Crafter.Graphics/interfaces/Crafter.Graphics-UI.cppm

431 lines
20 KiB
Text
Raw Normal View History

2026-03-09 20:10:19 +01:00
/*
Crafter®.Graphics
Copyright (C) 2026 Catcrafts®
2026-05-01 23:35:37 +02:00
catcrafts.net
2026-03-09 20:10:19 +01:00
This library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
2026-05-01 23:35:37 +02:00
License version 3.0 as published by the Free Software Foundation;
2026-03-09 20:10:19 +01:00
This library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with this library; if not, write to the Free Software
Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
2026-05-02 21:08:20 +02:00
module;
2026-05-18 02:07:48 +02:00
#ifndef CRAFTER_GRAPHICS_WINDOW_DOM
2026-05-02 21:08:20 +02:00
#include "vulkan/vulkan.h"
2026-05-18 04:58:52 +02:00
#endif
2026-05-01 23:35:37 +02:00
export module Crafter.Graphics:UI;
2026-05-02 21:08:20 +02:00
import std;
import Crafter.Event;
import :Window;
import :RenderPass;
2026-05-18 04:58:52 +02:00
import :GraphicsTypes;
import :FontAtlas;
import :Font;
#ifndef CRAFTER_GRAPHICS_WINDOW_DOM
import :Device;
2026-05-02 21:08:20 +02:00
import :DescriptorHeapVulkan;
import :ImageVulkan;
import :VulkanBuffer;
import :ComputeShader;
2026-05-18 04:58:52 +02:00
#else
import :DescriptorHeapWebGPU;
import :WebGPU;
import :WebGPUBuffer;
import :WebGPUComputeShader;
#endif
2026-03-09 20:10:19 +01:00
2026-05-02 21:08:20 +02:00
export namespace Crafter {
// ─── push-constant header ───────────────────────────────────────────
// Mirrors shaders/ui-shared.glsl::UIDispatchHeader byte-for-byte. User
// shaders MUST embed this as the first member of their push-constant
// struct so UIRenderer::FillHeader works.
struct UIDispatchHeader {
std::uint32_t outImage;
std::uint32_t itemBuffer;
std::uint32_t surfaceWidth;
std::uint32_t surfaceHeight;
float clipX, clipY, clipW, clipH;
std::uint32_t itemCount;
std::uint32_t frameIdx;
std::uint32_t flags;
std::uint32_t _pad;
};
static_assert(sizeof(UIDispatchHeader) == 48);
// Reserved `flags` bit (mirrors shaders/ui-shared.glsl::UI_FLAG_CLIP).
// FillHeader sets it when the clip rect is narrower than the surface so
// the standard shaders can skip the per-pixel clip compares otherwise.
// The remaining low bits stay free for user-defined feature flags.
inline constexpr std::uint32_t kUIFlagClip = 0x80000000u;
perf(ui): add additive DispatchFused to collapse consecutive UI passes (#47) Add a fifth UI compute path — DispatchFused — alongside the four per-element Dispatch* calls. It composites quads → circles → images → text (canonical back-to-front order) in ONE dispatch that loads the destination image once and stores once, eliminating the per-pass load+store and the inter-pass VkMemoryBarriers a run of consecutive Dispatch* calls would pay. The win materialises at >= 2 non-empty categories; an absent category is a zero-trip, push-constant-uniform loop (~free). The new uber-kernel (shaders/ui-fused.comp.glsl) reuses each category's exact cooperative shared-memory tile-cull + per-pixel accumulate, so a fused category is pixel-identical to its standalone pass. Categories run sequentially per thread and share one set of shared-memory scratch, so the VGPR/shared high-water mark stays at the per-category level, not the sum. Additive and non-breaking: the per-element Dispatch* calls and the frozen 48-byte UIDispatchHeader are untouched. DispatchFused gets its own 128-byte push-constant layout (UIFusedHeader: four uvec4 headers + four per-category clip vec4s). To interleave a custom ui.Dispatch() between standard passes, bracket it with two DispatchFused calls — the dispatch boundary is the explicit, app-declared flush point. WebGPU (Vulkan-second): DispatchFused falls back to the per-element Dispatch* calls in canonical order — one texture/sampler per dispatch there can't feed the bindless image phase — so the result matches; only the load/store fusion is Vulkan-only. HelloUI now composites its background quads, mouse circle, and label text in a single DispatchFused. Verified rendering identical on a live Vulkan/Wayland session. Resolves #47 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-16 17:03:24 +00:00
// ─── fused-dispatch push-constant header (issue #47) ────────────────
// Mirrors the PC block in shaders/ui-fused.comp.glsl byte-for-byte: every
// member is vec4-aligned, no padding holes, exactly 128 bytes (the
// guaranteed push-constant minimum). This is a SEPARATE contract — the
// per-element Dispatch* path and the frozen 48-byte UIDispatchHeader are
// untouched. DispatchFused builds this; users never fill it directly.
