aboutsummaryrefslogtreecommitdiff
path: root/pyecsca/ec/mult/base.py
blob: 537f43ec4d86ccbcf959edeb92b4dd23e733cc84 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
"""Provides (mostly abstract) base classes for scalar multipliers, enums used to specify their parameters
and actions used in them."""

from abc import ABC, abstractmethod
from copy import copy
from enum import Enum
from public import public
from typing import Mapping, Tuple, Optional, ClassVar, Set, Type

from pyecsca.ec.context import ResultAction, Action
from pyecsca.ec.formula import Formula
from pyecsca.ec.params import DomainParameters
from pyecsca.ec.point import Point


@public
class ProcessingDirection(Enum):
    """Scalar processing direction."""

    LTR = "Left-to-right"
    RTL = "Right-to-left"


@public
class AccumulationOrder(Enum):
    """Accumulation order (makes a difference for the projective result)."""

    PeqPR = "P = P + R"
    PeqRP = "P = R + P"


@public
class ScalarMultiplicationAction(ResultAction):
    """A scalar multiplication of a point on a curve by a scalar."""

    point: Point
    scalar: int

    def __init__(self, point: Point, scalar: int):
        super().__init__()
        self.point = point
        self.scalar = scalar

    def __repr__(self):
        return f"{self.__class__.__name__}({self.point}, {self.scalar})"


@public
class PrecomputationAction(Action):
    """A precomputation of a point in scalar multiplication."""

    params: DomainParameters
    point: Point

    def __init__(self, params: DomainParameters, point: Point):
        super().__init__()
        self.params = params
        self.point = point

    def __repr__(self):
        return f"{self.__class__.__name__}({self.params}, {self.point})"


@public
class ScalarMultiplier(ABC):
    """
    A scalar multiplication algorithm.

    .. note::
        The __init__ method of all concrete subclasses needs to have type annotations so that
        configuration enumeration works.

    :param short_circuit: Whether the use of formulas will be guarded by short-circuit on inputs
                          of the point at infinity.
    :param formulas: Formulas this instance will use.
    """

    requires: ClassVar[Set[Type]]  # Type[Formula] but mypy has a false positive
    """The set of formula types that the multiplier requires."""
    optionals: ClassVar[Set[Type]]  # Type[Formula] but mypy has a false positive
    """The optional set of formula types that the multiplier can use."""
    short_circuit: bool
    """Whether the formulas will short-circuit upon input of the point at infinity."""
    formulas: Mapping[str, Formula]
    """All formulas the multiplier was initialized with."""
    _params: DomainParameters
    _point: Point
    _initialized: bool = False

    def __init__(self, short_circuit: bool = True, **formulas: Optional[Formula]):
        if (
            len(
                {
                    formula.coordinate_model
                    for formula in formulas.values()
                    if formula is not None
                }
            )
            != 1
        ):
            raise ValueError("Formulas need to belong to the same coordinate model.")
        self.short_circuit = short_circuit
        self.formulas = {k: v for k, v in formulas.items() if v is not None}
        found_required = set()
        for formula in self.formulas.values():
            for required in self.requires:
                if isinstance(formula, required):
                    found_required.add(required)
                    break
            else:
                for optional in self.optionals:
                    if isinstance(formula, optional):
                        break
                else:
                    raise ValueError("Not required or optional formulas provided.")
        if found_required != self.requires:
            raise ValueError("Required formulas missing.")

