arcane_core/renderer/

sprite.rs

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

use super::camera::Camera2D;
use super::gpu::GpuContext;
use super::lighting::LightingUniform;
use super::texture::TextureStore;

/// Blend mode constants. Matches TS enum order.
pub const BLEND_ALPHA: u8 = 0;
pub const BLEND_ADDITIVE: u8 = 1;
pub const BLEND_MULTIPLY: u8 = 2;
pub const BLEND_SCREEN: u8 = 3;

/// A sprite draw command queued from TypeScript.
#[derive(Debug, Clone)]
pub struct SpriteCommand {
    pub texture_id: u32,
    pub x: f32,
    pub y: f32,
    pub w: f32,
    pub h: f32,
    pub layer: i32,
    pub uv_x: f32,
    pub uv_y: f32,
    pub uv_w: f32,
    pub uv_h: f32,
    pub tint_r: f32,
    pub tint_g: f32,
    pub tint_b: f32,
    pub tint_a: f32,
    pub rotation: f32,
    pub origin_x: f32,
    pub origin_y: f32,
    pub flip_x: bool,
    pub flip_y: bool,
    pub opacity: f32,
    pub blend_mode: u8,
    pub shader_id: u32,
}

/// Per-vertex data for the unit quad.
#[repr(C)]
#[derive(Copy, Clone, Pod, Zeroable)]
struct QuadVertex {
    position: [f32; 2],
    uv: [f32; 2],
}

/// Per-instance data for each sprite.
#[repr(C)]
#[derive(Copy, Clone, Pod, Zeroable)]
struct SpriteInstance {
    world_pos: [f32; 2],
    size: [f32; 2],
    uv_offset: [f32; 2],
    uv_size: [f32; 2],
    tint: [f32; 4],
    /// [rotation_radians, origin_x (0-1), origin_y (0-1), padding]
    rotation_origin: [f32; 4],
}

/// Camera uniform buffer data.
#[repr(C)]
#[derive(Copy, Clone, Pod, Zeroable)]
struct CameraUniform {
    view_proj: [f32; 16],
}

// Unit quad: two triangles forming a 1x1 square at origin
const QUAD_VERTICES: &[QuadVertex] = &[
    QuadVertex { position: [0.0, 0.0], uv: [0.0, 0.0] }, // top-left
    QuadVertex { position: [1.0, 0.0], uv: [1.0, 0.0] }, // top-right
    QuadVertex { position: [1.0, 1.0], uv: [1.0, 1.0] }, // bottom-right
    QuadVertex { position: [0.0, 1.0], uv: [0.0, 1.0] }, // bottom-left
];

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

/// Get the wgpu BlendState for each blend mode.
fn blend_state_for(mode: u8) -> wgpu::BlendState {
    use wgpu::{BlendComponent, BlendFactor, BlendOperation};
    match mode {
        BLEND_ALPHA => wgpu::BlendState::ALPHA_BLENDING,
        BLEND_ADDITIVE => wgpu::BlendState {
            color: BlendComponent {
                src_factor: BlendFactor::SrcAlpha,
                dst_factor: BlendFactor::One,
                operation: BlendOperation::Add,
            },
            alpha: BlendComponent {
                src_factor: BlendFactor::One,
                dst_factor: BlendFactor::One,
                operation: BlendOperation::Add,
            },
        },
        BLEND_MULTIPLY => wgpu::BlendState {
            color: BlendComponent {
                src_factor: BlendFactor::Dst,
                dst_factor: BlendFactor::OneMinusSrcAlpha,
                operation: BlendOperation::Add,
            },
            alpha: BlendComponent {
                src_factor: BlendFactor::DstAlpha,
                dst_factor: BlendFactor::OneMinusSrcAlpha,
                operation: BlendOperation::Add,
            },
        },
        BLEND_SCREEN => wgpu::BlendState {
            color: BlendComponent {
                src_factor: BlendFactor::One,
                dst_factor: BlendFactor::OneMinusSrc,
                operation: BlendOperation::Add,
            },
            alpha: BlendComponent {
                src_factor: BlendFactor::One,
                dst_factor: BlendFactor::OneMinusSrcAlpha,
                operation: BlendOperation::Add,
            },
        },
        _ => wgpu::BlendState::ALPHA_BLENDING, // unknown → default to alpha
    }
}