struct UIFusedHeader {
std::uint32_t itemBuffers[4]; // heap slots: quads, circles, images, text
std::uint32_t itemCounts[4]; // item counts: quads, circles, images, text
std::uint32_t outImage; // swapchain image heap slot
std::uint32_t fontTexture; // font-atlas image slot (text phase)
std::uint32_t fontSampler; // font-atlas sampler slot (text phase)
std::uint32_t flags; // per-category clip-active bits (0x1 quads .. 0x8 text)
perf(ui): add additive DispatchFused to collapse consecutive UI passes (#47) Add a fifth UI compute path — DispatchFused — alongside the four per-element Dispatch* calls. It composites quads → circles → images → text (canonical back-to-front order) in ONE dispatch that loads the destination image once and stores once, eliminating the per-pass load+store and the inter-pass VkMemoryBarriers a run of consecutive Dispatch* calls would pay. The win materialises at >= 2 non-empty categories; an absent category is a zero-trip, push-constant-uniform loop (~free). The new uber-kernel (shaders/ui-fused.comp.glsl) reuses each category's exact cooperative shared-memory tile-cull + per-pixel accumulate, so a fused category is pixel-identical to its standalone pass. Categories run sequentially per thread and share one set of shared-memory scratch, so the VGPR/shared high-water mark stays at the per-category level, not the sum. Additive and non-breaking: the per-element Dispatch* calls and the frozen 48-byte UIDispatchHeader are untouched. DispatchFused gets its own 128-byte push-constant layout (UIFusedHeader: four uvec4 headers + four per-category clip vec4s). To interleave a custom ui.Dispatch() between standard passes, bracket it with two DispatchFused calls — the dispatch boundary is the explicit, app-declared flush point. WebGPU (Vulkan-second): DispatchFused falls back to the per-element Dispatch* calls in canonical order — one texture/sampler per dispatch there can't feed the bindless image phase — so the result matches; only the load/store fusion is Vulkan-only. HelloUI now composites its background quads, mouse circle, and label text in a single DispatchFused. Verified rendering identical on a live Vulkan/Wayland session. Resolves #47 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-16 17:03:24 +00:00
std::uint32_t surfaceWidth;
std::uint32_t surfaceHeight;
std::uint32_t frameIdx;
std::uint32_t _pad; // reserved — keep zeroed
float clipQuads[4]; // (x, y, w, h) clip per category
float clipCircles[4];
float clipImages[4];
float clipText[4];
};
static_assert(sizeof(UIFusedHeader) == 128);
// One category's input to UIRenderer::DispatchFused. A zero itemCount
// makes that category a free no-op (its kernel loop is zero-trip).
