@openspecui/web 0.9.3 → 1.0.0

package/dist-ssg/server/assets/WebGPURenderer-DCgUFny7.js

@@ -0,0 +1,2146 @@
+import { r as State, ad as getTextureBatchBindGroup, y as ExtensionType, ae as GCManagedHash, W as uid, af as fastCopy, X as DOMAdapter, K as STENCIL_MODES, Q as BufferUsage, P as Buffer, B as BindGroup, ag as createIdFromString, D as warn, ah as CLEAR, a6 as CanvasSource, ai as TextureSource, U as UniformGroup, aj as CanvasPool, M as Matrix, ak as compileHighShaderGpuProgram, al as colorBit, am as generateTextureBatchBit, an as roundPixelsBit, S as Shader, z as Texture, a7 as AbstractRenderer, R as RendererType, H as extensions } from "../entry-server.js";
+import { c as createUboSyncFunction, u as uboSyncFunctionsWGSL, U as UboSystem, B as BufferResource, G as GpuStencilModesToPixi, e as ensureAttributes, l as localUniformBitGroup2, a as localUniformBit, t as textureBit } from "./BufferResource-BXrsGVSz.js";
+import { R as RenderTargetSystem, S as SharedSystems, d as SharedRenderPipes } from "./RenderTargetSystem-CVz6i60H.js";
+import "util";
+import "crypto";
+import "async_hooks";
+import "stream";
+import "process";
+import "buffer";
+import "node:process";
+import "node:path";
+import "node:url";
+import "./Filter-ClU0-pLL.js";
+const tempState = State.for2d();
+class GpuBatchAdaptor {
+  start(batchPipe, geometry, shader) {
+    const renderer = batchPipe.renderer;
+    const encoder = renderer.encoder;
+    const program = shader.gpuProgram;
+    this._shader = shader;
+    this._geometry = geometry;
+    encoder.setGeometry(geometry, program);
+    tempState.blendMode = "normal";
+    renderer.pipeline.getPipeline(
+      geometry,
+      program,
+      tempState
+    );
+    const globalUniformsBindGroup = renderer.globalUniforms.bindGroup;
+    encoder.resetBindGroup(1);
+    encoder.setBindGroup(0, globalUniformsBindGroup, program);
+  }
+  execute(batchPipe, batch) {
+    const program = this._shader.gpuProgram;
+    const renderer = batchPipe.renderer;
+    const encoder = renderer.encoder;
+    if (!batch.bindGroup) {
+      const textureBatch = batch.textures;
+      batch.bindGroup = getTextureBatchBindGroup(
+        textureBatch.textures,
+        textureBatch.count,
+        renderer.limits.maxBatchableTextures
+      );
+    }
+    tempState.blendMode = batch.blendMode;
+    const gpuBindGroup = renderer.bindGroup.getBindGroup(
+      batch.bindGroup,
+      program,
+      1
+    );
+    const pipeline = renderer.pipeline.getPipeline(
+      this._geometry,
+      program,
+      tempState,
+      batch.topology
+    );
+    batch.bindGroup._touch(renderer.gc.now, renderer.tick);
+    encoder.setPipeline(pipeline);
+    encoder.renderPassEncoder.setBindGroup(1, gpuBindGroup);
+    encoder.renderPassEncoder.drawIndexed(batch.size, 1, batch.start);
+  }
+}
+GpuBatchAdaptor.extension = {
+  type: [
+    ExtensionType.WebGPUPipesAdaptor
+  ],
+  name: "batch"
+};
+class BindGroupSystem {
+  constructor(renderer) {
+    this._hash = /* @__PURE__ */ Object.create(null);
+    this._renderer = renderer;
+  }
+  contextChange(gpu) {
+    this._gpu = gpu;
+  }
+  getBindGroup(bindGroup, program, groupIndex) {
+    bindGroup._updateKey();
+    const gpuBindGroup = this._hash[bindGroup._key] || this._createBindGroup(bindGroup, program, groupIndex);
+    return gpuBindGroup;
+  }
+  _createBindGroup(group, program, groupIndex) {
+    const device = this._gpu.device;
+    const groupLayout = program.layout[groupIndex];
+    const entries = [];
+    const renderer = this._renderer;
+    for (const j in groupLayout) {
+      const resource = group.resources[j] ?? group.resources[groupLayout[j]];
+      let gpuResource;
+      if (resource._resourceType === "uniformGroup") {
+        const uniformGroup = resource;
+        renderer.ubo.updateUniformGroup(uniformGroup);
+        const buffer = uniformGroup.buffer;
+        gpuResource = {
+          buffer: renderer.buffer.getGPUBuffer(buffer),
+          offset: 0,
+          size: buffer.descriptor.size
+        };
+      } else if (resource._resourceType === "buffer") {
+        const buffer = resource;
+        gpuResource = {
+          buffer: renderer.buffer.getGPUBuffer(buffer),
+          offset: 0,
+          size: buffer.descriptor.size
+        };
+      } else if (resource._resourceType === "bufferResource") {
+        const bufferResource = resource;
+        gpuResource = {
+          buffer: renderer.buffer.getGPUBuffer(bufferResource.buffer),
+          offset: bufferResource.offset,
+          size: bufferResource.size
+        };
+      } else if (resource._resourceType === "textureSampler") {
+        const sampler = resource;
+        gpuResource = renderer.texture.getGpuSampler(sampler);
+      } else if (resource._resourceType === "textureSource") {
+        const texture = resource;
+        gpuResource = renderer.texture.getTextureView(texture);
+      }
+      entries.push({
+        binding: groupLayout[j],
+        resource: gpuResource
+      });
+    }
+    const layout = renderer.shader.getProgramData(program).bindGroups[groupIndex];
+    const gpuBindGroup = device.createBindGroup({
+      layout,
+      entries
+    });
+    this._hash[group._key] = gpuBindGroup;
+    return gpuBindGroup;
+  }
+  destroy() {
+    this._hash = null;
+    this._renderer = null;
+  }
+}
+BindGroupSystem.extension = {
+  type: [
+    ExtensionType.WebGPUSystem
+  ],
+  name: "bindGroup"
+};
+class GpuBufferData {
+  constructor(gpuBuffer) {
+    this.gpuBuffer = gpuBuffer;
+  }
+  destroy() {
+    this.gpuBuffer.destroy();
+    this.gpuBuffer = null;
+  }
+}
+class GpuBufferSystem {
+  constructor(renderer) {
+    this._renderer = renderer;
+    this._managedBuffers = new GCManagedHash({
+      renderer,
+      type: "resource",
+      onUnload: this.onBufferUnload.bind(this),
+      name: "gpuBuffer"
+    });
+  }
+  contextChange(gpu) {
+    this._gpu = gpu;
+  }
+  getGPUBuffer(buffer) {
+    buffer._gcLastUsed = this._renderer.gc.now;
+    return buffer._gpuData[this._renderer.uid]?.gpuBuffer || this.createGPUBuffer(buffer);
+  }
+  updateBuffer(buffer) {
+    const gpuBuffer = this.getGPUBuffer(buffer);
+    const data = buffer.data;
+    if (buffer._updateID && data) {
+      buffer._updateID = 0;
+      this._gpu.device.queue.writeBuffer(
+        gpuBuffer,
+        0,
+        data.buffer,
+        0,
+        // round to the nearest 4 bytes
+        (buffer._updateSize || data.byteLength) + 3 & -4
+      );
+    }
+    return gpuBuffer;
+  }
+  /** dispose all WebGL resources of all managed buffers */
+  destroyAll() {
+    this._managedBuffers.removeAll();
+  }
+  onBufferUnload(buffer) {
+    buffer.off("update", this.updateBuffer, this);
+    buffer.off("change", this.onBufferChange, this);
+  }
+  createGPUBuffer(buffer) {
+    const gpuBuffer = this._gpu.device.createBuffer(buffer.descriptor);
+    buffer._updateID = 0;
+    buffer._resourceId = uid("resource");
+    if (buffer.data) {
+      fastCopy(
+        buffer.data.buffer,
+        gpuBuffer.getMappedRange(),
+        buffer.data.byteOffset,
+        buffer.data.byteLength
+      );
+      gpuBuffer.unmap();
+    }
+    buffer._gpuData[this._renderer.uid] = new GpuBufferData(gpuBuffer);
+    if (this._managedBuffers.add(buffer)) {
+      buffer.on("update", this.updateBuffer, this);
+      buffer.on("change", this.onBufferChange, this);
+    }
+    return gpuBuffer;
+  }
+  onBufferChange(buffer) {
+    this._managedBuffers.remove(buffer);
+    buffer._updateID = 0;
+    this.createGPUBuffer(buffer);
+  }
+  destroy() {
+    this._managedBuffers.destroy();
+    this._renderer = null;
+    this._gpu = null;
+  }
+}
+GpuBufferSystem.extension = {
+  type: [
+    ExtensionType.WebGPUSystem
+  ],
+  name: "buffer"
+};
+class UboBatch {
+  constructor({ minUniformOffsetAlignment: minUniformOffsetAlignment2 }) {
+    this._minUniformOffsetAlignment = 256;
+    this.byteIndex = 0;
+    this._minUniformOffsetAlignment = minUniformOffsetAlignment2;
+    this.data = new Float32Array(65535);
+  }
+  clear() {
+    this.byteIndex = 0;
+  }
+  addEmptyGroup(size) {
+    if (size > this._minUniformOffsetAlignment / 4) {
+      throw new Error(`UniformBufferBatch: array is too large: ${size * 4}`);
+    }
+    const start = this.byteIndex;
+    let newSize = start + size * 4;
+    newSize = Math.ceil(newSize / this._minUniformOffsetAlignment) * this._minUniformOffsetAlignment;
+    if (newSize > this.data.length * 4) {
+      throw new Error("UniformBufferBatch: ubo batch got too big");
+    }
+    this.byteIndex = newSize;
+    return start;
+  }
+  addGroup(array) {
+    const offset = this.addEmptyGroup(array.length);
+    for (let i = 0; i < array.length; i++) {
+      this.data[offset / 4 + i] = array[i];
+    }
+    return offset;
+  }
+  destroy() {
+    this.data = null;
+  }
+}
+class GpuColorMaskSystem {
+  constructor(renderer) {
+    this._colorMaskCache = 15;
+    this._renderer = renderer;
+  }
+  setMask(colorMask) {
+    if (this._colorMaskCache === colorMask) return;
+    this._colorMaskCache = colorMask;
+    this._renderer.pipeline.setColorMask(colorMask);
+  }
+  destroy() {
+    this._renderer = null;
+    this._colorMaskCache = null;
+  }
+}
+GpuColorMaskSystem.extension = {
+  type: [
+    ExtensionType.WebGPUSystem
+  ],
+  name: "colorMask"
+};
+class GpuDeviceSystem {
+  /**
+   * @param {WebGPURenderer} renderer - The renderer this System works for.
