Source code for nvtripy.frontend.dimension_size

#
# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: Apache-2.0
#
# 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.
#

from typing import Optional

from nvtripy import export
from nvtripy.frontend.tensor import Tensor




@export.public_api(document_under="tensor")
class DimensionSize(Tensor):
    """
    A 0D, :class:`int32` tensor that represents a scalar value extracted from the shape of a tensor.
    """

    def __init__(self, data: int, name: Optional[str] = None) -> None:
        r"""
        Args:
            data: The value of the DimensionSize, which should be a scalar integer.
            name: An optional name.
        """
        super().__init__(data=data, name=name)

    def __int__(self) -> int:
        return self.tolist()

    def __repr__(self) -> str:
        return str(self)

    def __str__(self) -> str:
        val = self.tolist()
        assert isinstance(val, int)
        return str(val)



    def eval(self) -> "nvtripy.DimensionSize":
        """
        Immediately evaluates this ``DimensionSize`` object.

        .. note:: ``DimensionSize`` will always reside on host even after it is evaluated.

        Returns:
            The evaluated ``DimensionSize``.

        .. code-block:: python
            :linenos:


            dim_size = tp.ones((2, 2)).shape[0]
            dim_size.eval()
            print(dim_size.device)
            assert dim_size.device.kind == "cpu"

        """
        from nvtripy.backend.mlir import memref
        from nvtripy.trace.ops.constant import Constant
        from nvtripy.trace.ops.shape import GetDimensionSize, Shape

        # TODO (#593): Generalize this to any branchy graph:
        # If we find a pattern like Shape -> GetDimensionSize, we want to eval the Shape operation
        # so that we aren't evaluating the entire graph for each dimension.
        producer = self.trace_tensor.producer
        if isinstance(producer, GetDimensionSize) and isinstance(producer.inputs[0].producer, Shape):
            shape_tensor = producer.inputs[0].frontend_tensor
            if shape_tensor is None:
                shape_tensor = Tensor.from_trace_tensor(producer.inputs[0], preserve_existing_stack_info=True)
            shape_tensor.eval()

            # `eval()` above can rebind the frontend tensor to a Constant. Keep using the original
            # traced Shape output here so we rebuild only the cheap GetDimensionSize node.
            dim_size = GetDimensionSize([producer.inputs[0]], dim=producer.dim)
            dim_size.outputs[0].is_compile_tracer = self.trace_tensor.is_compile_tracer
            self.trace_tensor = dim_size.outputs[0]

        if not isinstance(producer, Constant):
            super().eval()
        dim_value = memref.tolist(self.trace_tensor.producer.data)
        dim_size = DimensionSize(data=int(dim_value), name=self.name)
        self.trace_tensor = dim_size.trace_tensor
        return self