// clipRectPx defaults to "no clip", matching the Dispatch* default.
struct FusedBatch {
std::uint32_t bufferSlot = 0;
std::uint32_t itemCount = 0;
std::array<float,4> clipRectPx = {0.0f, 0.0f, 1e9f, 1e9f};
};
2026-05-02 21:08:20 +02:00
// ─── standard item PODs (match GLSL std430) ─────────────────────────
struct QuadItem {
float x, y, w, h;
float r, g, b, a;
2026-05-18 04:58:52 +02:00
float cTL, cTR, cBR, cBL;
float outline, oR, oG, oB;
2026-05-02 21:08:20 +02:00
};
static_assert(sizeof(QuadItem) == 64);
struct CircleItem {
float cx, cy, radius, _p0;
float r, g, b, a;
float outline, oR, oG, oB;
};
static_assert(sizeof(CircleItem) == 48);
struct ImageItem {
float x, y, w, h;
float u0, v0, u1, v1;
float tR, tG, tB, tA;
std::uint32_t texSlot, sampSlot, _p1, _p2;
};
static_assert(sizeof(ImageItem) == 64);
struct GlyphItem {
float x, y, w, h;
float u0, v0, u1, v1;
float r, g, b, a;
};
static_assert(sizeof(GlyphItem) == 48);
// Where the X coordinate sits relative to the emitted glyph run. Used by
// UIRenderer::ShapeText and the Tier 3 text components (DrawText). Lives
// here (not in :UIComponents) so ShapeText can fold the alignment shift
// into its single pass.
enum class TextAlign : std::uint8_t { Left, Center, Right };
2026-05-02 21:08:20 +02:00
// ─── tiny rect-carving helper ───────────────────────────────────────
struct Rect {
float x = 0, y = 0, w = 0, h = 0;
enum class Anchor { Top, Bottom, Left, Right };
Rect SubRect(float size, Anchor a) const noexcept {
switch (a) {
case Anchor::Top: return { x, y, w, std::min(size, h) };
case Anchor::Bottom: return { x, y + h - std::min(size, h), w, std::min(size, h) };
case Anchor::Left: return { x, y, std::min(size, w), h };
case Anchor::Right: return { x + w - std::min(size, w), y, std::min(size, w), h };
}
return *this;
}
Rect Inset(float padding) const noexcept {
return Inset(padding, padding, padding, padding);
}
Rect Inset(float top, float right, float bottom, float left) const noexcept {
float nw = std::max(0.0f, w - left - right);
float nh = std::max(0.0f, h - top - bottom);
return { x + left, y + top, nw, nh };
}
bool Contains(float px, float py) const noexcept {
return px >= x && px < x + w && py >= y && py < y + h;
}
static Rect FromWindow(const Window& win) noexcept {
return { 0, 0, static_cast<float>(win.width), static_cast<float>(win.height) };
}
};
// ─── per-frame callback args ────────────────────────────────────────
struct UIBuildArgs {
2026-05-18 04:58:52 +02:00
GraphicsCommandBuffer cmd;
std::uint32_t frameIdx;
2026-05-02 21:08:20 +02:00
};
// ─── UIRenderer ─────────────────────────────────────────────────────
class UIRenderer : public RenderPass {
public:
2026-05-18 04:58:52 +02:00
GraphicsComputeShader drawQuads;
GraphicsComputeShader drawCircles;
GraphicsComputeShader drawImages;
GraphicsComputeShader drawText;
perf(ui): add additive DispatchFused to collapse consecutive UI passes (#47) Add a fifth UI compute path — DispatchFused — alongside the four per-element Dispatch* calls. It composites quads → circles → images → text (canonical back-to-front order) in ONE dispatch that loads the destination image once and stores once, eliminating the per-pass load+store and the inter-pass VkMemoryBarriers a run of consecutive Dispatch* calls would pay. The win materialises at >= 2 non-empty categories; an absent category is a zero-trip, push-constant-uniform loop (~free). The new uber-kernel (shaders/ui-fused.comp.glsl) reuses each category's exact cooperative shared-memory tile-cull + per-pixel accumulate, so a fused category is pixel-identical to its standalone pass. Categories run sequentially per thread and share one set of shared-memory scratch, so the VGPR/shared high-water mark stays at the per-category level, not the sum. Additive and non-breaking: the