+   */
+  constructor(renderer) {
+    this._renderer = renderer;
+  }
+  async init(options) {
+    if (this._initPromise) return this._initPromise;
+    this._initPromise = (options.gpu ? Promise.resolve(options.gpu) : this._createDeviceAndAdaptor(options)).then((gpu) => {
+      this.gpu = gpu;
+      this._renderer.runners.contextChange.emit(this.gpu);
+    });
+    return this._initPromise;
+  }
+  /**
+   * Handle the context change event
+   * @param gpu
+   */
+  contextChange(gpu) {
+    this._renderer.gpu = gpu;
+  }
+  /**
+   * Helper class to create a WebGL Context
+   * @param {object} options - An options object that gets passed in to the canvas element containing the
+   *    context attributes
+   * @see https://developer.mozilla.org/en/docs/Web/API/HTMLCanvasElement/getContext
+   * @returns {WebGLRenderingContext} the WebGL context
+   */
+  async _createDeviceAndAdaptor(options) {
+    const adapter = await DOMAdapter.get().getNavigator().gpu.requestAdapter({
+      powerPreference: options.powerPreference,
+      forceFallbackAdapter: options.forceFallbackAdapter
+    });
+    const requiredFeatures = [
+      "texture-compression-bc",
+      "texture-compression-astc",
+      "texture-compression-etc2"
+    ].filter((feature) => adapter.features.has(feature));
+    const device = await adapter.requestDevice({
+      requiredFeatures
+    });
+    return { adapter, device };
+  }
+  destroy() {
+    this.gpu = null;
+    this._renderer = null;
+  }
+}
+GpuDeviceSystem.extension = {
+  type: [
+    ExtensionType.WebGPUSystem
+  ],
+  name: "device"
+};
+GpuDeviceSystem.defaultOptions = {
+  /**
+   * {@link WebGPUOptions.powerPreference}
+   * @default default
+   */
+  powerPreference: void 0,
+  /**
+   * Force the use of the fallback adapter
+   * @default false
+   */
+  forceFallbackAdapter: false
+};
+class GpuEncoderSystem {
+  constructor(renderer) {
+    this._boundBindGroup = /* @__PURE__ */ Object.create(null);
+    this._boundVertexBuffer = /* @__PURE__ */ Object.create(null);
+    this._renderer = renderer;
+  }
+  renderStart() {
+    this.commandFinished = new Promise((resolve) => {
+      this._resolveCommandFinished = resolve;
+    });
+    this.commandEncoder = this._renderer.gpu.device.createCommandEncoder();
+  }
+  beginRenderPass(gpuRenderTarget) {
+    this.endRenderPass();
+    this._clearCache();
+    this.renderPassEncoder = this.commandEncoder.beginRenderPass(gpuRenderTarget.descriptor);
+  }
+  endRenderPass() {
+    if (this.renderPassEncoder) {
+      this.renderPassEncoder.end();
+    }
+    this.renderPassEncoder = null;
+  }
+  setViewport(viewport) {
+    this.renderPassEncoder.setViewport(viewport.x, viewport.y, viewport.width, viewport.height, 0, 1);
+  }
+  setPipelineFromGeometryProgramAndState(geometry, program, state, topology) {
+    const pipeline = this._renderer.pipeline.getPipeline(geometry, program, state, topology);
+    this.setPipeline(pipeline);
+  }
+  setPipeline(pipeline) {
+    if (this._boundPipeline === pipeline) return;
+    this._boundPipeline = pipeline;
+    this.renderPassEncoder.setPipeline(pipeline);
+  }
+  _setVertexBuffer(index, buffer) {
+    if (this._boundVertexBuffer[index] === buffer) return;
+    this._boundVertexBuffer[index] = buffer;
+    this.renderPassEncoder.setVertexBuffer(index, this._renderer.buffer.updateBuffer(buffer));
+  }
+  _setIndexBuffer(buffer) {
+    if (this._boundIndexBuffer === buffer) return;
+    this._boundIndexBuffer = buffer;
+    const indexFormat = buffer.data.BYTES_PER_ELEMENT === 2 ? "uint16" : "uint32";
+    this.renderPassEncoder.setIndexBuffer(this._renderer.buffer.updateBuffer(buffer), indexFormat);
+  }
+  resetBindGroup(index) {
+    this._boundBindGroup[index] = null;
+  }
+  setBindGroup(index, bindGroup, program) {
+    if (this._boundBindGroup[index] === bindGroup) return;
+    this._boundBindGroup[index] = bindGroup;
+    bindGroup._touch(this._renderer.gc.now, this._renderer.tick);
+    const gpuBindGroup = this._renderer.bindGroup.getBindGroup(bindGroup, program, index);
+    this.renderPassEncoder.setBindGroup(index, gpuBindGroup);
+  }
+  setGeometry(geometry, program) {
+    const buffersToBind = this._renderer.pipeline.getBufferNamesToBind(geometry, program);
+    for (const i in buffersToBind) {
+      this._setVertexBuffer(parseInt(i, 10), geometry.attributes[buffersToBind[i]].buffer);
+    }
+    if (geometry.indexBuffer) {
+      this._setIndexBuffer(geometry.indexBuffer);
+    }
+  }
+  _setShaderBindGroups(shader, skipSync) {
+    for (const i in shader.groups) {
+      const bindGroup = shader.groups[i];
+      if (!skipSync) {
+        this._syncBindGroup(bindGroup);
+      }
+      this.setBindGroup(i, bindGroup, shader.gpuProgram);
+    }
+  }
+  _syncBindGroup(bindGroup) {
+    for (const j in bindGroup.resources) {
+      const resource = bindGroup.resources[j];
+      if (resource.isUniformGroup) {
+        this._renderer.ubo.updateUniformGroup(resource);
+      }
+    }
+  }
+  draw(options) {
+    const { geometry, shader, state, topology, size, start, instanceCount, skipSync } = options;
+    this.setPipelineFromGeometryProgramAndState(geometry, shader.gpuProgram, state, topology);
+    this.setGeometry(geometry, shader.gpuProgram);
+    this._setShaderBindGroups(shader, skipSync);
+    if (geometry.indexBuffer) {
+      this.renderPassEncoder.drawIndexed(
+        size || geometry.indexBuffer.data.length,
+        instanceCount ?? geometry.instanceCount,
+        start || 0
+      );
+    } else {
+      this.renderPassEncoder.draw(size || geometry.getSize(), instanceCount ?? geometry.instanceCount, start || 0);
+    }
+  }
+  finishRenderPass() {
+    if (this.renderPassEncoder) {
+      this.renderPassEncoder.end();
+      this.renderPassEncoder = null;
+    }
+  }
+  postrender() {
+    this.finishRenderPass();
+    this._gpu.device.queue.submit([this.commandEncoder.finish()]);
+    this._resolveCommandFinished();
+    this.commandEncoder = null;
+  }
+  // restores a render pass if finishRenderPass was called
+  // not optimised as really used for debugging!
+  // used when we want to stop drawing and log a texture..
