#!/usr/bin/env python
# coding: utf-8
-# PyDERASN -- Python ASN.1 DER/BER codec with abstract structures
-# Copyright (C) 2017-2020 Sergey Matveev <stargrave@stargrave.org>
+# cython: language_level=3
+# pylint: disable=line-too-long,superfluous-parens,protected-access,too-many-lines
+# pylint: disable=too-many-return-statements,too-many-branches,too-many-statements
+# PyDERASN -- Python ASN.1 DER/CER/BER codec with abstract structures
+# Copyright (C) 2017-2022 Sergey Matveev <stargrave@stargrave.org>
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Lesser General Public License as
#
# You should have received a copy of the GNU Lesser General Public
# License along with this program. If not, see <http://www.gnu.org/licenses/>.
-"""Python ASN.1 DER/BER codec with abstract structures
+"""Python ASN.1 DER/CER/BER codec with abstract structures
-This library allows you to marshal various structures in ASN.1 DER
-format, unmarshal them in BER/CER/DER ones.
+This library allows you to marshal various structures in ASN.1 DER/CER
+format, unmarshal BER/CER/DER ones.
>>> i = Integer(123)
>>> raw = i.encode()
There are :py:func:`pyderasn.tag_ctxp` and :py:func:`pyderasn.tag_ctxc`
functions, allowing you to easily create ``CONTEXT``
``PRIMITIVE``/``CONSTRUCTED`` tags, by specifying only the required tag
-number. Pay attention that explicit tags always have *constructed* tag
-(``tag_ctxc``), but implicit tags for primitive types are primitive
-(``tag_ctxp``).
+number.
+
+.. note::
+
+ EXPLICIT tags always have **constructed** tag. PyDERASN does not
+ explicitly check correctness of schema input here.
+
+.. note::
+
+ Implicit tags have **primitive** (``tag_ctxp``) encoding for
+ primitive values.
::
* :ref:`allow_unordered_set <allow_unordered_set_ctx>`
* :ref:`bered <bered_ctx>`
* :ref:`defines_by_path <defines_by_path_ctx>`
+* :ref:`evgen_mode_upto <evgen_mode_upto_ctx>`
+* :ref:`keep_memoryview <keep_memoryview_ctx>`
.. _pprinting:
Only applicable to BER encoded data. If object has BER-specific
encoding, then ``BER`` will be shown. It does not depend on indefinite
length encoding. ``EOC``, ``BOOLEAN``, ``BIT STRING``, ``OCTET STRING``
- (and its derivatives), ``SET``, ``SET OF`` could be BERed.
+ (and its derivatives), ``SET``, ``SET OF``, ``UTCTime``, ``GeneralizedTime``
+ could be BERed.
+Also it could be helpful to add quick ASN.1 pprinting command in your
+pdb's configuration file::
+
+ alias pp1 import pyderasn ;; print(pyderasn.pprint(%1, oid_maps=(locals().get("OID_STR_TO_NAME", {}),)))
.. _definedby:
______________________________
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.
+in the schema. You can dynamically apply those definitions when calling
+:py:meth:`pyderasn.Obj.decode` method.
Specify ``defines_by_path`` key in the :ref:`decode context <ctx>`. Its
value must be sequence of following tuples::
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 <ctx>` argument to True. Indefinite lengths and
+By default PyDERASN accepts only DER encoded data. By default it encodes
+to DER. But you can optionally enable BER decoding with setting
+``bered`` :ref:`context <ctx>` 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``, ``OBJECT IDENTIFIER``, ``SEQUENCE``, ``SET``, ``SET OF``
- can contain it.
+ STRING``, ``OBJECT IDENTIFIER``, ``SEQUENCE``, ``SET``, ``SET OF``,
+ ``UTCTime``, ``GeneralizedTime`` 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
This option should be used only for skipping some decode errors, just
to see the decoded structure somehow.
+.. _streaming:
+
+Streaming and dealing with huge structures
+------------------------------------------
+
+.. _evgen_mode:
+
+evgen mode
+__________
+
+ASN.1 structures can be huge, they can hold millions of objects inside
+(for example Certificate Revocation Lists (CRL), holding revocation
+state for every previously issued X.509 certificate). CACert.org's 8 MiB
+CRL file takes more than half a gigabyte of memory to hold the decoded
+structure.
+
+If you just simply want to check the signature over the ``tbsCertList``,
+you can create specialized schema with that field represented as
+OctetString for example::
+
+ class TBSCertListFast(Sequence):
+ schema = (
+ [...]
+ ("revokedCertificates", OctetString(
+ impl=SequenceOf.tag_default,
+ optional=True,
+ )),
+ [...]
+ )
+
+This allows you to quickly decode a few fields and check the signature
+over the ``tbsCertList`` bytes.
+
+But how can you get all certificate's serial number from it, after you
+trust that CRL after signature validation? You can use so called
+``evgen`` (event generation) mode, to catch the events/facts of some
+successful object decoding. Let's use command line capabilities::
+
+ $ python -m pyderasn --schema tests.test_crl:CertificateList --evgen revoke.crl
+ 10 [1,1, 1] . . version: Version INTEGER v2 (01) OPTIONAL
+ 15 [1,1, 9] . . . algorithm: OBJECT IDENTIFIER 1.2.840.113549.1.1.13
+ 26 [0,0, 2] . . . parameters: [UNIV 5] ANY OPTIONAL
+ 13 [1,1, 13] . . signature: AlgorithmIdentifier SEQUENCE
+ 34 [1,1, 3] . . . . . . type: AttributeType OBJECT IDENTIFIER 2.5.4.10
+ 39 [0,0, 9] . . . . . . value: [UNIV 19] AttributeValue ANY
+ 32 [1,1, 14] . . . . . 0: AttributeTypeAndValue SEQUENCE
+ 30 [1,1, 16] . . . . 0: RelativeDistinguishedName SET OF
+ [...]
+ 188 [1,1, 1] . . . . userCertificate: CertificateSerialNumber INTEGER 17 (11)
+ 191 [1,1, 13] . . . . . utcTime: UTCTime UTCTime 2003-04-01T14:25:08
+ 191 [0,0, 15] . . . . revocationDate: Time CHOICE utcTime
+ 191 [1,1, 13] . . . . . utcTime: UTCTime UTCTime 2003-04-01T14:25:08
+ 186 [1,1, 18] . . . 0: RevokedCertificate SEQUENCE
+ 208 [1,1, 1] . . . . userCertificate: CertificateSerialNumber INTEGER 20 (14)
+ 211 [1,1, 13] . . . . . utcTime: UTCTime UTCTime 2002-10-01T02:18:01
+ 211 [0,0, 15] . . . . revocationDate: Time CHOICE utcTime
+ 211 [1,1, 13] . . . . . utcTime: UTCTime UTCTime 2002-10-01T02:18:01
+ 206 [1,1, 18] . . . 1: RevokedCertificate SEQUENCE
+ [...]
+ 9144992 [0,0, 15] . . . . revocationDate: Time CHOICE utcTime
+ 9144992 [1,1, 13] . . . . . utcTime: UTCTime UTCTime 2020-02-08T07:25:06
+ 9144985 [1,1, 20] . . . 415755: RevokedCertificate SEQUENCE
+ 181 [1,4,9144821] . . revokedCertificates: RevokedCertificates SEQUENCE OF OPTIONAL
+ 5 [1,4,9144997] . tbsCertList: TBSCertList SEQUENCE
+ 9145009 [1,1, 9] . . algorithm: OBJECT IDENTIFIER 1.2.840.113549.1.1.13
+ 9145020 [0,0, 2] . . parameters: [UNIV 5] ANY OPTIONAL
+ 9145007 [1,1, 13] . signatureAlgorithm: AlgorithmIdentifier SEQUENCE
+ 9145022 [1,3, 513] . signatureValue: BIT STRING 4096 bits
+ 0 [1,4,9145534] CertificateList SEQUENCE
+
+Here we see how decoder works: it decodes SEQUENCE's tag, length, then
+decodes underlying values. It can not tell if SEQUENCE is decoded, so
+the event of the upper level SEQUENCE is the last one we see.
+``version`` field is just a single INTEGER -- it is decoded and event is
+fired immediately. Then we see that ``algorithm`` and ``parameters``
+fields are decoded and only after them the ``signature`` SEQUENCE is
+fired as a successfully decoded. There are 4 events for each revoked
+certificate entry in that CRL: ``userCertificate`` serial number,
+``utcTime`` of ``revocationDate`` CHOICE, ``RevokedCertificate`` itself
+as a one of entity in ``revokedCertificates`` SEQUENCE OF.
+
+We can do that in our ordinary Python code and understand where we are
+by looking at deterministically generated decode paths (do not forget
+about useful ``--print-decode-path`` CLI option). We must use
+:py:meth:`pyderasn.Obj.decode_evgen` method, instead of ordinary
+:py:meth:`pyderasn.Obj.decode`. It is generator yielding ``(decode_path,
+obj, tail)`` tuples::
+
+ for decode_path, obj, _ in CertificateList().decode_evgen(crl_raw):
+ if (
+ len(decode_path) == 4 and
+ decode_path[:2] == ("tbsCertList", "revokedCertificates"),
+ decode_path[3] == "userCertificate"
+ ):
+ print("serial number:", int(obj))
+
+Virtually it does not take any memory except at least needed for single
+object storage. You can easily use that mode to determine required
+object ``.offset`` and ``.*len`` to be able to decode it separately, or
+maybe verify signature upon it just by taking bytes by ``.offset`` and
+``.tlvlen``.
+
+.. _evgen_mode_upto_ctx:
+
+evgen_mode_upto
+_______________
+
+There is full ability to get any kind of data from the CRL in the
+example above. However it is not too convenient to get the whole
+``RevokedCertificate`` structure, that is pretty lightweight and one may
+do not want to disassemble it. You can use ``evgen_mode_upto``
+:ref:`ctx <ctx>` option that semantically equals to
+:ref:`defines_by_path <defines_by_path_ctx>` -- list of decode paths
+mapped to any non-None value. If specified decode path is met, then any
+subsequent objects won't be decoded in evgen mode. That allows us to
+parse the CRL above with fully assembled ``RevokedCertificate``::
+
+ for decode_path, obj, _ in CertificateList().decode_evgen(
+ crl_raw,
+ ctx={"evgen_mode_upto": (
+ (("tbsCertList", "revokedCertificates", any), True),
+ )},
+ ):
+ if (
+ len(decode_path) == 3 and
+ decode_path[:2] == ("tbsCertList", "revokedCertificates"),
+ ):
+ print("serial number:", int(obj["userCertificate"]))
+
+.. note::
+
+ SEQUENCE/SET values with DEFAULT specified are automatically decoded
+ without evgen mode.
+
+.. _mmap:
+
+mmap-ed file
+____________
+
+POSIX compliant systems have ``mmap`` syscall, giving ability to work
+the memory mapped file. You can deal with the file like it was an
+ordinary binary string, allowing you not to load it to the memory first.
+Also you can use them as an input for OCTET STRING, taking no Python
+memory for their storage.
+
+There is convenient :py:func:`pyderasn.file_mmaped` function that
+creates read-only memoryview on the file contents::
+
+ with open("huge", "rb") as fd:
+ raw = file_mmaped(fd)
+ obj = Something.decode(raw)
+
+.. warning::
+
+ mmap maps the **whole** file. So it plays no role if you seek-ed it
+ before. Take the slice of the resulting memoryview with required
+ offset instead.
+
+.. note::
+
+ If you use ZFS as underlying storage, then pay attention that
+ currently most platforms does not deal good with ZFS ARC and ordinary
+ page cache used for mmaps. It can take twice the necessary size in
+ the memory: both in page cache and ZFS ARC.
+
+.. _keep_memoryview_ctx:
+
+That read-only memoryview could be safe to be used as a value inside
+decoded :py:class:`pyderasn.OctetString` and :py:class:`pyderasn.Any`
+objects. You can enable that by setting `"keep_memoryview": True` in
+:ref:`decode context <ctx>`. No OCTET STRING and ANY values will be
+copied to memory. Of course that works only in DER encoding, where the
+value is continuously encoded.
+
+CER encoding
+____________
+
+We can parse any kind of data now, but how can we produce files
+streamingly, without storing their encoded representation in memory?
+SEQUENCE by default encodes in memory all its values, joins them in huge
+binary string, just to know the exact size of SEQUENCE's value for
+encoding it in TLV. DER requires you to know all exact sizes of the
+objects.
+
+You can use CER encoding mode, that slightly differs from the DER, but
+does not require exact sizes knowledge, allowing streaming encoding
+directly to some writer/buffer. Just use
+:py:meth:`pyderasn.Obj.encode_cer` method, providing the writer where
+encoded data will flow::
+
+ with open("result", "wb") as fd:
+ obj.encode_cer(fd.write)
+
+::
+
+ buf = io.BytesIO()
+ obj.encode_cer(buf.write)
+
+If you do not want to create in-memory buffer every time, then you can
+use :py:func:`pyderasn.encode_cer` function::
+
+ data = encode_cer(obj)
+
+Remember that CER is **not valid** DER in most cases, so you **have to**
+use :ref:`bered <bered_ctx>` :ref:`ctx <ctx>` option during its
+decoding. Also currently there is **no** validation that provided CER is
+valid one -- you are sure that it has only valid BER encoding.
+
+.. warning::
+
+ SET OF values can not be streamingly encoded, because they are
+ required to be sorted byte-by-byte. Big SET OF values still will take
+ much memory. Use neither SET nor SET OF values, as modern ASN.1
+ also recommends too.
+
+Do not forget about using :ref:`mmap-ed <mmap>` memoryviews for your
+OCTET STRINGs! They will be streamingly copied from underlying file to
+the buffer using 1 KB chunks.
+
+Some structures require that some of the elements have to be forcefully
+DER encoded. For example ``SignedData`` CMS requires you to encode
+``SignedAttributes`` and X.509 certificates in DER form, allowing you to
+encode everything else in BER. You can tell any of the structures to be
+forcefully encoded in DER during CER encoding, by specifying
+``der_forced=True`` attribute::
+
+ class Certificate(Sequence):
+ schema = (...)
+ der_forced = True
+
+ class SignedAttributes(SetOf):
+ schema = Attribute()
+ bounds = (1, float("+inf"))
+ der_forced = True
+
+.. _agg_octet_string:
+
+agg_octet_string
+________________
+
+In most cases, huge quantity of binary data is stored as OCTET STRING.
+CER encoding splits it on 1 KB chunks. BER allows splitting on various
+levels of chunks inclusion::
+
+ SOME STRING[CONSTRUCTED]
+ OCTET STRING[CONSTRUCTED]
+ OCTET STRING[PRIMITIVE]
+ DATA CHUNK
+ OCTET STRING[PRIMITIVE]
+ DATA CHUNK
+ OCTET STRING[PRIMITIVE]
+ DATA CHUNK
+ OCTET STRING[PRIMITIVE]
+ DATA CHUNK
+ OCTET STRING[CONSTRUCTED]
+ OCTET STRING[PRIMITIVE]
+ DATA CHUNK
+ OCTET STRING[PRIMITIVE]
+ DATA CHUNK
+ OCTET STRING[CONSTRUCTED]
+ OCTET STRING[CONSTRUCTED]
+ OCTET STRING[PRIMITIVE]
+ DATA CHUNK
+
+You can not just take the offset and some ``.vlen`` of the STRING and
+treat it as the payload. If you decode it without
+:ref:`evgen mode <evgen_mode>`, then it will be automatically aggregated
+and ``bytes()`` will give the whole payload contents.
+
+You are forced to use :ref:`evgen mode <evgen_mode>` for decoding for
+small memory footprint. There is convenient
+:py:func:`pyderasn.agg_octet_string` helper for reconstructing the
+payload. Let's assume you have got BER/CER encoded ``ContentInfo`` with
+huge ``SignedData`` and ``EncapsulatedContentInfo``. Let's calculate the
+SHA512 digest of its ``eContent``::
+
+ fd = open("data.p7m", "rb")
+ raw = file_mmaped(fd)
+ ctx = {"bered": True}
+ for decode_path, obj, _ in ContentInfo().decode_evgen(raw, ctx=ctx):
+ if decode_path == ("content",):
+ content = obj
+ break
+ else:
+ raise ValueError("no content found")
+ hasher_state = sha512()
+ def hasher(data):
+ hasher_state.update(data)
+ return len(data)
+ evgens = SignedData().decode_evgen(
+ raw[content.offset:],
+ offset=content.offset,
+ ctx=ctx,
+ )
+ agg_octet_string(evgens, ("encapContentInfo", "eContent"), raw, hasher)
+ fd.close()
+ digest = hasher_state.digest()
+
+Simply replace ``hasher`` with some writeable file's ``fd.write`` to
+copy the payload (without BER/CER encoding interleaved overhead) in it.
+Virtually it won't take memory more than for keeping small structures
+and 1 KB binary chunks.
+
+.. _seqof-iterators:
+
+SEQUENCE OF iterators
+_____________________
+
+You can use iterators as a value in :py:class:`pyderasn.SequenceOf`
+classes. The only difference with providing the full list of objects, is
+that type and bounds checking is done during encoding process. Also
+sequence's value will be emptied after encoding, forcing you to set its
+value again.
+
+This is very useful when you have to create some huge objects, like
+CRLs, with thousands and millions of entities inside. You can write the
+generator taking necessary data from the database and giving the
+``RevokedCertificate`` objects. Only binary representation of that
+objects will take memory during DER encoding.
+
+2-pass DER encoding
+-------------------
+
+There is ability to do 2-pass encoding to DER, writing results directly
+to specified writer (buffer, file, whatever). It could be 1.5+ times
+slower than ordinary encoding, but it takes little memory for 1st pass
+state storing. For example, 1st pass state for CACert.org's CRL with
+~416K of certificate entries takes nearly 3.5 MB of memory.
+``SignedData`` with several gigabyte ``EncapsulatedContentInfo`` takes
+nearly 0.5 KB of memory.
+
+If you use :ref:`mmap-ed <mmap>` memoryviews, :ref:`SEQUENCE OF
+iterators <seqof-iterators>` and write directly to opened file, then
+there is very small memory footprint.
+
+1st pass traverses through all the objects of the structure and returns
+the size of DER encoded structure, together with 1st pass state object.
+That state contains precalculated lengths for various objects inside the
+structure.
+
+::
+
+ fulllen, state = obj.encode1st()
+
+2nd pass takes the writer and 1st pass state. It traverses through all
+the objects again, but writes their encoded representation to the writer.
+
+::
+
+ with open("result", "wb") as fd:
+ obj.encode2nd(fd.write, iter(state))
+
+.. warning::
+
+ You **MUST NOT** use 1st pass state if anything is changed in the
+ objects. It is intended to be used immediately after 1st pass is
+ done!
+
+If you use :ref:`SEQUENCE OF iterators <seqof-iterators>`, then you
+have to reinitialize the values after the 1st pass. And you **have to**
+be sure that the iterator gives exactly the same values as previously.
+Yes, you have to run your iterator twice -- because this is two pass
+encoding mode.
+
+If you want to encode to the memory, then you can use convenient
+:py:func:`pyderasn.encode2pass` helper.
