lenindef: + if v[:EOC_LEN].tobytes() != EOC: + raise DecodeError( + "no EOC", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + tail = v[EOC_LEN:] + vlen += EOC_LEN + elif len(v) > 0: raise DecodeError( "remaining data", klass=self.__class__, @@ -3855,9 +4727,11 @@ class Sequence(Obj): expl=self._expl, default=self.default, optional=self.optional, - _decoded=(offset, llen, l), + _decoded=(offset, llen, vlen), ) obj._value = values + obj.lenindef = lenindef + obj.ber_encoded = ber_encoded return obj, tail def __repr__(self): @@ -3867,8 +4741,8 @@ class Sequence(Obj): _value = self._value.get(name) if _value is None: continue - cols.append(repr(_value)) - return "%s[%s]" % (value, ", ".join(cols)) + cols.append("%s: %s" % (name, repr(_value))) + return "%s[%s]" % (value, "; ".join(cols)) def pps(self, decode_path=()): yield _pp( @@ -3887,18 +4761,32 @@ class Sequence(Obj): expl_tlen=self.expl_tlen if self.expled else None, expl_llen=self.expl_llen if self.expled else None, expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + lenindef=self.lenindef, + ber_encoded=self.ber_encoded, + bered=self.bered, ) for name in self.specs: value = self._value.get(name) if value is None: continue yield value.pps(decode_path=decode_path + (name,)) + for pp in self.pps_lenindef(decode_path): + yield pp class Set(Sequence): """``SET`` structure type Its usage is identical to :py:class:`pyderasn.Sequence`. + + .. _allow_unordered_set_ctx: + + DER prohibits unordered values encoding and will raise an error + during decode. If If :ref:`bered ` context option is set, + then no error will occure. Also you can disable strict values + ordering check by setting ``"allow_unordered_set": True`` + :ref:`context ` option. """ __slots__ = () tag_default = tag_encode(form=TagFormConstructed, num=17) @@ -3910,7 +4798,7 @@ class Set(Sequence): v = b"".join(raws) return b"".join((self.tag, len_encode(len(v)), v)) - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): + def _decode(self, tlv, offset, decode_path, ctx, tag_only): try: t, tlen, lv = tag_strip(tlv) except DecodeError as err: @@ -3926,8 +4814,22 @@ class Set(Sequence): decode_path=decode_path, offset=offset, ) + if tag_only: + return + lenindef = False + ctx_bered = ctx.get("bered", False) try: l, llen, v = len_decode(lv) + except LenIndefForm as err: + if not ctx_bered: + raise err.__class__( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + l, llen, v = 0, 1, lv[1:] + lenindef = True except DecodeError as err: raise err.__class__( msg=err.msg, @@ -3941,29 +4843,32 @@ class Set(Sequence): klass=self.__class__, offset=offset, ) - v, tail = v[:l], v[l:] + if not lenindef: + v, tail = v[:l], v[l:] + vlen = 0 sub_offset = offset + tlen + llen values = {} + ber_encoded = False + ctx_allow_default_values = ctx.get("allow_default_values", False) + ctx_allow_unordered_set = ctx.get("allow_unordered_set", False) + value_prev = memoryview(v[:0]) specs_items = self.specs.items while len(v) > 0: + if lenindef and v[:EOC_LEN].tobytes() == EOC: + break for name, spec in specs_items(): + sub_decode_path = decode_path + (name,) try: - value, v_tail = spec.decode( + spec.decode( v, sub_offset, leavemm=True, - decode_path=decode_path + (name,), - defines_by_path=defines_by_path, + decode_path=sub_decode_path, + ctx=ctx, + tag_only=True, ) except TagMismatch: continue - sub_offset += ( - value.expl_tlvlen if value.expled else value.tlvlen - ) - v = v_tail - if spec.default is None or value != spec.default: # pragma: no cover - # SeqMixing.test_encoded_default_accepted covers that place - values[name] = value break else: raise TagMismatch( @@ -3971,15 +4876,67 @@ class Set(Sequence): decode_path=decode_path, offset=offset, ) + value, v_tail = spec.decode( + v, + sub_offset, + leavemm=True, + decode_path=sub_decode_path, + ctx=ctx, + ) + value_len = value.fulllen + if value_prev.tobytes() > v[:value_len].tobytes(): + if ctx_bered or ctx_allow_unordered_set: + ber_encoded = True + else: + raise DecodeError( + "unordered " + self.asn1_type_name, + klass=self.