use crate::Transformation;
use iced_graphics::layer;
use iced_native::Rectangle;

use bytemuck::{Pod, Zeroable};
use std::mem;
use wgpu::util::DeviceExt;

#[derive(Debug)]
pub struct Pipeline {
    pipeline: wgpu::RenderPipeline,
    constants: wgpu::BindGroup,
    constants_buffer: wgpu::Buffer,
    vertices: wgpu::Buffer,
    indices: wgpu::Buffer,
    instances: wgpu::Buffer,
}

impl Pipeline {
    pub fn new(device: &wgpu::Device, format: wgpu::TextureFormat) -> Pipeline {
        let constant_layout =
            device.create_bind_group_layout(&wgpu::BindGroupLayoutDescriptor {
                label: Some("iced_wgpu::quad uniforms layout"),
                entries: &[wgpu::BindGroupLayoutEntry {
                    binding: 0,
                    visibility: wgpu::ShaderStages::VERTEX,
                    ty: wgpu::BindingType::Buffer {
                        ty: wgpu::BufferBindingType::Uniform,
                        has_dynamic_offset: false,
                        min_binding_size: wgpu::BufferSize::new(
                            mem::size_of::<Uniforms>() as wgpu::BufferAddress,
                        ),
                    },
                    count: None,
                }],
            });

        let constants_buffer = device.create_buffer(&wgpu::BufferDescriptor {
            label: Some("iced_wgpu::quad uniforms buffer"),
            size: mem::size_of::<Uniforms>() as wgpu::BufferAddress,
            usage: wgpu::BufferUsages::UNIFORM | wgpu::BufferUsages::COPY_DST,
            mapped_at_creation: false,
        });

        let constants = device.create_bind_group(&wgpu::BindGroupDescriptor {
            label: Some("iced_wgpu::quad uniforms bind group"),
            layout: &constant_layout,
            entries: &[wgpu::BindGroupEntry {
                binding: 0,
                resource: constants_buffer.as_entire_binding(),
            }],
        });

        let layout =
            device.create_pipeline_layout(&wgpu::PipelineLayoutDescriptor {
                label: Some("iced_wgpu::quad pipeline layout"),
                push_constant_ranges: &[],
                bind_group_layouts: &[&constant_layout],
            });

        let shader =
            device.create_shader_module(&wgpu::ShaderModuleDescriptor {
                label: Some("iced_wgpu::quad::shader"),
                source: wgpu::ShaderSource::Wgsl(std::borrow::Cow::Borrowed(
                    include_str!("shader/quad.wgsl"),
                )),
            });

        let pipeline =
            device.create_render_pipeline(&wgpu::RenderPipelineDescriptor {
                label: Some("iced_wgpu::quad pipeline"),
                layout: Some(&layout),
                vertex: wgpu::VertexState {
                    module: &shader,
                    entry_point: "vs_main",
                    buffers: &[
                        wgpu::VertexBufferLayout {
                            array_stride: mem::size_of::<Vertex>() as u64,
                            step_mode: wgpu::VertexStepMode::Vertex,
                            attributes: &[wgpu::VertexAttribute {
                                shader_location: 0,
                                format: wgpu::VertexFormat::Float32x2,
                                offset: 0,
                            }],
                        },
                        wgpu::VertexBufferLayout {
                            array_stride: mem::size_of::<layer::Quad>() as u64,
                            step_mode: wgpu::VertexStepMode::Instance,
                            attributes: &wgpu::vertex_attr_array!(
                                1 => Float32x2,
                                2 => Float32x2,
                                3 => Float32x4,
                                4 => Float32x4,
                                5 => Float32,
                                6 => Float32,
                            ),
                        },
                    ],
                },
                fragment: Some(wgpu::FragmentState {
                    module: &shader,
                    entry_point: "fs_main",
                    targets: &[wgpu::ColorTargetState {
                        format,
                        blend: Some(wgpu::BlendState {
                            color: wgpu::BlendComponent {
                                src_factor: wgpu::BlendFactor::SrcAlpha,
                                dst_factor: wgpu::BlendFactor::OneMinusSrcAlpha,
                                operation: wgpu::BlendOperation::Add,
                            },
                            alpha: wgpu::BlendComponent {
                                src_factor: wgpu::BlendFactor::One,
                                dst_factor: wgpu::BlendFactor::OneMinusSrcAlpha,
                                operation: wgpu::BlendOperation::Add,
                            },
                        }),
                        write_mask: wgpu::ColorWrites::ALL,
                    }],
                }),
                primitive: wgpu::PrimitiveState {
                    topology: wgpu::PrimitiveTopology::TriangleList,
                    front_face: wgpu::FrontFace::Cw,
                    ..Default::default()
                },
                depth_stencil: None,
                multisample: wgpu::MultisampleState {
                    count: 1,
                    mask: !0,
                    alpha_to_coverage_enabled: false,
                },
                multiview: None,
            });