per-element Dispatch* calls and the frozen 48-byte UIDispatchHeader are untouched. DispatchFused gets its own 128-byte push-constant layout (UIFusedHeader: four uvec4 headers + four per-category clip vec4s). To interleave a custom ui.Dispatch() between standard passes, bracket it with two DispatchFused calls — the dispatch boundary is the explicit, app-declared flush point. WebGPU (Vulkan-second): DispatchFused falls back to the per-element Dispatch* calls in canonical order — one texture/sampler per dispatch there can't feed the bindless image phase — so the result matches; only the load/store fusion is Vulkan-only. HelloUI now composites its background quads, mouse circle, and label text in a single DispatchFused. Verified rendering identical on a live Vulkan/Wayland session. Resolves #47 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-16 17:03:24 +00:00
GraphicsComputeShader drawFused;
2026-05-02 21:08:20 +02:00
FontAtlas* fontAtlas = nullptr;
Crafter::Event<UIBuildArgs> onBuild;
UIRenderer() = default;
UIRenderer(const UIRenderer&) = delete;
UIRenderer& operator=(const UIRenderer&) = delete;
2026-05-18 04:58:52 +02:00
void Initialize(Window& window, GraphicsDescriptorHeap& heap, GraphicsCommandBuffer initCmd,
2026-05-02 21:08:20 +02:00
std::filesystem::path quadsSpv = "ui-quads.comp.spv",
std::filesystem::path circlesSpv = "ui-circles.comp.spv",
std::filesystem::path imagesSpv = "ui-images.comp.spv",
perf(ui): add additive DispatchFused to collapse consecutive UI passes (#47) Add a fifth UI compute path — DispatchFused — alongside the four per-element Dispatch* calls. It composites quads → circles → images → text (canonical back-to-front order) in ONE dispatch that loads the destination image once and stores once, eliminating the per-pass load+store and the inter-pass VkMemoryBarriers a run of consecutive Dispatch* calls would pay. The win materialises at >= 2 non-empty categories; an absent category is a zero-trip, push-constant-uniform loop (~free). The new uber-kernel (shaders/ui-fused.comp.glsl) reuses each category's exact cooperative shared-memory tile-cull + per-pixel accumulate, so a fused category is pixel-identical to its standalone pass. Categories run sequentially per thread and share one set of shared-memory scratch, so the VGPR/shared high-water mark stays at the per-category level, not the sum. Additive and non-breaking: the per-element Dispatch* calls and the frozen 48-byte UIDispatchHeader are untouched. DispatchFused gets its own 128-byte push-constant layout (UIFusedHeader: four uvec4 headers + four per-category clip vec4s). To interleave a custom ui.Dispatch() between standard passes, bracket it with two DispatchFused calls — the dispatch boundary is the explicit, app-declared flush point. WebGPU (Vulkan-second): DispatchFused falls back to the per-element Dispatch* calls in canonical order — one texture/sampler per dispatch there can't feed the bindless image phase — so the result matches; only the load/store fusion is Vulkan-only. HelloUI now composites its background quads, mouse circle, and label text in a single DispatchFused. Verified rendering identical on a live Vulkan/Wayland session. Resolves #47 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-16 17:03:24 +00:00
std::filesystem::path textSpv = "ui-text.comp.spv",
std::filesystem::path fusedSpv = "ui-fused.comp.spv");
2026-05-02 21:08:20 +02:00
2026-05-18 04:58:52 +02:00
void Record(GraphicsCommandBuffer cmd, std::uint32_t frameIdx, Window& window) override;
2026-05-02 21:08:20 +02:00
#ifndef CRAFTER_GRAPHICS_WINDOW_DOM
// A UI pass draws into the swapchain image entirely from compute
// shaders (see Dispatch), so the frame loop's inter-pass / frame-edge
// barriers only need to synchronise the COMPUTE_SHADER stage for it.