+  restoreRenderPass() {
+    const descriptor = this._renderer.renderTarget.adaptor.getDescriptor(
+      this._renderer.renderTarget.renderTarget,
+      false,
+      [0, 0, 0, 1],
+      this._renderer.renderTarget.mipLevel,
+      this._renderer.renderTarget.layer
+    );
+    this.renderPassEncoder = this.commandEncoder.beginRenderPass(descriptor);
+    const boundPipeline = this._boundPipeline;
+    const boundVertexBuffer = { ...this._boundVertexBuffer };
+    const boundIndexBuffer = this._boundIndexBuffer;
+    const boundBindGroup = { ...this._boundBindGroup };
+    this._clearCache();
+    const viewport = this._renderer.renderTarget.viewport;
+    this.renderPassEncoder.setViewport(viewport.x, viewport.y, viewport.width, viewport.height, 0, 1);
+    this.setPipeline(boundPipeline);
+    for (const i in boundVertexBuffer) {
+      this._setVertexBuffer(i, boundVertexBuffer[i]);
+    }
+    for (const i in boundBindGroup) {
+      this.setBindGroup(i, boundBindGroup[i], null);
+    }
+    this._setIndexBuffer(boundIndexBuffer);
+  }
+  _clearCache() {
+    for (let i = 0; i < 16; i++) {
+      this._boundBindGroup[i] = null;
+      this._boundVertexBuffer[i] = null;
+    }
+    this._boundIndexBuffer = null;
+    this._boundPipeline = null;
+  }
+  destroy() {
+    this._renderer = null;
+    this._gpu = null;
+    this._boundBindGroup = null;
+    this._boundVertexBuffer = null;
+    this._boundIndexBuffer = null;
+    this._boundPipeline = null;
+  }
+  contextChange(gpu) {
+    this._gpu = gpu;
+  }
+}
+GpuEncoderSystem.extension = {
+  type: [ExtensionType.WebGPUSystem],
+  name: "encoder",
+  priority: 1
+};
+class GpuLimitsSystem {
+  constructor(renderer) {
+    this._renderer = renderer;
+  }
+  contextChange() {
+    this.maxTextures = this._renderer.device.gpu.device.limits.maxSampledTexturesPerShaderStage;
+    this.maxBatchableTextures = this.maxTextures;
+  }
+  destroy() {
+  }
+}
+GpuLimitsSystem.extension = {
+  type: [
+    ExtensionType.WebGPUSystem
+  ],
+  name: "limits"
+};
+class GpuStencilSystem {
+  constructor(renderer) {
+    this._renderTargetStencilState = /* @__PURE__ */ Object.create(null);
+    this._renderer = renderer;
+    renderer.renderTarget.onRenderTargetChange.add(this);
+  }
+  onRenderTargetChange(renderTarget) {
+    let stencilState = this._renderTargetStencilState[renderTarget.uid];
+    if (!stencilState) {
+      stencilState = this._renderTargetStencilState[renderTarget.uid] = {
+        stencilMode: STENCIL_MODES.DISABLED,
+        stencilReference: 0
+      };
+    }
+    this._activeRenderTarget = renderTarget;
+    this.setStencilMode(stencilState.stencilMode, stencilState.stencilReference);
+  }
+  setStencilMode(stencilMode, stencilReference) {
+    const stencilState = this._renderTargetStencilState[this._activeRenderTarget.uid];
+    stencilState.stencilMode = stencilMode;
+    stencilState.stencilReference = stencilReference;
+    const renderer = this._renderer;
+    renderer.pipeline.setStencilMode(stencilMode);
+    renderer.encoder.renderPassEncoder.setStencilReference(stencilReference);
+  }
+  destroy() {
+    this._renderer.renderTarget.onRenderTargetChange.remove(this);
+    this._renderer = null;
+    this._activeRenderTarget = null;
+    this._renderTargetStencilState = null;
+  }
+}
+GpuStencilSystem.extension = {
+  type: [
+    ExtensionType.WebGPUSystem
+  ],
+  name: "stencil"
+};
+const WGSL_ALIGN_SIZE_DATA = {
+  i32: { align: 4, size: 4 },
+  u32: { align: 4, size: 4 },
+  f32: { align: 4, size: 4 },
+  f16: { align: 2, size: 2 },
+  "vec2<i32>": { align: 8, size: 8 },
+  "vec2<u32>": { align: 8, size: 8 },
+  "vec2<f32>": { align: 8, size: 8 },
+  "vec2<f16>": { align: 4, size: 4 },
+  "vec3<i32>": { align: 16, size: 12 },
+  "vec3<u32>": { align: 16, size: 12 },
+  "vec3<f32>": { align: 16, size: 12 },
+  "vec3<f16>": { align: 8, size: 6 },
+  "vec4<i32>": { align: 16, size: 16 },
+  "vec4<u32>": { align: 16, size: 16 },
+  "vec4<f32>": { align: 16, size: 16 },
+  "vec4<f16>": { align: 8, size: 8 },
+  "mat2x2<f32>": { align: 8, size: 16 },
+  "mat2x2<f16>": { align: 4, size: 8 },
+  "mat3x2<f32>": { align: 8, size: 24 },
+  "mat3x2<f16>": { align: 4, size: 12 },
+  "mat4x2<f32>": { align: 8, size: 32 },
+  "mat4x2<f16>": { align: 4, size: 16 },
+  "mat2x3<f32>": { align: 16, size: 32 },
+  "mat2x3<f16>": { align: 8, size: 16 },
+  "mat3x3<f32>": { align: 16, size: 48 },
+  "mat3x3<f16>": { align: 8, size: 24 },
+  "mat4x3<f32>": { align: 16, size: 64 },
+  "mat4x3<f16>": { align: 8, size: 32 },
+  "mat2x4<f32>": { align: 16, size: 32 },
+  "mat2x4<f16>": { align: 8, size: 16 },
+  "mat3x4<f32>": { align: 16, size: 48 },
+  "mat3x4<f16>": { align: 8, size: 24 },
+  "mat4x4<f32>": { align: 16, size: 64 },
+  "mat4x4<f16>": { align: 8, size: 32 }
+};
+function createUboElementsWGSL(uniformData) {
+  const uboElements = uniformData.map((data) => ({
+    data,
+    offset: 0,
+    size: 0
+  }));
+  let offset = 0;
+  for (let i = 0; i < uboElements.length; i++) {
+    const uboElement = uboElements[i];
+    let size = WGSL_ALIGN_SIZE_DATA[uboElement.data.type].size;
+    const align = WGSL_ALIGN_SIZE_DATA[uboElement.data.type].align;
+    if (!WGSL_ALIGN_SIZE_DATA[uboElement.data.type]) {
+      throw new Error(`[Pixi.js] WebGPU UniformBuffer: Unknown type ${uboElement.data.type}`);
+    }
+    if (uboElement.data.size > 1) {
+      size = Math.max(size, align) * uboElement.data.size;
+    }
+    offset = Math.ceil(offset / align) * align;
+    uboElement.size = size;
+    uboElement.offset = offset;
+    offset += size;
+  }
+  offset = Math.ceil(offset / 16) * 16;
+  return { uboElements, size: offset };
+}
+function generateArraySyncWGSL(uboElement, offsetToAdd) {
+  const { size, align } = WGSL_ALIGN_SIZE_DATA[uboElement.data.type];
+  const remainder = (align - size) / 4;
+  const data = uboElement.data.type.indexOf("i32") >= 0 ? "dataInt32" : "data";
+  return `
+         v = uv.${uboElement.data.name};
+         ${offsetToAdd !== 0 ? `offset += ${offsetToAdd};` : ""}
+
+         arrayOffset = offset;
+
+         t = 0;
+
+         for(var i=0; i < ${uboElement.data.size * (size / 4)}; i++)
+         {
+             for(var j = 0; j < ${size / 4}; j++)
+             {
+                 ${data}[arrayOffset++] = v[t++];
+             }
+             ${remainder !== 0 ? `arrayOffset += ${remainder};` : ""}
+         }
+     `;
+}
+function createUboSyncFunctionWGSL(uboElements) {
+  return createUboSyncFunction(
+    uboElements,
+    "uboWgsl",
+    generateArraySyncWGSL,
+    uboSyncFunctionsWGSL
+  );
+}
+class GpuUboSystem extends UboSystem {
+  constructor() {
+    super({
+      createUboElements: createUboElementsWGSL,
+      generateUboSync: createUboSyncFunctionWGSL
+    });
+  }
+}
+GpuUboSystem.extension = {
+  type: [ExtensionType.WebGPUSystem],
+  name: "ubo"
+};
+const minUniformOffsetAlignment = 128;
+class GpuUniformBatchPipe {
+  constructor(renderer) {
+    this._bindGroupHash = /* @__PURE__ */ Object.create(null);
+    this._buffers = [];
+    this._bindGroups = [];
+    this._bufferResources = [];
+    this._renderer = renderer;
+    this._batchBuffer = new UboBatch({ minUniformOffsetAlignment });
+    const totalBuffers = 256 / minUniformOffsetAlignment;
+    for (let i = 0; i < totalBuffers; i++) {
+      let usage = BufferUsage.UNIFORM | BufferUsage.COPY_DST;
+      if (i === 0) usage |= BufferUsage.COPY_SRC;
+      this._buffers.push(new Buffer({
+        data: this._batchBuffer.data,
+        usage
+      }));
+    }
+  }
+  renderEnd() {
+    this._uploadBindGroups();
+    this._resetBindGroups();
+  }
+  _resetBindGroups() {
+    this._bindGroupHash = /* @__PURE__ */ Object.create(null);
+    this._batchBuffer.clear();
+  }
+  // just works for single bind groups for now
+  getUniformBindGroup(group, duplicate) {
+    if (!duplicate && this._bindGroupHash[group.uid]) {
+      return this._bindGroupHash[group.uid];
+    }
+    this._renderer.ubo.ensureUniformGroup(group);
+    const data = group.buffer.data;
+    const offset = this._batchBuffer.addEmptyGroup(data.length);
+    this._renderer.ubo.syncUniformGroup(group, this._batchBuffer.data, offset / 4);
+    this._bindGroupHash[group.uid] = this._getBindGroup(offset / minUniformOffsetAlignment);
+    return this._bindGroupHash[group.uid];
+  }
+  getUboResource(group) {
+    this._renderer.ubo.updateUniformGroup(group);
+    const data = group.buffer.data;
+    const offset = this._batchBuffer.addGroup(data);
+    return this._getBufferResource(offset / minUniformOffsetAlignment);
+  }
+  getArrayBindGroup(data) {
+    const offset = this._batchBuffer.addGroup(data);
+    return this._getBindGroup(offset / minUniformOffsetAlignment);
+  }
+  getArrayBufferResource(data) {
+    const offset = this._batchBuffer.addGroup(data);
+    const index = offset / minUniformOffsetAlignment;
+    return this._getBufferResource(index);
+  }
+  _getBufferResource(index) {
+    if (!this._bufferResources[index]) {
+      const buffer = this._buffers[index % 2];
+      this._bufferResources[index] = new BufferResource({
+        buffer,
+        offset: (index / 2 | 0) * 256,
+        size: minUniformOffsetAlignment
+      });
+    }
+    return this._bufferResources[index];
+  }
+  _getBindGroup(index) {
+    if (!this._bindGroups[index]) {
+      const bindGroup = new BindGroup({
+        0: this._getBufferResource(index)
+      });
+      this._bindGroups[index] = bindGroup;
+    }
+    return this._bindGroups[index];
+  }
+  _uploadBindGroups() {
+    const bufferSystem = this._renderer.buffer;
+    const firstBuffer = this._buffers[0];
+    firstBuffer.update(this._batchBuffer.byteIndex);
+    bufferSystem.updateBuffer(firstBuffer);
+    const commandEncoder = this._renderer.gpu.device.createCommandEncoder();
+    for (let i = 1; i < this._buffers.length; i++) {
+      const buffer = this._buffers[i];