+
+.. _browser:
+
+ASN.1 browser
+-------------
+.. autofunction:: pyderasn.browse
+
Base Obj
--------
.. autoclass:: pyderasn.Obj
Integer
_______
.. autoclass:: pyderasn.Integer
- :members: __init__
+ :members: __init__, named, tohex
BitString
_________
.. autoclass:: pyderasn.BitString
- :members: __init__
+ :members: __init__, bit_len, named
OctetString
___________
PrintableString
_______________
.. autoclass:: pyderasn.PrintableString
- :members: __init__
+ :members: __init__, allow_asterisk, allow_ampersand
+
+IA5String
+_________
+.. autoclass:: pyderasn.IA5String
+
+VisibleString
+_____________
+.. autoclass:: pyderasn.VisibleString
UTCTime
_______
.. autoclass:: pyderasn.UTCTime
- :members: __init__, todatetime
+ :members: __init__, todatetime, totzdatetime
GeneralizedTime
_______________
.. autoclass:: pyderasn.GeneralizedTime
+ :members: __init__, todatetime, totzdatetime
Special types
-------------
Choice
______
.. autoclass:: pyderasn.Choice
- :members: __init__
+ :members: __init__, choice, value
PrimitiveTypes
______________
-------
.. autofunction:: pyderasn.abs_decode_path
+.. autofunction:: pyderasn.agg_octet_string
+.. autofunction:: pyderasn.ascii_visualize
.. autofunction:: pyderasn.colonize_hex
+.. autofunction:: pyderasn.encode2pass
+.. autofunction:: pyderasn.encode_cer
+.. autofunction:: pyderasn.file_mmaped
.. autofunction:: pyderasn.hexenc
.. autofunction:: pyderasn.hexdec
+.. autofunction:: pyderasn.hexdump
.. autofunction:: pyderasn.tag_encode
.. autofunction:: pyderasn.tag_decode
.. autofunction:: pyderasn.tag_ctxp
.. autoclass:: pyderasn.ObjNotReady
.. autoclass:: pyderasn.InvalidValueType
.. autoclass:: pyderasn.BoundsError
+
+.. _cmdline:
+
+Command-line usage
+------------------
+
+You can decode DER/BER files using command line abilities::
+
+ $ python -m pyderasn --schema tests.test_crts:Certificate path/to/file
+
+If there is no schema for your file, then you can try parsing it without,
+but of course IMPLICIT tags will often make it impossible. But result is
+good enough for the certificate above::
+
+ $ python -m pyderasn path/to/file
+ 0 [1,3,1604] . >: SEQUENCE OF
+ 4 [1,3,1453] . . >: SEQUENCE OF
+ 8 [0,0, 5] . . . . >: [0] ANY
+ . . . . . A0:03:02:01:02
+ 13 [1,1, 3] . . . . >: INTEGER 61595
+ 18 [1,1, 13] . . . . >: SEQUENCE OF
+ 20 [1,1, 9] . . . . . . >: OBJECT IDENTIFIER 1.2.840.113549.1.1.5
+ 31 [1,1, 0] . . . . . . >: NULL
+ 33 [1,3, 274] . . . . >: SEQUENCE OF
+ 37 [1,1, 11] . . . . . . >: SET OF
+ 39 [1,1, 9] . . . . . . . . >: SEQUENCE OF
+ 41 [1,1, 3] . . . . . . . . . . >: OBJECT IDENTIFIER 2.5.4.6
+ 46 [1,1, 2] . . . . . . . . . . >: PrintableString PrintableString ES
+ [...]
+ 1409 [1,1, 50] . . . . . . >: SEQUENCE OF
+ 1411 [1,1, 8] . . . . . . . . >: OBJECT IDENTIFIER 1.3.6.1.5.5.7.1.1
+ 1421 [1,1, 38] . . . . . . . . >: OCTET STRING 38 bytes
+ . . . . . . . . . 30:24:30:22:06:08:2B:06:01:05:05:07:30:01:86:16
+ . . . . . . . . . 68:74:74:70:3A:2F:2F:6F:63:73:70:2E:69:70:73:63
+ . . . . . . . . . 61:2E:63:6F:6D:2F
+ 1461 [1,1, 13] . . >: SEQUENCE OF
+ 1463 [1,1, 9] . . . . >: OBJECT IDENTIFIER 1.2.840.113549.1.1.5
+ 1474 [1,1, 0] . . . . >: NULL
+ 1476 [1,2, 129] . . >: BIT STRING 1024 bits
+ . . . 68:EE:79:97:97:DD:3B:EF:16:6A:06:F2:14:9A:6E:CD
+ . . . 9E:12:F7:AA:83:10:BD:D1:7C:98:FA:C7:AE:D4:0E:2C
+ [...]
+
+Human readable OIDs
+___________________
+
+If you have got dictionaries with ObjectIdentifiers, like example one
+from ``tests/test_crts.py``::
+
+ stroid2name = {
+ "1.2.840.113549.1.1.1": "id-rsaEncryption",
+ "1.2.840.113549.1.1.5": "id-sha1WithRSAEncryption",
+ [...]
+ "2.5.4.10": "id-at-organizationName",
+ "2.5.4.11": "id-at-organizationalUnitName",
+ }
+
+then you can pass it to pretty printer to see human readable OIDs::
+
+ $ python -m pyderasn --oids tests.test_crts:stroid2name path/to/file
+ [...]
+ 37 [1,1, 11] . . . . . . >: SET OF
+ 39 [1,1, 9] . . . . . . . . >: SEQUENCE OF
+ 41 [1,1, 3] . . . . . . . . . . >: OBJECT IDENTIFIER id-at-countryName (2.5.4.6)
+ 46 [1,1, 2] . . . . . . . . . . >: PrintableString PrintableString ES
+ 50 [1,1, 18] . . . . . . >: SET OF
+ 52 [1,1, 16] . . . . . . . . >: SEQUENCE OF
+ 54 [1,1, 3] . . . . . . . . . . >: OBJECT IDENTIFIER id-at-stateOrProvinceName (2.5.4.8)
+ 59 [1,1, 9] . . . . . . . . . . >: PrintableString PrintableString Barcelona
+ 70 [1,1, 18] . . . . . . >: SET OF
+ 72 [1,1, 16] . . . . . . . . >: SEQUENCE OF
+ 74 [1,1, 3] . . . . . . . . . . >: OBJECT IDENTIFIER id-at-localityName (2.5.4.7)
+ 79 [1,1, 9] . . . . . . . . . . >: PrintableString PrintableString Barcelona
+ [...]
+
+Decode paths
+____________
+
+Each decoded element has so-called decode path: sequence of structure
+names it is passing during the decode process. Each element has its own
+unique path inside the whole ASN.1 tree. You can print it out with
+``--print-decode-path`` option::
+
+ $ python -m pyderasn --schema path.to:Certificate --print-decode-path path/to/file
+ 0 [1,3,1604] Certificate SEQUENCE []
+ 4 [1,3,1453] . tbsCertificate: TBSCertificate SEQUENCE [tbsCertificate]
+ 10-2 [1,1, 1] . . version: [0] EXPLICIT Version INTEGER v3 OPTIONAL [tbsCertificate:version]
+ 13 [1,1, 3] . . serialNumber: CertificateSerialNumber INTEGER 61595 [tbsCertificate:serialNumber]
+ 18 [1,1, 13] . . signature: AlgorithmIdentifier SEQUENCE [tbsCertificate:signature]
+ 20 [1,1, 9] . . . algorithm: OBJECT IDENTIFIER 1.2.840.113549.1.1.5 [tbsCertificate:signature:algorithm]
+ 31 [0,0, 2] . . . parameters: [UNIV 5] ANY OPTIONAL [tbsCertificate:signature:parameters]
+ . . . . 05:00
+ 33 [0,0, 278] . . issuer: Name CHOICE rdnSequence [tbsCertificate:issuer]
+ 33 [1,3, 274] . . . rdnSequence: RDNSequence SEQUENCE OF [tbsCertificate:issuer:rdnSequence]
+ 37 [1,1, 11] . . . . 0: RelativeDistinguishedName SET OF [tbsCertificate:issuer:rdnSequence:0]
+ 39 [1,1, 9] . . . . . 0: AttributeTypeAndValue SEQUENCE [tbsCertificate:issuer:rdnSequence:0:0]
+ 41 [1,1, 3] . . . . . . type: AttributeType OBJECT IDENTIFIER 2.5.4.6 [tbsCertificate:issuer:rdnSequence:0:0:type]
+ 46 [0,0, 4] . . . . . . value: [UNIV 19] AttributeValue ANY [tbsCertificate:issuer:rdnSequence:0:0:value]
+ . . . . . . . 13:02:45:53
+ 46 [1,1, 2] . . . . . . . DEFINED BY 2.5.4.6: CountryName PrintableString ES [tbsCertificate:issuer:rdnSequence:0:0:value:DEFINED BY 2.5.4.6]
+ [...]
+
+Now you can print only the specified tree, for example signature algorithm::
+
+ $ python -m pyderasn --schema path.to:Certificate --decode-path-only tbsCertificate:signature path/to/file
+ 18 [1,1, 13] AlgorithmIdentifier SEQUENCE
+ 20 [1,1, 9] . algorithm: OBJECT IDENTIFIER 1.2.840.113549.1.1.5
+ 31 [0,0, 2] . parameters: [UNIV 5] ANY OPTIONAL
+ . . 05:00
"""
-from codecs import getdecoder
-from codecs import getencoder
+from array import array
from collections import namedtuple
from collections import OrderedDict
from copy import copy
from datetime import datetime
+from datetime import timedelta
+from io import BytesIO
from math import ceil
-from os import environ
+from operator import attrgetter
from string import ascii_letters
from string import digits
+from struct import Struct as struct_Struct
+from sys import maxsize as sys_maxsize
+from sys import version_info
from unicodedata import category as unicat
-from six import add_metaclass
-from six import binary_type
-from six import byte2int
-from six import indexbytes
-from six import int2byte
-from six import integer_types
-from six import iterbytes
-from six import iteritems
-from six import itervalues
-from six import PY2
-from six import string_types
-from six import text_type
-from six import unichr as six_unichr
-from six.moves import xrange as six_xrange
-
-
try:
from termcolor import colored
except ImportError: # pragma: no cover
def colored(what, *args, **kwargs):
return what
-__version__ = "6.1"
+try:
+ from dateutil.tz import UTC as tzUTC
+except ImportError: # pragma: no cover
+ tzUTC = "missing"
+
+
+__version__ = "9.3"
__all__ = (
+ "agg_octet_string",
"Any",
"BitString",
"BMPString",
"Boolean",
"BoundsError",
"Choice",
+ "colonize_hex",
"DecodeError",
"DecodePathDefBy",
+ "encode2pass",
+ "encode_cer",
"Enumerated",
"ExceedingData",
+ "file_mmaped",
"GeneralizedTime",
"GeneralString",
"GraphicString",
EOC = b"\x00\x00"
EOC_LEN = len(EOC)
LENINDEF = b"\x80" # length indefinite mark
-LENINDEF_PP_CHAR = "I" if PY2 else "∞"
+LENINDEF_PP_CHAR = "∞"
+NAMEDTUPLE_KWARGS = {} if version_info < (3, 6) else {"module": __name__}
+SET01 = frozenset("01")
+DECIMALS = frozenset(digits)
+DECIMAL_SIGNS = ".,"
+NEXT_ATTR_NAME = "__next__"
+
+
+def file_mmaped(fd):
+ """Make mmap-ed memoryview for reading from file
+
+ :param fd: file object
+ :returns: memoryview over read-only mmap-ing of the whole file
+
+ .. warning::
+
+ It does not work under Windows.
+ """
+ import mmap
+ return memoryview(mmap.mmap(fd.fileno(), length=0, prot=mmap.PROT_READ))
+
+
+def pureint(value):
+ if not set(value) <= DECIMALS:
+ raise ValueError("non-pure integer")
+ return int(value)
+
+
+def fractions2float(fractions_raw):
+ pureint(fractions_raw)
+ return float("0." + fractions_raw)
+
+
+def get_def_by_path(defines_by_path, sub_decode_path):
+ """Get define by decode path
+ """
+ for path, define in defines_by_path:
+ if len(path) != len(sub_decode_path):
+ continue
+ for p1, p2 in zip(path, sub_decode_path):
+ if (p1 is not any) and (p1 != p2):
+ break
+ else:
+ return define
########################################################################
decoding process has passed
:param int offset: binary offset where failure happened
"""
- super(DecodeError, self).__init__()
+ super().__init__()
self.msg = msg
self.klass = klass
self.decode_path = decode_path
class ExceedingData(ASN1Error):
def __init__(self, nbytes):
- super(ExceedingData, self).__init__()
+ super().__init__()
self.nbytes = nbytes
def __str__(self):
class ObjUnknown(ASN1Error):
def __init__(self, name):
- super(ObjUnknown, self).__init__()
+ super().__init__()
self.name = name
def __str__(self):
class ObjNotReady(ASN1Error):
def __init__(self, name):
- super(ObjNotReady, self).__init__()
+ super().__init__()
self.name = name
def __str__(self):
class InvalidValueType(ASN1Error):
def __init__(self, expected_types):
- super(InvalidValueType, self).__init__()
+ super().__init__()
self.expected_types = expected_types
def __str__(self):
class BoundsError(ASN1Error):
def __init__(self, bound_min, value, bound_max):
- super(BoundsError, self).__init__()
+ super().__init__()
self.bound_min = bound_min
self.value = value
self.bound_max = bound_max
# Basic coders
########################################################################
-_hexdecoder = getdecoder("hex")
-_hexencoder = getencoder("hex")
-
-
def hexdec(data):
"""Binary data to hexadecimal string convert
"""
- return _hexdecoder(data)[0]
+ return bytes.fromhex(data)
def hexenc(data):
"""Hexadecimal string to binary data convert
"""
- return _hexencoder(data)[0].decode("ascii")
+ return data.hex()
def int_bytes_len(num, byte_len=8):
return bytes(octets)
+int2byte = struct_Struct(">B").pack
+
+
def tag_encode(num, klass=TagClassUniversal, form=TagFormPrimitive):
"""Encode tag to binary form
It returns tuple with three integers, as
:py:func:`pyderasn.tag_encode` accepts.
"""
- first_octet = byte2int(tag)
+ first_octet = tag[0]
klass = first_octet & 0xC0
form = first_octet & 0x20
if first_octet & 0x1F < 0x1F:
return (klass, form, first_octet & 0x1F)
num = 0
- for octet in iterbytes(tag[1:]):
+ for octet in tag[1:]:
num <<= 7
num |= octet & 0x7F
return (klass, form, num)
"""
if len(data) == 0:
raise NotEnoughData("no data at all")
- if byte2int(data) & 0x1F < 31:
+ if data[0] & 0x1F < 31:
return data[:1], 1, data[1:]
i = 0
while True:
i += 1
if i == len(data):
raise DecodeError("unfinished tag")
- if indexbytes(data, i) & 0x80 == 0:
+ if data[i] & 0x80 == 0:
break
+ if i == 1 and data[1] < 0x1F:
+ raise DecodeError("unexpected long form")
+ if i > 1 and data[1] & 0x7F == 0:
+ raise DecodeError("leading zero byte in tag value")
i += 1
return data[:i], i, data[i:]
return int2byte(l)
octets = bytearray(int_bytes_len(l) + 1)
octets[0] = 0x80 | (len(octets) - 1)
- for i in six_xrange(len(octets) - 1, 0, -1):
+ for i in range(len(octets) - 1, 0, -1):
octets[i] = l & 0xFF
l >>= 8
return bytes(octets)
"""
if len(data) == 0:
raise NotEnoughData("no data at all")
- first_octet = byte2int(data)
+ first_octet = data[0]
if first_octet & 0x80 == 0:
return first_octet, 1, data[1:]
octets_num = first_octet & 0x7F
raise NotEnoughData("encoded length is longer than data")
if octets_num == 0:
raise LenIndefForm()
- if byte2int(data[1:]) == 0:
+ if data[1] == 0:
raise DecodeError("leading zeros")
l = 0
- for v in iterbytes(data[1:1 + octets_num]):
+ for v in data[1:1 + octets_num]:
l = (l << 8) | v
if l <= 127:
raise DecodeError("long form instead of short one")
return l, 1 + octets_num, data[1 + octets_num:]
+LEN0 = len_encode(0)
+LEN1 = len_encode(1)
+LEN1K = len_encode(1000)
+
+
+def len_size(l):
+ """How many bytes length field will take
+ """
+ if l < 128:
+ return 1
+ if l < 256: # 1 << 8
+ return 2
+ if l < 65536: # 1 << 16
+ return 3
+ if l < 16777216: # 1 << 24
+ return 4
+ if l < 4294967296: # 1 << 32
+ return 5
+ if l < 1099511627776: # 1 << 40
+ return 6
+ if l < 281474976710656: # 1 << 48
+ return 7
+ if l < 72057594037927936: # 1 << 56
+ return 8
+ raise OverflowError("too big length")
+
+
+def write_full(writer, data):
+ """Fully write provided data
+
+ :param writer: must comply with ``io.RawIOBase.write`` behaviour
+
+ BytesIO does not guarantee that the whole data will be written at
+ once. That function write everything provided, raising an error if
+ ``writer`` returns None.
+ """
+ data = memoryview(data)
+ written = 0
+ while written != len(data):
+ n = writer(data[written:])
+ if n is None:
+ raise ValueError("can not write to buf")
+ written += n
+
+
+# If it is 64-bit system, then use compact 64-bit array of unsigned
+# longs. Use an ordinary list with universal integers otherwise, that
+# is slower.
+if sys_maxsize > 2 ** 32:
+ def state_2pass_new():
+ return array("L")
+else:
+ def state_2pass_new():
+ return []
+
+
########################################################################
# Base class
########################################################################
class AutoAddSlots(type):
def __new__(cls, name, bases, _dict):
_dict["__slots__"] = _dict.get("__slots__", ())
- return type.__new__(cls, name, bases, _dict)
+ return super().__new__(cls, name, bases, _dict)
+
+
+BasicState = namedtuple("BasicState", (
+ "version",
+ "tag",
+ "tag_order",
+ "expl",
+ "default",
+ "optional",
+ "offset",
+ "llen",
+ "vlen",
+ "expl_lenindef",
+ "lenindef",
+ "ber_encoded",
+), **NAMEDTUPLE_KWARGS)
-@add_metaclass(AutoAddSlots)
-class Obj(object):
+class Obj(metaclass=AutoAddSlots):
"""Common ASN.1 object class
All ASN.1 types are inherited from it. It has metaclass that
"""
__slots__ = (
"tag",
+ "_tag_order",
"_value",
"_expl",
"default",
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")
+ if self.tag is None:
+ self._tag_order = None
+ else:
+ tag_class, _, tag_num = tag_decode(
+ self.tag if self._expl is None else self._expl
+ )
+ self._tag_order = (tag_class, tag_num)
if default is not None:
optional = True
self.optional = optional
def __setstate__(self, state):
if state.version != __version__:
raise ValueError("data is pickled by different PyDERASN version")
- self.tag = self.tag_default
- self._value = None
- self._expl = None
- self.default = None
- self.optional = False
- self.offset = 0
- self.llen = 0
- self.vlen = 0
- self.expl_lenindef = False
- self.lenindef = False
- self.ber_encoded = False
+ self.tag = state.tag
+ self._tag_order = state.tag_order
+ self._expl = state.expl
+ self.default = state.default
+ self.optional = state.optional
+ self.offset = state.offset
+ self.llen = state.llen
+ self.vlen = state.vlen
+ self.expl_lenindef = state.expl_lenindef
+ self.lenindef = state.lenindef
+ self.ber_encoded = state.ber_encoded
+
+ @property
+ def tag_order(self):
+ """Tag's (class, number) used for DER/CER sorting
+ """
+ return self._tag_order
+
+ @property
+ def tag_order_cer(self):
+ return self.tag_order
@property
def tlen(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
return len(self.tag)
@property
def tlvlen(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
return self.tlen + self.llen + self.vlen
def __str__(self): # pragma: no cover
- return self.__bytes__() if PY2 else self.__unicode__()
+ return self.__unicode__()
def __ne__(self, their):
return not(self == their)
def _encode(self): # pragma: no cover
raise NotImplementedError()
- def _decode(self, tlv, offset, decode_path, ctx, tag_only): # pragma: no cover
+ def _encode_cer(self, writer):
+ write_full(writer, self._encode())
+
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode): # pragma: no cover
+ yield NotImplemented
+
+ def _encode1st(self, state):
+ raise NotImplementedError()
+
+ def _encode2nd(self, writer, state_iter):
raise NotImplementedError()
def encode(self):
- """Encode the structure
+ """DER encode the structure
:returns: DER representation
"""
return raw
return b"".join((self._expl, len_encode(len(raw)), raw))
+ def encode1st(self, state=None):
+ """Do the 1st pass of 2-pass encoding
+
+ :rtype: (int, array("L"))
+ :returns: full length of encoded data and precalculated various
+ objects lengths
+ """
+ if state is None:
+ state = state_2pass_new()
+ if self._expl is None:
+ return self._encode1st(state)
+ state.append(0)
+ idx = len(state) - 1
+ vlen, _ = self._encode1st(state)
+ state[idx] = vlen
+ fulllen = len(self._expl) + len_size(vlen) + vlen
+ return fulllen, state
+
+ def encode2nd(self, writer, state_iter):
+ """Do the 2nd pass of 2-pass encoding
+
+ :param writer: must comply with ``io.RawIOBase.write`` behaviour
+ :param state_iter: iterator over the 1st pass state (``iter(state)``)
+ """
+ if self._expl is None:
+ self._encode2nd(writer, state_iter)
+ else:
+ write_full(writer, self._expl + len_encode(next(state_iter)))
+ self._encode2nd(writer, state_iter)
+
+ def encode_cer(self, writer):
+ """CER encode the structure to specified writer
+
+ :param writer: must comply with ``io.RawIOBase.write``
+ behaviour. It takes slice to be written and
+ returns number of bytes processed. If it returns
+ None, then exception will be raised
+ """
+ if self._expl is not None:
+ write_full(writer, self._expl + LENINDEF)
+ if getattr(self, "der_forced", False):
+ write_full(writer, self._encode())
+ else:
+ self._encode_cer(writer)
+ if self._expl is not None:
+ write_full(writer, EOC)
+
def hexencode(self):
"""Do hexadecimal encoded :py:meth:`pyderasn.Obj.encode`
"""
: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 decode_path: current decode path (tuples of strings,
+ possibly with DecodePathDefBy) with will be
+ the root for all underlying objects
:param ctx: optional :ref:`context <ctx>` 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)
- :param _ctx_immutable: do we need to ``copy.copy()`` ``ctx``
- before using it?