VkPipelineStageFlags SwapchainStage() const override {
return VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT;
}
#endif // !CRAFTER_GRAPHICS_WINDOW_DOM
2026-05-02 21:08:20 +02:00
UIDispatchHeader FillHeader(std::uint32_t itemBufferSlot,
std::uint32_t itemCount,
std::array<float,4> clipRectPx = {0.0f, 0.0f, 1e9f, 1e9f},
std::uint32_t flags = 0) const noexcept;
// Builds the header `flags` word: the caller's user flags with the
// reserved kUIFlagClip bit set iff `clipRectPx` does not already cover
// the whole surface. When the clip rect spans the surface, the four
// per-pixel clip compares in uiResolveScreenPixel can never reject an
// in-surface pixel, so the bit stays clear and the shaders skip them.
// Static + dimension-parameterised so the decision is unit-testable
// without a live Window; FillHeader feeds it the current window size.
static std::uint32_t ClipFlags(std::array<float,4> clipRectPx,
std::uint32_t surfaceWidth,
std::uint32_t surfaceHeight,
std::uint32_t flags) noexcept;
2026-05-18 04:58:52 +02:00
void DispatchQuads(GraphicsCommandBuffer cmd, std::uint32_t bufferSlot, std::uint32_t itemCount,
2026-05-02 21:08:20 +02:00
std::array<float,4> clipRectPx = {0.0f, 0.0f, 1e9f, 1e9f});
2026-05-18 04:58:52 +02:00
void DispatchCircles(GraphicsCommandBuffer cmd, std::uint32_t bufferSlot, std::uint32_t itemCount,
2026-05-02 21:08:20 +02:00
std::array<float,4> clipRectPx = {0.0f, 0.0f, 1e9f, 1e9f});
2026-05-18 04:58:52 +02:00
void DispatchImages(GraphicsCommandBuffer cmd, std::uint32_t bufferSlot, std::uint32_t itemCount,
2026-05-02 21:08:20 +02:00
std::array<float,4> clipRectPx = {0.0f, 0.0f, 1e9f, 1e9f});
2026-05-24 13:32:08 +02:00
#ifdef CRAFTER_GRAPHICS_WINDOW_DOM
// WebGPU-only overload. WebGPU bind groups can only carry one
// texture/sampler per dispatch, so all items in `bufferSlot`
// share the same texture (`imageSlot`) and sampler (`samplerSlot`).
// The per-item `slots` field in ImageItem is ignored on this
// backend. On Vulkan the bindless heap resolves per-item slots,
// so the cross-backend path is to call the 4-arg overload above
// on native and this 6-arg overload on DOM.