pub struct SpritePipeline {
    /// One pipeline per blend mode: [alpha, additive, multiply, screen]
    pipelines: [wgpu::RenderPipeline; 4],
    vertex_buffer: wgpu::Buffer,
    index_buffer: wgpu::Buffer,
    camera_buffer: wgpu::Buffer,
    camera_bind_group: wgpu::BindGroup,
    pub texture_bind_group_layout: wgpu::BindGroupLayout,
    lighting_buffer: wgpu::Buffer,
    lighting_bind_group: wgpu::BindGroup,
}

impl SpritePipeline {
    /// Create a sprite pipeline for headless testing.
    /// Takes raw GPU components instead of GpuContext (which requires a surface).
    pub fn new_headless(
        device: &wgpu::Device,
        queue: &wgpu::Queue,
        format: wgpu::TextureFormat,
    ) -> Self {
        Self::new_internal(device, queue, format)
    }

    pub fn new(gpu: &GpuContext) -> Self {
        Self::new_internal(&gpu.device, &gpu.queue, gpu.config.format)
    }

    fn new_internal(
        device: &wgpu::Device,
        _queue: &wgpu::Queue,
        surface_format: wgpu::TextureFormat,
    ) -> Self {
        let shader = device.create_shader_module(wgpu::ShaderModuleDescriptor {
            label: Some("sprite_shader"),
            source: wgpu::ShaderSource::Wgsl(
                include_str!("shaders/sprite.wgsl").into(),
            ),
        });

        // Camera uniform bind group layout (group 0)
        let camera_bind_group_layout =
            device
                .create_bind_group_layout(&wgpu::BindGroupLayoutDescriptor {
                    label: Some("camera_bind_group_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: None,
                        },
                        count: None,
                    }],
                });

        // Texture bind group layout (group 1)
        let texture_bind_group_layout =
            device
                .create_bind_group_layout(&wgpu::BindGroupLayoutDescriptor {
                    label: Some("texture_bind_group_layout"),
                    entries: &[
                        wgpu::BindGroupLayoutEntry {
                            binding: 0,
                            visibility: wgpu::ShaderStages::FRAGMENT,
                            ty: wgpu::BindingType::Texture {
                                multisampled: false,
                                view_dimension: wgpu::TextureViewDimension::D2,
                                sample_type: wgpu::TextureSampleType::Float { filterable: true },
                            },
                            count: None,
                        },
                        wgpu::BindGroupLayoutEntry {
                            binding: 1,
                            visibility: wgpu::ShaderStages::FRAGMENT,
                            ty: wgpu::BindingType::Sampler(wgpu::SamplerBindingType::Filtering),
                            count: None,
                        },
                    ],
                });

        // Lighting uniform bind group layout (group 2)
        let lighting_bind_group_layout =
            device
                .create_bind_group_layout(&wgpu::BindGroupLayoutDescriptor {
                    label: Some("lighting_bind_group_layout"),
                    entries: &[wgpu::BindGroupLayoutEntry {
                        binding: 0,
                        visibility: wgpu::ShaderStages::FRAGMENT,
                        ty: wgpu::BindingType::Buffer {
                            ty: wgpu::BufferBindingType::Uniform,
                            has_dynamic_offset: false,
                            min_binding_size: None,
                        },
                        count: None,
                    }],
                });

        let pipeline_layout =
            device
                .create_pipeline_layout(&wgpu::PipelineLayoutDescriptor {
                    label: Some("sprite_pipeline_layout"),
                    bind_group_layouts: &[
                        &camera_bind_group_layout,
                        &texture_bind_group_layout,
                        &lighting_bind_group_layout,
                    ],
                    push_constant_ranges: &[],
                });