+ :param bool tag_only: decode only the tag, without length and
+ contents (used only in Choice and Set
+ structures, trying to determine if tag satisfies
+ the schema)
+ :param bool _ctx_immutable: do we need to ``copy.copy()`` ``ctx``
+ before using it?
:returns: (Obj, remaining data)
.. seealso:: :ref:`decoding`
"""
+ result = next(self.decode_evgen(
+ data,
+ offset,
+ leavemm,
+ decode_path,
+ ctx,
+ tag_only,
+ _ctx_immutable,
+ _evgen_mode=False,
+ ))
+ if result is None:
+ return None
+ _, obj, tail = result
+ return obj, tail
+
+ def decode_evgen(
+ self,
+ data,
+ offset=0,
+ leavemm=False,
+ decode_path=(),
+ ctx=None,
+ tag_only=False,
+ _ctx_immutable=True,
+ _evgen_mode=True,
+ ):
+ """Decode with evgen mode on
+
+ That method is identical to :py:meth:`pyderasn.Obj.decode`, but
+ it returns the generator producing ``(decode_path, obj, tail)``
+ values.
+ .. seealso:: :ref:`evgen mode <evgen_mode>`.
+ """
if ctx is None:
ctx = {}
elif _ctx_immutable:
ctx = copy(ctx)
tlv = memoryview(data)
+ if (
+ _evgen_mode and
+ get_def_by_path(ctx.get("evgen_mode_upto", ()), decode_path) is not None
+ ):
+ _evgen_mode = False
if self._expl is None:
- result = self._decode(
- tlv,
- offset,
- decode_path=decode_path,
- ctx=ctx,
- tag_only=tag_only,
- )
- if tag_only:
- return None
- obj, tail = result
+ for result in self._decode(
+ tlv,
+ offset=offset,
+ decode_path=decode_path,
+ ctx=ctx,
+ tag_only=tag_only,
+ evgen_mode=_evgen_mode,
+ ):
+ if tag_only:
+ yield None
+ return
+ _decode_path, obj, tail = result
+ if _decode_path is not decode_path:
+ yield result
else:
try:
t, tlen, lv = tag_strip(tlv)
)
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 None
- obj, tail = result
+ for result in self._decode(
+ v,
+ offset=offset,
+ decode_path=decode_path,
+ ctx=ctx,
+ tag_only=tag_only,
+ evgen_mode=_evgen_mode,
+ ):
+ if tag_only: # pragma: no cover
+ yield None
+ return
+ _decode_path, obj, tail = result
+ if _decode_path is not decode_path:
+ yield result
eoc_expected, tail = tail[:EOC_LEN], tail[EOC_LEN:]
if eoc_expected.tobytes() != EOC:
raise DecodeError(
decode_path=decode_path,
offset=offset,
)
- result = self._decode(
- v,
- offset=offset + tlen + llen,
- decode_path=decode_path,
- ctx=ctx,
- tag_only=tag_only,
- )
- if tag_only: # pragma: no cover
- return None
- 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",
+ for result in self._decode(
+ v,
+ offset=offset + tlen + llen,
+ decode_path=decode_path,
+ ctx=ctx,
+ tag_only=tag_only,
+ evgen_mode=_evgen_mode,
+ ):
+ if tag_only: # pragma: no cover
+ yield None
+ return
+ _decode_path, obj, tail = result
+ if _decode_path is not decode_path:
+ yield 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())
+ yield decode_path, obj, (tail if leavemm else tail.tobytes())
def decod(self, data, offset=0, decode_path=(), ctx=None):
"""Decode the data, check that tail is empty
@property
def expled(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
return self._expl is not None
@property
def expl_tag(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
return self._expl
@property
def expl_tlen(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
return len(self._expl)
@property
def expl_llen(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
if self.expl_lenindef:
return 1
@property
def expl_offset(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
return self.offset - self.expl_tlen - self.expl_llen
@property
def expl_vlen(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
return self.tlvlen
@property
def expl_tlvlen(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
return self.expl_tlen + self.expl_llen + self.expl_vlen
@property
def fulloffset(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
return self.expl_offset if self.expled else self.offset
@property
def fulllen(self):
- """See :ref:`decoding`
+ """.. seealso:: :ref:`decoding`
"""
return self.expl_tlvlen if self.expled else self.tlvlen
)
-class DecodePathDefBy(object):
+def encode_cer(obj):
+ """Encode to CER in memory buffer
+
+ :returns bytes: memory buffer contents
+ """
+ buf = BytesIO()
+ obj.encode_cer(buf.write)
+ return buf.getvalue()
+
+
+def encode2pass(obj):
+ """Encode (2-pass mode) to DER in memory buffer
+
+ :returns bytes: memory buffer contents
+ """
+ buf = BytesIO()
+ _, state = obj.encode1st()
+ obj.encode2nd(buf.write, iter(state))
+ return buf.getvalue()
+
+
+class DecodePathDefBy:
"""DEFINED BY representation inside decode path
"""
__slots__ = ("defined_by",)
"lenindef",
"ber_encoded",
"bered",
-))
+), **NAMEDTUPLE_KWARGS)
def _pp(
def colonize_hex(hexed):
"""Separate hexadecimal string with colons
"""
- return ":".join(hexed[i:i + 2] for i in six_xrange(0, len(hexed), 2))
+ return ":".join(hexed[i:i + 2] for i in range(0, len(hexed), 2))
+
+
+def find_oid_name(asn1_type_name, oid_maps, value):
+ if len(oid_maps) > 0 and asn1_type_name == ObjectIdentifier.asn1_type_name:
+ for oid_map in oid_maps:
+ oid_name = oid_map.get(value)
+ if oid_name is not None:
+ return oid_name
+ return None
def pp_console_row(
col += _colourize("B", "red", with_colours) if pp.bered else " "
cols.append(col)
col = "[%d,%d,%4d]%s" % (
- pp.tlen,
- pp.llen,
- pp.vlen,
+ pp.tlen, pp.llen, pp.vlen,
LENINDEF_PP_CHAR if pp.lenindef else " "
)
col = _colourize(col, "green", with_colours, ())
if isinstance(ent, DecodePathDefBy):
cols.append(_colourize("DEFINED BY", "red", with_colours, ("reverse",)))
value = str(ent.defined_by)
- oid_name = None
- if (
- len(oid_maps) > 0 and
- ent.defined_by.asn1_type_name ==
- ObjectIdentifier.asn1_type_name
- ):
- for oid_map in oid_maps:
- oid_name = oid_map.get(value)
- if oid_name is not None:
- cols.append(_colourize("%s:" % oid_name, "green", with_colours))
- break
+ oid_name = find_oid_name(ent.defined_by.asn1_type_name, oid_maps, value)
if oid_name is None:
cols.append(_colourize("%s:" % value, "white", with_colours, ("reverse",)))
+ else:
+ cols.append(_colourize("%s:" % oid_name, "green", with_colours))
else:
cols.append(_colourize("%s:" % ent, "yellow", with_colours, ("reverse",)))
if pp.expl is not None:
if pp.value is not None:
value = pp.value
cols.append(_colourize(value, "white", with_colours, ("reverse",)))
- if (
- len(oid_maps) > 0 and
- pp.asn1_type_name == ObjectIdentifier.asn1_type_name
- ):
- for oid_map in oid_maps:
- oid_name = oid_map.get(value)
- if oid_name is not None:
- cols.append(_colourize("(%s)" % oid_name, "green", with_colours))
- break
+ oid_name = find_oid_name(pp.asn1_type_name, oid_maps, pp.value)
+ if oid_name is not None:
+ cols.append(_colourize("(%s)" % oid_name, "green", with_colours))
if pp.asn1_type_name == Integer.asn1_type_name:
- hex_repr = hex(int(pp.obj._value))[2:].upper()
- if len(hex_repr) % 2 != 0:
- hex_repr = "0" + hex_repr
cols.append(_colourize(
- "(%s)" % colonize_hex(hex_repr),
- "green",
- with_colours,
+ "(%s)" % colonize_hex(pp.obj.tohex()), "green", with_colours,
))
if with_blob:
- if isinstance(pp.blob, binary_type):
+ if pp.blob.__class__ == bytes:
cols.append(hexenc(pp.blob))
- elif isinstance(pp.blob, tuple):
+ elif pp.blob.__class__ == tuple:
cols.append(", ".join(pp.blob))
if pp.optional:
cols.append(_colourize("OPTIONAL", "red", with_colours))
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):
+ if pp.blob.__class__ == bytes:
blob = hexenc(pp.blob).upper()
- for i in six_xrange(0, len(blob), 32):
+ for i in range(0, len(blob), 32):
chunk = blob[i:i + 32]
yield " ".join(cols + [colonize_hex(chunk)])
- elif isinstance(pp.blob, tuple):
+ elif pp.blob.__class__ == tuple:
yield " ".join(cols + [", ".join(pp.blob)])
with_colours=False,
with_decode_path=False,
decode_path_only=(),
+ decode_path=(),
):
"""Pretty print object
:param Obj obj: object you want to pretty print
- :param oid_maps: list of ``str(OID) <-> human readable string`` dictionary.
+ :param oid_maps: list of ``str(OID) <-> human readable string`` dictionaries.
Its human readable form is printed when OID is met
:param big_blobs: if large binary objects are met (like OctetString
values), do we need to print them too, on separate
else:
for row in _pprint_pps(pp):
yield row
- return "\n".join(_pprint_pps(obj.pps()))
+ return "\n".join(_pprint_pps(obj.pps(decode_path)))
########################################################################
# ASN.1 primitive types
########################################################################
-BooleanState = namedtuple("BooleanState", (
- "version",
- "value",
- "tag",
- "expl",
- "default",
- "optional",
- "offset",
- "llen",
- "vlen",
- "expl_lenindef",
- "lenindef",
- "ber_encoded",
-))
+BooleanState = namedtuple(
+ "BooleanState",
+ BasicState._fields + ("value",),
+ **NAMEDTUPLE_KWARGS
+)
class Boolean(Obj):
:param default: set default value. Type same as in ``value``
:param bool optional: is object ``OPTIONAL`` in sequence
"""
- super(Boolean, self).__init__(impl, expl, default, optional, _decoded)
+ super().__init__(impl, expl, default, optional, _decoded)
self._value = None if value is None else self._value_sanitize(value)
if default is not None:
default = self._value_sanitize(default)
self._value = default
def _value_sanitize(self, value):
- if isinstance(value, bool):
+ if value.__class__ == bool:
return value
if issubclass(value.__class__, Boolean):
return value._value
def __getstate__(self):
return BooleanState(
__version__,
- self._value,
self.tag,
+ self._tag_order,
self._expl,
self.default,
self.optional,
self.expl_lenindef,
self.lenindef,
self.ber_encoded,
+ self._value,
)
def __setstate__(self, state):
- super(Boolean, self).__setstate__(state)
+ super().__setstate__(state)
self._value = state.value
- self.tag = state.tag
- self._expl = state.expl
- self.default = state.default
- self.optional = state.optional
- self.offset = state.offset
- self.llen = state.llen
- self.vlen = state.vlen
- self.expl_lenindef = state.expl_lenindef
- self.lenindef = state.lenindef
- self.ber_encoded = state.ber_encoded
def __nonzero__(self):
self._assert_ready()
return self._value
def __eq__(self, their):
- if isinstance(their, bool):
+ if their.__class__ == bool:
return self._value == their
if not issubclass(their.__class__, Boolean):
return False
def _encode(self):
self._assert_ready()
- return b"".join((
- self.tag,
- len_encode(1),
- (b"\xFF" if self._value else b"\x00"),
- ))
+ return b"".join((self.tag, LEN1, (b"\xFF" if self._value else b"\x00")))
+
+ def _encode1st(self, state):
+ return len(self.tag) + 2, state
- def _decode(self, tlv, offset, decode_path, ctx, tag_only):
+ def _encode2nd(self, writer, state_iter):
+ self._assert_ready()
+ write_full(writer, self._encode())
+
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
try:
t, _, lv = tag_strip(tlv)
except DecodeError as err:
offset=offset,
)
if tag_only:
- return None
+ yield None
+ return
try:
l, _, v = len_decode(lv)
except DecodeError as err:
decode_path=decode_path,
offset=offset,
)
- first_octet = byte2int(v)
+ first_octet = v[0]
ber_encoded = False
if first_octet == 0:
value = False
_decoded=(offset, 1, 1),
)
obj.ber_encoded = ber_encoded
- return obj, v[1:]
+ yield decode_path, obj, v[1:]
def __repr__(self):
return pp_console_row(next(self.pps()))
yield pp
-IntegerState = namedtuple("IntegerState", (
- "version",
- "specs",
- "value",
- "bound_min",
- "bound_max",
- "tag",
- "expl",
- "default",
- "optional",
- "offset",
- "llen",
- "vlen",
- "expl_lenindef",
- "lenindef",
- "ber_encoded",
-))
+IntegerState = namedtuple(
+ "IntegerState",
+ BasicState._fields + ("specs", "value", "bound_min", "bound_max"),
+ **NAMEDTUPLE_KWARGS
+)
class Integer(Obj):
:param default: set default value. Type same as in ``value``
:param bool optional: is object ``OPTIONAL`` in sequence
"""
- super(Integer, self).__init__(impl, expl, default, optional, _decoded)
+ super().__init__(impl, expl, default, optional, _decoded)
self._value = value
specs = getattr(self, "schema", {}) if _specs is None else _specs
- self.specs = specs if isinstance(specs, dict) else dict(specs)
+ self.specs = specs if specs.__class__ == dict else dict(specs)
self._bound_min, self._bound_max = getattr(
self,
"bounds",
self._value = default
def _value_sanitize(self, value):
- if isinstance(value, integer_types):
+ if isinstance(value, int):
pass
elif issubclass(value.__class__, Integer):
value = value._value
- elif isinstance(value, str):
+ elif value.__class__ == str:
value = self.specs.get(value)
if value is None:
raise ObjUnknown("integer value: %s" % value)
def __getstate__(self):
return IntegerState(
__version__,
- self.specs,
- self._value,
- self._bound_min,
- self._bound_max,
self.tag,
+ self._tag_order,
self._expl,
self.default,
self.optional,
self.expl_lenindef,
self.lenindef,
self.ber_encoded,
+ self.specs,
+ self._value,
+ self._bound_min,
+ self._bound_max,
)
def __setstate__(self, state):
- super(Integer, self).__setstate__(state)
+ super().__setstate__(state)
self.specs = state.specs
self._value = state.value
self._bound_min = state.bound_min
self._bound_max = state.bound_max
- self.tag = state.tag
- self._expl = state.expl
- self.default = state.default
- self.optional = state.optional
- self.offset = state.offset
- self.llen = state.llen
- self.vlen = state.vlen
- self.expl_lenindef = state.expl_lenindef
- self.lenindef = state.lenindef
- self.ber_encoded = state.ber_encoded
def __int__(self):
self._assert_ready()
return int(self._value)
+ def tohex(self):
+ """Hexadecimal representation
+
+ Use :py:func:`pyderasn.colonize_hex` for colonizing it.
+ """
+ hex_repr = hex(int(self))[2:].upper()
+ if len(hex_repr) % 2 != 0:
+ hex_repr = "0" + hex_repr
+ return hex_repr
+
def __hash__(self):
self._assert_ready()
- return hash(
- self.tag +
- bytes(self._expl or b"") +
+ return hash(b"".join((
+ self.tag,
+ bytes(self._expl or b""),
str(self._value).encode("ascii"),
- )
+ )))
def __eq__(self, their):
- if isinstance(their, integer_types):
+ if isinstance(their, int):
return self._value == their
if not issubclass(their.__class__, Integer):
return False
@property
def named(self):
- for name, value in iteritems(self.specs):
+ """Return named representation (if exists) of the value
+ """
+ for name, value in self.specs.items():
if value == self._value:
return name
return None
_specs=self.specs,
)
- def _encode(self):
+ def _encode_payload(self):
self._assert_ready()
value = self._value
- if PY2:
- if value == 0:
- octets = bytearray([0])
- elif value < 0:
- value = -value
- value -= 1
- octets = bytearray()
- while value > 0:
- octets.append((value & 0xFF) ^ 0xFF)
- value >>= 8
- if len(octets) == 0 or octets[-1] & 0x80 == 0:
- octets.append(0xFF)
+ bytes_len = ceil(value.bit_length() / 8) or 1
+ while True:
+ try:
+ octets = value.to_bytes(bytes_len, byteorder="big", signed=True)
+ except OverflowError:
+ bytes_len += 1
else:
- octets = bytearray()
- while value > 0:
- octets.append(value & 0xFF)
- value >>= 8
- if octets[-1] & 0x80 > 0:
- octets.append(0x00)
- octets.reverse()
- octets = bytes(octets)
- else:
- bytes_len = ceil(value.bit_length() / 8) or 1
- while True:
- try:
- octets = value.to_bytes(
- bytes_len,
- byteorder="big",
- signed=True,
- )
- except OverflowError:
- bytes_len += 1
- else:
- break
+ break
+ return octets
+
+ def _encode(self):
+ octets = self._encode_payload()
return b"".join((self.tag, len_encode(len(octets)), octets))
- def _decode(self, tlv, offset, decode_path, ctx, tag_only):
+ def _encode1st(self, state):
+ l = len(self._encode_payload())
+ return len(self.tag) + len_size(l) + l, state
+
+ def _encode2nd(self, writer, state_iter):
+ write_full(writer, self._encode())
+
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
try:
t, _, lv = tag_strip(tlv)
except DecodeError as err:
offset=offset,
)
if tag_only:
- return None
+ yield None
+ return
try:
l, llen, v = len_decode(lv)
except DecodeError as err:
offset=offset,
)
v, tail = v[:l], v[l:]
- first_octet = byte2int(v)
+ first_octet = v[0]
if l > 1:
- second_octet = byte2int(v[1:])
+ second_octet = v[1]
if (
((first_octet == 0x00) and (second_octet & 0x80 == 0)) or
((first_octet == 0xFF) and (second_octet & 0x80 != 0))
decode_path=decode_path,
offset=offset,
)
- if PY2:
- value = 0
- if first_octet & 0x80 > 0:
- octets = bytearray()
- for octet in bytearray(v):
- octets.append(octet ^ 0xFF)
- for octet in octets:
- value = (value << 8) | octet
- value += 1
- value = -value
- else:
- for octet in bytearray(v):
- value = (value << 8) | octet
- else:
- value = int.from_bytes(v, byteorder="big", signed=True)
+ value = int.from_bytes(v, byteorder="big", signed=True)
try:
obj = self.__class__(
value=value,
decode_path=decode_path,
offset=offset,
)
- return obj, tail
+ yield decode_path, obj, tail
def __repr__(self):
return pp_console_row(next(self.pps()))
yield pp
-SET01 = frozenset(("0", "1"))
-BitStringState = namedtuple("BitStringState", (
- "version",
- "specs",
- "value",
- "tag",
- "expl",
- "default",
- "optional",
- "offset",
- "llen",
- "vlen",
- "expl_lenindef",
- "lenindef",
- "ber_encoded",
- "tag_constructed",
- "defined",
-))
+BitStringState = namedtuple(
+ "BitStringState",
+ BasicState._fields + ("specs", "value", "tag_constructed", "defined"),
+ **NAMEDTUPLE_KWARGS
+)
class BitString(Obj):
:param default: set default value. Type same as in ``value``
:param bool optional: is object ``OPTIONAL`` in sequence
"""
- super(BitString, self).__init__(impl, expl, default, optional, _decoded)
+ super().__init__(impl, expl, default, optional, _decoded)
specs = getattr(self, "schema", {}) if _specs is None else _specs
- self.specs = specs if isinstance(specs, dict) else dict(specs)
+ self.specs = specs if specs.__class__ == dict else dict(specs)
self._value = None if value is None else self._value_sanitize(value)
if default is not None:
default = self._value_sanitize(default)
bit_len = len(bits)
bits += "0" * ((8 - (bit_len % 8)) % 8)
octets = bytearray(len(bits) // 8)
- for i in six_xrange(len(octets)):
+ for i in range(len(octets)):
octets[i] = int(bits[i * 8:(i * 8) + 8], 2)
return bit_len, bytes(octets)
def _value_sanitize(self, value):
- if isinstance(value, (string_types, binary_type)):
+ if isinstance(value, (str, bytes)):
if (
- isinstance(value, string_types) and
+ isinstance(value, str) and
value.startswith("'")
):
if value.endswith("'B"):
len(value) * 4,
hexdec(value + ("" if len(value) % 2 == 0 else "0")),
)
- if isinstance(value, binary_type):
+ if value.__class__ == bytes:
return (len(value) * 8, value)
- raise InvalidValueType((self.__class__, string_types, binary_type))
- if isinstance(value, tuple):
+ raise InvalidValueType((self.__class__, str, bytes))
+ if value.__class__ == tuple:
if (
len(value) == 2 and
- isinstance(value[0], integer_types) and
- isinstance(value[1], binary_type)
+ isinstance(value[0], int) and
+ value[1].__class__ == bytes
):
return value
bits = []
bits = frozenset(bits)
return self._bits2octets("".join(
("1" if bit in bits else "0")
- for bit in six_xrange(max(bits) + 1)
+ for bit in range(max(bits) + 1)
))
if issubclass(value.__class__, BitString):
return value._value
- raise InvalidValueType((self.__class__, binary_type, string_types))
+ raise InvalidValueType((self.__class__, bytes, str))
@property
def ready(self):
def __getstate__(self):
return BitStringState(
__version__,
- self.specs,
- self._value,
self.tag,
+ self._tag_order,
self._expl,
self.default,
self.optional,
self.expl_lenindef,
self.lenindef,
self.ber_encoded,
+ self.specs,
+ self._value,
self.tag_constructed,
self.defined,
)
def __setstate__(self, state):
- super(BitString, self).__setstate__(state)
+ super().__setstate__(state)
self.specs = state.specs
self._value = state.value
- self.tag = state.tag
- self._expl = state.expl
- self.default = state.default
- self.optional = state.optional
- self.offset = state.offset
- self.llen = state.llen
- self.vlen = state.vlen
- self.expl_lenindef = state.expl_lenindef
- self.lenindef = state.lenindef
- self.ber_encoded = state.ber_encoded
self.tag_constructed = state.tag_constructed
self.defined = state.defined
def __iter__(self):
self._assert_ready()
- for i in six_xrange(self._value[0]):
+ for i in range(self._value[0]):
yield self[i]
@property
def bit_len(self):
+ """Returns number of bits in the string
+ """
self._assert_ready()
return self._value[0]
return self._value[1]
def __eq__(self, their):
- if isinstance(their, bytes):
+ if their.__class__ == bytes:
return self._value[1] == their
if not issubclass(their.__class__, BitString):
return False
@property
def named(self):
- return [name for name, bit in iteritems(self.specs) if self[bit]]
+ """Named representation (if exists) of the bits
+
+ :returns: [str(name), ...]