void DispatchImages(GraphicsCommandBuffer cmd, std::uint32_t bufferSlot, std::uint32_t itemCount,
std::uint16_t imageSlot, std::uint16_t samplerSlot,
std::array<float,4> clipRectPx = {0.0f, 0.0f, 1e9f, 1e9f});
#endif
2026-05-18 04:58:52 +02:00
void DispatchText(GraphicsCommandBuffer cmd, std::uint32_t bufferSlot, std::uint32_t itemCount,
2026-05-02 21:08:20 +02:00
std::array<float,4> clipRectPx = {0.0f, 0.0f, 1e9f, 1e9f});
perf(ui): add additive DispatchFused to collapse consecutive UI passes (#47) Add a fifth UI compute path — DispatchFused — alongside the four per-element Dispatch* calls. It composites quads → circles → images → text (canonical back-to-front order) in ONE dispatch that loads the destination image once and stores once, eliminating the per-pass load+store and the inter-pass VkMemoryBarriers a run of consecutive Dispatch* calls would pay. The win materialises at >= 2 non-empty categories; an absent category is a zero-trip, push-constant-uniform loop (~free). The new uber-kernel (shaders/ui-fused.comp.glsl) reuses each category's exact cooperative shared-memory tile-cull + per-pixel accumulate, so a fused category is pixel-identical to its standalone pass. Categories run sequentially per thread and share one set of shared-memory scratch, so the VGPR/shared high-water mark stays at the per-category level, not the sum. Additive and non-breaking: the per-element Dispatch* calls and the frozen 48-byte UIDispatchHeader are untouched. DispatchFused gets its own 128-byte push-constant layout (UIFusedHeader: four uvec4 headers + four per-category clip vec4s). To interleave a custom ui.Dispatch() between standard passes, bracket it with two DispatchFused calls — the dispatch boundary is the explicit, app-declared flush point. WebGPU (Vulkan-second): DispatchFused falls back to the per-element Dispatch* calls in canonical order — one texture/sampler per dispatch there can't feed the bindless image phase — so the result matches; only the load/store fusion is Vulkan-only. HelloUI now composites its background quads, mouse circle, and label text in a single DispatchFused. Verified rendering identical on a live Vulkan/Wayland session. Resolves #47 Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-16 17:03:24 +00:00
// Fused UI dispatch (issue #47). Composites up to four standard
// categories — quads → circles → images → text, canonical
// back-to-front order — in ONE compute dispatch that loads the
// destination image once and stores once, eliminating the per-pass
// load+store and the inter-pass memory barriers a run of consecutive
// Dispatch* calls would pay. The win materialises at >= 2 non-empty
// categories; an empty category (itemCount 0) is a free no-op.
//
// Additive: the per-element Dispatch* calls and UIDispatchHeader are
// untouched. Keep using those for single-category or fine-grained
// work. To interleave a custom ui.Dispatch() between standard passes,
// bracket it with two DispatchFused calls — the dispatch boundary is
// the explicit, app-declared flush point.
//
// The text category uses the registered font atlas (set fontAtlas
// before Initialize); passing a non-empty text batch without one
// throws, exactly like DispatchText. On the WebGPU backend this falls
// back to issuing the per-element Dispatch* calls in canonical order
// (one texture/sampler per dispatch there — see DispatchImages),
// so the result matches; the load/store fusion is Vulkan-only.
void DispatchFused(GraphicsCommandBuffer cmd,
const FusedBatch& quads,
const FusedBatch& circles,
const FusedBatch& images,
const FusedBatch& text);
2026-05-18 05:39:17 +02:00
// Generic dispatch for user-authored shaders. On Vulkan, `shader` is
// a SPIR-V compute pipeline (bindless via VK_EXT_descriptor_heap, so
// any resource indices baked into push data resolve through the
// global heap). On DOM, `shader` carries a UICustomBinding list
// declared at Load time; the renderer reads the listed slot uints
// out of `push`, resolves them against heap.bufferTable /
// imageTable / samplerTable, and builds the bind groups before
// dispatching.
void Dispatch(GraphicsCommandBuffer cmd, const GraphicsComputeShader& shader,
2026-05-02 21:08:20 +02:00
const void* push, std::uint32_t pushBytes,
std::uint32_t gx, std::uint32_t gy = 1, std::uint32_t gz = 1);
2026-05-18 04:58:52 +02:00
// Allocates a heap slot for the buffer and registers the GPU handle.
// Returns a move-only BufferSlot RAII handle.