+      commandEncoder.copyBufferToBuffer(
+        bufferSystem.getGPUBuffer(firstBuffer),
+        minUniformOffsetAlignment,
+        bufferSystem.getGPUBuffer(buffer),
+        0,
+        this._batchBuffer.byteIndex
+      );
+    }
+    this._renderer.gpu.device.queue.submit([commandEncoder.finish()]);
+  }
+  destroy() {
+    for (let i = 0; i < this._bindGroups.length; i++) {
+      this._bindGroups[i]?.destroy();
+    }
+    this._bindGroups = null;
+    this._bindGroupHash = null;
+    for (let i = 0; i < this._buffers.length; i++) {
+      this._buffers[i].destroy();
+    }
+    this._buffers = null;
+    for (let i = 0; i < this._bufferResources.length; i++) {
+      this._bufferResources[i].destroy();
+    }
+    this._bufferResources = null;
+    this._batchBuffer.destroy();
+    this._renderer = null;
+  }
+}
+GpuUniformBatchPipe.extension = {
+  type: [
+    ExtensionType.WebGPUPipes
+  ],
+  name: "uniformBatch"
+};
+const topologyStringToId = {
+  "point-list": 0,
+  "line-list": 1,
+  "line-strip": 2,
+  "triangle-list": 3,
+  "triangle-strip": 4
+};
+function getGraphicsStateKey(geometryLayout, shaderKey, state, blendMode, topology) {
+  return geometryLayout << 24 | shaderKey << 16 | state << 10 | blendMode << 5 | topology;
+}
+function getGlobalStateKey(stencilStateId, multiSampleCount, colorMask, renderTarget, colorTargetCount) {
+  return colorMask << 8 | stencilStateId << 5 | renderTarget << 3 | colorTargetCount << 1 | multiSampleCount;
+}
+class PipelineSystem {
+  constructor(renderer) {
+    this._moduleCache = /* @__PURE__ */ Object.create(null);
+    this._bufferLayoutsCache = /* @__PURE__ */ Object.create(null);
+    this._bindingNamesCache = /* @__PURE__ */ Object.create(null);
+    this._pipeCache = /* @__PURE__ */ Object.create(null);
+    this._pipeStateCaches = /* @__PURE__ */ Object.create(null);
+    this._colorMask = 15;
+    this._multisampleCount = 1;
+    this._colorTargetCount = 1;
+    this._renderer = renderer;
+  }
+  contextChange(gpu) {
+    this._gpu = gpu;
+    this.setStencilMode(STENCIL_MODES.DISABLED);
+    this._updatePipeHash();
+  }
+  setMultisampleCount(multisampleCount) {
+    if (this._multisampleCount === multisampleCount) return;
+    this._multisampleCount = multisampleCount;
+    this._updatePipeHash();
+  }
+  setRenderTarget(renderTarget) {
+    this._multisampleCount = renderTarget.msaaSamples;
+    this._depthStencilAttachment = renderTarget.descriptor.depthStencilAttachment ? 1 : 0;
+    this._colorTargetCount = renderTarget.colorTargetCount;
+    this._updatePipeHash();
+  }
+  setColorMask(colorMask) {
+    if (this._colorMask === colorMask) return;
+    this._colorMask = colorMask;
+    this._updatePipeHash();
+  }
+  setStencilMode(stencilMode) {
+    if (this._stencilMode === stencilMode) return;
+    this._stencilMode = stencilMode;
+    this._stencilState = GpuStencilModesToPixi[stencilMode];
+    this._updatePipeHash();
+  }
+  setPipeline(geometry, program, state, passEncoder) {
+    const pipeline = this.getPipeline(geometry, program, state);
+    passEncoder.setPipeline(pipeline);
+  }
+  getPipeline(geometry, program, state, topology) {
+    if (!geometry._layoutKey) {
+      ensureAttributes(geometry, program.attributeData);
+      this._generateBufferKey(geometry);
+    }
+    topology || (topology = geometry.topology);
+    const key = getGraphicsStateKey(
+      geometry._layoutKey,
+      program._layoutKey,
+      state.data,
+      state._blendModeId,
+      topologyStringToId[topology]
+    );
+    if (this._pipeCache[key]) return this._pipeCache[key];
+    this._pipeCache[key] = this._createPipeline(geometry, program, state, topology);
+    return this._pipeCache[key];
+  }
+  _createPipeline(geometry, program, state, topology) {
+    const device = this._gpu.device;
+    const buffers = this._createVertexBufferLayouts(geometry, program);
+    const blendModes = this._renderer.state.getColorTargets(state, this._colorTargetCount);
+    const writeMask = this._stencilMode === STENCIL_MODES.RENDERING_MASK_ADD ? 0 : this._colorMask;
+    for (let i = 0; i < blendModes.length; i++) {
+      blendModes[i].writeMask = writeMask;
+    }
+    const layout = this._renderer.shader.getProgramData(program).pipeline;
+    const descriptor = {
+      // TODO later check if its helpful to create..
+      // layout,
+      vertex: {
+        module: this._getModule(program.vertex.source),
+        entryPoint: program.vertex.entryPoint,
+        // geometry..
+        buffers
+      },
+      fragment: {
+        module: this._getModule(program.fragment.source),
+        entryPoint: program.fragment.entryPoint,
+        targets: blendModes
+      },
+      primitive: {
+        topology,
+        cullMode: state.cullMode
+      },
+      layout,
+      multisample: {
+        count: this._multisampleCount
+      },
+      // depthStencil,
+      label: `PIXI Pipeline`
+    };
+    if (this._depthStencilAttachment) {
+      descriptor.depthStencil = {
+        ...this._stencilState,
+        format: "depth24plus-stencil8",
+        depthWriteEnabled: state.depthTest,
+        depthCompare: state.depthTest ? "less" : "always"
+      };
+    }
+    const pipeline = device.createRenderPipeline(descriptor);
+    return pipeline;
+  }
+  _getModule(code) {
+    return this._moduleCache[code] || this._createModule(code);
+  }
+  _createModule(code) {
+    const device = this._gpu.device;
+    this._moduleCache[code] = device.createShaderModule({
+      code
+    });
+    return this._moduleCache[code];
+  }
+  _generateBufferKey(geometry) {
+    const keyGen = [];
+    let index = 0;
+    const attributeKeys = Object.keys(geometry.attributes).sort();
+    for (let i = 0; i < attributeKeys.length; i++) {
+      const attribute = geometry.attributes[attributeKeys[i]];
+      keyGen[index++] = attribute.offset;
+      keyGen[index++] = attribute.format;
+      keyGen[index++] = attribute.stride;
+      keyGen[index++] = attribute.instance;
+    }
+    const stringKey = keyGen.join("|");
+    geometry._layoutKey = createIdFromString(stringKey, "geometry");
+    return geometry._layoutKey;
+  }
+  _generateAttributeLocationsKey(program) {
+    const keyGen = [];
+    let index = 0;
+    const attributeKeys = Object.keys(program.attributeData).sort();
+    for (let i = 0; i < attributeKeys.length; i++) {
+      const attribute = program.attributeData[attributeKeys[i]];
+      keyGen[index++] = attribute.location;
+    }
+    const stringKey = keyGen.join("|");
+    program._attributeLocationsKey = createIdFromString(stringKey, "programAttributes");
+    return program._attributeLocationsKey;
+  }
+  /**
+   * Returns a hash of buffer names mapped to bind locations.
+   * This is used to bind the correct buffer to the correct location in the shader.
+   * @param geometry - The geometry where to get the buffer names
+   * @param program - The program where to get the buffer names
+   * @returns An object of buffer names mapped to the bind location.
+   */
+  getBufferNamesToBind(geometry, program) {
+    const key = geometry._layoutKey << 16 | program._attributeLocationsKey;
+    if (this._bindingNamesCache[key]) return this._bindingNamesCache[key];
+    const data = this._createVertexBufferLayouts(geometry, program);
+    const bufferNamesToBind = /* @__PURE__ */ Object.create(null);
+    const attributeData = program.attributeData;
+    for (let i = 0; i < data.length; i++) {
+      const attributes = Object.values(data[i].attributes);
+      const shaderLocation = attributes[0].shaderLocation;
+      for (const j in attributeData) {
+        if (attributeData[j].location === shaderLocation) {
+          bufferNamesToBind[i] = j;
+          break;
+        }
+      }
+    }
+    this._bindingNamesCache[key] = bufferNamesToBind;
+    return bufferNamesToBind;
+  }
+  _createVertexBufferLayouts(geometry, program) {
+    if (!program._attributeLocationsKey) this._generateAttributeLocationsKey(program);
+    const key = geometry._layoutKey << 16 | program._attributeLocationsKey;
+    if (this._bufferLayoutsCache[key]) {
+      return this._bufferLayoutsCache[key];
+    }
+    const vertexBuffersLayout = [];
+    geometry.buffers.forEach((buffer) => {
+      const bufferEntry = {
+        arrayStride: 0,
+        stepMode: "vertex",
+        attributes: []
+      };
+      const bufferEntryAttributes = bufferEntry.attributes;
+      for (const i in program.attributeData) {
+        const attribute = geometry.attributes[i];
+        if ((attribute.divisor ?? 1) !== 1) {
+          warn(`Attribute ${i} has an invalid divisor value of '${attribute.divisor}'. WebGPU only supports a divisor value of 1`);
+        }
+        if (attribute.buffer === buffer) {
+          bufferEntry.arrayStride = attribute.stride;
+          bufferEntry.stepMode = attribute.instance ? "instance" : "vertex";
+          bufferEntryAttributes.push({
+            shaderLocation: program.attributeData[i].location,
+            offset: attribute.offset,
+            format: attribute.format
+          });
+        }
+      }
+      if (bufferEntryAttributes.length) {
+        vertexBuffersLayout.push(bufferEntry);
+      }
+    });
+    this._bufferLayoutsCache[key] = vertexBuffersLayout;
+    return vertexBuffersLayout;
+  }
+  _updatePipeHash() {
+    const key = getGlobalStateKey(
+      this._stencilMode,
+      this._multisampleCount,
+      this._colorMask,
+      this._depthStencilAttachment,
+      this._colorTargetCount
+    );
+    if (!this._pipeStateCaches[key]) {
+      this._pipeStateCaches[key] = /* @__PURE__ */ Object.create(null);
+    }
+    this._pipeCache = this._pipeStateCaches[key];
+  }
+  destroy() {
+    this._renderer = null;
+    this._bufferLayoutsCache = null;
+  }
+}
+PipelineSystem.extension = {
+  type: [ExtensionType.WebGPUSystem],
+  name: "pipeline"
+};
+class GpuRenderTarget {
+  constructor() {
+    this.contexts = [];