+ """
+ return [name for name, bit in self.specs.items() if self[bit]]
def __call__(
self,
)
def __getitem__(self, key):
- if isinstance(key, int):
+ if key.__class__ == int:
bit_len, octets = self._value
if key >= bit_len:
return False
- return (
- byte2int(memoryview(octets)[key // 8:]) >>
- (7 - (key % 8))
- ) & 1 == 1
- if isinstance(key, string_types):
+ return memoryview(octets)[key // 8] >> (7 - (key % 8)) & 1 == 1
+ if isinstance(key, str):
value = self.specs.get(key)
if value is None:
raise ObjUnknown("BitString value: %s" % key)
octets,
))
- def _decode_chunk(self, lv, offset, decode_path):
- try:
- l, llen, v = len_decode(lv)
- 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 l == 0:
- raise NotEnoughData(
- "zero length",
- klass=self.__class__,
- decode_path=decode_path,
- offset=offset,
- )
- pad_size = byte2int(v)
- if l == 1 and pad_size != 0:
- raise DecodeError(
- "invalid empty value",
- klass=self.__class__,
- decode_path=decode_path,
- offset=offset,
- )
- if pad_size > 7:
- raise DecodeError(
- "too big pad",
- klass=self.__class__,
- decode_path=decode_path,
- offset=offset,
- )
- if byte2int(v[l - 1:l]) & ((1 << pad_size) - 1) != 0:
- raise DecodeError(
- "invalid pad",
- klass=self.__class__,
- decode_path=decode_path,
- offset=offset,
- )
- v, tail = v[:l], v[l:]
- obj = self.__class__(
- value=((len(v) - 1) * 8 - pad_size, v[1:].tobytes()),
- impl=self.tag,
- expl=self._expl,
- default=self.default,
- optional=self.optional,
- _specs=self.specs,
- _decoded=(offset, llen, l),
- )
- return obj, tail
+ def _encode1st(self, state):
+ self._assert_ready()
+ _, octets = self._value
+ l = len(octets) + 1
+ return len(self.tag) + len_size(l) + l, state
+
+ def _encode2nd(self, writer, state_iter):
+ bit_len, octets = self._value
+ write_full(writer, b"".join((
+ self.tag,
+ len_encode(len(octets) + 1),
+ int2byte((8 - bit_len % 8) % 8),
+ )))
+ write_full(writer, octets)
- def _decode(self, tlv, offset, decode_path, ctx, tag_only):
+ def _encode_cer(self, writer):
+ bit_len, octets = self._value
+ if len(octets) + 1 <= 1000:
+ write_full(writer, self._encode())
+ return
+ write_full(writer, self.tag_constructed)
+ write_full(writer, LENINDEF)
+ for offset in range(0, (len(octets) // 999) * 999, 999):
+ write_full(writer, b"".join((
+ BitString.tag_default,
+ LEN1K,
+ b"\x00",
+ octets[offset:offset + 999],
+ )))
+ tail = octets[offset + 999:]
+ if len(tail) > 0:
+ tail = int2byte((8 - bit_len % 8) % 8) + tail
+ write_full(writer, b"".join((
+ BitString.tag_default,
+ len_encode(len(tail)),
+ tail,
+ )))
+ write_full(writer, EOC)
+
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
try:
t, tlen, lv = tag_strip(tlv)
except DecodeError as err:
)
if t == self.tag:
if tag_only: # pragma: no cover
- return None
- return self._decode_chunk(lv, offset, decode_path)
- 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 None
- lenindef = False
+ yield None
+ return
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,
decode_path=decode_path,
offset=offset,
)
- if not lenindef and l == 0:
+ if 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,
- _ctx_immutable=False,
- )
- 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:
+ pad_size = v[0]
+ if l == 1 and pad_size != 0:
raise DecodeError(
- "no chunks",
+ "invalid empty value",
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
+ if pad_size > 7:
+ raise DecodeError(
+ "too big pad",
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
+ if v[l - 1] & ((1 << pad_size) - 1) != 0:
+ raise DecodeError(
+ "invalid pad",
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
+ v, tail = v[:l], v[l:]
+ bit_len = (len(v) - 1) * 8 - pad_size
obj = self.__class__(
- value=(bit_len, b"".join(values)),
+ value=None if evgen_mode else (bit_len, v[1:].tobytes()),
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)),
+ _decoded=(offset, llen, l),
)
- 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,
+ if evgen_mode:
+ obj._value = (bit_len, None)
+ yield decode_path, obj, tail
+ return
+ if t != self.tag_constructed:
+ raise TagMismatch(
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
+ 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
+ yield None
+ 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:
+ if evgen_mode:
+ for _decode_path, chunk, v_tail in BitString().decode_evgen(
+ v,
+ offset=sub_offset,
+ decode_path=sub_decode_path,
+ leavemm=True,
+ ctx=ctx,
+ _ctx_immutable=False,
+ ):
+ yield _decode_path, chunk, v_tail
+ else:
+ _, chunk, v_tail = next(BitString().decode_evgen(
+ v,
+ offset=sub_offset,
+ decode_path=sub_decode_path,
+ leavemm=True,
+ ctx=ctx,
+ _ctx_immutable=False,
+ _evgen_mode=False,
+ ))
+ 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,
+ )
+ if not evgen_mode:
+ values.append(bytes(chunk))
+ bit_len += chunk.bit_len
+ chunk_last = chunks[-1]
+ if not evgen_mode:
+ values.append(bytes(chunk_last))
+ bit_len += chunk_last.bit_len
+ obj = self.__class__(
+ value=None if evgen_mode else (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)),
)
+ if evgen_mode:
+ obj._value = (bit_len, None)
+ obj.lenindef = lenindef
+ obj.ber_encoded = True
+ yield decode_path, obj, (v[EOC_LEN:] if lenindef else v)
def __repr__(self):
return pp_console_row(next(self.pps()))
if self.ready:
bit_len, blob = self._value
value = "%d bits" % bit_len
- if len(self.specs) > 0:
+ if len(self.specs) > 0 and blob is not None:
blob = tuple(self.named)
yield _pp(
obj=self,
yield pp
-OctetStringState = namedtuple("OctetStringState", (
- "version",
- "value",
- "bound_min",
- "bound_max",
- "tag",
- "expl",
- "default",
- "optional",
- "offset",
- "llen",
- "vlen",
- "expl_lenindef",
- "lenindef",
- "ber_encoded",
- "tag_constructed",
- "defined",
-))
+OctetStringState = namedtuple(
+ "OctetStringState",
+ BasicState._fields + (
+ "value",
+ "bound_min",
+ "bound_max",
+ "tag_constructed",
+ "defined",
+ ),
+ **NAMEDTUPLE_KWARGS
+)
class OctetString(Obj):
>>> 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 <bered_ctx>`, then decoder will fail, because
- of DER restrictions.
+ Memoryviews can be used as a values. If memoryview is made on
+ mmap-ed file, then it does not take storage inside OctetString
+ itself. In CER encoding mode it will be streamed to the specified
+ writer, copying 1 KB chunks.
"""
__slots__ = ("tag_constructed", "_bound_min", "_bound_max", "defined")
tag_default = tag_encode(4)
asn1_type_name = "OCTET STRING"
+ evgen_mode_skip_value = True
+ memoryview_safe = True
def __init__(
self,
default=None,
optional=False,
_decoded=(0, 0, 0),
+ ctx=None,
):
"""
:param value: set the value. Either binary type, or
:param default: set default value. Type same as in ``value``
:param bool optional: is object ``OPTIONAL`` in sequence
"""
- super(OctetString, self).__init__(impl, expl, default, optional, _decoded)
+ super().__init__(impl, expl, default, optional, _decoded)
self._value = value
self._bound_min, self._bound_max = getattr(
self,
)
def _value_sanitize(self, value):
- if isinstance(value, binary_type):
+ if value.__class__ == bytes or value.__class__ == memoryview:
pass
elif issubclass(value.__class__, OctetString):
value = value._value
else:
- raise InvalidValueType((self.__class__, bytes))
+ raise InvalidValueType((self.__class__, bytes, memoryview))
if not self._bound_min <= len(value) <= self._bound_max:
raise BoundsError(self._bound_min, len(value), self._bound_max)
return value
def __getstate__(self):
return OctetStringState(
__version__,
- self._value,
- self._bound_min,
- self._bound_max,
self.tag,
+ self._tag_order,
self._expl,
self.default,
self.optional,
self.expl_lenindef,
self.lenindef,
self.ber_encoded,
+ self._value,
+ self._bound_min,
+ self._bound_max,
self.tag_constructed,
self.defined,
)
def __setstate__(self, state):
- super(OctetString, self).__setstate__(state)
+ super().__setstate__(state)
self._value = state.value
self._bound_min = state.bound_min
self._bound_max = state.bound_max
- self.tag = state.tag
- self._expl = state.expl
- self.default = state.default
- self.optional = state.optional
- self.offset = state.offset
- self.llen = state.llen
- self.vlen = state.vlen
- self.expl_lenindef = state.expl_lenindef
- self.lenindef = state.lenindef
- self.ber_encoded = state.ber_encoded
self.tag_constructed = state.tag_constructed
self.defined = state.defined
def __bytes__(self):
self._assert_ready()
- return self._value
+ return bytes(self._value)
+
+ def memoryview(self):
+ self._assert_ready()
+ return memoryview(self._value)
def __eq__(self, their):
- if isinstance(their, binary_type):
+ if their.__class__ == bytes:
return self._value == their
if not issubclass(their.__class__, OctetString):
return False
self._value,
))
- def _decode_chunk(self, lv, offset, decode_path):
- try:
- l, llen, v = len_decode(lv)
- 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,
- )
- v, tail = v[:l], v[l:]
- try:
- obj = self.__class__(
- value=v.tobytes(),
- bounds=(self._bound_min, self._bound_max),
- impl=self.tag,
- expl=self._expl,
- default=self.default,
- 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),
- klass=self.__class__,
- decode_path=decode_path,
- offset=offset,
- )
- return obj, tail
+ def _encode1st(self, state):
+ self._assert_ready()
+ l = len(self._value)
+ return len(self.tag) + len_size(l) + l, state
- def _decode(self, tlv, offset, decode_path, ctx, tag_only):
+ def _encode2nd(self, writer, state_iter):
+ value = self._value
+ write_full(writer, self.tag + len_encode(len(value)))
+ write_full(writer, value)
+
+ def _encode_cer(self, writer):
+ octets = self._value
+ if len(octets) <= 1000:
+ write_full(writer, self._encode())
+ return
+ write_full(writer, self.tag_constructed)
+ write_full(writer, LENINDEF)
+ for offset in range(0, (len(octets) // 1000) * 1000, 1000):
+ write_full(writer, b"".join((
+ OctetString.tag_default,
+ LEN1K,
+ octets[offset:offset + 1000],
+ )))
+ tail = octets[offset + 1000:]
+ if len(tail) > 0:
+ write_full(writer, b"".join((
+ OctetString.tag_default,
+ len_encode(len(tail)),
+ tail,
+ )))
+ write_full(writer, EOC)
+
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
try:
t, tlen, lv = tag_strip(tlv)
except DecodeError as err:
)
if t == self.tag:
if tag_only:
- return None
- return self._decode_chunk(lv, offset, decode_path)
- 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 None
- lenindef = False
+ yield None
+ return
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,
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,
- _ctx_immutable=False,
- )
- 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
+ v, tail = v[:l], v[l:]
+ if evgen_mode and not self._bound_min <= len(v) <= self._bound_max:
+ raise DecodeError(
+ msg=str(BoundsError(self._bound_min, len(v), self._bound_max)),
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
+ if evgen_mode and self.evgen_mode_skip_value:
+ value = None
+ elif self.memoryview_safe and ctx.get("keep_memoryview", False):
+ value = v
+ else:
+ value = v.tobytes()
try:
obj = self.__class__(
- value=b"".join(bytes(chunk) for chunk in chunks),
+ value=value,
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)),
+ _decoded=(offset, llen, l),
+ ctx=ctx,
)
except DecodeError as err:
raise DecodeError(
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,
- )
+ yield decode_path, obj, tail
+ return
+ if t != self.tag_constructed:
+ raise TagMismatch(
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
+ if not ctx.get("bered", False):
+ raise DecodeError(
+ "unallowed BER constructed encoding",
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
+ if tag_only:
+ yield None
+ 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 = []
+ chunks_count = 0
+ sub_offset = offset + tlen + llen
+ vlen = 0
+ payload_len = 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,
+ )
+ try:
+ if evgen_mode:
+ sub_decode_path = decode_path + (str(chunks_count),)
+ for _decode_path, chunk, v_tail in OctetString().decode_evgen(
+ v,
+ offset=sub_offset,
+ decode_path=sub_decode_path,
+ leavemm=True,
+ ctx=ctx,
+ _ctx_immutable=False,
+ ):
+ yield _decode_path, chunk, v_tail
+ if not chunk.ber_encoded:
+ payload_len += chunk.vlen
+ chunks_count += 1
+ else:
+ sub_decode_path = decode_path + (str(len(chunks)),)
+ _, chunk, v_tail = next(OctetString().decode_evgen(
+ v,
+ offset=sub_offset,
+ decode_path=sub_decode_path,
+ leavemm=True,
+ ctx=ctx,
+ _ctx_immutable=False,
+ _evgen_mode=False,
+ ))
+ chunks.append(chunk)
+ except TagMismatch:
+ raise DecodeError(
+ "expected OctetString encoded chunk",
+ klass=self.__class__,
+ decode_path=sub_decode_path,
+ offset=sub_offset,
+ )
+ sub_offset += chunk.tlvlen
+ vlen += chunk.tlvlen
+ v = v_tail
+ if evgen_mode and not self._bound_min <= payload_len <= self._bound_max:
+ raise DecodeError(
+ msg=str(BoundsError(self._bound_min, payload_len, self._bound_max)),
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
+ try:
+ obj = self.__class__(
+ value=(
+ None if evgen_mode else
+ 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)),
+ ctx=ctx,
+ )
+ 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
+ yield decode_path, obj, (v[EOC_LEN:] if lenindef else v)
def __repr__(self):
return pp_console_row(next(self.pps()))
yield pp
-NullState = namedtuple("NullState", (
- "version",
- "tag",
- "expl",
- "default",
- "optional",
- "offset",
- "llen",
- "vlen",
- "expl_lenindef",
- "lenindef",
- "ber_encoded",
-))
+def agg_octet_string(evgens, decode_path, raw, writer):
+ """Aggregate constructed string (OctetString and its derivatives)
+
+ :param evgens: iterator of generated events
+ :param decode_path: points to the string we want to decode
+ :param raw: slicebable (memoryview, bytearray, etc) with
+ the data evgens are generated on
+ :param writer: buffer.write where string is going to be saved
+ :param writer: where string is going to be saved. Must comply
+ with ``io.RawIOBase.write`` behaviour
+
+ .. seealso:: :ref:`agg_octet_string`
+ """
+ decode_path_len = len(decode_path)
+ for dp, obj, _ in evgens:
+ if dp[:decode_path_len] != decode_path:
+ continue
+ if not obj.ber_encoded:
+ write_full(writer, raw[
+ obj.offset + obj.tlen + obj.llen:
+ obj.offset + obj.tlen + obj.llen + obj.vlen -
+ (EOC_LEN if obj.expl_lenindef else 0)
+ ])
+ if len(dp) == decode_path_len:
+ break
+
+
+NullState = namedtuple("NullState", BasicState._fields, **NAMEDTUPLE_KWARGS)
class Null(Obj):
:param bytes expl: override default tag with ``EXPLICIT`` one
:param bool optional: is object ``OPTIONAL`` in sequence
"""
- super(Null, self).__init__(impl, expl, None, optional, _decoded)
+ super().__init__(impl, expl, None, optional, _decoded)
self.default = None
@property
return NullState(
__version__,
self.tag,
+ self._tag_order,
self._expl,
self.default,
self.optional,
self.ber_encoded,
)
- def __setstate__(self, state):
- super(Null, self).__setstate__(state)
- self.tag = state.tag
- self._expl = state.expl
- self.default = state.default
- self.optional = state.optional
- self.offset = state.offset
- self.llen = state.llen
- self.vlen = state.vlen
- self.expl_lenindef = state.expl_lenindef
- self.lenindef = state.lenindef
- self.ber_encoded = state.ber_encoded
-
def __eq__(self, their):
if not issubclass(their.__class__, Null):
return False
)
def _encode(self):
- return self.tag + len_encode(0)
+ return self.tag + LEN0
- def _decode(self, tlv, offset, decode_path, ctx, tag_only):
+ def _encode1st(self, state):
+ return len(self.tag) + 1, state
+
+ def _encode2nd(self, writer, state_iter):
+ write_full(writer, self.tag + LEN0)
+
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
try:
t, _, lv = tag_strip(tlv)
except DecodeError as err:
offset=offset,
)
if tag_only: # pragma: no cover
- return None
+ yield None
+ return
try:
l, _, v = len_decode(lv)
except DecodeError as err:
optional=self.optional,
_decoded=(offset, 1, 0),
)
- return obj, v
+ yield decode_path, obj, v
def __repr__(self):
return pp_console_row(next(self.pps()))
yield pp
-ObjectIdentifierState = namedtuple("ObjectIdentifierState", (
- "version",
- "value",
- "tag",
- "expl",
- "default",
- "optional",
- "offset",
- "llen",
- "vlen",
- "expl_lenindef",
- "lenindef",
- "ber_encoded",
- "defines",
-))
-
-
-def pureint(value):
- i = int(value)
- if (value[0] in "+- ") or (value[-1] == " "):
- raise ValueError("non-pure integer")
- return i
+ObjectIdentifierState = namedtuple(
+ "ObjectIdentifierState",
+ BasicState._fields + ("value", "defines"),
+ **NAMEDTUPLE_KWARGS
+)
class ObjectIdentifier(Obj):
tuple element is ``{OID: pyderasn.Obj()}``
dictionary, mapping between current OID value
and structure applied to defined field.