2026-05-02 21:08:20 +02:00
template<typename T, bool Mapped>
2026-05-18 04:58:52 +02:00
BufferSlot RegisterBuffer(GraphicsBuffer<T, Mapped>& buffer);
2026-05-02 21:08:20 +02:00
2026-05-18 04:58:52 +02:00
#ifndef CRAFTER_GRAPHICS_WINDOW_DOM
2026-05-02 21:08:20 +02:00
template<typename Pixel>
2026-05-05 00:02:04 +02:00
ImageSlot RegisterImage(ImageVulkan<Pixel>& image, VkFormat format,
VkImageLayout layout = VK_IMAGE_LAYOUT_SHADER_READ_ONLY_OPTIMAL);
2026-05-02 21:08:20 +02:00
2026-05-05 00:02:04 +02:00
SamplerSlot RegisterSampler(const VkSamplerCreateInfo& info);
2026-05-18 04:58:52 +02:00
#endif
2026-05-05 00:02:04 +02:00
SamplerSlot RegisterLinearClampSampler();
2026-05-02 21:08:20 +02:00
// Shapes `utf8` into `out` (up to `outCapacity` glyphs), positioning
// the run at `(x, baselineY)` with horizontal `align`. Returns the
// number of glyphs written; `*outAdvance` (if non-null) receives the
// full advance width regardless of truncation.
//
// The origin-relative shaped run is memoized in a CPU cache keyed by
// (Font*, pxSize, color, utf8); a hit skips UTF-8 decode, glyph
// lookup, and layout, translating the cached run into `out` instead.
// Caching is byte-equivalent — the glyph buffer is fully rewritten
// and re-flushed every frame regardless. Call InvalidateFont before
// destroying a Font (the cache keys on the Font pointer).
2026-05-02 21:08:20 +02:00
std::uint32_t ShapeText(Font& font, float pxSize,
float x, float baselineY,
std::string_view utf8,
std::array<float,4> color,
GlyphItem* out, std::uint32_t outCapacity,
float* outAdvance = nullptr,
TextAlign align = TextAlign::Left);
// Drop every cached shaped run that references `font`. Must be called
// before a Font is destroyed: the run cache (and FontAtlas's glyph
// cache) key on the Font pointer, so a Font freed and a new one
// reallocated at the same address would otherwise alias stale runs.
void InvalidateFont(const Font& font) noexcept;
2026-05-02 21:08:20 +02:00
std::uint16_t FontAtlasImageSlot() const noexcept { return fontAtlasImageSlot_; }
std::uint16_t FontAtlasSamplerSlot() const noexcept { return fontAtlasSamplerSlot_; }
2026-05-05 23:49:29 +02:00
std::uint16_t OutImageSlot() const noexcept { return outImageSlot_; }
2026-05-02 21:08:20 +02:00
private:
Window* window_ = nullptr;
2026-05-18 04:58:52 +02:00
GraphicsDescriptorHeap* heap_ = nullptr;
2026-05-02 21:08:20 +02:00
// ─── shaped-run cache ───────────────────────────────────────────
// Glyphs are stored origin-relative (pen at x=0, baseline=0); a hit
// only needs a translate by (x + alignShift, baselineY). The full
// string is part of the key (compared on lookup) so hash collisions
// can't return the wrong run.
struct ShapedRunKey {
const Font* font;
float pxSize;
std::array<float, 4> color;
std::string text;
bool operator==(const ShapedRunKey&) const = default;
};
struct ShapedRunKeyHash {
std::size_t operator()(const ShapedRunKey& k) const noexcept {
std::size_t h = std::hash<const void*>{}(k.font);
auto mix = [&h](std::size_t v) noexcept {
h ^= v + 0x9e3779b97f4a7c15ULL + (h << 6) + (h >> 2);
};
mix(std::hash<std::uint32_t>{}(std::bit_cast<std::uint32_t>(k.pxSize)));
for (float c : k.color)
mix(std::hash<std::uint32_t>{}(std::bit_cast<std::uint32_t>(c)));
mix(std::hash<std::string_view>{}(k.text));
return h;
}
};
struct ShapedRun {
std::vector<GlyphItem> glyphs; // origin-relative
float advance = 0;
};
// Soft cap. A pathological caller drawing a fresh string every frame
// would grow this without bound; on overflow we drop everything and
// rebuild (correctness is unaffected, only the hit rate after a flush).