        // Vertex buffer layouts
        let vertex_layout = wgpu::VertexBufferLayout {
            array_stride: std::mem::size_of::<QuadVertex>() as wgpu::BufferAddress,
            step_mode: wgpu::VertexStepMode::Vertex,
            attributes: &[
                wgpu::VertexAttribute {
                    offset: 0,
                    shader_location: 0,
                    format: wgpu::VertexFormat::Float32x2,
                },
                wgpu::VertexAttribute {
                    offset: 8,
                    shader_location: 1,
                    format: wgpu::VertexFormat::Float32x2,
                },
            ],
        };

        let instance_layout = wgpu::VertexBufferLayout {
            array_stride: std::mem::size_of::<SpriteInstance>() as wgpu::BufferAddress,
            step_mode: wgpu::VertexStepMode::Instance,
            attributes: &[
                wgpu::VertexAttribute {
                    offset: 0,
                    shader_location: 2,
                    format: wgpu::VertexFormat::Float32x2, // world_pos
                },
                wgpu::VertexAttribute {
                    offset: 8,
                    shader_location: 3,
                    format: wgpu::VertexFormat::Float32x2, // size
                },
                wgpu::VertexAttribute {
                    offset: 16,
                    shader_location: 4,
                    format: wgpu::VertexFormat::Float32x2, // uv_offset
                },
                wgpu::VertexAttribute {
                    offset: 24,
                    shader_location: 5,
                    format: wgpu::VertexFormat::Float32x2, // uv_size
                },
                wgpu::VertexAttribute {
                    offset: 32,
                    shader_location: 6,
                    format: wgpu::VertexFormat::Float32x4, // tint
                },
                wgpu::VertexAttribute {
                    offset: 48,
                    shader_location: 7,
                    format: wgpu::VertexFormat::Float32x4, // rotation_origin
                },
            ],
        };

        // Create one pipeline per blend mode
        let blend_names = ["alpha", "additive", "multiply", "screen"];
        let pipelines: Vec<wgpu::RenderPipeline> = (0..4u8)
            .map(|mode| {
                device.create_render_pipeline(&wgpu::RenderPipelineDescriptor {
                    label: Some(&format!("sprite_pipeline_{}", blend_names[mode as usize])),
                    layout: Some(&pipeline_layout),
                    vertex: wgpu::VertexState {
                        module: &shader,
                        entry_point: Some("vs_main"),
                        buffers: &[vertex_layout.clone(), instance_layout.clone()],
                        compilation_options: Default::default(),
                    },
                    fragment: Some(wgpu::FragmentState {
                        module: &shader,
                        entry_point: Some("fs_main"),
                        targets: &[Some(wgpu::ColorTargetState {
                            format: surface_format,
                            blend: Some(blend_state_for(mode)),
                            write_mask: wgpu::ColorWrites::ALL,
                        })],
                        compilation_options: Default::default(),
                    }),
                    primitive: wgpu::PrimitiveState {
                        topology: wgpu::PrimitiveTopology::TriangleList,
                        strip_index_format: None,
                        front_face: wgpu::FrontFace::Ccw,
                        cull_mode: None,
                        polygon_mode: wgpu::PolygonMode::Fill,
                        unclipped_depth: false,
                        conservative: false,
                    },
                    depth_stencil: None,
                    multisample: wgpu::MultisampleState::default(),
                    multiview: None,
                    cache: None,
                })
            })
            .collect();

        let pipelines: [wgpu::RenderPipeline; 4] = pipelines.try_into().unwrap();

        let vertex_buffer = device.create_buffer_init(&wgpu::util::BufferInitDescriptor {
            label: Some("quad_vertex_buffer"),
            contents: bytemuck::cast_slice(QUAD_VERTICES),
            usage: wgpu::BufferUsages::VERTEX,
        });

        let index_buffer = device.create_buffer_init(&wgpu::util::BufferInitDescriptor {
            label: Some("quad_index_buffer"),
            contents: bytemuck::cast_slice(QUAD_INDICES),
            usage: wgpu::BufferUsages::INDEX,
        });