- :ref:`Read about DEFINED BY <definedby>`
+
+ .. seealso:: :ref:`definedby`
+
:param bytes impl: override default tag with ``IMPLICIT`` one
:param bytes expl: override default tag with ``EXPLICIT`` one
:param default: set default value. Type same as in ``value``
:param bool optional: is object ``OPTIONAL`` in sequence
"""
- super(ObjectIdentifier, self).__init__(impl, expl, default, optional, _decoded)
+ super().__init__(impl, expl, default, optional, _decoded)
self._value = value
if value is not None:
self._value = self._value_sanitize(value)
self.defines = defines
def __add__(self, their):
+ if their.__class__ == tuple:
+ return self.__class__(self._value + array("L", their))
if isinstance(their, self.__class__):
return self.__class__(self._value + their._value)
- if isinstance(their, tuple):
- return self.__class__(self._value + their)
raise InvalidValueType((self.__class__, tuple))
def _value_sanitize(self, value):
if issubclass(value.__class__, ObjectIdentifier):
return value._value
- if isinstance(value, string_types):
+ if isinstance(value, str):
try:
- value = tuple(pureint(arc) for arc in value.split("."))
+ value = array("L", (pureint(arc) for arc in value.split(".")))
except ValueError:
raise InvalidOID("unacceptable arcs values")
- if isinstance(value, tuple):
+ if value.__class__ == tuple:
+ try:
+ value = array("L", value)
+ except OverflowError as err:
+ raise InvalidOID(repr(err))
+ if value.__class__ is array:
if len(value) < 2:
raise InvalidOID("less than 2 arcs")
first_arc = value[0]
def __getstate__(self):
return ObjectIdentifierState(
__version__,
- self._value,
self.tag,
+ self._tag_order,
self._expl,
self.default,
self.optional,
self.expl_lenindef,
self.lenindef,
self.ber_encoded,
+ self._value,
self.defines,
)
def __setstate__(self, state):
- super(ObjectIdentifier, self).__setstate__(state)
+ super().__setstate__(state)
self._value = state.value
- self.tag = state.tag
- self._expl = state.expl
- self.default = state.default
- self.optional = state.optional
- self.offset = state.offset
- self.llen = state.llen
- self.vlen = state.vlen
- self.expl_lenindef = state.expl_lenindef
- self.lenindef = state.lenindef
- self.ber_encoded = state.ber_encoded
self.defines = state.defines
def __iter__(self):
def __hash__(self):
self._assert_ready()
- return hash(
- self.tag +
- bytes(self._expl or b"") +
+ return hash(b"".join((
+ self.tag,
+ bytes(self._expl or b""),
str(self._value).encode("ascii"),
- )
+ )))
def __eq__(self, their):
- if isinstance(their, tuple):
- return self._value == their
+ if their.__class__ == tuple:
+ return self._value == array("L", their)
if not issubclass(their.__class__, ObjectIdentifier):
return False
return (
optional=self.optional if optional is None else optional,
)
- def _encode(self):
+ def _encode_octets(self):
self._assert_ready()
value = self._value
first_value = value[1]
octets = [zero_ended_encode(first_value)]
for arc in value[2:]:
octets.append(zero_ended_encode(arc))
- v = b"".join(octets)
+ return b"".join(octets)
+
+ def _encode(self):
+ v = self._encode_octets()
return b"".join((self.tag, len_encode(len(v)), v))
- def _decode(self, tlv, offset, decode_path, ctx, tag_only):
+ def _encode1st(self, state):
+ l = len(self._encode_octets())
+ return len(self.tag) + len_size(l) + l, state
+
+ def _encode2nd(self, writer, state_iter):
+ write_full(writer, self._encode())
+
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
try:
t, _, lv = tag_strip(tlv)
except DecodeError as err:
offset=offset,
)
if tag_only: # pragma: no cover
- return None
+ yield None
+ return
try:
l, llen, v = len_decode(lv)
except DecodeError as err:
offset=offset,
)
v, tail = v[:l], v[l:]
- arcs = []
+ arcs = array("L")
ber_encoded = False
while len(v) > 0:
i = 0
arc = 0
while True:
- octet = indexbytes(v, i)
+ octet = v[i]
if i == 0 and octet == 0x80:
if ctx.get("bered", False):
ber_encoded = True
else:
- raise DecodeError("non normalized arc encoding")
+ raise DecodeError(
+ "non normalized arc encoding",
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
arc = (arc << 7) | (octet & 0x7F)
if octet & 0x80 == 0:
- arcs.append(arc)
+ try:
+ arcs.append(arc)
+ except OverflowError:
+ raise DecodeError(
+ "too huge value for local unsigned long",
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
v = v[i + 1:]
break
i += 1
first_arc = 2
second_arc -= 80
obj = self.__class__(
- value=tuple([first_arc, second_arc] + arcs[1:]),
+ value=array("L", (first_arc, second_arc)) + arcs[1:],
impl=self.tag,
expl=self._expl,
default=self.default,
)
if ber_encoded:
obj.ber_encoded = True
- return obj, tail
+ yield decode_path, obj, tail
def __repr__(self):
return pp_console_row(next(self.pps()))
_decoded=(0, 0, 0),
bounds=None, # dummy argument, workability for Integer.decode
):
- super(Enumerated, self).__init__(
+ super().__init__(
value, bounds, impl, expl, default, optional, _specs, _decoded,
)
if len(self.specs) == 0:
def _value_sanitize(self, value):
if isinstance(value, self.__class__):
value = value._value
- elif isinstance(value, integer_types):
- for _value in itervalues(self.specs):
+ elif isinstance(value, int):
+ for _value in self.specs.values():
if _value == value:
break
else:
"unknown integer value: %s" % value,
klass=self.__class__,
)
- elif isinstance(value, string_types):
+ elif isinstance(value, str):
value = self.specs.get(value)
if value is None:
raise ObjUnknown("integer value: %s" % value)
def escape_control_unicode(c):
- if unicat(c).startswith("C"):
+ if unicat(c)[0] == "C":
c = repr(c).lstrip("u").strip("'")
return c
:header-rows: 1
* - Class
- - Text Encoding
+ - Text Encoding, validation
* - :py:class:`pyderasn.UTF8String`
- utf-8
* - :py:class:`pyderasn.NumericString`
- - ascii
+ - proper alphabet validation
* - :py:class:`pyderasn.PrintableString`
- - ascii
+ - proper alphabet validation
* - :py:class:`pyderasn.TeletexString`
- - ascii
+ - iso-8859-1
* - :py:class:`pyderasn.T61String`
- - ascii
+ - iso-8859-1
* - :py:class:`pyderasn.VideotexString`
- iso-8859-1
* - :py:class:`pyderasn.IA5String`
- - ascii
+ - proper alphabet validation
* - :py:class:`pyderasn.GraphicString`
- iso-8859-1
- * - :py:class:`pyderasn.VisibleString`
- - ascii
- * - :py:class:`pyderasn.ISO646String`
- - ascii
+ * - :py:class:`pyderasn.VisibleString`, :py:class:`pyderasn.ISO646String`
+ - proper alphabet validation
* - :py:class:`pyderasn.GeneralString`
- iso-8859-1
* - :py:class:`pyderasn.UniversalString`
- utf-16-be
"""
__slots__ = ()
+ memoryview_safe = False
def _value_sanitize(self, value):
value_raw = None
value_decoded = None
if isinstance(value, self.__class__):
value_raw = value._value
- elif isinstance(value, text_type):
+ elif value.__class__ == str:
value_decoded = value
- elif isinstance(value, binary_type):
+ elif value.__class__ == bytes:
value_raw = value
else:
- raise InvalidValueType((self.__class__, text_type, binary_type))
+ raise InvalidValueType((self.__class__, str, bytes))
try:
value_raw = (
value_decoded.encode(self.encoding)
return value_raw
def __eq__(self, their):
- if isinstance(their, binary_type):
+ if their.__class__ == bytes:
return self._value == their
- if isinstance(their, text_type):
+ if their.__class__ == str:
return self._value == their.encode(self.encoding)
if not isinstance(their, self.__class__):
return False
def __unicode__(self):
if self.ready:
return self._value.decode(self.encoding)
- return text_type(self._value)
+ return str(self._value)
+
+ def memoryview(self):
+ raise ValueError("CommonString does not support .memoryview()")
def __repr__(self):
- return pp_console_row(next(self.pps(no_unicode=PY2)))
+ return pp_console_row(next(self.pps()))
- def pps(self, decode_path=(), no_unicode=False):
+ def pps(self, decode_path=()):
value = None
if self.ready:
- value = (
- hexenc(bytes(self)) if no_unicode else
- "".join(escape_control_unicode(c) for c in self.__unicode__())
- )
+ value = "".join(escape_control_unicode(c) for c in self.__unicode__())
yield _pp(
obj=self,
asn1_type_name=self.asn1_type_name,
asn1_type_name = "UTF8String"
-class AllowableCharsMixin(object):
+class AllowableCharsMixin:
+ __slots__ = ()
+
@property
def allowable_chars(self):
- if PY2:
- return self._allowable_chars
- return frozenset(six_unichr(c) for c in self._allowable_chars)
+ return frozenset(chr(c) for c in self._allowable_chars)
+
+ def _value_sanitize(self, value):
+ value = super()._value_sanitize(value)
+ if not frozenset(value) <= self._allowable_chars:
+ raise DecodeError("non satisfying alphabet value")
+ return value
+
+
+NUMERIC_ALLOWABLE_CHARS = frozenset(digits.encode("ascii") + b" ")
class NumericString(AllowableCharsMixin, CommonString):
>>> NumericString().allowable_chars
frozenset(['0', '1', '2', '3', '4', '5', '6', '7', '8', '9', ' '])
"""
- __slots__ = ()
+ __slots__ = ("_allowable_chars",)
tag_default = tag_encode(18)
encoding = "ascii"
asn1_type_name = "NumericString"
- _allowable_chars = frozenset(digits.encode("ascii") + b" ")
- def _value_sanitize(self, value):
- value = super(NumericString, self)._value_sanitize(value)
- if not frozenset(value) <= self._allowable_chars:
- raise DecodeError("non-numeric value")
- return value
+ def __init__(self, *args, **kwargs):
+ self._allowable_chars = NUMERIC_ALLOWABLE_CHARS
+ super().__init__(*args, **kwargs)
PrintableStringState = namedtuple(
"PrintableStringState",
OctetStringState._fields + ("allowable_chars",),
+ **NAMEDTUPLE_KWARGS
+)
+
+
+PRINTABLE_ALLOWABLE_CHARS = frozenset(
+ (ascii_letters + digits + " '()+,-./:=?").encode("ascii")
)
>>> obj.allow_asterisk, obj.allow_ampersand
(True, False)
"""
- __slots__ = ()
+ __slots__ = ("_allowable_chars",)
tag_default = tag_encode(19)
encoding = "ascii"
asn1_type_name = "PrintableString"
- _allowable_chars = frozenset(
- (ascii_letters + digits + " '()+,-./:=?").encode("ascii")
- )
_asterisk = frozenset("*".encode("ascii"))
_ampersand = frozenset("&".encode("ascii"))
default=None,
optional=False,
_decoded=(0, 0, 0),
+ ctx=None,
allow_asterisk=False,
allow_ampersand=False,
):
:param allow_asterisk: allow asterisk character
:param allow_ampersand: allow ampersand character
"""
+ allowable_chars = PRINTABLE_ALLOWABLE_CHARS
if allow_asterisk:
- self._allowable_chars |= self._asterisk
+ allowable_chars |= self._asterisk
if allow_ampersand:
- self._allowable_chars |= self._ampersand
- super(PrintableString, self).__init__(
- value, bounds, impl, expl, default, optional, _decoded,
+ allowable_chars |= self._ampersand
+ self._allowable_chars = allowable_chars
+ super().__init__(
+ value, bounds, impl, expl, default, optional, _decoded, ctx,
)
@property
"""
return self._ampersand <= self._allowable_chars
- def _value_sanitize(self, value):
- value = super(PrintableString, self)._value_sanitize(value)
- if not frozenset(value) <= self._allowable_chars:
- raise DecodeError("non-printable value")
- return value
-
def __getstate__(self):
return PrintableStringState(
- *super(PrintableString, self).__getstate__(),
+ *super().__getstate__(),
**{"allowable_chars": self._allowable_chars}
)
def __setstate__(self, state):
- super(PrintableString, self).__setstate__(state)
+ super().__setstate__(state)
self._allowable_chars = state.allowable_chars
def __call__(
class TeletexString(CommonString):
__slots__ = ()
tag_default = tag_encode(20)
- encoding = "ascii"
+ encoding = "iso-8859-1"
asn1_type_name = "TeletexString"
asn1_type_name = "VideotexString"
-class IA5String(CommonString):
- __slots__ = ()
+IA5_ALLOWABLE_CHARS = frozenset(b"".join(
+ chr(c).encode("ascii") for c in range(128)
+))
+
+
+class IA5String(AllowableCharsMixin, CommonString):
+ """IA5 string
+
+ Its value is properly sanitized: it is a mix of
+
+ * http://www.itscj.ipsj.or.jp/iso-ir/006.pdf (G)
+ * http://www.itscj.ipsj.or.jp/iso-ir/001.pdf (C0)
+ * DEL character (0x7F)
+
+ It is just 7-bit ASCII.
+
+ >>> IA5String().allowable_chars
+ frozenset(["NUL", ... "DEL"])
+ """
+ __slots__ = ("_allowable_chars",)
tag_default = tag_encode(22)
encoding = "ascii"
asn1_type_name = "IA5"
+ def __init__(self, *args, **kwargs):
+ self._allowable_chars = IA5_ALLOWABLE_CHARS
+ super().__init__(*args, **kwargs)
+
LEN_YYMMDDHHMMSSZ = len("YYMMDDHHMMSSZ")
+LEN_LEN_YYMMDDHHMMSSZ = len_encode(LEN_YYMMDDHHMMSSZ)
+LEN_YYMMDDHHMMSSZ_WITH_LEN = len(LEN_LEN_YYMMDDHHMMSSZ) + LEN_YYMMDDHHMMSSZ
LEN_YYYYMMDDHHMMSSDMZ = len("YYYYMMDDHHMMSSDMZ")
LEN_YYYYMMDDHHMMSSZ = len("YYYYMMDDHHMMSSZ")
+LEN_LEN_YYYYMMDDHHMMSSZ = len_encode(LEN_YYYYMMDDHHMMSSZ)
+
+
+VISIBLE_ALLOWABLE_CHARS = frozenset(b"".join(
+ chr(c).encode("ascii") for c in range(ord(" "), ord("~") + 1)
+))
+
+
+class VisibleString(AllowableCharsMixin, CommonString):
+ """Visible string
+
+ Its value is properly sanitized. ASCII subset from space to tilde is
+ allowed: http://www.itscj.ipsj.or.jp/iso-ir/006.pdf
+
+ >>> VisibleString().allowable_chars
+ frozenset([" ", ... "~"])
+ """
+ __slots__ = ("_allowable_chars",)
+ tag_default = tag_encode(26)
+ encoding = "ascii"
+ asn1_type_name = "VisibleString"
+
+ def __init__(self, *args, **kwargs):
+ self._allowable_chars = VISIBLE_ALLOWABLE_CHARS
+ super().__init__(*args, **kwargs)
+
+
+class ISO646String(VisibleString):
+ __slots__ = ()
+ asn1_type_name = "ISO646String"
-class VisibleString(CommonString):
- __slots__ = ()
- tag_default = tag_encode(26)
- encoding = "ascii"
- asn1_type_name = "VisibleString"
+UTCTimeState = namedtuple(
+ "UTCTimeState",
+ OctetStringState._fields + ("ber_raw",),
+ **NAMEDTUPLE_KWARGS
+)
+
+
+def str_to_time_fractions(value):
+ v = pureint(value)
+ year, v = (v // 10**10), (v % 10**10)
+ month, v = (v // 10**8), (v % 10**8)
+ day, v = (v // 10**6), (v % 10**6)
+ hour, v = (v // 10**4), (v % 10**4)
+ minute, second = (v // 100), (v % 100)
+ return year, month, day, hour, minute, second
class UTCTime(VisibleString):
datetime.datetime(2017, 9, 30, 22, 7, 50)
>>> UTCTime(datetime(2057, 9, 30, 22, 7, 50)).todatetime()
datetime.datetime(1957, 9, 30, 22, 7, 50)
+ >>> UTCTime(datetime(2057, 9, 30, 22, 7, 50)).totzdatetime()
+ datetime.datetime(1957, 9, 30, 22, 7, 50, tzinfo=tzutc())
+
+ If BER encoded value was met, then ``ber_raw`` attribute will hold
+ its raw representation.
.. warning::
- BER encoding is unsupported.
+ Only **naive** ``datetime`` objects are supported.
+ Library assumes that all work is done in UTC.
+
+ .. warning::
+
+ Pay attention that ``UTCTime`` can not hold full year, so all years
+ having < 50 years are treated as 20xx, 19xx otherwise, according to
+ X.509 recommendation. Use ``GeneralizedTime`` instead for
+ removing ambiguity.
+
+ .. warning::
+
+ No strict validation of UTC offsets are made (only applicable to
+ **BER**), but very crude:
+
+ * minutes are not exceeding 60
+ * offset value is not exceeding 14 hours
"""
- __slots__ = ()
+ __slots__ = ("ber_raw",)
tag_default = tag_encode(23)
encoding = "ascii"
asn1_type_name = "UTCTime"
+ evgen_mode_skip_value = False
def __init__(
self,
optional=False,
_decoded=(0, 0, 0),
bounds=None, # dummy argument, workability for OctetString.decode
+ ctx=None,
):
"""
:param value: set the value. Either datetime type, or
:param default: set default value. Type same as in ``value``
:param bool optional: is object ``OPTIONAL`` in sequence
"""
- super(UTCTime, self).__init__(
- None, None, impl, expl, default, optional, _decoded,
- )
+ super().__init__(None, None, impl, expl, None, optional, _decoded, ctx)
self._value = value
+ self.ber_raw = None
if value is not None:
- self._value = self._value_sanitize(value)
+ self._value, self.ber_raw = self._value_sanitize(value, ctx)
+ self.ber_encoded = self.ber_raw is not None
if default is not None:
- default = self._value_sanitize(default)
+ default, _ = self._value_sanitize(default)
self.default = self.__class__(
value=default,
impl=self.tag,
)
if self._value is None:
self._value = default
+ optional = True
+ self.optional = optional
+
+ def _strptime_bered(self, value):
+ year, month, day, hour, minute, _ = str_to_time_fractions(value[:10] + "00")
+ value = value[10:]
+ if len(value) == 0:
+ raise ValueError("no timezone")
+ year += 2000 if year < 50 else 1900
+ decoded = datetime(year, month, day, hour, minute)
+ offset = 0
+ if value[-1] == "Z":
+ value = value[:-1]
+ else:
+ if len(value) < 5:
+ raise ValueError("invalid UTC offset")
+ if value[-5] == "-":
+ sign = -1
+ elif value[-5] == "+":
+ sign = 1
+ else:
+ raise ValueError("invalid UTC offset")
+ v = pureint(value[-4:])
+ offset, v = (60 * (v % 100)), v // 100
+ if offset >= 3600:
+ raise ValueError("invalid UTC offset minutes")
+ offset += 3600 * v
+ if offset > 14 * 3600:
+ raise ValueError("too big UTC offset")
+ offset *= sign
+ value = value[:-5]
+ if len(value) == 0:
+ return offset, decoded
+ if len(value) != 2:
+ raise ValueError("invalid UTC offset seconds")
+ seconds = pureint(value)
+ if seconds >= 60:
+ raise ValueError("invalid seconds value")
+ return offset, decoded + timedelta(seconds=seconds)
def _strptime(self, value):
# datetime.strptime's format: %y%m%d%H%M%SZ
raise ValueError("invalid UTCTime length")
if value[-1] != "Z":
raise ValueError("non UTC timezone")
- return datetime(
- 2000 + int(value[:2]), # %y
- int(value[2:4]), # %m
- int(value[4:6]), # %d
- int(value[6:8]), # %H
- int(value[8:10]), # %M
- int(value[10:12]), # %S
- )
+ year, month, day, hour, minute, second = str_to_time_fractions(value[:-1])
+ year += 2000 if year < 50 else 1900
+ return datetime(year, month, day, hour, minute, second)
- def _value_sanitize(self, value):
- if isinstance(value, binary_type):
+ def _dt_sanitize(self, value):
+ if value.year < 1950 or value.year > 2049:
+ raise ValueError("UTCTime can hold only 1950-2049 years")
+ return value.replace(microsecond=0)
+
+ def _value_sanitize(self, value, ctx=None):
+ if value.__class__ == bytes:
try:
value_decoded = value.decode("ascii")
except (UnicodeEncodeError, UnicodeDecodeError) as err:
raise DecodeError("invalid UTCTime encoding: %r" % err)
+ err = None
try:
- self._strptime(value_decoded)
- except (TypeError, ValueError) as err:
- raise DecodeError("invalid UTCTime format: %r" % err)
- return value
+ return self._strptime(value_decoded), None
+ except (TypeError, ValueError) as _err:
+ err = _err
+ if (ctx is not None) and ctx.get("bered", False):
+ try:
+ offset, _value = self._strptime_bered(value_decoded)
+ _value = _value - timedelta(seconds=offset)
+ return self._dt_sanitize(_value), value
+ except (TypeError, ValueError, OverflowError) as _err:
+ err = _err
+ raise DecodeError(
+ "invalid %s format: %r" % (self.asn1_type_name, err),
+ klass=self.__class__,
+ )
if isinstance(value, self.__class__):
- return value._value
- if isinstance(value, datetime):
- return value.strftime("%y%m%d%H%M%SZ").encode("ascii")
+ return value._value, None
+ if value.__class__ == datetime:
+ if value.tzinfo is not None:
+ raise ValueError("only naive datetime supported")
+ return self._dt_sanitize(value), None
raise InvalidValueType((self.__class__, datetime))
+ def _pp_value(self):
+ if self.ready:
+ value = self._value.isoformat()
+ if self.ber_encoded:
+ value += " (%s)" % self.ber_raw
+ return value
+ return None
+
+ def __unicode__(self):
+ if self.ready:
+ value = self._value.isoformat()
+ if self.ber_encoded:
+ value += " (%s)" % self.ber_raw
+ return value
+ return str(self._pp_value())
+
+ def __getstate__(self):
+ return UTCTimeState(*super().__getstate__(), **{"ber_raw": self.ber_raw})
+
+ def __setstate__(self, state):
+ super().__setstate__(state)
+ self.ber_raw = state.ber_raw
+
+ def __bytes__(self):
+ self._assert_ready()
+ return self._encode_time()
+
def __eq__(self, their):
- if isinstance(their, binary_type):
- return self._value == their
- if isinstance(their, datetime):
+ if their.__class__ == bytes:
+ return self._encode_time() == their
+ if their.__class__ == datetime:
return self.todatetime() == their
if not isinstance(their, self.__class__):
return False
self._expl == their._expl
)
- def todatetime(self):
- """Convert to datetime
+ def _encode_time(self):
+ return self._value.strftime("%y%m%d%H%M%SZ").encode("ascii")
+
+ def _encode(self):
+ self._assert_ready()
+ return b"".join((self.tag, LEN_LEN_YYMMDDHHMMSSZ, self._encode_time()))
- :returns: datetime
+ def _encode1st(self, state):
+ return len(self.tag) + LEN_YYMMDDHHMMSSZ_WITH_LEN, state
- Pay attention that UTCTime can not hold full year, so all years
- having < 50 years are treated as 20xx, 19xx otherwise, according
- to X.509 recomendation.
