# cborutil.py - CBOR extensions

#

# Copyright 2018 Gregory Szorc <gregory.szorc@gmail.com>

#

# This software may be used and distributed according to the terms of the

# GNU General Public License version 2 or any later version.

import struct

# Very short very of RFC 7049...

#

# Each item begins with a byte. The 3 high bits of that byte denote the

# "major type." The lower 5 bits denote the "subtype." Each major type

# has its own encoding mechanism.

#

# Most types have lengths. However, bytestring, string, array, and map

# can be indefinite length. These are denotes by a subtype with value 31.

# Sub-components of those types then come afterwards and are terminated

# by a "break" byte.

MAJOR_TYPE_UINT = 0

MAJOR_TYPE_NEGINT = 1

MAJOR_TYPE_BYTESTRING = 2

MAJOR_TYPE_STRING = 3

MAJOR_TYPE_ARRAY = 4

MAJOR_TYPE_MAP = 5

MAJOR_TYPE_SEMANTIC = 6

MAJOR_TYPE_SPECIAL = 7

SUBTYPE_MASK = 0b00011111

SUBTYPE_FALSE = 20

SUBTYPE_TRUE = 21

SUBTYPE_NULL = 22

SUBTYPE_HALF_FLOAT = 25

SUBTYPE_SINGLE_FLOAT = 26

SUBTYPE_DOUBLE_FLOAT = 27

SUBTYPE_INDEFINITE = 31

SEMANTIC_TAG_FINITE_SET = 258

# Indefinite types begin with their major type ORd with information value 31.

BEGIN_INDEFINITE_BYTESTRING = struct.pack(

    '>B', MAJOR_TYPE_BYTESTRING << 5 | SUBTYPE_INDEFINITE

)

BEGIN_INDEFINITE_ARRAY = struct.pack(

    '>B', MAJOR_TYPE_ARRAY << 5 | SUBTYPE_INDEFINITE

)

BEGIN_INDEFINITE_MAP = struct.pack(

    '>B', MAJOR_TYPE_MAP << 5 | SUBTYPE_INDEFINITE

)

ENCODED_LENGTH_1 = struct.Struct('>B')

ENCODED_LENGTH_2 = struct.Struct('>BB')

ENCODED_LENGTH_3 = struct.Struct('>BH')

ENCODED_LENGTH_4 = struct.Struct('>BL')

ENCODED_LENGTH_5 = struct.Struct('>BQ')

# The break ends an indefinite length item.

BREAK = b'\xff'

BREAK_INT = 255

def encodelength(majortype, length):

    """Obtain a value encoding the major type and its length."""

    if length < 24:

        return ENCODED_LENGTH_1.pack(majortype << 5 | length)

    elif length < 256:

        return ENCODED_LENGTH_2.pack(majortype << 5 | 24, length)

    elif length < 65536:

        return ENCODED_LENGTH_3.pack(majortype << 5 | 25, length)

    elif length < 4294967296:

        return ENCODED_LENGTH_4.pack(majortype << 5 | 26, length)

    else:

        return ENCODED_LENGTH_5.pack(majortype << 5 | 27, length)

def streamencodebytestring(v):

    yield encodelength(MAJOR_TYPE_BYTESTRING, len(v))

    yield v

def streamencodebytestringfromiter(it):

    """Convert an iterator of chunks to an indefinite bytestring.

    Given an input that is iterable and each element in the iterator is

    representable as bytes, emit an indefinite length bytestring.

    """

    yield BEGIN_INDEFINITE_BYTESTRING

    for chunk in it:

        yield encodelength(MAJOR_TYPE_BYTESTRING, len(chunk))

        yield chunk

    yield BREAK

def streamencodeindefinitebytestring(source, chunksize=65536):

    """Given a large source buffer, emit as an indefinite length bytestring.

    This is a generator of chunks constituting the encoded CBOR data.

    """

    yield BEGIN_INDEFINITE_BYTESTRING

    i = 0

    l = len(source)

    while True:

        chunk = source[i : i + chunksize]

        i += len(chunk)

        yield encodelength(MAJOR_TYPE_BYTESTRING, len(chunk))

        yield chunk

        if i >= l:

            break

    yield BREAK

def streamencodeint(v):

    if v >= 18446744073709551616 or v < -18446744073709551616:

        raise ValueError(b'big integers not supported')

    if v >= 0:

        yield encodelength(MAJOR_TYPE_UINT, v)

    else:

        yield encodelength(MAJOR_TYPE_NEGINT, abs(v) - 1)

def streamencodearray(l):