        let camera_uniform = CameraUniform {
            view_proj: Camera2D::default().view_proj(),
        };

        let camera_buffer = device.create_buffer_init(&wgpu::util::BufferInitDescriptor {
            label: Some("camera_uniform_buffer"),
            contents: bytemuck::cast_slice(&[camera_uniform]),
            usage: wgpu::BufferUsages::UNIFORM | wgpu::BufferUsages::COPY_DST,
        });

        let camera_bind_group = device.create_bind_group(&wgpu::BindGroupDescriptor {
            label: Some("camera_bind_group"),
            layout: &camera_bind_group_layout,
            entries: &[wgpu::BindGroupEntry {
                binding: 0,
                resource: camera_buffer.as_entire_binding(),
            }],
        });

        // Lighting uniform buffer (272 bytes = LightingUniform size)
        let default_lighting = LightingUniform {
            ambient: [1.0, 1.0, 1.0],
            light_count: 0,
            lights: [super::lighting::LightData {
                pos_radius: [0.0; 4],
                color_intensity: [0.0; 4],
            }; super::lighting::MAX_LIGHTS],
        };

        let lighting_buffer = device.create_buffer_init(&wgpu::util::BufferInitDescriptor {
            label: Some("lighting_uniform_buffer"),
            contents: bytemuck::cast_slice(&[default_lighting]),
            usage: wgpu::BufferUsages::UNIFORM | wgpu::BufferUsages::COPY_DST,
        });

        let lighting_bind_group = device.create_bind_group(&wgpu::BindGroupDescriptor {
            label: Some("lighting_bind_group"),
            layout: &lighting_bind_group_layout,
            entries: &[wgpu::BindGroupEntry {
                binding: 0,
                resource: lighting_buffer.as_entire_binding(),
            }],
        });

        Self {
            pipelines,
            vertex_buffer,
            index_buffer,
            camera_buffer,
            camera_bind_group,
            texture_bind_group_layout,
            lighting_buffer,
            lighting_bind_group,
        }
    }

    /// Return the camera uniform bind group (group 0).
    /// Used by other pipelines (e.g. GeometryBatch) that share the same view-proj matrix.
    pub fn camera_bind_group(&self) -> &wgpu::BindGroup {
        &self.camera_bind_group
    }

    /// Write camera and lighting uniforms to GPU buffers. Call once per frame
    /// before any `render()` calls to avoid redundant buffer writes.
    pub fn prepare(
        &self,
        device: &wgpu::Device,
        queue: &wgpu::Queue,
        camera: &Camera2D,
        lighting: &LightingUniform,
    ) {
        let _ = device; // device not needed for buffer writes, but kept for consistency
        let camera_uniform = CameraUniform {
            view_proj: camera.view_proj(),
        };
        queue.write_buffer(
            &self.camera_buffer,
            0,
            bytemuck::cast_slice(&[camera_uniform]),
        );
        queue.write_buffer(
            &self.lighting_buffer,
            0,
            bytemuck::cast_slice(&[*lighting]),
        );
    }

    /// Render a sorted list of sprite commands.
    /// Commands should be sorted by layer → shader_id → blend_mode → texture_id.
    ///
    /// `clear_color`: `Some(color)` → `LoadOp::Clear(color)` (first pass),
    ///                 `None` → `LoadOp::Load` (subsequent passes).
    pub fn render(
        &self,
        device: &wgpu::Device,
        _queue: &wgpu::Queue,
        textures: &TextureStore,
        shaders: &super::shader::ShaderStore,
        commands: &[SpriteCommand],
        target: &wgpu::TextureView,
        encoder: &mut wgpu::CommandEncoder,
        clear_color: Option<wgpu::Color>,
    ) {
        let load_op = match clear_color {
            Some(color) => wgpu::LoadOp::Clear(color),
            None => wgpu::LoadOp::Load,
        };