        let vertices =
            device.create_buffer_init(&wgpu::util::BufferInitDescriptor {
                label: Some("iced_wgpu::quad vertex buffer"),
                contents: bytemuck::cast_slice(&QUAD_VERTS),
                usage: wgpu::BufferUsages::VERTEX,
            });

        let indices =
            device.create_buffer_init(&wgpu::util::BufferInitDescriptor {
                label: Some("iced_wgpu::quad index buffer"),
                contents: bytemuck::cast_slice(&QUAD_INDICES),
                usage: wgpu::BufferUsages::INDEX,
            });

        let instances = device.create_buffer(&wgpu::BufferDescriptor {
            label: Some("iced_wgpu::quad instance buffer"),
            size: mem::size_of::<layer::Quad>() as u64 * MAX_INSTANCES as u64,
            usage: wgpu::BufferUsages::VERTEX | wgpu::BufferUsages::COPY_DST,
            mapped_at_creation: false,
        });

        Pipeline {
            pipeline,
            constants,
            constants_buffer,
            vertices,
            indices,
            instances,
        }
    }

    pub fn draw(
        &mut self,
        device: &wgpu::Device,
        staging_belt: &mut wgpu::util::StagingBelt,
        encoder: &mut wgpu::CommandEncoder,
        instances: &[layer::Quad],
        transformation: Transformation,
        scale: f32,
        bounds: Rectangle<u32>,
        target: &wgpu::TextureView,
    ) {
        let uniforms = Uniforms::new(transformation, scale);

        {
            let mut constants_buffer = staging_belt.write_buffer(
                encoder,
                &self.constants_buffer,
                0,
                wgpu::BufferSize::new(mem::size_of::<Uniforms>() as u64)
                    .unwrap(),
                device,
            );

            constants_buffer.copy_from_slice(bytemuck::bytes_of(&uniforms));
        }

        let mut i = 0;
        let total = instances.len();

        while i < total {
            let end = (i + MAX_INSTANCES).min(total);
            let amount = end - i;

            let instance_bytes = bytemuck::cast_slice(&instances[i..end]);

            let mut instance_buffer = staging_belt.write_buffer(
                encoder,
                &self.instances,
                0,
                wgpu::BufferSize::new(instance_bytes.len() as u64).unwrap(),
                device,
            );

            instance_buffer.copy_from_slice(instance_bytes);

            {
                let mut render_pass =
                    encoder.begin_render_pass(&wgpu::RenderPassDescriptor {
                        label: Some("iced_wgpu::quad render pass"),
                        color_attachments: &[wgpu::RenderPassColorAttachment {
                            view: target,
                            resolve_target: None,
                            ops: wgpu::Operations {
                                load: wgpu::LoadOp::Load,
                                store: true,
                            },
                        }],
                        depth_stencil_attachment: None,
                    });

                render_pass.set_pipeline(&self.pipeline);
                render_pass.set_bind_group(0, &self.constants, &[]);
                render_pass.set_index_buffer(
                    self.indices.slice(..),
                    wgpu::IndexFormat::Uint16,
                );
                render_pass.set_vertex_buffer(0, self.vertices.slice(..));
                render_pass.set_vertex_buffer(1, self.instances.slice(..));

                render_pass.set_scissor_rect(
                    bounds.x,
                    bounds.y,
                    bounds.width,
                    // TODO: Address anti-aliasing adjustments properly
                    bounds.height,
                );

                render_pass.draw_indexed(
                    0..QUAD_INDICES.len() as u32,
                    0,
                    0..amount as u32,
                );
            }

            i += MAX_INSTANCES;
        }
    }
}

#[repr(C)]
#[derive(Clone, Copy, Zeroable, Pod)]
pub struct Vertex {
    _position: [f32; 2],
}

const QUAD_INDICES: [u16; 6] = [0, 1, 2, 0, 2, 3];

const QUAD_VERTS: [Vertex; 4] = [
    Vertex {
        _position: [0.0, 0.0],
    },
    Vertex {
        _position: [1.0, 0.0],
    },
    Vertex {
        _position: [1.0, 1.0],
    },
    Vertex {
        _position: [0.0, 1.0],
    },
];

const MAX_INSTANCES: usize = 100_000;

#[repr(C)]
#[derive(Debug, Clone, Copy, Zeroable, Pod)]
struct Uniforms {
    transform: [f32; 16],
    scale: f32,
    // Uniforms must be aligned to their largest member,
    // this uses a mat4x4<f32> which aligns to 16, so align to that
    _padding: [f32; 3],
}

impl Uniforms {
    fn new(transformation: Transformation, scale: f32) -> Uniforms {
        Self {
            transform: *transformation.as_ref(),
            scale,
            _padding: [0.0; 3],
        }
    }
}

impl Default for Uniforms {
    fn default() -> Self {
        Self {
            transform: *Transformation::identity().as_ref(),
            scale: 1.0,
            _padding: [0.0; 3],
        }
    }
}