    """Encode a known size iterable to an array."""

    yield encodelength(MAJOR_TYPE_ARRAY, len(l))

    for i in l:

        for chunk in streamencode(i):

            yield chunk

def streamencodearrayfromiter(it):

    """Encode an iterator of items to an indefinite length array."""

    yield BEGIN_INDEFINITE_ARRAY

    for i in it:

        for chunk in streamencode(i):

            yield chunk

    yield BREAK

def _mixedtypesortkey(v):

    return type(v).__name__, v

def streamencodeset(s):

    # https://www.iana.org/assignments/cbor-tags/cbor-tags.xhtml defines

    # semantic tag 258 for finite sets.

    yield encodelength(MAJOR_TYPE_SEMANTIC, SEMANTIC_TAG_FINITE_SET)

    for chunk in streamencodearray(sorted(s, key=_mixedtypesortkey)):

        yield chunk

def streamencodemap(d):

    """Encode dictionary to a generator.

    Does not supporting indefinite length dictionaries.

    """

    yield encodelength(MAJOR_TYPE_MAP, len(d))

    for key, value in sorted(d.items(), key=lambda x: _mixedtypesortkey(x[0])):

        for chunk in streamencode(key):

            yield chunk

        for chunk in streamencode(value):

            yield chunk

def streamencodemapfromiter(it):

    """Given an iterable of (key, value), encode to an indefinite length map."""

    yield BEGIN_INDEFINITE_MAP

    for key, value in it:

        for chunk in streamencode(key):

            yield chunk

        for chunk in streamencode(value):

            yield chunk

    yield BREAK

def streamencodebool(b):

    # major type 7, simple value 20 and 21.

    yield b'\xf5' if b else b'\xf4'

def streamencodenone(v):

    # major type 7, simple value 22.

    yield b'\xf6'

STREAM_ENCODERS = {

    bytes: streamencodebytestring,

    int: streamencodeint,

    int: streamencodeint,

    list: streamencodearray,

    tuple: streamencodearray,

    dict: streamencodemap,

    set: streamencodeset,

    bool: streamencodebool,

    type(None): streamencodenone,

}

def streamencode(v):

    """Encode a value in a streaming manner.

    Given an input object, encode it to CBOR recursively.

    Returns a generator of CBOR encoded bytes. There is no guarantee

    that each emitted chunk fully decodes to a value or sub-value.

    Encoding is deterministic - unordered collections are sorted.

    """

    fn = STREAM_ENCODERS.get(v.__class__)

    if not fn:

        # handle subtypes such as encoding.localstr and util.sortdict

        for ty in STREAM_ENCODERS:

            if not isinstance(v, ty):

                continue

            fn = STREAM_ENCODERS[ty]

            break

    if not fn:

        raise ValueError(b'do not know how to encode %s' % type(v))

    return fn(v)

class CBORDecodeError(Exception):

    """Represents an error decoding CBOR."""

def _elementtointeger(b, i):

    return b[i]

STRUCT_BIG_UBYTE = struct.Struct('>B')

STRUCT_BIG_USHORT = struct.Struct(b'>H')

STRUCT_BIG_ULONG = struct.Struct(b'>L')

STRUCT_BIG_ULONGLONG = struct.Struct(b'>Q')

SPECIAL_NONE = 0

SPECIAL_START_INDEFINITE_BYTESTRING = 1

SPECIAL_START_ARRAY = 2

SPECIAL_START_MAP = 3

SPECIAL_START_SET = 4

SPECIAL_INDEFINITE_BREAK = 5

def decodeitem(b, offset=0):

    """Decode a new CBOR value from a buffer at offset.

    This function attempts to decode up to one complete CBOR value

    from ``b`` starting at offset ``offset``.

    The beginning of a collection (such as an array, map, set, or

    indefinite length bytestring) counts as a single value. For these

    special cases, a state flag will indicate that a special value was seen.

    When called, the function either returns a decoded value or gives

    a hint as to how many more bytes are needed to do so. By calling

    the function repeatedly given a stream of bytes, the caller can

    build up the original values.

    Returns a tuple with the following elements:

    * Bool indicating whether a complete value was decoded.

    * A decoded value if first value is True otherwise None

    * Integer number of bytes. If positive, the number of bytes

      read. If negative, the number of bytes we need to read to

      decode this value or the next chunk in this value.

    * One of the ``SPECIAL_*`` constants indicating special treatment

      for this value. ``SPECIAL_NONE`` means this is a fully decoded

      simple value (such as an integer or bool).