__class__, + decode_path=sub_decode_path, + offset=sub_offset, + ) + if spec.default is None or value != spec.default: + pass + elif ctx_bered or ctx_allow_default_values: + ber_encoded = True + else: + raise DecodeError( + "DEFAULT value met", + klass=self.__class__, + decode_path=sub_decode_path, + offset=sub_offset, + ) + values[name] = value + value_prev = v[:value_len] + sub_offset += value_len + vlen += value_len + v = v_tail obj = self.__class__( schema=self.specs, impl=self.tag, expl=self._expl, default=self.default, optional=self.optional, - _decoded=(offset, llen, l), + _decoded=(offset, llen, vlen + (EOC_LEN if lenindef else 0)), ) + if lenindef: + if v[:EOC_LEN].tobytes() != EOC: + raise DecodeError( + "no EOC", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + tail = v[EOC_LEN:] + obj.lenindef = True obj._value = values + if not obj.ready: + raise DecodeError( + "not all values are ready", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + obj.ber_encoded = ber_encoded return obj, tail @@ -4078,6 +5035,12 @@ class SequenceOf(Obj): def ready(self): return all(v.ready for v in self._value) + @property + def bered(self): + if self.expl_lenindef or self.lenindef or self.ber_encoded: + return True + return any(v.bered for v in self._value) + def copy(self): obj = self.__class__(schema=self.spec) obj._bound_min = self._bound_min @@ -4163,7 +5126,7 @@ class SequenceOf(Obj): v = b"".join(self._encoded_values()) return b"".join((self.tag, len_encode(len(v)), v)) - def _decode(self, tlv, offset=0, decode_path=(), defines_by_path=None): + def _decode(self, tlv, offset, decode_path, ctx, tag_only, ordering_check=False): try: t, tlen, lv = tag_strip(tlv) except DecodeError as err: @@ -4179,8 +5142,22 @@ class SequenceOf(Obj): decode_path=decode_path, offset=offset, ) + if tag_only: + return + lenindef = False + ctx_bered = ctx.get("bered", False) try: l, llen, v = len_decode(lv) + except LenIndefForm as err: + if not ctx_bered: + raise err.__class__( + msg=err.msg, + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + l, llen, v = 0, 1, lv[1:] + lenindef = True except DecodeError as err: raise err.__class__( msg=err.msg, @@ -4195,31 +5172,72 @@ class SequenceOf(Obj): decode_path=decode_path, offset=offset, ) - v, tail = v[:l], v[l:] + if not lenindef: + v, tail = v[:l], v[l:] + vlen = 0 sub_offset = offset + tlen + llen _value = [] + ctx_allow_unordered_set = ctx.get("allow_unordered_set", False) + value_prev = memoryview(v[:0]) + ber_encoded = False spec = self.spec while len(v) > 0: + if lenindef and v[:EOC_LEN].tobytes() == EOC: + break + sub_decode_path = decode_path + (str(len(_value)),) value, v_tail = spec.decode( v, sub_offset, leavemm=True, - decode_path=decode_path + (str(len(_value)),), - defines_by_path=defines_by_path, + decode_path=sub_decode_path, + ctx=ctx, ) - sub_offset += (value.expl_tlvlen if value.expled else value.tlvlen) - v = v_tail + value_len = value.fulllen + if ordering_check: + if value_prev.tobytes() > v[:value_len].tobytes(): + if ctx_bered or ctx_allow_unordered_set: + ber_encoded = True + else: + raise DecodeError( + "unordered " + self.asn1_type_name, + klass=self.__class__, + decode_path=sub_decode_path, + offset=sub_offset, + ) + value_prev = v[:value_len] _value.append(value) - obj = self.__class__( - value=_value, - schema=spec, - bounds=(self._bound_min, self._bound_max), - impl=self.tag, - expl=self._expl, - default=self.default, - optional=self.optional, - _decoded=(offset, llen, l), - ) + sub_offset += value_len + vlen += value_len + v = v_tail + try: + obj = self.__class__( + value=_value, + schema=spec, + bounds=(self._bound_min, self._bound_max), + impl=self.tag, + expl=self._expl, + default=self.default, + optional=self.optional, + _decoded=(offset, llen, vlen + (EOC_LEN if lenindef else 0)), + ) + except BoundsError as err: + raise DecodeError( + msg=str(err), + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + if lenindef: + if v[:EOC_LEN].tobytes() != EOC: + raise DecodeError( + "no EOC", + klass=self.