- """
- value = self._strptime(self._value.decode("ascii"))
- year = value.year % 100
- return datetime(
- year=(2000 + year) if year < 50 else (1900 + year),
- month=value.month,
- day=value.day,
- hour=value.hour,
- minute=value.minute,
- second=value.second,
- )
+ def _encode2nd(self, writer, state_iter):
+ self._assert_ready()
+ write_full(writer, self._encode())
+
+ def _encode_cer(self, writer):
+ write_full(writer, self._encode())
+
+ def todatetime(self):
+ return self._value
+
+ def totzdatetime(self):
+ try:
+ return self._value.replace(tzinfo=tzUTC)
+ except TypeError as err:
+ raise NotImplementedError("Missing dateutil.tz") from err
def __repr__(self):
return pp_console_row(next(self.pps()))
asn1_type_name=self.asn1_type_name,
obj_name=self.__class__.__name__,
decode_path=decode_path,
- value=self.todatetime().isoformat() if self.ready else None,
+ value=self._pp_value(),
optional=self.optional,
default=self == self.default,
impl=None if self.tag == self.tag_default else tag_decode(self.tag),
.. warning::
- BER encoding is unsupported.
+ Only **naive** datetime objects are supported.
+ Library assumes that all work is done in UTC.
.. warning::
- Only microsecond fractions are supported.
+ Only **microsecond** fractions are supported in DER encoding.
:py:exc:`pyderasn.DecodeError` will be raised during decoding of
higher precision values.
.. warning::
- Zero year is unsupported.
+ **BER** encoded data can loss information (accuracy) during
+ decoding because of float transformations.
+
+ .. warning::
+
+ **Zero** year is unsupported.
"""
__slots__ = ()
tag_default = tag_encode(24)
asn1_type_name = "GeneralizedTime"
+ def _dt_sanitize(self, value):
+ return value
+
+ def _strptime_bered(self, value):
+ if len(value) < 4 + 3 * 2:
+ raise ValueError("invalid GeneralizedTime")
+ year, month, day, hour, _, _ = str_to_time_fractions(value[:10] + "0000")
+ decoded = datetime(year, month, day, hour)
+ offset, value = 0, value[10:]
+ if len(value) == 0:
+ return offset, decoded
+ if value[-1] == "Z":
+ value = value[:-1]
+ else:
+ for char, sign in (("-", -1), ("+", 1)):
+ idx = value.rfind(char)
+ if idx == -1:
+ continue
+ offset_raw, value = value[idx + 1:].replace(":", ""), value[:idx]
+ v = pureint(offset_raw)
+ if len(offset_raw) == 4:
+ offset, v = (60 * (v % 100)), v // 100
+ if offset >= 3600:
+ raise ValueError("invalid UTC offset minutes")
+ elif len(offset_raw) == 2:
+ pass
+ else:
+ raise ValueError("invalid UTC offset")
+ offset += 3600 * v
+ if offset > 14 * 3600:
+ raise ValueError("too big UTC offset")
+ offset *= sign
+ break
+ if len(value) == 0:
+ return offset, decoded
+ if value[0] in DECIMAL_SIGNS:
+ return offset, (
+ decoded + timedelta(seconds=3600 * fractions2float(value[1:]))
+ )
+ if len(value) < 2:
+ raise ValueError("stripped minutes")
+ decoded += timedelta(seconds=60 * pureint(value[:2]))
+ value = value[2:]
+ if len(value) == 0:
+ return offset, decoded
+ if value[0] in DECIMAL_SIGNS:
+ return offset, (
+ decoded + timedelta(seconds=60 * fractions2float(value[1:]))
+ )
+ if len(value) < 2:
+ raise ValueError("stripped seconds")
+ decoded += timedelta(seconds=pureint(value[:2]))
+ value = value[2:]
+ if len(value) == 0:
+ return offset, decoded
+ if value[0] not in DECIMAL_SIGNS:
+ raise ValueError("invalid format after seconds")
+ return offset, (
+ decoded + timedelta(microseconds=10**6 * fractions2float(value[1:]))
+ )
+
def _strptime(self, value):
l = len(value)
if l == LEN_YYYYMMDDHHMMSSZ:
# datetime.strptime's format: %Y%m%d%H%M%SZ
if value[-1] != "Z":
raise ValueError("non UTC timezone")
- return datetime(
- int(value[:4]), # %Y
- int(value[4:6]), # %m
- int(value[6:8]), # %d
- int(value[8:10]), # %H
- int(value[10:12]), # %M
- int(value[12:14]), # %S
- )
+ return datetime(*str_to_time_fractions(value[:-1]))
if l >= LEN_YYYYMMDDHHMMSSDMZ:
# datetime.strptime's format: %Y%m%d%H%M%S.%fZ
if value[-1] != "Z":
us_len = len(us)
if us_len > 6:
raise ValueError("only microsecond fractions are supported")
- us = int(us + ("0" * (6 - us_len)))
- decoded = datetime(
- int(value[:4]), # %Y
- int(value[4:6]), # %m
- int(value[6:8]), # %d
- int(value[8:10]), # %H
- int(value[10:12]), # %M
- int(value[12:14]), # %S
- us, # %f
- )
- return decoded
+ us = pureint(us + ("0" * (6 - us_len)))
+ year, month, day, hour, minute, second = str_to_time_fractions(value[:14])
+ return datetime(year, month, day, hour, minute, second, us)
raise ValueError("invalid GeneralizedTime length")
- def _value_sanitize(self, value):
- if isinstance(value, binary_type):
- try:
- value_decoded = value.decode("ascii")
- except (UnicodeEncodeError, UnicodeDecodeError) as err:
- raise DecodeError("invalid GeneralizedTime encoding: %r" % err)
- try:
- self._strptime(value_decoded)
- except (TypeError, ValueError) as err:
- raise DecodeError(
- "invalid GeneralizedTime format: %r" % err,
- klass=self.__class__,
- )
- return value
- if isinstance(value, self.__class__):
- return value._value
- if isinstance(value, datetime):
- encoded = value.strftime("%Y%m%d%H%M%S")
- if value.microsecond > 0:
- encoded = encoded + (".%06d" % value.microsecond).rstrip("0")
- return (encoded + "Z").encode("ascii")
- raise InvalidValueType((self.__class__, datetime))
+ def _encode_time(self):
+ value = self._value
+ encoded = value.strftime("%Y%m%d%H%M%S")
+ if value.microsecond > 0:
+ encoded += (".%06d" % value.microsecond).rstrip("0")
+ return (encoded + "Z").encode("ascii")
- def todatetime(self):
- return self._strptime(self._value.decode("ascii"))
+ def _encode(self):
+ self._assert_ready()
+ value = self._value
+ if value.microsecond > 0:
+ encoded = self._encode_time()
+ return b"".join((self.tag, len_encode(len(encoded)), encoded))
+ return b"".join((self.tag, LEN_LEN_YYYYMMDDHHMMSSZ, self._encode_time()))
+
+ def _encode1st(self, state):
+ self._assert_ready()
+ vlen = len(self._encode_time())
+ return len(self.tag) + len_size(vlen) + vlen, state
+
+ def _encode2nd(self, writer, state_iter):
+ write_full(writer, self._encode())
class GraphicString(CommonString):
asn1_type_name = "GraphicString"
-class ISO646String(VisibleString):
- __slots__ = ()
- asn1_type_name = "ISO646String"
-
-
class GeneralString(CommonString):
__slots__ = ()
tag_default = tag_encode(27)
asn1_type_name = "BMPString"
-ChoiceState = namedtuple("ChoiceState", (
- "version",
- "specs",
- "value",
- "tag",
- "expl",
- "default",
- "optional",
- "offset",
- "llen",
- "vlen",
- "expl_lenindef",
- "lenindef",
- "ber_encoded",
-))
+ChoiceState = namedtuple(
+ "ChoiceState",
+ BasicState._fields + ("specs", "value",),
+ **NAMEDTUPLE_KWARGS
+)
class Choice(Obj):
"""
if impl is not None:
raise ValueError("no implicit tag allowed for CHOICE")
- super(Choice, self).__init__(None, expl, default, optional, _decoded)
+ super().__init__(None, expl, default, optional, _decoded)
if schema is None:
schema = getattr(self, "schema", ())
if len(schema) == 0:
raise ValueError("schema must be specified")
self.specs = (
- schema if isinstance(schema, OrderedDict) else OrderedDict(schema)
+ schema if schema.__class__ == OrderedDict else OrderedDict(schema)
)
self._value = None
if value is not None:
self.default = default_obj
if value is None:
self._value = copy(default_obj._value)
+ if self._expl is not None:
+ tag_class, _, tag_num = tag_decode(self._expl)
+ self._tag_order = (tag_class, tag_num)
def _value_sanitize(self, value):
- if isinstance(value, tuple) and len(value) == 2:
+ if (value.__class__ == tuple) and len(value) == 2:
choice, obj = value
spec = self.specs.get(choice)
if spec is None:
def __getstate__(self):
return ChoiceState(
__version__,
- self.specs,
- copy(self._value),
self.tag,
+ self._tag_order,
self._expl,
self.default,
self.optional,
self.expl_lenindef,
self.lenindef,
self.ber_encoded,
+ self.specs,
+ copy(self._value),
)
def __setstate__(self, state):
- super(Choice, self).__setstate__(state)
+ super().__setstate__(state)
self.specs = state.specs
self._value = state.value
- self._expl = state.expl
- self.default = state.default
- self.optional = state.optional
- self.offset = state.offset
- self.llen = state.llen
- self.vlen = state.vlen
- self.expl_lenindef = state.expl_lenindef
- self.lenindef = state.lenindef
- self.ber_encoded = state.ber_encoded
def __eq__(self, their):
- if isinstance(their, tuple) and len(their) == 2:
+ if (their.__class__ == tuple) and len(their) == 2:
return self._value == their
if not isinstance(their, self.__class__):
return False
@property
def choice(self):
+ """Name of the choice
+ """
self._assert_ready()
return self._value[0]
@property
def value(self):
+ """Value of underlying choice
+ """
self._assert_ready()
return self._value[1]
+ @property
+ def tag_order(self):
+ self._assert_ready()
+ return self._value[1].tag_order if self._tag_order is None else self._tag_order
+
+ @property
+ def tag_order_cer(self):
+ return min(v.tag_order_cer for v in self.specs.values())
+
def __getitem__(self, key):
if key not in self.specs:
raise ObjUnknown(key)
self._assert_ready()
return self._value[1].encode()
- def _decode(self, tlv, offset, decode_path, ctx, tag_only):
- for choice, spec in iteritems(self.specs):
+ def _encode1st(self, state):
+ self._assert_ready()
+ return self._value[1].encode1st(state)
+
+ def _encode2nd(self, writer, state_iter):
+ self._value[1].encode2nd(writer, state_iter)
+
+ def _encode_cer(self, writer):
+ self._assert_ready()
+ self._value[1].encode_cer(writer)
+
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
+ for choice, spec in self.specs.items():
sub_decode_path = decode_path + (choice,)
try:
spec.decode(
offset=offset,
)
if tag_only: # pragma: no cover
- return None
- value, tail = spec.decode(
- tlv,
- offset=offset,
- leavemm=True,
- decode_path=sub_decode_path,
- ctx=ctx,
- _ctx_immutable=False,
- )
+ yield None
+ return
+ if evgen_mode:
+ for _decode_path, value, tail in spec.decode_evgen(
+ tlv,
+ offset=offset,
+ leavemm=True,
+ decode_path=sub_decode_path,
+ ctx=ctx,
+ _ctx_immutable=False,
+ ):
+ yield _decode_path, value, tail
+ else:
+ _, value, tail = next(spec.decode_evgen(
+ tlv,
+ offset=offset,
+ leavemm=True,
+ decode_path=sub_decode_path,
+ ctx=ctx,
+ _ctx_immutable=False,
+ _evgen_mode=False,
+ ))
obj = self.__class__(
schema=self.specs,
expl=self._expl,
_decoded=(offset, 0, value.fulllen),
)
obj._value = (choice, value)
- return obj, tail
+ yield decode_path, obj, tail
def __repr__(self):
value = pp_console_row(next(self.pps()))
))
-AnyState = namedtuple("AnyState", (
- "version",
- "value",
- "tag",
- "expl",
- "optional",
- "offset",
- "llen",
- "vlen",
- "expl_lenindef",
- "lenindef",
- "ber_encoded",
- "defined",
-))
+AnyState = namedtuple(
+ "AnyState",
+ BasicState._fields + ("value", "defined"),
+ **NAMEDTUPLE_KWARGS
+)
class Any(Obj):
"""``ANY`` special type
>>> Any(Integer(-123))
- ANY 020185
+ ANY INTEGER -123 (0X:7B)
>>> a = Any(OctetString(b"hello world").encode())
ANY 040b68656c6c6f20776f726c64
>>> hexenc(bytes(a))
"""
:param value: set the value. Either any kind of pyderasn's
**ready** object, or bytes. Pay attention that
- **no** validation is performed is raw binary value
- is valid TLV
+ **no** validation is performed if raw binary value
+ is valid TLV, except just tag decoding
:param bytes expl: override default tag with ``EXPLICIT`` one
:param bool optional: is object ``OPTIONAL`` in sequence
"""
- super(Any, self).__init__(None, expl, None, optional, _decoded)
- self._value = None if value is None else self._value_sanitize(value)
+ super().__init__(None, expl, None, optional, _decoded)
+ if value is None:
+ self._value = None
+ else:
+ value = self._value_sanitize(value)
+ self._value = value
+ if self._expl is None:
+ if value.__class__ == bytes or value.__class__ == memoryview:
+ tag_class, _, tag_num = tag_decode(tag_strip(value)[0])
+ else:
+ tag_class, tag_num = value.tag_order
+ else:
+ tag_class, _, tag_num = tag_decode(self._expl)
+ self._tag_order = (tag_class, tag_num)
self.defined = None
def _value_sanitize(self, value):
- if isinstance(value, binary_type):
+ if value.__class__ == bytes or value.__class__ == memoryview:
+ if len(value) == 0:
+ raise ValueError("%s value can not be empty" % self.__class__.__name__)
return value
if isinstance(value, self.__class__):
return value._value
- if isinstance(value, Obj):
- return value.encode()
- raise InvalidValueType((self.__class__, Obj, binary_type))
+ if not isinstance(value, Obj):
+ raise InvalidValueType((self.__class__, Obj, bytes))
+ return value
@property
def ready(self):
return self._value is not None
+ @property
+ def tag_order(self):
+ self._assert_ready()
+ return self._tag_order
+
@property
def bered(self):
if self.expl_lenindef or self.lenindef:
def __getstate__(self):
return AnyState(
__version__,
- self._value,
self.tag,
+ self._tag_order,
self._expl,
+ None,
self.optional,
self.offset,
self.llen,
self.expl_lenindef,
self.lenindef,
self.ber_encoded,
+ self._value,
self.defined,
)
def __setstate__(self, state):
- super(Any, self).__setstate__(state)
+ super().__setstate__(state)
self._value = state.value
- self.tag = state.tag
- self._expl = state.expl
- self.optional = state.optional
- self.offset = state.offset
- self.llen = state.llen
- self.vlen = state.vlen
- self.expl_lenindef = state.expl_lenindef
- self.lenindef = state.lenindef
- self.ber_encoded = state.ber_encoded
self.defined = state.defined
def __eq__(self, their):
- if isinstance(their, binary_type):
- return self._value == their
+ if their.__class__ == bytes or their.__class__ == memoryview:
+ if self._value.__class__ == bytes or their.__class__ == memoryview:
+ return self._value == their
+ return self._value.encode() == their
if issubclass(their.__class__, Any):
- return self._value == their._value
+ if self.ready and their.ready:
+ return self.memoryview() == their.memoryview()
+ return self.ready == their.ready
return False
def __call__(
def __bytes__(self):
self._assert_ready()
- return self._value
+ value = self._value
+ if value.__class__ == bytes:
+ return value
+ if value.__class__ == memoryview:
+ return bytes(value)
+ return self._value.encode()
+
+ def memoryview(self):
+ self._assert_ready()
+ value = self._value
+ if value.__class__ == memoryview:
+ return memoryview(value)
+ return memoryview(bytes(self))
@property
def tlen(self):
def _encode(self):
self._assert_ready()
- return self._value
+ value = self._value
+ if value.__class__ == bytes or value.__class__ == memoryview:
+ return bytes(self)
+ return value.encode()
+
+ def _encode1st(self, state):
+ self._assert_ready()
+ value = self._value
+ if value.__class__ == bytes or value.__class__ == memoryview:
+ return len(value), state
+ return value.encode1st(state)
+
+ def _encode2nd(self, writer, state_iter):
+ value = self._value
+ if value.__class__ == bytes or value.__class__ == memoryview:
+ write_full(writer, value)
+ else:
+ value.encode2nd(writer, state_iter)
+
+ def _encode_cer(self, writer):
+ self._assert_ready()
+ value = self._value
+ if value.__class__ == bytes or value.__class__ == memoryview:
+ write_full(writer, value)
+ else:
+ value.encode_cer(writer)
- def _decode(self, tlv, offset, decode_path, ctx, tag_only):
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
try:
t, tlen, lv = tag_strip(tlv)
except DecodeError as err:
chunk_i += 1
tlvlen = tlen + llen + vlen + EOC_LEN
obj = self.__class__(
- value=tlv[:tlvlen].tobytes(),
+ value=None if evgen_mode else tlv[:tlvlen].tobytes(),
expl=self._expl,
optional=self.optional,
_decoded=(offset, 0, tlvlen),
)
obj.lenindef = True
obj.tag = t.tobytes()
- return obj, v[EOC_LEN:]
+ yield decode_path, obj, v[EOC_LEN:]
+ return
except DecodeError as err:
raise err.__class__(
msg=err.msg,
)
tlvlen = tlen + llen + l
v, tail = tlv[:tlvlen], v[l:]
+ if evgen_mode:
+ value = None
+ elif ctx.get("keep_memoryview", False):
+ value = v
+ else:
+ value = v.tobytes()
obj = self.__class__(
- value=v.tobytes(),
+ value=value,
expl=self._expl,
optional=self.optional,
_decoded=(offset, 0, tlvlen),
)
obj.tag = t.tobytes()
- return obj, tail
+ yield decode_path, obj, tail
def __repr__(self):
return pp_console_row(next(self.pps()))
def pps(self, decode_path=()):
+ value = self._value
+ if value is None:
+ pass
+ elif value.__class__ == bytes or value.__class__ == memoryview:
+ value = None
+ else:
+ value = repr(value)
yield _pp(
obj=self,
asn1_type_name=self.asn1_type_name,
obj_name=self.__class__.__name__,
decode_path=decode_path,
- blob=self._value if self.ready else None,
+ value=value,
+ blob=self._value if (
+ self._value.__class__ == bytes or
+ value.__class__ == memoryview
+ ) else None,
optional=self.optional,
default=self == self.default,
impl=None if self.tag == self.tag_default else tag_decode(self.tag),
# ASN.1 constructed types
########################################################################
-def get_def_by_path(defines_by_path, sub_decode_path):
- """Get define by decode path
- """
- for path, define in defines_by_path:
- if len(path) != len(sub_decode_path):
- continue
- for p1, p2 in zip(path, sub_decode_path):
- if (p1 != any) and (p1 != p2):
- break
- else:
- return define
-
-
def abs_decode_path(decode_path, rel_path):
"""Create an absolute decode path from current and relative ones
"""
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
-
-
-SequenceState = namedtuple("SequenceState", (
- "version",
- "specs",
- "value",
- "tag",
- "expl",
- "default",
- "optional",
- "offset",
- "llen",
- "vlen",
- "expl_lenindef",
- "lenindef",
- "ber_encoded",
-))
+ if rel_path[0] == "..":
+ return abs_decode_path(decode_path[:-1], rel_path[1:])
+ return decode_path + rel_path
+
+
+SequenceState = namedtuple(
+ "SequenceState",
+ BasicState._fields + ("specs", "value",),
+ **NAMEDTUPLE_KWARGS
+)
+
+
+class SequenceEncode1stMixin:
+ __slots__ = ()
+
+ def _encode1st(self, state):
+ state.append(0)
+ idx = len(state) - 1
+ vlen = 0
+ for v in self._values_for_encoding():
+ l, _ = v.encode1st(state)
+ vlen += l
+ state[idx] = vlen
+ return len(self.tag) + len_size(vlen) + vlen, state
-class Sequence(Obj):
+class Sequence(SequenceEncode1stMixin, Obj):
"""``SEQUENCE`` structure type
You have to make specification of sequence::
defaulted values existence validation by setting
``"allow_default_values": True`` :ref:`context <ctx>` option.