+    this.msaaTextures = [];
+    this.msaaSamples = 1;
+  }
+}
+class GpuRenderTargetAdaptor {
+  init(renderer, renderTargetSystem) {
+    this._renderer = renderer;
+    this._renderTargetSystem = renderTargetSystem;
+  }
+  copyToTexture(sourceRenderSurfaceTexture, destinationTexture, originSrc, size, originDest) {
+    const renderer = this._renderer;
+    const baseGpuTexture = this._getGpuColorTexture(
+      sourceRenderSurfaceTexture
+    );
+    const backGpuTexture = renderer.texture.getGpuSource(
+      destinationTexture.source
+    );
+    renderer.encoder.commandEncoder.copyTextureToTexture(
+      {
+        texture: baseGpuTexture,
+        origin: originSrc
+      },
+      {
+        texture: backGpuTexture,
+        origin: originDest
+      },
+      size
+    );
+    return destinationTexture;
+  }
+  startRenderPass(renderTarget, clear = true, clearColor, viewport, mipLevel = 0, layer = 0) {
+    const renderTargetSystem = this._renderTargetSystem;
+    const gpuRenderTarget = renderTargetSystem.getGpuRenderTarget(renderTarget);
+    if (layer !== 0 && gpuRenderTarget.msaaTextures?.length) {
+      throw new Error("[RenderTargetSystem] Rendering to array layers is not supported with MSAA render targets.");
+    }
+    if (mipLevel > 0 && gpuRenderTarget.msaaTextures?.length) {
+      throw new Error("[RenderTargetSystem] Rendering to mip levels is not supported with MSAA render targets.");
+    }
+    const descriptor = this.getDescriptor(renderTarget, clear, clearColor, mipLevel, layer);
+    gpuRenderTarget.descriptor = descriptor;
+    this._renderer.pipeline.setRenderTarget(gpuRenderTarget);
+    this._renderer.encoder.beginRenderPass(gpuRenderTarget);
+    this._renderer.encoder.setViewport(viewport);
+  }
+  finishRenderPass() {
+    this._renderer.encoder.endRenderPass();
+  }
+  /**
+   * returns the gpu texture for the first color texture in the render target
+   * mainly used by the filter manager to get copy the texture for blending
+   * @param renderTarget
+   * @returns a gpu texture
+   */
+  _getGpuColorTexture(renderTarget) {
+    const gpuRenderTarget = this._renderTargetSystem.getGpuRenderTarget(renderTarget);
+    if (gpuRenderTarget.contexts[0]) {
+      return gpuRenderTarget.contexts[0].getCurrentTexture();
+    }
+    return this._renderer.texture.getGpuSource(
+      renderTarget.colorTextures[0].source
+    );
+  }
+  getDescriptor(renderTarget, clear, clearValue, mipLevel = 0, layer = 0) {
+    if (typeof clear === "boolean") {
+      clear = clear ? CLEAR.ALL : CLEAR.NONE;
+    }
+    const renderTargetSystem = this._renderTargetSystem;
+    const gpuRenderTarget = renderTargetSystem.getGpuRenderTarget(renderTarget);
+    const colorAttachments = renderTarget.colorTextures.map(
+      (texture, i) => {
+        const context = gpuRenderTarget.contexts[i];
+        let view;
+        let resolveTarget;
+        if (context) {
+          if (layer !== 0) {
+            throw new Error("[RenderTargetSystem] Rendering to array layers is not supported for canvas targets.");
+          }
+          const currentTexture = context.getCurrentTexture();
+          const canvasTextureView = currentTexture.createView();
+          view = canvasTextureView;
+        } else {
+          view = this._renderer.texture.getGpuSource(texture).createView({
+            // Render attachments must be 2d views; for array/cube textures we select a single layer.
+            dimension: "2d",
+            baseMipLevel: mipLevel,
+            mipLevelCount: 1,
+            baseArrayLayer: layer,
+            arrayLayerCount: 1
+          });
+        }
+        if (gpuRenderTarget.msaaTextures[i]) {
+          resolveTarget = view;
+          view = this._renderer.texture.getTextureView(
+            gpuRenderTarget.msaaTextures[i]
+          );
+        }
+        const loadOp = clear & CLEAR.COLOR ? "clear" : "load";
+        clearValue ?? (clearValue = renderTargetSystem.defaultClearColor);
+        return {
+          view,
+          resolveTarget,
+          clearValue,
+          storeOp: "store",
+          loadOp
+        };
+      }
+    );
+    let depthStencilAttachment;
+    if ((renderTarget.stencil || renderTarget.depth) && !renderTarget.depthStencilTexture) {
+      renderTarget.ensureDepthStencilTexture();
+      renderTarget.depthStencilTexture.source.sampleCount = gpuRenderTarget.msaa ? 4 : 1;
+    }
+    if (renderTarget.depthStencilTexture) {
+      const stencilLoadOp = clear & CLEAR.STENCIL ? "clear" : "load";
+      const depthLoadOp = clear & CLEAR.DEPTH ? "clear" : "load";
+      depthStencilAttachment = {
+        view: this._renderer.texture.getGpuSource(renderTarget.depthStencilTexture.source).createView({
+          dimension: "2d",
+          baseMipLevel: mipLevel,
+          mipLevelCount: 1,
+          baseArrayLayer: layer,
+          arrayLayerCount: 1
+        }),
+        stencilStoreOp: "store",
+        stencilLoadOp,
+        depthClearValue: 1,
+        depthLoadOp,
+        depthStoreOp: "store"
+      };
+    }
+    const descriptor = {
+      colorAttachments,
+      depthStencilAttachment
+    };
+    return descriptor;
+  }
+  clear(renderTarget, clear = true, clearColor, viewport, mipLevel = 0, layer = 0) {
+    if (!clear) return;
+    const { gpu, encoder } = this._renderer;
+    const device = gpu.device;
+    const standAlone = encoder.commandEncoder === null;
+    if (standAlone) {
+      const commandEncoder = device.createCommandEncoder();
+      const renderPassDescriptor = this.getDescriptor(renderTarget, clear, clearColor, mipLevel, layer);
+      const passEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
+      passEncoder.setViewport(viewport.x, viewport.y, viewport.width, viewport.height, 0, 1);
+      passEncoder.end();
+      const gpuCommands = commandEncoder.finish();
+      device.queue.submit([gpuCommands]);
+    } else {
+      this.startRenderPass(renderTarget, clear, clearColor, viewport, mipLevel, layer);
+    }
+  }
+  initGpuRenderTarget(renderTarget) {
+    renderTarget.isRoot = true;
+    const gpuRenderTarget = new GpuRenderTarget();
+    gpuRenderTarget.colorTargetCount = renderTarget.colorTextures.length;
+    renderTarget.colorTextures.forEach((colorTexture, i) => {
+      if (colorTexture instanceof CanvasSource) {
+        const context = colorTexture.resource.getContext(
+          "webgpu"
+        );
+        const alphaMode = colorTexture.transparent ? "premultiplied" : "opaque";
+        try {
+          context.configure({
+            device: this._renderer.gpu.device,
+            usage: GPUTextureUsage.TEXTURE_BINDING | GPUTextureUsage.COPY_DST | GPUTextureUsage.RENDER_ATTACHMENT | GPUTextureUsage.COPY_SRC,
+            format: "bgra8unorm",
+            alphaMode
+          });
+        } catch (e) {
+          console.error(e);
+        }
+        gpuRenderTarget.contexts[i] = context;
+      }
+      gpuRenderTarget.msaa = colorTexture.source.antialias;
+      if (colorTexture.source.antialias) {
+        const msaaTexture = new TextureSource({
+          width: 0,
+          height: 0,
+          sampleCount: 4,
+          arrayLayerCount: colorTexture.source.arrayLayerCount
+        });
+        gpuRenderTarget.msaaTextures[i] = msaaTexture;
+      }
+    });
+    if (gpuRenderTarget.msaa) {
+      gpuRenderTarget.msaaSamples = 4;
+      if (renderTarget.depthStencilTexture) {
+        renderTarget.depthStencilTexture.source.sampleCount = 4;
+      }
+    }
+    return gpuRenderTarget;
+  }
+  destroyGpuRenderTarget(gpuRenderTarget) {
+    gpuRenderTarget.contexts.forEach((context) => {
+      context.unconfigure();
+    });
+    gpuRenderTarget.msaaTextures.forEach((texture) => {
+      texture.destroy();
+    });
+    gpuRenderTarget.msaaTextures.length = 0;
+    gpuRenderTarget.contexts.length = 0;
+  }
+  ensureDepthStencilTexture(renderTarget) {
+    const gpuRenderTarget = this._renderTargetSystem.getGpuRenderTarget(renderTarget);
+    if (renderTarget.depthStencilTexture && gpuRenderTarget.msaa) {
+      renderTarget.depthStencilTexture.source.sampleCount = 4;
+    }
+  }
+  resizeGpuRenderTarget(renderTarget) {
+    const gpuRenderTarget = this._renderTargetSystem.getGpuRenderTarget(renderTarget);
+    gpuRenderTarget.width = renderTarget.width;
+    gpuRenderTarget.height = renderTarget.height;
+    if (gpuRenderTarget.msaa) {
+      renderTarget.colorTextures.forEach((colorTexture, i) => {
+        const msaaTexture = gpuRenderTarget.msaaTextures[i];
+        msaaTexture?.resize(
+          colorTexture.source.width,
+          colorTexture.source.height,
+          colorTexture.source._resolution
+        );
+      });
+    }
+  }
+}
+class GpuRenderTargetSystem extends RenderTargetSystem {
+  constructor(renderer) {
+    super(renderer);
+    this.adaptor = new GpuRenderTargetAdaptor();
+    this.adaptor.init(renderer, this);
+  }
+}
+GpuRenderTargetSystem.extension = {
+  type: [ExtensionType.WebGPUSystem],
+  name: "renderTarget"
+};
+class GpuShaderSystem {
+  constructor() {
+    this._gpuProgramData = /* @__PURE__ */ Object.create(null);
+  }
+  contextChange(gpu) {
+    this._gpu = gpu;
+  }
+  getProgramData(program) {
+    return this._gpuProgramData[program._layoutKey] || this._createGPUProgramData(program);
+  }
+  _createGPUProgramData(program) {
+    const device = this._gpu.device;
+    const bindGroups = program.gpuLayout.map((group) => device.createBindGroupLayout({ entries: group }));
+    const pipelineLayoutDesc = { bindGroupLayouts: bindGroups };
+    this._gpuProgramData[program._layoutKey] = {
+      bindGroups,
+      pipeline: device.createPipelineLayout(pipelineLayoutDesc)
+    };
+    return this._gpuProgramData[program._layoutKey];
+  }
+  destroy() {
+    this._gpu = null;
+    this._gpuProgramData = null;
+  }
+}
+GpuShaderSystem.extension = {
+  type: [
+    ExtensionType.WebGPUSystem
+  ],
