The GPUTextureView interface of the WebGPU API represents a view into a subset of the texture resources defined by a particular GPUTexture.
A GPUTextureView object instance is created using the GPUTexture.createView() method.
In the WebGPU Samples Cubemap demo, you will see multiple examples of how GPUTextureViews (created by GPUTexture.createView() calls) are used, both as a resource in a GPUDevice.createBindGroup() call, and as a provided view in the depthStencilAttachment object of a GPUCommandEncoder.beginRenderPass() descriptor.
const uniformBindGroup = device.createBindGroup({
layout: pipeline.getBindGroupLayout(0),
entries: [
{
binding: 0,
resource: {
buffer: uniformBuffer,
offset: 0,
size: uniformBufferSize,
},
},
{
binding: 1,
resource: sampler,
},
{
binding: 2,
resource: cubemapTexture.createView({
dimension: "cube",
}),
},
],
});
const renderPassDescriptor: GPURenderPassDescriptor = {
colorAttachments: [
{
view: undefined,
loadOp: "clear",
storeOp: "store",
},
],
depthStencilAttachment: {
view: depthTexture.createView(),
depthClearValue: 1.0,
depthLoadOp: "clear",
depthStoreOp: "store",
},
};
const commandEncoder = device.createCommandEncoder();
const passEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);