static constexpr std::size_t kMaxShapedRuns = 8192;
std::unordered_map<ShapedRunKey, ShapedRun, ShapedRunKeyHash> shapedRuns_;
2026-05-18 04:58:52 +02:00
ImageSlot outImageSlot_;
2026-05-05 00:02:04 +02:00
ImageSlot fontAtlasImageSlot_;
SamplerSlot fontAtlasSamplerSlot_;
2026-05-02 21:08:20 +02:00
2026-05-18 04:58:52 +02:00
#ifndef CRAFTER_GRAPHICS_WINDOW_DOM
VkImageViewCreateInfo atlasViewCreateInfo_{};
2026-05-02 21:08:20 +02:00
bool firstDispatchThisFrame_ = true;
2026-05-18 04:58:52 +02:00
#endif
2026-05-02 21:08:20 +02:00
2026-05-12 00:24:48 +02:00
Crafter::EventListener<void> resizeSub_;
2026-05-18 04:58:52 +02:00
#ifndef CRAFTER_GRAPHICS_WINDOW_DOM
2026-05-02 21:08:20 +02:00
void WriteSwapchainDescriptors();
void WriteFontAtlasDescriptor();
void WriteBufferDescriptor(std::uint16_t slot, VkDeviceAddress address, std::uint32_t size);
void WriteSampledImageDescriptor(std::uint16_t slot,
const VkImageViewCreateInfo& viewInfo,
VkImageLayout layout);
2026-05-18 04:58:52 +02:00
#endif
2026-05-02 21:08:20 +02:00
};
// ─── template-method implementations ────────────────────────────────
template<typename T, bool Mapped>
2026-05-18 04:58:52 +02:00
BufferSlot UIRenderer::RegisterBuffer(GraphicsBuffer<T, Mapped>& buffer) {
2026-05-02 21:08:20 +02:00
auto range = heap_->AllocateBufferSlots(1);
2026-05-18 04:58:52 +02:00
#ifndef CRAFTER_GRAPHICS_WINDOW_DOM
2026-05-02 21:08:20 +02:00
WriteBufferDescriptor(range.firstElement, buffer.address, buffer.size);
2026-05-18 04:58:52 +02:00
#else
heap_->bufferTable[range.firstElement] = buffer.handle;
#endif
2026-05-05 00:02:04 +02:00
return BufferSlot{heap_, range.firstElement};
2026-05-02 21:08:20 +02:00
}
2026-05-18 04:58:52 +02:00
#ifndef CRAFTER_GRAPHICS_WINDOW_DOM
2026-05-02 21:08:20 +02:00
template<typename Pixel>
2026-05-05 00:02:04 +02:00
ImageSlot UIRenderer::RegisterImage(ImageVulkan<Pixel>& image, VkFormat format,
VkImageLayout layout) {
2026-05-02 21:08:20 +02:00
auto range = heap_->AllocateImageSlots(1);
VkImageViewCreateInfo info {
.sType = VK_STRUCTURE_TYPE_IMAGE_VIEW_CREATE_INFO,
.image = image.image,
.viewType = VK_IMAGE_VIEW_TYPE_2D,
.format = format,
.components = {
VK_COMPONENT_SWIZZLE_IDENTITY, VK_COMPONENT_SWIZZLE_IDENTITY,
VK_COMPONENT_SWIZZLE_IDENTITY, VK_COMPONENT_SWIZZLE_IDENTITY,
},
.subresourceRange = {
.aspectMask = VK_IMAGE_ASPECT_COLOR_BIT,
.baseMipLevel = 0,
.levelCount = image.mipLevels,
.baseArrayLayer = 0,
.layerCount = 1,
},
};
WriteSampledImageDescriptor(range.firstElement, info, layout);
2026-05-05 00:02:04 +02:00
return ImageSlot{heap_, range.firstElement};
2026-05-02 21:08:20 +02:00
}
2026-05-18 04:58:52 +02:00
#endif
2026-05-02 21:08:20 +02:00
}