# coding=utf-8
# Copyright 2025 The HuggingFace Inc. team. 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.
"""Fast Image processor class for Perceiver."""

from typing import Optional, Union

import torch
from torchvision.transforms.v2 import functional as F

from ...image_processing_utils_fast import BaseImageProcessorFast, BatchFeature
from ...image_transforms import group_images_by_shape, reorder_images
from ...image_utils import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD, PILImageResampling, SizeDict
from ...utils import (
    TensorType,
    auto_docstring,
)


@auto_docstring
class PerceiverImageProcessorFast(BaseImageProcessorFast):
    resample = PILImageResampling.BICUBIC
    image_mean = IMAGENET_DEFAULT_MEAN
    image_std = IMAGENET_DEFAULT_STD
    size = {"height": 224, "width": 224}
    crop_size = {"height": 256, "width": 256}
    do_resize = True
    do_center_crop = True
    do_rescale = True
    do_normalize = True

    def center_crop(
        self,
        image: "torch.Tensor",
        crop_size: dict[str, int],
        size: dict[str, int],
        **kwargs,
    ) -> "torch.Tensor":
        """
        Center crop an image to `(size["height"] / crop_size["height"] * min_dim, size["width"] / crop_size["width"] *
        min_dim)`. Where `min_dim = min(size["height"], size["width"])`.

        If the input size is smaller than `crop_size` along any edge, the image will be padded with zeros and then
        center cropped.

        Args:
            image (`"torch.Tensor"`):
                Image to center crop.
            crop_size (`dict[str, int]`):
                Desired output size after applying the center crop.
            size (`dict[str, int]`):
                Size of the output image.

        Returns:
            `torch.Tensor`: The center cropped image.
        """
        if size.height is None or size.width is None:
            raise ValueError(f"The size dictionary must have keys 'height' and 'width'. Got {size.keys()}")
        height, width = image.shape[-2:]
        min_dim = min(height, width)
        cropped_height = int((size.height / crop_size.height) * min_dim)
        cropped_width = int((size.width / crop_size.width) * min_dim)
        return super().center_crop(image, SizeDict(height=cropped_height, width=cropped_width))

    def _preprocess(
        self,
        images: list["torch.Tensor"],
        do_resize: bool,
        size: SizeDict,
        interpolation: Optional["F.InterpolationMode"],
        do_center_crop: bool,
        crop_size: SizeDict,
        do_rescale: bool,
        rescale_factor: float,
        do_normalize: bool,
        image_mean: Optional[Union[float, list[float]]],
        image_std: Optional[Union[float, list[float]]],
        disable_grouping: Optional[bool],
        return_tensors: Optional[Union[str, TensorType]],
        **kwargs,
    ) -> BatchFeature:
        # Group images by size for batched resizing
        grouped_images, grouped_images_index = group_images_by_shape(images, disable_grouping=disable_grouping)
        resized_images_grouped = {}
        for shape, stacked_images in grouped_images.items():
            if do_center_crop:
                stacked_images = self.center_crop(stacked_images, size=size, crop_size=crop_size)
            if do_resize:
                stacked_images = self.resize(image=stacked_images, size=size, interpolation=interpolation)
            resized_images_grouped[shape] = stacked_images
        resized_images = reorder_images(resized_images_grouped, grouped_images_index)

        # Group images by size for further processing
        # Needed in case do_resize is False, or resize returns images with different sizes
        grouped_images, grouped_images_index = group_images_by_shape(resized_images, disable_grouping=disable_grouping)
        processed_images_grouped = {}
        for shape, stacked_images in grouped_images.items():
            # Fused rescale and normalize
            stacked_images = self.rescale_and_normalize(
                stacked_images, do_rescale, rescale_factor, do_normalize, image_mean, image_std
            )
            processed_images_grouped[shape] = stacked_images

        processed_images = reorder_images(processed_images_grouped, grouped_images_index)
        processed_images = torch.stack(processed_images, dim=0) if return_tensors else processed_images

        return BatchFeature(data={"pixel_values": processed_images}, tensor_type=return_tensors)


__all__ = ["PerceiverImageProcessorFast"]