• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1# Copyright 2020 The Pigweed Authors
2#
3# Licensed under the Apache License, Version 2.0 (the "License"); you may not
4# use this file except in compliance with the License. You may obtain a copy of
5# the License at
6#
7#     https://www.apache.org/licenses/LICENSE-2.0
8#
9# Unless required by applicable law or agreed to in writing, software
10# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
11# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
12# License for the specific language governing permissions and limitations under
13# the License.
14"""Provides functionality for encoding tokenized messages."""
15
16import argparse
17import base64
18import struct
19import sys
20from typing import Sequence, Union
21
22from pw_tokenizer import tokens
23
24_INT32_MAX = 2**31 - 1
25_UINT32_MAX = 2**32 - 1
26BASE64_PREFIX = '$'
27
28
29def _zig_zag_encode(value: int) -> int:
30    """Encodes signed integers to give a compact varint encoding."""
31    return value << 1 if value >= 0 else (value << 1) ^ (~0)
32
33
34def _little_endian_base128_encode(integer: int) -> bytearray:
35    data = bytearray()
36
37    while True:
38        # Grab 7 bits; the eighth bit is set to 1 to indicate more data coming.
39        data.append((integer & 0x7F) | 0x80)
40        integer >>= 7
41
42        if not integer:
43            break
44
45    data[-1] &= 0x7F  # clear the top bit of the last byte
46    return data
47
48
49def _encode_int32(arg: int) -> bytearray:
50    # Convert large unsigned numbers into their corresponding signed values.
51    if arg > _INT32_MAX:
52        arg -= 2**32
53
54    return _little_endian_base128_encode(_zig_zag_encode(arg))
55
56
57def _encode_string(arg: bytes) -> bytes:
58    size_byte = len(arg) if len(arg) < 128 else 0xFF
59    return struct.pack('B', size_byte) + arg[:127]
60
61
62def encode_args(*args: Union[int, float, bytes, str]) -> bytes:
63    """Encodes a list of arguments to their on-wire representation."""
64
65    data = bytearray(b'')
66    for arg in args:
67        if isinstance(arg, int):
68            if arg.bit_length() > 32:
69                raise ValueError(
70                    f'Cannot encode {arg}: only 32-bit integers may be encoded'
71                )
72            data += _encode_int32(arg)
73        elif isinstance(arg, float):
74            data += struct.pack('<f', arg)
75        elif isinstance(arg, str):
76            data += _encode_string(arg.encode())
77        elif isinstance(arg, bytes):
78            data += _encode_string(arg)
79        else:
80            raise ValueError(
81                f'{arg} has type {type(arg)}, which is not supported'
82            )
83    return bytes(data)
84
85
86def encode_token_and_args(
87    token: int, *args: Union[int, float, bytes, str]
88) -> bytes:
89    """Encodes a tokenized message given its token and arguments.
90
91    This function assumes that the token represents a format string with
92    conversion specifiers that correspond with the provided argument types.
93    Currently, only 32-bit integers are supported.
94    """
95
96    if token < 0 or token > _UINT32_MAX:
97        raise ValueError(
98            f'The token ({token}) must be an unsigned 32-bit integer'
99        )
100
101    return struct.pack('<I', token) + encode_args(*args)
102
103
104def prefixed_base64(data: bytes, prefix: str = '$') -> str:
105    """Encodes a tokenized message as prefixed Base64."""
106    return prefix + base64.b64encode(data).decode()
107
108
109def _parse_user_input(string: str):
110    """Evaluates a string as Python code or returns it as a literal string."""
111    try:
112        value = eval(string, dict(__builtins__={}))  # pylint: disable=eval-used
113    except (NameError, SyntaxError):
114        return string
115
116    return value if isinstance(value, (int, float)) else string
117
118
119def _main(format_string_list: Sequence[str], raw_args: Sequence[str]) -> int:
120    (format_string,) = format_string_list
121    token = tokens.pw_tokenizer_65599_hash(format_string)
122    args = tuple(_parse_user_input(a) for a in raw_args)
123
124    data = encode_token_and_args(token, *args)
125    token = int.from_bytes(data[:4], 'little')
126    binary = ' '.join(f'{b:02x}' for b in data)
127
128    print(f'      Raw input: {format_string!r} % {args!r}')
129    print(f'Formatted input: {format_string % args}')
130    print(f'          Token: 0x{token:08x}')
131    print(f'        Encoded: {data!r} ({binary}) [{len(data)} bytes]')
132    print(f'Prefixed Base64: {prefixed_base64(data)}')
133
134    return 0
135
136
137def _parse_args() -> dict:
138    parser = argparse.ArgumentParser(
139        description=__doc__,
140        formatter_class=argparse.RawDescriptionHelpFormatter,
141    )
142    parser.add_argument(
143        'format_string_list',
144        metavar='FORMAT_STRING',
145        nargs=1,
146        help='Format string with optional %%-style arguments.',
147    )
148    parser.add_argument(
149        'raw_args',
150        metavar='ARG',
151        nargs='*',
152        help=(
153            'Arguments for the format string, if any. Arguments are parsed '
154            'as Python expressions, with no builtins (e.g. 9 is the number '
155            '9 and \'"9"\' is the string "9"). Arguments that are not valid '
156            'Python are treated as string literals.'
157        ),
158    )
159    return vars(parser.parse_args())
160
161
162if __name__ == '__main__':
163    sys.exit(_main(**_parse_args()))
164