    """

    initial = _elementtointeger(b, offset)

    offset += 1

    majortype = initial >> 5

    subtype = initial & SUBTYPE_MASK

    if majortype == MAJOR_TYPE_UINT:

        complete, value, readcount = decodeuint(subtype, b, offset)

        if complete:

            return True, value, readcount + 1, SPECIAL_NONE

        else:

            return False, None, readcount, SPECIAL_NONE

    elif majortype == MAJOR_TYPE_NEGINT:

        # Negative integers are the same as UINT except inverted minus 1.

        complete, value, readcount = decodeuint(subtype, b, offset)

        if complete:

            return True, -value - 1, readcount + 1, SPECIAL_NONE

        else:

            return False, None, readcount, SPECIAL_NONE

    elif majortype == MAJOR_TYPE_BYTESTRING:

        # Beginning of bytestrings are treated as uints in order to

        # decode their length, which may be indefinite.

        complete, size, readcount = decodeuint(

            subtype, b, offset, allowindefinite=True

        )

        # We don't know the size of the bytestring. It must be a definitive

        # length since the indefinite subtype would be encoded in the initial

        # byte.

        if not complete:

            return False, None, readcount, SPECIAL_NONE

        # We know the length of the bytestring.

        if size is not None:

            # And the data is available in the buffer.

            if offset + readcount + size <= len(b):

                value = b[offset + readcount : offset + readcount + size]

                return True, value, readcount + size + 1, SPECIAL_NONE

            # And we need more data in order to return the bytestring.

            else:

                wanted = len(b) - offset - readcount - size

                return False, None, wanted, SPECIAL_NONE

        # It is an indefinite length bytestring.

        else:

            return True, None, 1, SPECIAL_START_INDEFINITE_BYTESTRING

    elif majortype == MAJOR_TYPE_STRING:

        raise CBORDecodeError(b'string major type not supported')

    elif majortype == MAJOR_TYPE_ARRAY:

        # Beginning of arrays are treated as uints in order to decode their

        # length. We don't allow indefinite length arrays.

        complete, size, readcount = decodeuint(subtype, b, offset)

        if complete:

            return True, size, readcount + 1, SPECIAL_START_ARRAY

        else:

            return False, None, readcount, SPECIAL_NONE

    elif majortype == MAJOR_TYPE_MAP:

        # Beginning of maps are treated as uints in order to decode their

        # number of elements. We don't allow indefinite length arrays.

        complete, size, readcount = decodeuint(subtype, b, offset)

        if complete:

            return True, size, readcount + 1, SPECIAL_START_MAP

        else:

            return False, None, readcount, SPECIAL_NONE

    elif majortype == MAJOR_TYPE_SEMANTIC:

        # Semantic tag value is read the same as a uint.

        complete, tagvalue, readcount = decodeuint(subtype, b, offset)

        if not complete:

            return False, None, readcount, SPECIAL_NONE

        # This behavior here is a little wonky. The main type being "decorated"

        # by this semantic tag follows. A more robust parser would probably emit

        # a special flag indicating this as a semantic tag and let the caller

        # deal with the types that follow. But since we don't support many

        # semantic tags, it is easier to deal with the special cases here and

        # hide complexity from the caller. If we add support for more semantic

        # tags, we should probably move semantic tag handling into the caller.

        if tagvalue == SEMANTIC_TAG_FINITE_SET:

            if offset + readcount >= len(b):

                return False, None, -1, SPECIAL_NONE

            complete, size, readcount2, special = decodeitem(

                b, offset + readcount

            )

            if not complete:

                return False, None, readcount2, SPECIAL_NONE

            if special != SPECIAL_START_ARRAY:

                raise CBORDecodeError(

                    b'expected array after finite set semantic tag'

                )

            return True, size, readcount + readcount2 + 1, SPECIAL_START_SET

        else:

            raise CBORDecodeError(b'semantic tag %d not allowed' % tagvalue)

    elif majortype == MAJOR_TYPE_SPECIAL:

        # Only specific values for the information field are allowed.

        if subtype == SUBTYPE_FALSE:

            return True, False, 1, SPECIAL_NONE

        elif subtype == SUBTYPE_TRUE:

            return True, True, 1, SPECIAL_NONE

        elif subtype == SUBTYPE_NULL:

            return True, None, 1, SPECIAL_NONE

        elif subtype == SUBTYPE_INDEFINITE:

            return True, None, 1, SPECIAL_INDEFINITE_BREAK

        # If value is 24, subtype is in next byte.

        else:

            raise CBORDecodeError(b'special type %d not allowed' % subtype)

    else:

        assert False