+  name: "shader"
+};
+const GpuBlendModesToPixi = {};
+GpuBlendModesToPixi.normal = {
+  alpha: {
+    srcFactor: "one",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  },
+  color: {
+    srcFactor: "one",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  }
+};
+GpuBlendModesToPixi.add = {
+  alpha: {
+    srcFactor: "src-alpha",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  },
+  color: {
+    srcFactor: "one",
+    dstFactor: "one",
+    operation: "add"
+  }
+};
+GpuBlendModesToPixi.multiply = {
+  alpha: {
+    srcFactor: "one",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  },
+  color: {
+    srcFactor: "dst",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  }
+};
+GpuBlendModesToPixi.screen = {
+  alpha: {
+    srcFactor: "one",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  },
+  color: {
+    srcFactor: "one",
+    dstFactor: "one-minus-src",
+    operation: "add"
+  }
+};
+GpuBlendModesToPixi.overlay = {
+  alpha: {
+    srcFactor: "one",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  },
+  color: {
+    srcFactor: "one",
+    dstFactor: "one-minus-src",
+    operation: "add"
+  }
+};
+GpuBlendModesToPixi.none = {
+  alpha: {
+    srcFactor: "one",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  },
+  color: {
+    srcFactor: "zero",
+    dstFactor: "zero",
+    operation: "add"
+  }
+};
+GpuBlendModesToPixi["normal-npm"] = {
+  alpha: {
+    srcFactor: "one",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  },
+  color: {
+    srcFactor: "src-alpha",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  }
+};
+GpuBlendModesToPixi["add-npm"] = {
+  alpha: {
+    srcFactor: "one",
+    dstFactor: "one",
+    operation: "add"
+  },
+  color: {
+    srcFactor: "src-alpha",
+    dstFactor: "one",
+    operation: "add"
+  }
+};
+GpuBlendModesToPixi["screen-npm"] = {
+  alpha: {
+    srcFactor: "one",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  },
+  color: {
+    srcFactor: "src-alpha",
+    dstFactor: "one-minus-src",
+    operation: "add"
+  }
+};
+GpuBlendModesToPixi.erase = {
+  alpha: {
+    srcFactor: "zero",
+    dstFactor: "one-minus-src-alpha",
+    operation: "add"
+  },
+  color: {
+    srcFactor: "zero",
+    dstFactor: "one-minus-src",
+    operation: "add"
+  }
+};
+GpuBlendModesToPixi.min = {
+  alpha: {
+    srcFactor: "one",
+    dstFactor: "one",
+    operation: "min"
+  },
+  color: {
+    srcFactor: "one",
+    dstFactor: "one",
+    operation: "min"
+  }
+};
+GpuBlendModesToPixi.max = {
+  alpha: {
+    srcFactor: "one",
+    dstFactor: "one",
+    operation: "max"
+  },
+  color: {
+    srcFactor: "one",
+    dstFactor: "one",
+    operation: "max"
+  }
+};
+class GpuStateSystem {
+  constructor() {
+    this.defaultState = new State();
+    this.defaultState.blend = true;
+  }
+  contextChange(gpu) {
+    this.gpu = gpu;
+  }
+  /**
+   * Gets the blend mode data for the current state
+   * @param state - The state to get the blend mode from
+   * @param count - The number of color targets to create
+   */
+  getColorTargets(state, count) {
+    const blend = GpuBlendModesToPixi[state.blendMode] || GpuBlendModesToPixi.normal;
+    const targets = [];
+    const target = {
+      format: "bgra8unorm",
+      writeMask: 0,
+      blend
+    };
+    for (let i = 0; i < count; i++) {
+      targets[i] = target;
+    }
+    return targets;
+  }
+  destroy() {
+    this.gpu = null;
+  }
+}
+GpuStateSystem.extension = {
+  type: [
+    ExtensionType.WebGPUSystem
+  ],
+  name: "state"
+};
+const gpuUploadBufferImageResource = {
+  type: "image",
+  upload(source, gpuTexture, gpu, originZOverride = 0) {
+    const resource = source.resource;
+    const total = (source.pixelWidth | 0) * (source.pixelHeight | 0);
+    const bytesPerPixel = resource.byteLength / total;
+    gpu.device.queue.writeTexture(
+      { texture: gpuTexture, origin: { x: 0, y: 0, z: originZOverride } },
+      resource,
+      {
+        offset: 0,
+        rowsPerImage: source.pixelHeight,
+        bytesPerRow: source.pixelWidth * bytesPerPixel
+      },
+      {
+        width: source.pixelWidth,
+        height: source.pixelHeight,
+        depthOrArrayLayers: 1
+      }
+    );
+  }
+};
+const blockDataMap = {
+  "bc1-rgba-unorm": { blockBytes: 8, blockWidth: 4, blockHeight: 4 },
+  "bc2-rgba-unorm": { blockBytes: 16, blockWidth: 4, blockHeight: 4 },
+  "bc3-rgba-unorm": { blockBytes: 16, blockWidth: 4, blockHeight: 4 },
+  "bc7-rgba-unorm": { blockBytes: 16, blockWidth: 4, blockHeight: 4 },
+  "etc1-rgb-unorm": { blockBytes: 8, blockWidth: 4, blockHeight: 4 },
+  "etc2-rgba8unorm": { blockBytes: 16, blockWidth: 4, blockHeight: 4 },
+  "astc-4x4-unorm": { blockBytes: 16, blockWidth: 4, blockHeight: 4 }
+};
+const defaultBlockData = { blockBytes: 4, blockWidth: 1, blockHeight: 1 };
+const gpuUploadCompressedTextureResource = {
+  type: "compressed",
+  upload(source, gpuTexture, gpu, originZOverride = 0) {
+    let mipWidth = source.pixelWidth;
+    let mipHeight = source.pixelHeight;
+    const blockData = blockDataMap[source.format] || defaultBlockData;
+    for (let i = 0; i < source.resource.length; i++) {
+      const levelBuffer = source.resource[i];
+      const bytesPerRow = Math.ceil(mipWidth / blockData.blockWidth) * blockData.blockBytes;
+      gpu.device.queue.writeTexture(
+        {
+          texture: gpuTexture,
+          mipLevel: i,
+          origin: { x: 0, y: 0, z: originZOverride }
+        },
+        levelBuffer,
+        {
+          offset: 0,
+          bytesPerRow
+        },
+        {
+          width: Math.ceil(mipWidth / blockData.blockWidth) * blockData.blockWidth,
+          height: Math.ceil(mipHeight / blockData.blockHeight) * blockData.blockHeight,
+          depthOrArrayLayers: 1
+        }
+      );
+      mipWidth = Math.max(mipWidth >> 1, 1);
+      mipHeight = Math.max(mipHeight >> 1, 1);
+    }
+  }
+};
+const FACE_ORDER = ["right", "left", "top", "bottom", "front", "back"];
+function createGpuUploadCubeTextureResource(uploaders) {
+  return {
+    type: "cube",
+    upload(source, gpuTexture, gpu) {
+      const faces = source.faces;
+      for (let i = 0; i < FACE_ORDER.length; i++) {
+        const key = FACE_ORDER[i];
+        const face = faces[key];
+        const uploader = uploaders[face.uploadMethodId] || uploaders.image;
+        uploader.upload(face, gpuTexture, gpu, i);
+      }
+    }
+  };
+}
+const gpuUploadImageResource = {
+  type: "image",
+  upload(source, gpuTexture, gpu, originZOverride = 0) {
+    const resource = source.resource;
+    if (!resource) return;
+    if (globalThis.HTMLImageElement && resource instanceof HTMLImageElement) {
+      const canvas = DOMAdapter.get().createCanvas(resource.width, resource.height);
+      const context = canvas.getContext("2d");
+      context.drawImage(resource, 0, 0, resource.width, resource.height);
+      source.resource = canvas;
+      warn("ImageSource: Image element passed, converting to canvas and replacing resource.");
+    }
+    const width = Math.min(gpuTexture.width, source.resourceWidth || source.pixelWidth);
+    const height = Math.min(gpuTexture.height, source.resourceHeight || source.pixelHeight);
+    const premultipliedAlpha = source.alphaMode === "premultiply-alpha-on-upload";
+    gpu.device.queue.copyExternalImageToTexture(
+      { source: resource },
+      { texture: gpuTexture, origin: { x: 0, y: 0, z: originZOverride }, premultipliedAlpha },
+      {
+        width,
+        height
+      }
+    );
+  }
+};
+const gpuUploadVideoResource = {
+  type: "video",
+  upload(source, gpuTexture, gpu, originZOverride) {
+    gpuUploadImageResource.upload(source, gpuTexture, gpu, originZOverride);
+  }
+};
+class GpuMipmapGenerator {
+  constructor(device) {
+    this.device = device;
+    this.sampler = device.createSampler({ minFilter: "linear" });
+    this.pipelines = {};
+  }
+  _getMipmapPipeline(format) {
+    let pipeline = this.pipelines[format];
+    if (!pipeline) {
+      if (!this.mipmapShaderModule) {
+        this.mipmapShaderModule = this.device.createShaderModule({
+          code: (
+            /* wgsl */
+            `
+                        var<private> pos : array<vec2<f32>, 3> = array<vec2<f32>, 3>(
+                        vec2<f32>(-1.0, -1.0), vec2<f32>(-1.0, 3.0), vec2<f32>(3.0, -1.0));
+
+                        struct VertexOutput {
+                        @builtin(position) position : vec4<f32>,
+                        @location(0) texCoord : vec2<f32>,
+                        };
+
+                        @vertex
+                        fn vertexMain(@builtin(vertex_index) vertexIndex : u32) -> VertexOutput {
+                        var output : VertexOutput;
+                        output.texCoord = pos[vertexIndex] * vec2<f32>(0.5, -0.5) + vec2<f32>(0.5);
+                        output.position = vec4<f32>(pos[vertexIndex], 0.0, 1.0);
+                        return output;
+                        }
+
+                        @group(0) @binding(0) var imgSampler : sampler;
+                        @group(0) @binding(1) var img : texture_2d<f32>;
+
+                        @fragment
+                        fn fragmentMain(@location(0) texCoord : vec2<f32>) -> @location(0) vec4<f32> {
+                        return textureSample(img, imgSampler, texCoord);
+                        }
+                    `
+          )
+        });
+      }
+      pipeline = this.device.createRenderPipeline({
+        layout: "auto",
+        vertex: {
+          module: this.mipmapShaderModule,
+          entryPoint: "vertexMain"
+        },
+        fragment: {
+          module: this.mipmapShaderModule,
+          entryPoint: "fragmentMain",
+          targets: [{ format }]
+        }
+      });
+      this.pipelines[format] = pipeline;
+    }
+    return pipeline;
+  }
+  /**
+   * Generates mipmaps for the given GPUTexture from the data in level 0.
