impl/python/decode_jwt.py

import json
import math
from typing import Any, Dict, List, Mapping, Optional, Sequence, Tuple

from .encoding_base64_base64_url_decode import base64_url_decode

#: Containers nested deeper than this are refused, in every language, rather
#: than left to each parser's own limit.
MAX_DEPTH = 32
MAX_SAFE = 9007199254740991


def _sextet(ch: str) -> int:
    if "A" <= ch <= "Z":
        return ord(ch) - 65
    if "a" <= ch <= "z":
        return ord(ch) - 71
    if "0" <= ch <= "9":
        return ord(ch) + 4
    if ch == "-":
        return 62
    if ch == "_":
        return 63
    return -1


def _is_segment(text: str, allow_empty: bool) -> bool:
    """One canonical, unpadded base64url segment? Checked before decoding so a
    hostile token is an answer (None), never an exception."""
    if len(text) == 0:
        return allow_empty
    if len(text) % 4 == 1:
        return False
    last = 0
    for ch in text:
        last = _sextet(ch)
        if last < 0:
            return False
    if len(text) % 4 == 2:
        return last & 15 == 0
    if len(text) % 4 == 3:
        return last & 3 == 0
    return True


def _reject_constant(name: str) -> Any:
    # json.loads accepts NaN and Infinity, which are not JSON.
    raise ValueError("not JSON: %s" % name)


def _has_surrogate(text: str) -> bool:
    return any(0xD800 <= ord(ch) <= 0xDFFF for ch in text)


def _is_portable(value: Any, depth: int) -> bool:
    """Refuse what the three languages' parsers would answer differently:
    integers beyond 2^53, non-finite numbers, lone surrogates, deep nesting."""
    if isinstance(value, bool) or value is None:
        return True
    if isinstance(value, int):
        return abs(value) <= MAX_SAFE
    if isinstance(value, float):
        return math.isfinite(value) and (not value.is_integer() or abs(value) <= MAX_SAFE)
    if isinstance(value, str):
        return not _has_surrogate(value)
    if depth > MAX_DEPTH:
        return False
    if isinstance(value, list):
        return all(_is_portable(item, depth + 1) for item in value)
    return all(not _has_surrogate(k) and _is_portable(v, depth + 1) for k, v in value.items())


def _parse_object(segment: str) -> Optional[Dict[str, Any]]:
    try:
        text = bytes(base64_url_decode(segment)).decode("utf-8")
        parsed = json.loads(text, parse_constant=_reject_constant)
    except (ValueError, RecursionError):
        return None
    if not isinstance(parsed, dict):
        return None
    return parsed if _is_portable(parsed, 1) else None


def split_jwt(token: str) -> Optional[Tuple[Dict[str, Any], Dict[str, Any], str, List[int]]]:
    """(header, claims, signing input, signature bytes) of a compact JWT, or
    None when it is not three canonical base64url segments whose first two
    are JSON objects. Nothing is verified here; verify_jwt builds on it."""
    if not isinstance(token, str):
        return None
    parts = token.split(".")
    if len(parts) != 3:
        return None
    if not _is_segment(parts[0], False) or not _is_segment(parts[1], False) or not _is_segment(parts[2], True):
        return None
    header = _parse_object(parts[0])
    claims = _parse_object(parts[1])
    if header is None or claims is None:
        return None
    return header, claims, parts[0] + "." + parts[1], base64_url_decode(parts[2])


def check_jwt_secret(secret: Sequence[int]) -> None:
    """RFC 7518 section 3.2: an HS256 key MUST be at least 256 bits."""
    if isinstance(secret, (str, dict)) or not hasattr(secret, "__len__"):
        raise TypeError("secret must be a list of integers from 0 to 255")
    for b in secret:
        if type(b) is not int or b < 0 or b > 255:
            raise ValueError("secret must be a list of integers from 0 to 255")
    if len(secret) < 32:
        raise ValueError("secret must be at least 32 bytes (RFC 7518 section 3.2), received %d" % len(secret))


def decode_jwt(token: str) -> Optional[Mapping[str, Any]]:
    """The claims of a JWT WITHOUT checking its signature, or None when it is
    malformed. For reading exp or name from a token you hold; never for
    deciding whether to trust one."""
    parts = split_jwt(token)
    return None if parts is None else parts[1]