+ All values with DEFAULT specified are decoded atomically in
+ :ref:`evgen mode <evgen_mode>`. If DEFAULT value is some kind of
+ SEQUENCE, then it will be yielded as a single element, not
+ disassembled. That is required for DEFAULT existence check.
+
Two sequences are equal if they have equal specification (schema),
implicit/explicit tagging and the same values.
"""
optional=False,
_decoded=(0, 0, 0),
):
- super(Sequence, self).__init__(impl, expl, default, optional, _decoded)
+ super().__init__(impl, expl, default, optional, _decoded)
if schema is None:
schema = getattr(self, "schema", ())
self.specs = (
- schema if isinstance(schema, OrderedDict) else OrderedDict(schema)
+ schema if schema.__class__ == OrderedDict else OrderedDict(schema)
)
self._value = {}
if value is not None:
@property
def ready(self):
- for name, spec in iteritems(self.specs):
+ for name, spec in self.specs.items():
value = self._value.get(name)
if value is None:
if spec.optional:
def bered(self):
if self.expl_lenindef or self.lenindef or self.ber_encoded:
return True
- return any(value.bered for value in itervalues(self._value))
+ return any(value.bered for value in self._value.values())
def __getstate__(self):
return SequenceState(
__version__,
- self.specs,
- {k: copy(v) for k, v in iteritems(self._value)},
self.tag,
+ self._tag_order,
self._expl,
self.default,
self.optional,
self.expl_lenindef,
self.lenindef,
self.ber_encoded,
+ self.specs,
+ {k: copy(v) for k, v in self._value.items()},
)
def __setstate__(self, state):
- super(Sequence, self).__setstate__(state)
+ super().__setstate__(state)
self.specs = state.specs
self._value = state.value
- self.tag = state.tag
- self._expl = state.expl
- self.default = state.default
- self.optional = state.optional
- self.offset = state.offset
- self.llen = state.llen
- self.vlen = state.vlen
- self.expl_lenindef = state.expl_lenindef
- self.lenindef = state.lenindef
- self.ber_encoded = state.ber_encoded
def __eq__(self, their):
if not isinstance(their, self.__class__):
return spec.default
return None
- def _encoded_values(self):
- raws = []
- for name, spec in iteritems(self.specs):
+ def _values_for_encoding(self):
+ for name, spec in self.specs.items():
value = self._value.get(name)
if value is None:
if spec.optional:
continue
raise ObjNotReady(name)
- raws.append(value.encode())
- return raws
+ yield value
def _encode(self):
- v = b"".join(self._encoded_values())
+ v = b"".join(v.encode() for v in self._values_for_encoding())
return b"".join((self.tag, len_encode(len(v)), v))
- def _decode(self, tlv, offset, decode_path, ctx, tag_only):
+ def _encode2nd(self, writer, state_iter):
+ write_full(writer, self.tag + len_encode(next(state_iter)))
+ for v in self._values_for_encoding():
+ v.encode2nd(writer, state_iter)
+
+ def _encode_cer(self, writer):
+ write_full(writer, self.tag + LENINDEF)
+ for v in self._values_for_encoding():
+ v.encode_cer(writer)
+ write_full(writer, EOC)
+
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
try:
t, tlen, lv = tag_strip(tlv)
except DecodeError as err:
offset=offset,
)
if tag_only: # pragma: no cover
- return None
+ yield None
+ return
lenindef = False
ctx_bered = ctx.get("bered", False)
try:
values = {}
ber_encoded = False
ctx_allow_default_values = ctx.get("allow_default_values", False)
- for name, spec in iteritems(self.specs):
+ for name, spec in self.specs.items():
if spec.optional and (
(lenindef and v[:EOC_LEN].tobytes() == EOC) or
len(v) == 0
):
continue
+ spec_defaulted = spec.default is not None
sub_decode_path = decode_path + (name,)
try:
- value, v_tail = spec.decode(
- v,
- sub_offset,
- leavemm=True,
- decode_path=sub_decode_path,
- ctx=ctx,
- _ctx_immutable=False,
- )
+ if evgen_mode and not spec_defaulted:
+ for _decode_path, value, v_tail in spec.decode_evgen(
+ v,
+ sub_offset,
+ leavemm=True,
+ decode_path=sub_decode_path,
+ ctx=ctx,
+ _ctx_immutable=False,
+ ):
+ yield _decode_path, value, v_tail
+ else:
+ _, value, v_tail = next(spec.decode_evgen(
+ v,
+ sub_offset,
+ leavemm=True,
+ decode_path=sub_decode_path,
+ ctx=ctx,
+ _ctx_immutable=False,
+ _evgen_mode=False,
+ ))
except TagMismatch as err:
if (len(err.decode_path) == len(decode_path) + 1) and spec.optional:
continue
raise
defined = get_def_by_path(ctx.get("_defines", ()), sub_decode_path)
- if defined is not None:
+ if not evgen_mode and defined is not None:
defined_by, defined_spec = defined
if issubclass(value.__class__, SequenceOf):
for i, _value in enumerate(value):
vlen += value_len
sub_offset += value_len
v = v_tail
- if spec.default is not None and value == spec.default:
- 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", ())
- 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 spec_defaulted:
+ if evgen_mode:
+ yield sub_decode_path, value, v_tail
+ if value == spec.default:
+ 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,
+ )
+ if not evgen_mode:
+ values[name] = value
+ 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(
obj._value = values
obj.lenindef = lenindef
obj.ber_encoded = ber_encoded
- return obj, tail
+ yield decode_path, obj, tail
def __repr__(self):
value = pp_console_row(next(self.pps()))
yield pp
-class Set(Sequence):
+class Set(Sequence, SequenceEncode1stMixin):
"""``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 <bered_ctx>` context option is set,
- then no error will occure. Also you can disable strict values
+ during decode. If :ref:`bered <bered_ctx>` context option is set,
+ then no error will occur. Also you can disable strict values
ordering check by setting ``"allow_unordered_set": True``
:ref:`context <ctx>` option.
"""
tag_default = tag_encode(form=TagFormConstructed, num=17)
asn1_type_name = "SET"
- def _encode(self):
- raws = self._encoded_values()
- raws.sort()
- v = b"".join(raws)
- return b"".join((self.tag, len_encode(len(v)), v))
+ def _values_for_encoding(self):
+ return sorted(super()._values_for_encoding(), key=attrgetter("tag_order"))
- def _specs_items(self):
- return iteritems(self.specs)
+ def _encode_cer(self, writer):
+ write_full(writer, self.tag + LENINDEF)
+ for v in sorted(
+ super()._values_for_encoding(),
+ key=attrgetter("tag_order_cer"),
+ ):
+ v.encode_cer(writer)
+ write_full(writer, EOC)
- def _decode(self, tlv, offset, decode_path, ctx, tag_only):
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
try:
t, tlen, lv = tag_strip(tlv)
except DecodeError as err:
offset=offset,
)
if tag_only:
- return None
+ yield None
+ return
lenindef = False
ctx_bered = ctx.get("bered", False)
try:
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])
+ tag_order_prev = (0, 0)
+ _specs_items = copy(self.specs)
while len(v) > 0:
if lenindef and v[:EOC_LEN].tobytes() == EOC:
break
- for name, spec in self._specs_items():
+ for name, spec in _specs_items.items():
sub_decode_path = decode_path + (name,)
try:
spec.decode(
decode_path=decode_path,
offset=offset,
)
- value, v_tail = spec.decode(
- v,
- sub_offset,
- leavemm=True,
- decode_path=sub_decode_path,
- ctx=ctx,
- _ctx_immutable=False,
- )
+ spec_defaulted = spec.default is not None
+ if evgen_mode and not spec_defaulted:
+ for _decode_path, value, v_tail in spec.decode_evgen(
+ v,
+ sub_offset,
+ leavemm=True,
+ decode_path=sub_decode_path,
+ ctx=ctx,
+ _ctx_immutable=False,
+ ):
+ yield _decode_path, value, v_tail
+ else:
+ _, value, v_tail = next(spec.decode_evgen(
+ v,
+ sub_offset,
+ leavemm=True,
+ decode_path=sub_decode_path,
+ ctx=ctx,
+ _ctx_immutable=False,
+ _evgen_mode=False,
+ ))
+ value_tag_order = value.tag_order
value_len = value.fulllen
- if value_prev.tobytes() > v[:value_len].tobytes():
+ if tag_order_prev >= value_tag_order:
if ctx_bered or ctx_allow_unordered_set:
ber_encoded = True
else:
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,
- )
+ if spec_defaulted:
+ if evgen_mode:
+ yield sub_decode_path, value, v_tail
+ if 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]
+ del _specs_items[name]
+ tag_order_prev = value_tag_order
sub_offset += value_len
vlen += value_len
v = v_tail
+
obj = self.__class__(
schema=self.specs,
impl=self.tag,
)
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,
- )
+ for name, spec in self.specs.items():
+ if name not in values and not spec.optional:
+ raise DecodeError(
+ "%s value is not ready" % name,
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
+ if not evgen_mode:
+ obj._value = values
obj.ber_encoded = ber_encoded
- return obj, tail
+ yield decode_path, obj, tail
-SequenceOfState = namedtuple("SequenceOfState", (
- "version",
- "spec",
- "value",
- "bound_min",
- "bound_max",
- "tag",
- "expl",
- "default",
- "optional",
- "offset",
- "llen",
- "vlen",
- "expl_lenindef",
- "lenindef",
- "ber_encoded",
-))
+SequenceOfState = namedtuple(
+ "SequenceOfState",
+ BasicState._fields + ("spec", "value", "bound_min", "bound_max"),
+ **NAMEDTUPLE_KWARGS
+)
-class SequenceOf(Obj):
+class SequenceOf(SequenceEncode1stMixin, Obj):
"""``SEQUENCE OF`` sequence type
For that kind of type you must specify the object it will carry on
>>> ints
Ints SEQUENCE OF[INTEGER 123, INTEGER 345]
- Also you can initialize sequence with preinitialized values:
+ You can initialize sequence with preinitialized values:
>>> ints = Ints([Integer(123), Integer(234)])
+
+ Also you can use iterator as a value:
+
+ >>> ints = Ints(iter(Integer(i) for i in range(1000000)))
+
+ And it won't be iterated until encoding process. Pay attention that
+ bounds and required schema checks are done only during the encoding
+ process in that case! After encode was called, then value is zeroed
+ back to empty list and you have to set it again. That mode is useful
+ mainly with CER encoding mode, where all objects from the iterable
+ will be streamed to the buffer, without copying all of them to
+ memory first.
"""
__slots__ = ("spec", "_bound_min", "_bound_max")
tag_default = tag_encode(form=TagFormConstructed, num=16)
optional=False,
_decoded=(0, 0, 0),
):
- super(SequenceOf, self).__init__(impl, expl, default, optional, _decoded)
+ super().__init__(impl, expl, default, optional, _decoded)
if schema is None:
schema = getattr(self, "schema", None)
if schema is None:
self._value = copy(default_obj._value)
def _value_sanitize(self, value):
+ iterator = False
if issubclass(value.__class__, SequenceOf):
value = value._value
+ elif hasattr(value, NEXT_ATTR_NAME):
+ iterator = True
elif hasattr(value, "__iter__"):
value = list(value)
else:
- raise InvalidValueType((self.__class__, iter))
- if not self._bound_min <= len(value) <= self._bound_max:
- raise BoundsError(self._bound_min, len(value), self._bound_max)
- for v in value:
- if not isinstance(v, self.spec.__class__):
- raise InvalidValueType((self.spec.__class__,))
+ raise InvalidValueType((self.__class__, iter, "iterator"))
+ if not iterator:
+ if not self._bound_min <= len(value) <= self._bound_max:
+ raise BoundsError(self._bound_min, len(value), self._bound_max)
+ class_expected = self.spec.__class__
+ for v in value:
+ if not isinstance(v, class_expected):
+ raise InvalidValueType((class_expected,))
return value
@property
def ready(self):
+ if hasattr(self._value, NEXT_ATTR_NAME):
+ return True
+ if self._bound_min > 0 and len(self._value) == 0:
+ return False
return all(v.ready for v in self._value)
@property
return any(v.bered for v in self._value)
def __getstate__(self):
+ if hasattr(self._value, NEXT_ATTR_NAME):
+ raise ValueError("can not pickle SequenceOf with iterator")
return SequenceOfState(
__version__,
- self.spec,
- [copy(v) for v in self._value],
- self._bound_min,
- self._bound_max,
self.tag,
+ self._tag_order,
self._expl,
self.default,
self.optional,
self.expl_lenindef,
self.lenindef,
self.ber_encoded,
+ self.spec,
+ [copy(v) for v in self._value],
+ self._bound_min,
+ self._bound_max,
)
def __setstate__(self, state):
- super(SequenceOf, self).__setstate__(state)
+ super().__setstate__(state)
self.spec = state.spec
self._value = state.value
self._bound_min = state.bound_min
self._bound_max = state.bound_max
- self.tag = state.tag
- self._expl = state.expl
- self.default = state.default
- self.optional = state.optional
- self.offset = state.offset
- self.llen = state.llen
- self.vlen = state.vlen
- self.expl_lenindef = state.expl_lenindef
- self.lenindef = state.lenindef
- self.ber_encoded = state.ber_encoded
def __eq__(self, their):
if isinstance(their, self.__class__):
self._value.append(value)
def __iter__(self):
- self._assert_ready()
return iter(self._value)
def __len__(self):
- self._assert_ready()
return len(self._value)
def __setitem__(self, key, value):
def __getitem__(self, key):
return self._value[key]
- def _encoded_values(self):
- return [v.encode() for v in self._value]
+ def _values_for_encoding(self):
+ return iter(self._value)
def _encode(self):
- v = b"".join(self._encoded_values())
- return b"".join((self.tag, len_encode(len(v)), v))
+ iterator = hasattr(self._value, NEXT_ATTR_NAME)
+ if iterator:
+ values = []
+ values_append = values.append
+ class_expected = self.spec.__class__
+ values_for_encoding = self._values_for_encoding()
+ self._value = []
+ for v in values_for_encoding:
+ if not isinstance(v, class_expected):
+ raise InvalidValueType((class_expected,))
+ values_append(v.encode())
+ if not self._bound_min <= len(values) <= self._bound_max:
+ raise BoundsError(self._bound_min, len(values), self._bound_max)
+ value = b"".join(values)
+ else:
+ value = b"".join(v.encode() for v in self._values_for_encoding())
+ return b"".join((self.tag, len_encode(len(value)), value))
+
+ def _encode1st(self, state):
+ state = super()._encode1st(state)
+ if hasattr(self._value, NEXT_ATTR_NAME):
+ self._value = []
+ return state
+
+ def _encode2nd(self, writer, state_iter):
+ write_full(writer, self.tag + len_encode(next(state_iter)))
+ iterator = hasattr(self._value, NEXT_ATTR_NAME)
+ if iterator:
+ values_count = 0
+ class_expected = self.spec.__class__
+ values_for_encoding = self._values_for_encoding()
+ self._value = []
+ for v in values_for_encoding:
+ if not isinstance(v, class_expected):
+ raise InvalidValueType((class_expected,))
+ v.encode2nd(writer, state_iter)
+ values_count += 1
+ if not self._bound_min <= values_count <= self._bound_max:
+ raise BoundsError(self._bound_min, values_count, self._bound_max)
+ else:
+ for v in self._values_for_encoding():
+ v.encode2nd(writer, state_iter)
+
+ def _encode_cer(self, writer):
+ write_full(writer, self.tag + LENINDEF)
+ iterator = hasattr(self._value, NEXT_ATTR_NAME)
+ if iterator:
+ class_expected = self.spec.__class__
+ values_count = 0
+ values_for_encoding = self._values_for_encoding()
+ self._value = []
+ for v in values_for_encoding:
+ if not isinstance(v, class_expected):
+ raise InvalidValueType((class_expected,))
+ v.encode_cer(writer)
+ values_count += 1
+ if not self._bound_min <= values_count <= self._bound_max:
+ raise BoundsError(self._bound_min, values_count, self._bound_max)
+ else:
+ for v in self._values_for_encoding():
+ v.encode_cer(writer)
+ write_full(writer, EOC)
- def _decode(self, tlv, offset, decode_path, ctx, tag_only, ordering_check=False):
+ def _decode(
+ self,
+ tlv,
+ offset,
+ decode_path,
+ ctx,
+ tag_only,
+ evgen_mode,
+ ordering_check=False,
+ ):
try:
t, tlen, lv = tag_strip(tlv)
except DecodeError as err:
offset=offset,
)
if tag_only:
- return None
+ yield None
+ return
lenindef = False
ctx_bered = ctx.get("bered", False)
try:
vlen = 0
sub_offset = offset + tlen + llen
_value = []
+ _value_count = 0
ctx_allow_unordered_set = ctx.get("allow_unordered_set", False)
value_prev = memoryview(v[:0])
ber_encoded = False
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=sub_decode_path,
- ctx=ctx,
- _ctx_immutable=False,
- )
+ sub_decode_path = decode_path + (str(_value_count),)
+ if evgen_mode:
+ for _decode_path, value, v_tail in spec.decode_evgen(
+ v,
+ sub_offset,
+ leavemm=True,
+ decode_path=sub_decode_path,
+ ctx=ctx,
+ _ctx_immutable=False,
+ ):
+ yield _decode_path, value, v_tail
+ else:
+ _, value, v_tail = next(spec.decode_evgen(
+ v,
+ sub_offset,
+ leavemm=True,
+ decode_path=sub_decode_path,
+ ctx=ctx,
+ _ctx_immutable=False,
+ _evgen_mode=False,
+ ))
value_len = value.fulllen
if ordering_check:
if value_prev.tobytes() > v[:value_len].tobytes():
offset=sub_offset,
)
value_prev = v[:value_len]
- _value.append(value)
+ _value_count += 1
+ if not evgen_mode:
+ _value.append(value)
sub_offset += value_len
vlen += value_len
v = v_tail
+ if evgen_mode and not self._bound_min <= _value_count <= self._bound_max:
+ raise DecodeError(
+ msg=str(BoundsError(self._bound_min, _value_count, self._bound_max)),
+ klass=self.__class__,
+ decode_path=decode_path,
+ offset=offset,
+ )
try:
obj = self.__class__(
- value=_value,
+ value=None if evgen_mode else _value,
schema=spec,
bounds=(self._bound_min, self._bound_max),
impl=self.tag,
obj.lenindef = True
tail = v[EOC_LEN:]
obj.ber_encoded = ber_encoded
- return obj, tail
+ yield decode_path, obj, tail
def __repr__(self):
return "%s[%s]" % (
tag_default = tag_encode(form=TagFormConstructed, num=17)
asn1_type_name = "SET OF"
+ def _value_sanitize(self, value):
+ value = super()._value_sanitize(value)
+ if hasattr(value, NEXT_ATTR_NAME):
+ raise ValueError(
+ "SetOf does not support iterator values, as no sense in them"
+ )
+ return value
+
def _encode(self):
- raws = self._encoded_values()
- raws.sort()
- v = b"".join(raws)
+ v = b"".join(sorted(v.encode() for v in self._values_for_encoding()))
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(
+ def _encode2nd(self, writer, state_iter):
+ write_full(writer, self.tag + len_encode(next(state_iter)))
+ values = []
+ for v in self._values_for_encoding():
+ buf = BytesIO()
+ v.encode2nd(buf.write, state_iter)
+ values.append(buf.getvalue())
+ values.sort()
+ for v in values:
+ write_full(writer, v)
+
+ def _encode_cer(self, writer):
+ write_full(writer, self.tag + LENINDEF)
+ for v in sorted(encode_cer(v) for v in self._values_for_encoding()):
+ write_full(writer, v)
+ write_full(writer, EOC)
+
+ def _decode(self, tlv, offset, decode_path, ctx, tag_only, evgen_mode):
+ return super()._decode(
tlv,
offset,
decode_path,
ctx,
tag_only,
+ evgen_mode,
ordering_check=True,
)
choice = PrimitiveTypes()
choice.specs["SequenceOf"] = SequenceOf(schema=choice)
choice.specs["SetOf"] = SetOf(schema=choice)
- for i in six_xrange(31):
+ for i in range(31):
choice.specs["SequenceOf%d" % i] = SequenceOf(
schema=choice,
expl=tag_ctxc(i),
with_colours=False,
with_decode_path=False,
decode_path_only=(),
+ decode_path=(),
):
def _pprint_pps(pps):
for pp in pps:
else:
for row in _pprint_pps(pp):
yield row
- return "\n".join(_pprint_pps(obj.pps()))
+ return "\n".join(_pprint_pps(obj.pps(decode_path)))
return SEQUENCEOF(), pprint_any
+def ascii_visualize(ba):
+ """Output only ASCII printable characters, like in hexdump -C
+
+ Example output for given binary string (right part)::
+
+ 92 2b 39 20 65 91 e6 8e 95 93 1a 58 df 02 78 ea |.+9 e......X..x.|
+ ^^^^^^^^^^^^^^^^
+ """
+ return "".join((chr(b) if 0x20 <= b <= 0x7E else ".") for b in ba)
+
+
+def hexdump(raw):
+ """Generate ``hexdump -C`` like output
+
+ Rendered example::
+
+ 00000000 30 80 30 80 a0 80 02 01 02 00 00 02 14 54 a5 18 |0.0..........T..|
+ 00000010 69 ef 8b 3f 15 fd ea ad bd 47 e0 94 81 6b 06 6a |i..?.....G...k.j|
+
+ Result of that function is a generator of lines, where each line is
+ a list of columns::
+
+ [
+ [...],
+ ["00000010 ", " 69", " ef", " 8b", " 3f", " 15", " fd", " ea", " ad ",
+ " bd", " 47", " e0", " 94", " 81", " 6b", " 06", " 6a ",
+ " |i..?.....G...k.j|"]
+ [...],
+ ]
+ """
+ hexed = hexenc(raw).upper()
+ addr, cols = 0, ["%08x " % 0]
+ for i in range(0, len(hexed), 2):
+ if i != 0 and i // 2 % 8 == 0:
+ cols[-1] += " "
+ if i != 0 and i // 2 % 16 == 0:
+ cols.append(" |%s|" % ascii_visualize(bytes(raw[addr:addr + 16])))
+ yield cols
+ addr += 16
+ cols = ["%08x " % addr]
+ cols.append(" " + hexed[i:i + 2])
+ if len(cols) > 0:
+ cols.append(" |%s|" % ascii_visualize(bytes(raw[addr:])))
+ yield cols
+
+
+def browse(raw, obj, oid_maps=()):
+ """Interactive browser
+
+ :param bytes raw: binary data you decoded
+ :param obj: decoded :py:class:`pyderasn.Obj`
+ :param oid_maps: list of ``str(OID) <-> human readable string`` dictionaries.
