+import collections
+import struct
+
+VOID = 0x00
+TRUE = 0x01
+FALSE = 0x02
+BOOL = (TRUE, FALSE)
+INT8 = 0x10
+INT16 = 0x11
+INT32 = 0x12
+INT64 = 0x13
+FLOAT = 0x20
+DOUBLE = 0x21
+BINARY = 0x30
+UTF8 = 0x31
+UTF16 = 0x32
+UTF32 = 0x33
+LIST = 0x40
+DICTIONARY = 0x41
+
+DEFAULT_INTEGER_ENCODING = INT32
+DEFAULT_DECIMAL_ENCODING = DOUBLE
+DEFAULT_STRING_ENCODING = UTF8
+
+TaggedObject = collections.namedtuple('TaggedObject', ['tag', 'value'])
+
+_TYPES_TO_TAGS = {
+ int: DEFAULT_INTEGER_ENCODING,
+ float: DEFAULT_DECIMAL_ENCODING,
+ bytes: BINARY,
+ str: DEFAULT_STRING_ENCODING,
+ list: LIST,
+ dict: DICTIONARY,
+ collections.OrderedDict: DICTIONARY,
+}
+
+def tag(o):
+ if isinstance(o, TaggedObject):
+ return o
+
+ if o is None:
+ return TaggedObject(tag = VOID, value = o)
+
+ if o is True:
+ return TaggedObject(tag = TRUE, value = o)
+
+ if o is False:
+ return TaggedObject(tag = FALSE, value = o)
+
+ return TaggedObject(tag = _TYPES_TO_TAGS[type(o)], value = o)
+
+def serialize_tag_only_type(o):
+ return b''
+
+def make_serializer_from_pack_format_string(pfs):
+ def serializer(i):
+ return struct.pack(pfs, i)
+ return serializer
+
+def make_string_serializer_from_encoder(e):
+ def serializer(s):
+ encoded = e(s)
+ return struct.pack('!I', len(encoded)) + encoded
+ return serializer
+
+def serialize_list(items):
+ # TODO Enforce that items are all the same type
+ items = [tag(i) for i in items]
+
+ if len(items) == 0:
+ item_tag = VOID
+ else:
+ item_tag = items[0].tag
+
+ item_serializer = _SERIALIZERS[item_tag]
+ items = [item_serializer(i.value) for i in items]
+ item_length = len(items)
+ items = b''.join(items)
+ byte_length = len(items)
+ return struct.pack('!BII', item_tag, byte_length, item_length) + items
+
+def serialize_dict(d):
+ item_length = 0
+ serialized = b''
+
+ key_serializer = _SERIALIZERS[UTF8]
+
+ for key, value in d.items():
+ assert isinstance(key, str)
+ item_length += 1
+ serialized += key_serializer(key) + serialize(value)
+
+ byte_length = len(serialized)
+ return struct.pack('!II', byte_length, item_length) + serialized
+
+_SERIALIZERS = {
+ VOID: serialize_tag_only_type,
+ TRUE: serialize_tag_only_type,
+ FALSE: serialize_tag_only_type,
+ INT8: make_serializer_from_pack_format_string('!b'),
+ INT16: make_serializer_from_pack_format_string('!h'),
+ INT32: make_serializer_from_pack_format_string('!i'),
+ FLOAT: make_serializer_from_pack_format_string('!f'),
+ DOUBLE: make_serializer_from_pack_format_string('!d'),
+ BINARY: make_string_serializer_from_encoder(lambda b: b),
+ UTF8: make_string_serializer_from_encoder(lambda s: s.encode('utf-8')),
+ UTF16: make_string_serializer_from_encoder(lambda s: s.encode('utf-16')),
+ UTF32: make_string_serializer_from_encoder(lambda s: s.encode('utf-32')),
+ LIST: serialize_list,
+ DICTIONARY: serialize_dict,
+}
+
+def serialize(o):
+ o = tag(o)
+ return struct.pack('!B', o.tag) + _SERIALIZERS[o.tag](o.value)
+
+ParseResult = collections.namedtuple(
+ 'ParseResult',
+ [
+ 'success',
+ 'value',
+ 'remaining',
+ ],
+)
+
+_FAILED_PARSE_RESULT = ParseResult(success = False, value = None, remaining = None)
+
