pw_tokenizer/py/pw_tokenizer/encode.py

# Copyright 2020 The Pigweed Authors
#
# Licensed under the Apache License, Version 2.0 (the "License"); you may not
# use this file except in compliance with the License. You may obtain a copy of
# the License at
#
#     https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
# License for the specific language governing permissions and limitations under
# the License.
"""Provides functionality for encoding tokenized messages."""

import argparse
import base64
import struct
import sys
from typing import Sequence

from pw_tokenizer import tokens

_INT32_MAX = 2**31 - 1
_UINT32_MAX = 2**32 - 1
NESTED_TOKEN_PREFIX = '$'
NESTED_TOKEN_BASE_PREFIX = '#'


def _zig_zag_encode(value: int) -> int:
    """Encodes signed integers to give a compact varint encoding."""
    return value << 1 if value >= 0 else (value << 1) ^ (~0)


def _little_endian_base128_encode(integer: int) -> bytearray:
    data = bytearray()

    while True:
        # Grab 7 bits; the eighth bit is set to 1 to indicate more data coming.
        data.append((integer & 0x7F) | 0x80)
        integer >>= 7

        if not integer:
            break

    data[-1] &= 0x7F  # clear the top bit of the last byte
    return data


def _encode_int32(arg: int) -> bytearray:
    # Convert large unsigned numbers into their corresponding signed values.
    if arg > _INT32_MAX:
        arg -= 2**32

    return _little_endian_base128_encode(_zig_zag_encode(arg))


def _encode_string(arg: bytes) -> bytes:
    size_byte = len(arg) if len(arg) < 128 else 0xFF
    return struct.pack('B', size_byte) + arg[:127]


def encode_args(*args: int | float | bytes | str) -> bytes:
    """Encodes a list of arguments to their on-wire representation."""

    data = bytearray(b'')
    for arg in args:
        if isinstance(arg, int):
            if arg.bit_length() > 32:
                raise ValueError(
                    f'Cannot encode {arg}: only 32-bit integers may be encoded'
                )
            data += _encode_int32(arg)
        elif isinstance(arg, float):
            data += struct.pack('<f', arg)
        elif isinstance(arg, str):
            data += _encode_string(arg.encode())
        elif isinstance(arg, bytes):
            data += _encode_string(arg)
        else:
            raise ValueError(
                f'{arg} has type {type(arg)}, which is not supported'
            )
    return bytes(data)


def encode_token_and_args(
    token: int, *args: int | float | bytes | str
) -> bytes:
    """Encodes a tokenized message given its token and arguments.

    This function assumes that the token represents a format string with
    conversion specifiers that correspond with the provided argument types.
    Currently, only 32-bit integers are supported.
    """

    if token < 0 or token > _UINT32_MAX:
        raise ValueError(
            f'The token ({token}) must be an unsigned 32-bit integer'
        )

    return struct.pack('<I', token) + encode_args(*args)


def prefixed_base64(data: bytes, prefix: str = '$') -> str:
    """Encodes a tokenized message as prefixed Base64."""
    return prefix + base64.b64encode(data).decode()


def _parse_user_input(string: str):
    """Evaluates a string as Python code or returns it as a literal string."""
    try:
        value = eval(string, dict(__builtins__={}))  # pylint: disable=eval-used
    except (NameError, SyntaxError):
        return string

    return value if isinstance(value, (int, float)) else string


def _main(format_string_list: Sequence[str], raw_args: Sequence[str]) -> int:
    (format_string,) = format_string_list
    token = tokens.pw_tokenizer_65599_hash(format_string)
    args = tuple(_parse_user_input(a) for a in raw_args)

    data = encode_token_and_args(token, *args)
    token = int.from_bytes(data[:4], 'little')
    binary = ' '.join(f'{b:02x}' for b in data)

    print(f'      Raw input: {format_string!r} % {args!r}')
    print(f'Formatted input: {format_string % args}')
    print(f'          Token: 0x{token:08x}')
    print(f'        Encoded: {data!r} ({binary}) [{len(data)} bytes]')
    print(f'Prefixed Base64: {prefixed_base64(data)}')

    return 0


def _parse_args() -> dict:
    parser = argparse.ArgumentParser(
        description=__doc__,
        formatter_class=argparse.RawDescriptionHelpFormatter,
    )
    parser.add_argument(
        'format_string_list',
        metavar='FORMAT_STRING',
        nargs=1,
        help='Format string with optional %%-style arguments.',
    )
    parser.add_argument(
        'raw_args',
        metavar='ARG',
        nargs='*',
        help=(
            'Arguments for the format string, if any. Arguments are parsed '
            'as Python expressions, with no builtins (e.g. 9 is the number '
            '9 and \'"9"\' is the string "9"). Arguments that are not valid '
            'Python are treated as string literals.'
        ),
    )
    return vars(parser.parse_args())


if __name__ == '__main__':
    sys.exit(_main(**_parse_args()))