    def _add(self, one: Point, other: Point) -> Point:
        if "add" not in self.formulas:
            raise NotImplementedError
        if self.short_circuit:
            if one == self._params.curve.neutral:
                return copy(other)
            if other == self._params.curve.neutral:
                return copy(one)
        return self.formulas["add"](
            self._params.curve.prime, one, other, **self._params.curve.parameters
        )[0]

    def _dbl(self, point: Point) -> Point:
        if "dbl" not in self.formulas:
            raise NotImplementedError
        if self.short_circuit and point == self._params.curve.neutral:
            return copy(point)
        return self.formulas["dbl"](
            self._params.curve.prime, point, **self._params.curve.parameters
        )[0]

    def _scl(self, point: Point) -> Point:
        if "scl" not in self.formulas:
            raise NotImplementedError
        return self.formulas["scl"](
            self._params.curve.prime, point, **self._params.curve.parameters
        )[0]

    def _ladd(self, start: Point, to_dbl: Point, to_add: Point) -> Tuple[Point, ...]:
        if "ladd" not in self.formulas:
            raise NotImplementedError
        if self.short_circuit:
            if to_dbl == self._params.curve.neutral:
                return to_dbl, to_add
            if to_add == self._params.curve.neutral:
                return self._dbl(to_dbl), to_dbl
        return self.formulas["ladd"](
            self._params.curve.prime,
            start,
            to_dbl,
            to_add,
            **self._params.curve.parameters,
        )

    def _dadd(self, start: Point, one: Point, other: Point) -> Point:
        if "dadd" not in self.formulas:
            raise NotImplementedError
        if self.short_circuit:
            if one == self._params.curve.neutral:
                return copy(other)
            if other == self._params.curve.neutral:
                return copy(one)
        return self.formulas["dadd"](
            self._params.curve.prime, start, one, other, **self._params.curve.parameters
        )[0]

    def _neg(self, point: Point) -> Point:
        if "neg" not in self.formulas:
            raise NotImplementedError
        return self.formulas["neg"](
            self._params.curve.prime, point, **self._params.curve.parameters
        )[0]

    def __hash__(self):
        return hash(
            (
                ScalarMultiplier,
                tuple(self.formulas.keys()),
                tuple(self.formulas.values()),
                self.short_circuit,
            )
        )

    def __eq__(self, other):
        if not isinstance(other, ScalarMultiplier):
            return False
        return (
            self.formulas == other.formulas
            and self.short_circuit == other.short_circuit
        )

    def __repr__(self):
        return f"{self.__class__.__name__}({', '.join(map(str, self.formulas.values()))}, short_circuit={self.short_circuit})"

    def init(self, params: DomainParameters, point: Point):
        """
        Initialize the scalar multiplier with :paramref:`~.init.params` and a :paramref:`~.init.point`.

        .. warning::
            The point is not verified to be on the curve represented in the domain parameters.

        :param params: The domain parameters to initialize the multiplier with.
        :param point: The point to initialize the multiplier with.
        """
        coord_model = set(self.formulas.values()).pop().coordinate_model
        if (
            params.curve.coordinate_model != coord_model
            or point.coordinate_model != coord_model
        ):
            raise ValueError(
                "Coordinate models of the parameters, formulas or point are not compatible."
            )
        self._params = params
        self._point = point
        self._initialized = True

    @abstractmethod
    def multiply(self, scalar: int) -> Point:
        """
        Multiply the point with the scalar.

        .. note::
            The multiplier needs to be initialized by a call to the :py:meth:`.init` method.

        :param scalar: The scalar to use.
        :return: The resulting multiple.
        """
        raise NotImplementedError


@public
class AccumulatorMultiplier(ScalarMultiplier, ABC):
    """
    A scalar multiplication algorithm mix-in class for a multiplier that accumulates.

    :param accumulation_order: The order of accumulation of points.
    """

    accumulation_order: AccumulationOrder
    """The order of accumulation of points."""

    def __init__(
        self,
        *args,
        accumulation_order: AccumulationOrder = AccumulationOrder.PeqPR,
        **kwargs,
    ):
        super().__init__(*args, **kwargs)
        self.accumulation_order = accumulation_order

    def _accumulate(self, p: Point, r: Point) -> Point:
        if self.accumulation_order is AccumulationOrder.PeqPR:
            p = self._add(p, r)
        elif self.accumulation_order is AccumulationOrder.PeqRP:
            p = self._add(r, p)
        return p