+   * @param {module:External.GPUTexture} texture - Texture to generate mipmaps for.
+   * @returns {module:External.GPUTexture} - The originally passed texture
+   */
+  generateMipmap(texture) {
+    const pipeline = this._getMipmapPipeline(texture.format);
+    if (texture.dimension === "3d" || texture.dimension === "1d") {
+      throw new Error("Generating mipmaps for non-2d textures is currently unsupported!");
+    }
+    let mipTexture = texture;
+    const arrayLayerCount = texture.depthOrArrayLayers || 1;
+    const renderToSource = texture.usage & GPUTextureUsage.RENDER_ATTACHMENT;
+    if (!renderToSource) {
+      const mipTextureDescriptor = {
+        size: {
+          width: Math.ceil(texture.width / 2),
+          height: Math.ceil(texture.height / 2),
+          depthOrArrayLayers: arrayLayerCount
+        },
+        format: texture.format,
+        usage: GPUTextureUsage.TEXTURE_BINDING | GPUTextureUsage.COPY_SRC | GPUTextureUsage.RENDER_ATTACHMENT,
+        mipLevelCount: texture.mipLevelCount - 1
+      };
+      mipTexture = this.device.createTexture(mipTextureDescriptor);
+    }
+    const commandEncoder = this.device.createCommandEncoder({});
+    const bindGroupLayout = pipeline.getBindGroupLayout(0);
+    for (let arrayLayer = 0; arrayLayer < arrayLayerCount; ++arrayLayer) {
+      let srcView = texture.createView({
+        baseMipLevel: 0,
+        mipLevelCount: 1,
+        dimension: "2d",
+        baseArrayLayer: arrayLayer,
+        arrayLayerCount: 1
+      });
+      let dstMipLevel = renderToSource ? 1 : 0;
+      for (let i = 1; i < texture.mipLevelCount; ++i) {
+        const dstView = mipTexture.createView({
+          baseMipLevel: dstMipLevel++,
+          mipLevelCount: 1,
+          dimension: "2d",
+          baseArrayLayer: arrayLayer,
+          arrayLayerCount: 1
+        });
+        const passEncoder = commandEncoder.beginRenderPass({
+          colorAttachments: [{
+            view: dstView,
+            storeOp: "store",
+            loadOp: "clear",
+            clearValue: { r: 0, g: 0, b: 0, a: 0 }
+          }]
+        });
+        const bindGroup = this.device.createBindGroup({
+          layout: bindGroupLayout,
+          entries: [{
+            binding: 0,
+            resource: this.sampler
+          }, {
+            binding: 1,
+            resource: srcView
+          }]
+        });
+        passEncoder.setPipeline(pipeline);
+        passEncoder.setBindGroup(0, bindGroup);
+        passEncoder.draw(3, 1, 0, 0);
+        passEncoder.end();
+        srcView = dstView;
+      }
+    }
+    if (!renderToSource) {
+      const mipLevelSize = {
+        width: Math.ceil(texture.width / 2),
+        height: Math.ceil(texture.height / 2),
+        depthOrArrayLayers: arrayLayerCount
+      };
+      for (let i = 1; i < texture.mipLevelCount; ++i) {
+        commandEncoder.copyTextureToTexture({
+          texture: mipTexture,
+          mipLevel: i - 1
+        }, {
+          texture,
+          mipLevel: i
+        }, mipLevelSize);
+        mipLevelSize.width = Math.ceil(mipLevelSize.width / 2);
+        mipLevelSize.height = Math.ceil(mipLevelSize.height / 2);
+      }
+    }
+    this.device.queue.submit([commandEncoder.finish()]);
+    if (!renderToSource) {
+      mipTexture.destroy();
+    }
+    return texture;
+  }
+}
+class GPUTextureGpuData {
+  constructor(gpuTexture) {
+    this.textureView = null;
+    this.gpuTexture = gpuTexture;
+  }
+  /** Destroys this GPU data instance. */
+  destroy() {
+    this.gpuTexture.destroy();
+    this.textureView = null;
+    this.gpuTexture = null;
+  }
+}
+class GpuTextureSystem {
+  constructor(renderer) {
+    this._gpuSamplers = /* @__PURE__ */ Object.create(null);
+    this._bindGroupHash = /* @__PURE__ */ Object.create(null);
+    this._renderer = renderer;
+    renderer.gc.addCollection(this, "_bindGroupHash", "hash");
+    this._managedTextures = new GCManagedHash({
+      renderer,
+      type: "resource",
+      onUnload: this.onSourceUnload.bind(this),
+      name: "gpuTextureSource"
+    });
+    const baseUploaders = {
+      image: gpuUploadImageResource,
+      buffer: gpuUploadBufferImageResource,
+      video: gpuUploadVideoResource,
+      compressed: gpuUploadCompressedTextureResource
+    };
+    this._uploads = {
+      ...baseUploaders,
+      cube: createGpuUploadCubeTextureResource(baseUploaders)
+    };
+  }
+  /**
+   * @deprecated since 8.15.0
+   */
+  get managedTextures() {
+    return Object.values(this._managedTextures.items);
+  }
+  contextChange(gpu) {
+    this._gpu = gpu;
+  }
+  /**
+   * Initializes a texture source, if it has already been initialized nothing will happen.
+   * @param source - The texture source to initialize.
+   * @returns The initialized texture source.
+   */
+  initSource(source) {
+    return source._gpuData[this._renderer.uid]?.gpuTexture || this._initSource(source);
+  }
+  _initSource(source) {
+    if (source.autoGenerateMipmaps) {
+      const biggestDimension = Math.max(source.pixelWidth, source.pixelHeight);
+      source.mipLevelCount = Math.floor(Math.log2(biggestDimension)) + 1;
+    }
+    let usage = GPUTextureUsage.TEXTURE_BINDING | GPUTextureUsage.COPY_DST;
+    if (source.uploadMethodId !== "compressed") {
+      usage |= GPUTextureUsage.RENDER_ATTACHMENT;
+      usage |= GPUTextureUsage.COPY_SRC;
+    }
+    const blockData = blockDataMap[source.format] || { blockWidth: 1, blockHeight: 1 };
+    const width = Math.ceil(source.pixelWidth / blockData.blockWidth) * blockData.blockWidth;
+    const height = Math.ceil(source.pixelHeight / blockData.blockHeight) * blockData.blockHeight;
+    const textureDescriptor = {
+      label: source.label,
+      size: { width, height, depthOrArrayLayers: source.arrayLayerCount },
+      format: source.format,
+      sampleCount: source.sampleCount,
+      mipLevelCount: source.mipLevelCount,
+      dimension: source.dimension,
+      usage
+    };
+    const gpuTexture = this._gpu.device.createTexture(textureDescriptor);
+    source._gpuData[this._renderer.uid] = new GPUTextureGpuData(gpuTexture);
+    const added = this._managedTextures.add(source);
+    if (added) {
+      source.on("update", this.onSourceUpdate, this);
+      source.on("resize", this.onSourceResize, this);
+      source.on("updateMipmaps", this.onUpdateMipmaps, this);
+    }
+    this.onSourceUpdate(source);
+    return gpuTexture;
+  }
+  onSourceUpdate(source) {
+    const gpuTexture = this.getGpuSource(source);
+    if (!gpuTexture) return;
+    if (this._uploads[source.uploadMethodId]) {
+      this._uploads[source.uploadMethodId].upload(source, gpuTexture, this._gpu);
+    }
+    if (source.autoGenerateMipmaps && source.mipLevelCount > 1) {
+      this.onUpdateMipmaps(source);
+    }
+  }
+  onUpdateMipmaps(source) {
+    if (!this._mipmapGenerator) {
+      this._mipmapGenerator = new GpuMipmapGenerator(this._gpu.device);
+    }
+    const gpuTexture = this.getGpuSource(source);
+    this._mipmapGenerator.generateMipmap(gpuTexture);
+  }
+  onSourceUnload(source) {
+    source.off("update", this.onSourceUpdate, this);
+    source.off("resize", this.onSourceResize, this);
+    source.off("updateMipmaps", this.onUpdateMipmaps, this);
+  }
+  onSourceResize(source) {
+    source._gcLastUsed = this._renderer.gc.now;
+    const gpuData = source._gpuData[this._renderer.uid];
+    const gpuTexture = gpuData?.gpuTexture;
+    if (!gpuTexture) {
+      this.initSource(source);
+    } else if (gpuTexture.width !== source.pixelWidth || gpuTexture.height !== source.pixelHeight) {
+      gpuData.destroy();
+      this._bindGroupHash[source.uid] = null;
+      source._gpuData[this._renderer.uid] = null;
+      this.initSource(source);
+    }
+  }
+  _initSampler(sampler) {
+    this._gpuSamplers[sampler._resourceId] = this._gpu.device.createSampler(sampler);
+    return this._gpuSamplers[sampler._resourceId];
+  }
+  getGpuSampler(sampler) {
+    return this._gpuSamplers[sampler._resourceId] || this._initSampler(sampler);
+  }
+  getGpuSource(source) {
+    source._gcLastUsed = this._renderer.gc.now;
+    return source._gpuData[this._renderer.uid]?.gpuTexture || this.initSource(source);
+  }
+  /**
+   * this returns s bind group for a specific texture, the bind group contains
+   * - the texture source
+   * - the texture style
+   * - the texture matrix
+   * This is cached so the bind group should only be created once per texture
+   * @param texture - the texture you want the bindgroup for
