# Copyright 2022 The GPflow Contributors. All Rights Reserved.
#
# 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
#
# http://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.
"""
Code for specifying expectations around shapes.
"""
from dataclasses import dataclass
from typing import Optional, Tuple
from .argument_ref import ArgumentRef
[docs]@dataclass(frozen=True)
class ParsedDimensionSpec:
constant: Optional[int]
variable_name: Optional[str]
variable_rank: bool
def __post_init__(self) -> None:
assert (self.constant is None) != (
self.variable_name is None
), "Argument must be either constant or variable."
if self.variable_rank:
assert (
self.variable_rank is not None
), "Variable-rank dimensions must be bound to a variable."
assert self.constant is None, "Constants cannot have a variable rank."
def __repr__(self) -> str:
if self.constant is not None:
return str(self.constant)
else:
assert self.variable_name is not None
suffix = "..." if self.variable_rank else ""
return self.variable_name + suffix
[docs]@dataclass(frozen=True)
class ParsedShapeSpec:
dims: Tuple[ParsedDimensionSpec, ...]
def __post_init__(self) -> None:
n_variable_rank = sum(dim.variable_rank for dim in self.dims)
assert (
n_variable_rank <= 1
), f"At most one variable-rank dimension allowed. Found {n_variable_rank} in {self}."
def __repr__(self) -> str:
dims = [repr(dim) for dim in self.dims]
return f"({', '.join(dims)})"
[docs]@dataclass(frozen=True)
class ParsedArgumentSpec:
argument_ref: ArgumentRef
shape: ParsedShapeSpec
def __repr__(self) -> str:
return f"{self.argument_ref}: {self.shape}"