+_BYTE_SIZES_TO_UNPACK_FORMATS = {
+ 1: '!b',
+ 2: '!h',
+ 4: '!i',
+ 8: '!q',
+}
+
+def make_integer_parser(size_in_bytes):
+ unpack_format = _BYTE_SIZES_TO_UNPACK_FORMATS[size_in_bytes]
+
+ def integer_parser(source):
+ value = struct.unpack(unpack_format, source[:size_in_bytes])[0]
+ remaining = source[size_in_bytes:]
+
+ return ParseResult(success = True, value = value, remaining = remaining)
+
+ return integer_parser
+
+def binary64_parser(source):
+ return ParseResult(
+ success = True,
+ value = struct.unpack('!d', source[:8])[0],
+ remaining = source[8:],
+ )
+
+def make_string_parser(decoder):
+ def string_parser(source):
+ length = struct.unpack('!I', source[:4])[0]
+ source = source[4:]
+ return ParseResult(
+ success = True,
+ value = decoder(source[:length]),
+ remaining = source[length:],
+ )
+
+ return string_parser
+
+def list_parser(source):
+ tag = source[0]
+ parser = _TAGS_TO_PARSERS[tag]
+
+ source = source[1:]
+ byte_length, items_length = struct.unpack('!II', source[:8])
+ source = source[8:]
+
+ remaining = source[byte_length:]
+ source = source[:byte_length]
+
+ def item_iterator(source):
+ count = 0
+
+ while len(source) > 0:
+ parse_result = parser(source)
+
+ if parse_result.success:
+ count += 1
+ yield parse_result.value
+ source = parse_result.remaining
+
+ assert count == items_length
+
+ return ParseResult(
+ success = True,
+ value = item_iterator(source),
+ remaining = remaining,
+ )
+
+def dictionary_parser(source):
+ key_parser = _TAGS_TO_PARSERS[UTF8]
+
+ byte_length, item_length = struct.unpack('!II', source[:8])
+ source = source[8:]
+
+ remaining = source[byte_length:]
+ source = source[:byte_length]
+
+ def kvp_iterator(source):
+ count = 0
+
+ while len(source) > 0:
+ count += 1
+ key_parse_result = key_parser(source)
+ key, source = key_parse_result.value, key_parse_result.remaining
+ value_parse_result = _object_parser(source)
+ value, source = value_parse_result.value, value_parse_result.remaining
+
+ yield key, value
+
+ assert count == item_length
+
+ return ParseResult(
+ success = True,
+ value = collections.OrderedDict(kvp_iterator(source)),
+ remaining = remaining,
+ )
+
+
+_TAGS_TO_PARSERS = {
+ VOID: lambda r: ParseResult(True, None, r),
+ TRUE: lambda r: ParseResult(True, True, r),
+ FALSE: lambda r: ParseResult(True, False, r),
+ INT8: make_integer_parser(1),
+ INT16: make_integer_parser(2),
+ INT32: make_integer_parser(4),
+ INT64: make_integer_parser(8),
+ DOUBLE: binary64_parser,
+ BINARY: make_string_parser(lambda b : b),
+ UTF8: make_string_parser(lambda b : b.decode('utf-8')),
+ UTF16: make_string_parser(lambda b : b.decode('utf-16')),
+ UTF32: make_string_parser(lambda b : b.decode('utf-32')),
+ LIST: list_parser,
+ DICTIONARY: dictionary_parser,
+}
+
+def _object_parser(source):
+ return _TAGS_TO_PARSERS[source[0]](source[1:])
+
+def _parse(parser, source, consume_all = True):
+ result = parser(source)
+
+ if result.success and result.remaining == b'':
+ return result.value
+
+ raise Exception('Unparsed trailing bytes: {}'.format(result.remaining))
+
+def deserialize(b):
+ return _parse(_object_parser, b)
projects / ton / commitdiff