        let mut render_pass = encoder.begin_render_pass(&wgpu::RenderPassDescriptor {
            label: Some("sprite_render_pass"),
            color_attachments: &[Some(wgpu::RenderPassColorAttachment {
                view: target,
                resolve_target: None,
                ops: wgpu::Operations {
                    load: load_op,
                    store: wgpu::StoreOp::Store,
                },
            })],
            depth_stencil_attachment: None,
            timestamp_writes: None,
            occlusion_query_set: None,
        });

        render_pass.set_bind_group(0, &self.camera_bind_group, &[]);
        render_pass.set_bind_group(2, &self.lighting_bind_group, &[]);
        render_pass.set_vertex_buffer(0, self.vertex_buffer.slice(..));
        render_pass.set_index_buffer(self.index_buffer.slice(..), wgpu::IndexFormat::Uint16);

        // Batch by shader_id + blend_mode + texture_id (commands pre-sorted)
        let mut current_shader: Option<u32> = None;
        let mut current_blend: Option<u8> = None;
        let mut i = 0;
        while i < commands.len() {
            let shader = commands[i].shader_id;
            let blend = commands[i].blend_mode.min(3);
            let tex_id = commands[i].texture_id;
            let batch_start = i;
            while i < commands.len()
                && commands[i].shader_id == shader
                && commands[i].blend_mode.min(3) == blend
                && commands[i].texture_id == tex_id
            {
                i += 1;
            }
            let batch = &commands[batch_start..i];

            // Switch pipeline: built-in (shader_id 0) vs custom
            if shader == 0 {
                if current_shader != Some(0) || current_blend != Some(blend) {
                    render_pass.set_pipeline(&self.pipelines[blend as usize]);
                    current_shader = Some(0);
                    current_blend = Some(blend);
                }
            } else if current_shader != Some(shader) {
                if let Some(pipeline) = shaders.get_pipeline(shader) {
                    render_pass.set_pipeline(pipeline);
                    if let Some(bg) = shaders.get_bind_group(shader) {
                        render_pass.set_bind_group(3, bg, &[]);
                    }
                    current_shader = Some(shader);
                    current_blend = None;
                } else {
                    continue; // skip batch if shader not loaded
                }
            }

            // Get texture bind group
            let bind_group = match textures.get_bind_group(tex_id) {
                Some(bg) => bg,
                None => continue, // skip if texture not loaded
            };

            // Build instance buffer for this batch
            let instances: Vec<SpriteInstance> = batch
                .iter()
                .map(|cmd| {
                    // Apply flip by negating UV and shifting offset
                    let mut uv_x = cmd.uv_x;
                    let mut uv_y = cmd.uv_y;
                    let mut uv_w = cmd.uv_w;
                    let mut uv_h = cmd.uv_h;
                    if cmd.flip_x {
                        uv_x += uv_w;
                        uv_w = -uv_w;
                    }
                    if cmd.flip_y {
                        uv_y += uv_h;
                        uv_h = -uv_h;
                    }
                    SpriteInstance {
                        world_pos: [cmd.x, cmd.y],
                        size: [cmd.w, cmd.h],
                        uv_offset: [uv_x, uv_y],
                        uv_size: [uv_w, uv_h],
                        tint: [cmd.tint_r, cmd.tint_g, cmd.tint_b, cmd.tint_a * cmd.opacity],
                        rotation_origin: [cmd.rotation, cmd.origin_x, cmd.origin_y, 0.0],
                    }
                })
                .collect();

            let instance_buffer =
                device
                    .create_buffer_init(&wgpu::util::BufferInitDescriptor {
                        label: Some("sprite_instance_buffer"),
                        contents: bytemuck::cast_slice(&instances),
                        usage: wgpu::BufferUsages::VERTEX,
                    });

            render_pass.set_bind_group(1, bind_group, &[]);
            render_pass.set_vertex_buffer(1, instance_buffer.slice(..));
            render_pass.draw_indexed(0..6, 0, 0..instances.len() as u32);
        }
    }
}