+ Its human readable form is printed when OID is met
+
+ .. note:: `urwid <http://urwid.org/>`__ dependency required
+
+ This browser is an interactive terminal application for browsing
+ structures of your decoded ASN.1 objects. You can quit it with **q**
+ key. It consists of three windows:
+
+ :tree:
+ View of ASN.1 elements hierarchy. You can navigate it using **Up**,
+ **Down**, **PageUp**, **PageDown**, **Home**, **End** keys.
+ **Left** key goes to constructed element above. **Plus**/**Minus**
+ keys collapse/uncollapse constructed elements. **Space** toggles it
+ :info:
+ window with various information about element. You can scroll it
+ with **h**/**l** (down, up) (**H**/**L** for triple speed) keys
+ :hexdump:
+ window with raw data hexdump and highlighted current element's
+ contents. It automatically focuses on element's data. You can
+ scroll it with **j**/**k** (down, up) (**J**/**K** for triple
+ speed) keys. If element has explicit tag, then it also will be
+ highlighted with different colour
+
+ Window's header contains current decode path and progress bars with
+ position in *info* and *hexdump* windows.
+
+ If you press **d**, then current element will be saved in the
+ current directory under its decode path name (adding ".0", ".1", etc
+ suffix if such file already exists). **D** will save it with explicit tag.
+
+ You can also invoke it with ``--browse`` command line argument.
+ """
+ from copy import deepcopy
+ from os.path import exists as path_exists
+ import urwid
+
+ class TW(urwid.TreeWidget):
+ def __init__(self, state, *args, **kwargs):
+ self.state = state
+ self.scrolled = {"info": False, "hexdump": False}
+ super().__init__(*args, **kwargs)
+
+ def _get_pp(self):
+ pp = self.get_node().get_value()
+ constructed = len(pp) > 1
+ return (pp if hasattr(pp, "_fields") else pp[0]), constructed
+
+ def _state_update(self):
+ pp, _ = self._get_pp()
+ self.state["decode_path"].set_text(
+ ":".join(str(p) for p in pp.decode_path)
+ )
+ lines = deepcopy(self.state["hexed"])
+
+ def attr_set(i, attr):
+ line = lines[i // 16]
+ idx = 1 + (i - 16 * (i // 16))
+ line[idx] = (attr, line[idx])
+
+ if pp.expl_offset is not None:
+ for i in range(
+ pp.expl_offset,
+ pp.expl_offset + pp.expl_tlen + pp.expl_llen,
+ ):
+ attr_set(i, "select-expl")
+ for i in range(pp.offset, pp.offset + pp.tlen + pp.llen + pp.vlen):
+ attr_set(i, "select-value")
+ self.state["hexdump"]._set_body([urwid.Text(line) for line in lines])
+ self.state["hexdump"].set_focus(pp.offset // 16)
+ self.state["hexdump"].set_focus_valign("middle")
+ self.state["hexdump_bar"].set_completion(
+ (100 * pp.offset // 16) //
+ len(self.state["hexdump"]._body.positions())
+ )
+
+ lines = [
+ [("header", "Name: "), pp.obj_name],
+ [("header", "Type: "), pp.asn1_type_name],
+ [("header", "Offset: "), "%d (0x%x)" % (pp.offset, pp.offset)],
+ [("header", "[TLV]len: "), "%d/%d/%d" % (
+ pp.tlen, pp.llen, pp.vlen,
+ )],
+ [("header", "TLVlen: "), "%d" % sum((
+ pp.tlen, pp.llen, pp.vlen,
+ ))],
+ [("header", "Slice: "), "[%d:%d]" % (
+ pp.offset, pp.offset + pp.tlen + pp.llen + pp.vlen,
+ )],
+ ]
+ if pp.lenindef:
+ lines.append([("warning", "LENINDEF")])
+ if pp.ber_encoded:
+ lines.append([("warning", "BER encoded")])
+ if pp.bered:
+ lines.append([("warning", "BERed")])
+ if pp.expl is not None:
+ lines.append([("header", "EXPLICIT")])
+ klass, _, num = pp.expl
+ lines.append([" Tag: %s%d" % (TagClassReprs[klass], num)])
+ if pp.expl_offset is not None:
+ lines.append([" Offset: %d" % pp.expl_offset])
+ lines.append([" [TLV]len: %d/%d/%d" % (
+ pp.expl_tlen, pp.expl_llen, pp.expl_vlen,
+ )])
+ lines.append([" TLVlen: %d" % sum((
+ pp.expl_tlen, pp.expl_llen, pp.expl_vlen,
+ ))])
+ lines.append([" Slice: [%d:%d]" % (
+ pp.expl_offset,
+ pp.expl_offset + pp.expl_tlen + pp.expl_llen + pp.expl_vlen,
+ )])
+ if pp.impl is not None:
+ klass, _, num = pp.impl
+ lines.append([
+ ("header", "IMPLICIT: "), "%s%d" % (TagClassReprs[klass], num),
+ ])
+ if pp.optional:
+ lines.append(["OPTIONAL"])
+ if pp.default:
+ lines.append(["DEFAULT"])
+ if len(pp.decode_path) > 0:
+ ent = pp.decode_path[-1]
+ if isinstance(ent, DecodePathDefBy):
+ lines.append([""])
+ value = str(ent.defined_by)
+ oid_name = find_oid_name(
+ ent.defined_by.asn1_type_name, oid_maps, value,
+ )
+ lines.append([("header", "DEFINED BY: "), "%s" % (
+ value if oid_name is None
+ else "%s (%s)" % (oid_name, value)
+ )])
+ lines.append([""])
+ if pp.value is not None:
+ lines.append([("header", "Value: "), pp.value])
+ if (
+ len(oid_maps) > 0 and
+ pp.asn1_type_name == ObjectIdentifier.asn1_type_name
+ ):
+ for oid_map in oid_maps:
+ oid_name = oid_map.get(pp.value)
+ if oid_name is not None:
+ lines.append([("header", "Human: "), oid_name])
+ break
+ if pp.asn1_type_name == Integer.asn1_type_name:
+ lines.append([
+ ("header", "Decimal: "), "%d" % int(pp.obj),
+ ])
+ lines.append([
+ ("header", "Hexadecimal: "), colonize_hex(pp.obj.tohex()),
+ ])
+ if pp.blob.__class__ == bytes:
+ blob = hexenc(pp.blob).upper()
+ for i in range(0, len(blob), 32):
+ lines.append([colonize_hex(blob[i:i + 32])])
+ elif pp.blob.__class__ == tuple:
+ lines.append([", ".join(pp.blob)])
+ self.state["info"]._set_body([urwid.Text(line) for line in lines])
+ self.state["info_bar"].set_completion(0)
+
+ def selectable(self):
+ if self.state["widget_current"] != self:
+ self.state["widget_current"] = self
+ self.scrolled["info"] = False
+ self.scrolled["hexdump"] = False
+ self._state_update()
+ return super().selectable()
+
+ def _get_display_text_without_offset(self):
+ pp, constructed = self._get_pp()
+ style = "constructed" if constructed else ""
+ if len(pp.decode_path) == 0:
+ return (style, pp.obj_name)
+ if pp.asn1_type_name == "EOC":
+ return ("eoc", "EOC")
+ ent = pp.decode_path[-1]
+ if isinstance(ent, DecodePathDefBy):
+ value = str(ent.defined_by)
+ oid_name = find_oid_name(
+ ent.defined_by.asn1_type_name, oid_maps, value,
+ )
+ return ("defby", "DEFBY:" + (
+ value if oid_name is None else oid_name
+ ))
+ return (style, ent)
+
+ def get_display_text(self):
+ pp, _ = self._get_pp()
+ style, ent = self._get_display_text_without_offset()
+ return [(style, ent), " [%d]" % pp.offset]
+
+ def _scroll(self, what, step):
+ self.state[what]._invalidate()
+ pos = self.state[what].focus_position
+ if not self.scrolled[what]:
+ self.scrolled[what] = True
+ pos -= 2
+ pos = max(0, pos + step)
+ pos = min(pos, len(self.state[what]._body.positions()) - 1)
+ self.state[what].set_focus(pos)
+ self.state[what].set_focus_valign("top")
+ self.state[what + "_bar"].set_completion(
+ (100 * pos) // len(self.state[what]._body.positions())
+ )
+
+ def keypress(self, size, key):
+ if key == "q":
+ raise urwid.ExitMainLoop()
+
+ if key == " ":
+ self.expanded = not self.expanded
+ self.update_expanded_icon()
+ return None
+
+ hexdump_steps = {"j": 1, "k": -1, "J": 5, "K": -5}
+ if key in hexdump_steps:
+ self._scroll("hexdump", hexdump_steps[key])
+ return None
+
+ info_steps = {"h": 1, "l": -1, "H": 5, "L": -5}
+ if key in info_steps:
+ self._scroll("info", info_steps[key])
+ return None
+
+ if key in ("d", "D"):
+ pp, _ = self._get_pp()
+ dp = ":".join(str(p) for p in pp.decode_path)
+ dp = dp.replace(" ", "_")
+ if dp == "":
+ dp = "root"
+ if key == "d" or pp.expl_offset is None:
+ data = self.state["raw"][pp.offset:(
+ pp.offset + pp.tlen + pp.llen + pp.vlen
+ )]
+ else:
+ data = self.state["raw"][pp.expl_offset:(
+ pp.expl_offset + pp.expl_tlen + pp.expl_llen + pp.expl_vlen
+ )]
+ ctr = 0
+
+ def duplicate_path(dp, ctr):
+ if ctr == 0:
+ return dp
+ return "%s.%d" % (dp, ctr)
+
+ while True:
+ if not path_exists(duplicate_path(dp, ctr)):
+ break
+ ctr += 1
+ dp = duplicate_path(dp, ctr)
+ with open(dp, "wb") as fd:
+ fd.write(data)
+ self.state["decode_path"].set_text(
+ ("warning", "Saved to: " + dp)
+ )
+ return None
+ return super().keypress(size, key)
+
+ class PN(urwid.ParentNode):
+ def __init__(self, state, value, *args, **kwargs):
+ self.state = state
+ if not hasattr(value, "_fields"):
+ value = list(value)
+ super().__init__(value, *args, **kwargs)
+
+ def load_widget(self):
+ return TW(self.state, self)
+
+ def load_child_keys(self):
+ value = self.get_value()
+ if hasattr(value, "_fields"):
+ return []
+ return range(len(value[1:]))
+
+ def load_child_node(self, key):
+ return PN(
+ self.state,
+ self.get_value()[key + 1],
+ parent=self,
+ key=key,
+ depth=self.get_depth() + 1,
+ )
+
+ class LabeledPG(urwid.ProgressBar):
+ def __init__(self, label, *args, **kwargs):
+ self.label = label
+ super().__init__(*args, **kwargs)
+
+ def get_text(self):
+ return "%s: %s" % (self.label, super().get_text())
+
+ WinHexdump = urwid.ListBox([urwid.Text("")])
+ WinInfo = urwid.ListBox([urwid.Text("")])
+ WinDecodePath = urwid.Text("", "center")
+ WinInfoBar = LabeledPG("info", "pg-normal", "pg-complete")
+ WinHexdumpBar = LabeledPG("hexdump", "pg-normal", "pg-complete")
+ WinTree = urwid.TreeListBox(urwid.TreeWalker(PN(
+ {
+ "raw": raw,
+ "hexed": list(hexdump(raw)),
+ "widget_current": None,
+ "info": WinInfo,
+ "info_bar": WinInfoBar,
+ "hexdump": WinHexdump,
+ "hexdump_bar": WinHexdumpBar,
+ "decode_path": WinDecodePath,
+ },
+ list(obj.pps()),
+ )))
+ help_text = " ".join((
+ "q:quit",
+ "space:(un)collapse",
+ "(pg)up/down/home/end:nav",
+ "jkJK:hexdump hlHL:info",
+ "dD:dump",
+ ))
+ urwid.MainLoop(
+ urwid.Frame(
+ urwid.Columns([
+ ("weight", 1, WinTree),
+ ("weight", 2, urwid.Pile([
+ urwid.LineBox(WinInfo),
+ urwid.LineBox(WinHexdump),
+ ])),
+ ]),
+ header=urwid.Columns([
+ ("weight", 2, urwid.AttrWrap(WinDecodePath, "header")),
+ ("weight", 1, WinInfoBar),
+ ("weight", 1, WinHexdumpBar),
+ ]),
+ footer=urwid.AttrWrap(urwid.Text(help_text), "help")
+ ),
+ [
+ ("header", "bold", ""),
+ ("constructed", "bold", ""),
+ ("help", "light magenta", ""),
+ ("warning", "light red", ""),
+ ("defby", "light red", ""),
+ ("eoc", "dark red", ""),
+ ("select-value", "light green", ""),
+ ("select-expl", "light red", ""),
+ ("pg-normal", "", "light blue"),
+ ("pg-complete", "black", "yellow"),
+ ],
+ ).run()
+
+
def main(): # pragma: no cover
import argparse
- parser = argparse.ArgumentParser(description="PyDERASN ASN.1 BER/DER decoder")
+ parser = argparse.ArgumentParser(description="PyDERASN ASN.1 BER/CER/DER decoder")
parser.add_argument(
"--skip",
type=int,
help="Allow explicit tag out-of-bound",
)
parser.add_argument(
- "DERFile",
+ "--evgen",
+ action="store_true",
+ help="Turn on event generation mode",
+ )
+ parser.add_argument(
+ "--browse",
+ action="store_true",
+ help="Start ASN.1 browser",
+ )
+ parser.add_argument(
+ "RAWFile",
type=argparse.FileType("rb"),
- help="Path to DER file you want to decode",
+ help="Path to BER/CER/DER file you want to decode",
)
args = parser.parse_args()
- args.DERFile.seek(args.skip)
- der = memoryview(args.DERFile.read())
- args.DERFile.close()
+ try:
+ raw = file_mmaped(args.RAWFile)[args.skip:]
+ except:
+ args.RAWFile.seek(args.skip)
+ raw = memoryview(args.RAWFile.read())
+ args.RAWFile.close()
oid_maps = (
[obj_by_path(_path) for _path in (args.oids or "").split(",")]
if args.oids else ()
)
+ from functools import partial
if args.schema:
schema = obj_by_path(args.schema)
- from functools import partial
pprinter = partial(pprint, big_blobs=True)
else:
schema, pprinter = generic_decoder()
}
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,
+ if args.browse:
+ obj, _ = schema().decode(raw, ctx=ctx)
+ browse(raw, obj, oid_maps)
+ from sys import exit as sys_exit
+ sys_exit(0)
+ from os import environ
+ pprinter = partial(
+ pprinter,
oid_maps=oid_maps,
with_colours=environ.get("NO_COLOR") is None,
with_decode_path=args.print_decode_path,
() if args.decode_path_only is None else
tuple(args.decode_path_only.split(":"))
),
- ))
+ )
+ if args.evgen:
+ for decode_path, obj, tail in schema().decode_evgen(raw, ctx=ctx):
+ print(pprinter(obj, decode_path=decode_path))
+ else:
+ obj, tail = schema().decode(raw, ctx=ctx)
+ print(pprinter(obj))
if tail != b"":
print("\nTrailing data: %s" % hexenc(tail))
if __name__ == "__main__":
+ from pyderasn import *
main()