__class__, + decode_path=decode_path, + offset=offset, + ) + obj.lenindef = True + tail = v[EOC_LEN:] + obj.ber_encoded = ber_encoded return obj, tail def __repr__(self): @@ -4245,9 +5263,15 @@ class SequenceOf(Obj): expl_tlen=self.expl_tlen if self.expled else None, expl_llen=self.expl_llen if self.expled else None, expl_vlen=self.expl_vlen if self.expled else None, + expl_lenindef=self.expl_lenindef, + lenindef=self.lenindef, + ber_encoded=self.ber_encoded, + bered=self.bered, ) for i, value in enumerate(self._value): yield value.pps(decode_path=decode_path + (str(i),)) + for pp in self.pps_lenindef(decode_path): + yield pp class SetOf(SequenceOf): @@ -4265,6 +5289,16 @@ class SetOf(SequenceOf): v = b"".join(raws) return b"".join((self.tag, len_encode(len(v)), v)) + def _decode(self, tlv, offset, decode_path, ctx, tag_only): + return super(SetOf, self)._decode( + tlv, + offset, + decode_path, + ctx, + tag_only, + ordering_check=True, + ) + def obj_by_path(pypath): # pragma: no cover """Import object specified as string Python path @@ -4301,10 +5335,21 @@ def generic_decoder(): # pragma: no cover __slots__ = () schema = choice - def pprint_any(obj, oids=None): + def pprint_any( + obj, + oids=None, + with_colours=False, + with_decode_path=False, + decode_path_only=(), + ): def _pprint_pps(pps): for pp in pps: if hasattr(pp, "_fields"): + if ( + decode_path_only != () and + pp.decode_path[:len(decode_path_only)] != decode_path_only + ): + continue if pp.asn1_type_name == Choice.asn1_type_name: continue pp_kwargs = pp._asdict() @@ -4315,8 +5360,14 @@ def generic_decoder(): # pragma: no cover oids=oids, with_offsets=True, with_blob=False, + with_colours=with_colours, + with_decode_path=with_decode_path, + decode_path_len_decrease=len(decode_path_only), ) - for row in pp_console_blob(pp): + for row in pp_console_blob( + pp, + decode_path_len_decrease=len(decode_path_only), + ): yield row else: for row in _pprint_pps(pp): @@ -4327,7 +5378,7 @@ def generic_decoder(): # pragma: no cover def main(): # pragma: no cover import argparse - parser = argparse.ArgumentParser(description="PyDERASN ASN.1 DER decoder") + parser = argparse.ArgumentParser(description="PyDERASN ASN.1 BER/DER decoder") parser.add_argument( "--skip", type=int, @@ -4346,6 +5397,25 @@ def main(): # pragma: no cover "--defines-by-path", help="Python path to decoder's defines_by_path", ) + parser.add_argument( + "--nobered", + action="store_true", + help="Disallow BER encoding", + ) + parser.add_argument( + "--print-decode-path", + action="store_true", + help="Print decode paths", + ) + parser.add_argument( + "--decode-path-only", + help="Print only specified decode path", + ) + parser.add_argument( + "--allow-expl-oob", + action="store_true", + help="Allow explicit tag out-of-bound", + ) parser.add_argument( "DERFile", type=argparse.FileType("rb"), @@ -4362,14 +5432,23 @@ def main(): # pragma: no cover pprinter = partial(pprint, big_blobs=True) else: schema, pprinter = generic_decoder() - obj, tail = schema().decode( - der, - defines_by_path=( - None if args.defines_by_path is None - else obj_by_path(args.defines_by_path) + ctx = { + "bered": not args.nobered, + "allow_expl_oob": args.allow_expl_oob, + } + if args.defines_by_path is not None: + ctx["defines_by_path"] = obj_by_path(args.defines_by_path) + obj, tail = schema().decode(der, ctx=ctx) + print(pprinter( + obj, + oids=oids, + with_colours=True if environ.get("NO_COLOR") is None else False, + with_decode_path=args.print_decode_path, + decode_path_only=( + () if args.decode_path_only is None else + tuple(args.decode_path_only.split(":")) ), - ) - print(pprinter(obj, oids=oids)) + )) if tail != b"": print("\nTrailing data: %s" % hexenc(tail))