+   * @returns the bind group for the texture
+   */
+  getTextureBindGroup(texture) {
+    return this._bindGroupHash[texture.uid] || this._createTextureBindGroup(texture);
+  }
+  _createTextureBindGroup(texture) {
+    const source = texture.source;
+    this._bindGroupHash[texture.uid] = new BindGroup({
+      0: source,
+      1: source.style,
+      2: new UniformGroup({
+        uTextureMatrix: { type: "mat3x3<f32>", value: texture.textureMatrix.mapCoord }
+      })
+    });
+    return this._bindGroupHash[texture.uid];
+  }
+  getTextureView(texture) {
+    const source = texture.source;
+    source._gcLastUsed = this._renderer.gc.now;
+    let gpuData = source._gpuData[this._renderer.uid];
+    if (!gpuData) {
+      this.initSource(source);
+      gpuData = source._gpuData[this._renderer.uid];
+    }
+    gpuData.textureView || (gpuData.textureView = gpuData.gpuTexture.createView({ dimension: source.viewDimension }));
+    return gpuData.textureView;
+  }
+  generateCanvas(texture) {
+    const renderer = this._renderer;
+    const commandEncoder = renderer.gpu.device.createCommandEncoder();
+    const canvas = DOMAdapter.get().createCanvas();
+    canvas.width = texture.source.pixelWidth;
+    canvas.height = texture.source.pixelHeight;
+    const context = canvas.getContext("webgpu");
+    context.configure({
+      device: renderer.gpu.device,
+      usage: GPUTextureUsage.COPY_DST | GPUTextureUsage.COPY_SRC,
+      format: DOMAdapter.get().getNavigator().gpu.getPreferredCanvasFormat(),
+      alphaMode: "premultiplied"
+    });
+    commandEncoder.copyTextureToTexture({
+      texture: renderer.texture.getGpuSource(texture.source),
+      origin: {
+        x: 0,
+        y: 0
+      }
+    }, {
+      texture: context.getCurrentTexture()
+    }, {
+      width: canvas.width,
+      height: canvas.height
+    });
+    renderer.gpu.device.queue.submit([commandEncoder.finish()]);
+    return canvas;
+  }
+  getPixels(texture) {
+    const webGPUCanvas = this.generateCanvas(texture);
+    const canvasAndContext = CanvasPool.getOptimalCanvasAndContext(webGPUCanvas.width, webGPUCanvas.height);
+    const context = canvasAndContext.context;
+    context.drawImage(webGPUCanvas, 0, 0);
+    const { width, height } = webGPUCanvas;
+    const imageData = context.getImageData(0, 0, width, height);
+    const pixels = new Uint8ClampedArray(imageData.data.buffer);
+    CanvasPool.returnCanvasAndContext(canvasAndContext);
+    return { pixels, width, height };
+  }
+  destroy() {
+    this._managedTextures.destroy();
+    for (const k of Object.keys(this._bindGroupHash)) {
+      const key = Number(k);
+      const bindGroup = this._bindGroupHash[key];
+      bindGroup?.destroy();
+    }
+    this._renderer = null;
+    this._gpu = null;
+    this._mipmapGenerator = null;
+    this._gpuSamplers = null;
+    this._bindGroupHash = null;
+  }
+}
+GpuTextureSystem.extension = {
+  type: [
+    ExtensionType.WebGPUSystem
+  ],
+  name: "texture"
+};
+class GpuGraphicsAdaptor {
+  constructor() {
+    this._maxTextures = 0;
+  }
+  contextChange(renderer) {
+    const localUniforms = new UniformGroup({
+      uTransformMatrix: { value: new Matrix(), type: "mat3x3<f32>" },
+      uColor: { value: new Float32Array([1, 1, 1, 1]), type: "vec4<f32>" },
+      uRound: { value: 0, type: "f32" }
+    });
+    this._maxTextures = renderer.limits.maxBatchableTextures;
+    const gpuProgram = compileHighShaderGpuProgram({
+      name: "graphics",
+      bits: [
+        colorBit,
+        generateTextureBatchBit(this._maxTextures),
+        localUniformBitGroup2,
+        roundPixelsBit
+      ]
+    });
+    this.shader = new Shader({
+      gpuProgram,
+      resources: {
+        // added on the fly!
+        localUniforms
+      }
+    });
+  }
+  execute(graphicsPipe, renderable) {
+    const context = renderable.context;
+    const shader = context.customShader || this.shader;
+    const renderer = graphicsPipe.renderer;
+    const contextSystem = renderer.graphicsContext;
+    const {
+      batcher,
+      instructions
+    } = contextSystem.getContextRenderData(context);
+    const encoder = renderer.encoder;
+    encoder.setGeometry(batcher.geometry, shader.gpuProgram);
+    const globalUniformsBindGroup = renderer.globalUniforms.bindGroup;
+    encoder.setBindGroup(0, globalUniformsBindGroup, shader.gpuProgram);
+    const localBindGroup = renderer.renderPipes.uniformBatch.getUniformBindGroup(shader.resources.localUniforms, true);
+    encoder.setBindGroup(2, localBindGroup, shader.gpuProgram);
+    const batches = instructions.instructions;
+    let topology = null;
+    for (let i = 0; i < instructions.instructionSize; i++) {
+      const batch = batches[i];
+      if (batch.topology !== topology) {
+        topology = batch.topology;
+        encoder.setPipelineFromGeometryProgramAndState(
+          batcher.geometry,
+          shader.gpuProgram,
+          graphicsPipe.state,
+          batch.topology
+        );
+      }
+      shader.groups[1] = batch.bindGroup;
+      if (!batch.gpuBindGroup) {
+        const textureBatch = batch.textures;
+        batch.bindGroup = getTextureBatchBindGroup(
+          textureBatch.textures,
+          textureBatch.count,
+          this._maxTextures
+        );
+        batch.gpuBindGroup = renderer.bindGroup.getBindGroup(
+          batch.bindGroup,
+          shader.gpuProgram,
+          1
+        );
+      }
+      encoder.setBindGroup(1, batch.bindGroup, shader.gpuProgram);
+      encoder.renderPassEncoder.drawIndexed(batch.size, 1, batch.start);
+    }
+  }
+  destroy() {
+    this.shader.destroy(true);
+    this.shader = null;
+  }
+}
+GpuGraphicsAdaptor.extension = {
+  type: [
+    ExtensionType.WebGPUPipesAdaptor
+  ],
+  name: "graphics"
+};
+class GpuMeshAdapter {
+  init() {
+    const gpuProgram = compileHighShaderGpuProgram({
+      name: "mesh",
+      bits: [
+        localUniformBit,
+        textureBit,
+        roundPixelsBit
+      ]
+    });
+    this._shader = new Shader({
+      gpuProgram,
+      resources: {
+        uTexture: Texture.EMPTY._source,
+        uSampler: Texture.EMPTY._source.style,
+        textureUniforms: {
+          uTextureMatrix: { type: "mat3x3<f32>", value: new Matrix() }
+        }
+      }
+    });
+  }
+  execute(meshPipe, mesh) {
+    const renderer = meshPipe.renderer;
+    let shader = mesh._shader;
+    if (!shader) {
+      shader = this._shader;
+      shader.groups[2] = renderer.texture.getTextureBindGroup(mesh.texture);
+    } else if (!shader.gpuProgram) {
+      warn("Mesh shader has no gpuProgram", mesh.shader);
+      return;
+    }
+    const gpuProgram = shader.gpuProgram;
+    if (gpuProgram.autoAssignGlobalUniforms) {
+      shader.groups[0] = renderer.globalUniforms.bindGroup;
+    }
+    if (gpuProgram.autoAssignLocalUniforms) {
+      const localUniforms = meshPipe.localUniforms;
+      shader.groups[1] = renderer.renderPipes.uniformBatch.getUniformBindGroup(localUniforms, true);
+    }
+    renderer.encoder.draw({
+      geometry: mesh._geometry,
+      shader,
+      state: mesh.state
+    });
+  }
+  destroy() {
+    this._shader.destroy(true);
+    this._shader = null;
+  }
+}
+GpuMeshAdapter.extension = {
+  type: [
+    ExtensionType.WebGPUPipesAdaptor
+  ],
+  name: "mesh"
+};
+const DefaultWebGPUSystems = [
+  ...SharedSystems,
+  GpuUboSystem,
+  GpuEncoderSystem,
+  GpuDeviceSystem,
+  GpuLimitsSystem,
+  GpuBufferSystem,
+  GpuTextureSystem,
+  GpuRenderTargetSystem,
+  GpuShaderSystem,
+  GpuStateSystem,
+  PipelineSystem,
+  GpuColorMaskSystem,
+  GpuStencilSystem,
+  BindGroupSystem
+];
+const DefaultWebGPUPipes = [...SharedRenderPipes, GpuUniformBatchPipe];
+const DefaultWebGPUAdapters = [GpuBatchAdaptor, GpuMeshAdapter, GpuGraphicsAdaptor];
+const systems = [];
+const renderPipes = [];
+const renderPipeAdaptors = [];
+extensions.handleByNamedList(ExtensionType.WebGPUSystem, systems);
+extensions.handleByNamedList(ExtensionType.WebGPUPipes, renderPipes);
+extensions.handleByNamedList(ExtensionType.WebGPUPipesAdaptor, renderPipeAdaptors);
+extensions.add(...DefaultWebGPUSystems, ...DefaultWebGPUPipes, ...DefaultWebGPUAdapters);
+class WebGPURenderer extends AbstractRenderer {
+  constructor() {
+    const systemConfig = {
+      name: "webgpu",
+      type: RendererType.WEBGPU,
+      systems,
+      renderPipes,
+      renderPipeAdaptors
+    };
+    super(systemConfig);
+  }
+}
+export {
+  WebGPURenderer
+};