Documentation / @ripl/webgpu / WebGPUContext3D

Class: WebGPUContext3D

Defined in: webgpu/src/context.ts:62

WebGPU-backed 3D rendering context with hardware depth testing and WGSL shaders.

Extends

• Context3D

Constructors

Constructor

new WebGPUContext3D(target, device, gpuContext, canvas, pipelineState, options?): WebGPUContext3D

Defined in: webgpu/src/context.ts:83

Parameters

Parameter	Type
target	string | HTMLElement
device	GPUDevice
gpuContext	GPUCanvasContext
canvas	HTMLCanvasElement
pipelineState	PipelineState
options?	WebGPUContextOptions

Returns

WebGPUContext3D

Overrides

Context3D.constructor

Properties

Property	Modifier	Type	Default value	Description	Inherited from	Defined in
cameraPosition	public	Vector3	undefined	The camera's world-space position, as last set by setCamera.	Context3D.cameraPosition	3d/src/core/context.ts:185
currentState	protected	BaseState	undefined	-	Context3D.currentState	core/src/context/context.ts:104
element	readonly	HTMLCanvasElement	undefined	The underlying DOM element the context renders into.	Context3D.element	core/src/context/context.ts:111
faceBuffer	public	ProjectedFace3D[]	[]	Faces accumulated since the last flush, sorted back-to-front and drained when they are drawn (painter's algorithm). A backend that never draws them — the base class, or a GPU context — leaves this empty by clearing it at the start of each pass.	Context3D.faceBuffer	3d/src/core/context.ts:191
far	protected	number	undefined	-	Context3D.far	3d/src/core/context.ts:195
fov	protected	number	undefined	-	Context3D.fov	3d/src/core/context.ts:193
height	public	number	undefined	Current height of the rendering surface, in logical pixels — CSS pixels, unaffected by the device pixel ratio.	Context3D.height	core/src/context/context.ts:124
hitTestHonorsTransform	public	boolean	false	Whether this context's hit testing natively accounts for element and ancestor group transforms (as SVG does, via the DOM). When false (e.g. canvas), callers map the hit point into the element's local space before testing. See Element.getWorldTransform.	Context3D.hitTestHonorsTransform	core/src/context/context.ts:120
lights	readonly	LightList	undefined	The lights illuminating the scene.	Context3D.lights	3d/src/core/context.ts:183
meta	readonly	Context3DMeta	undefined	Arbitrary metadata attached to the context.	Context3D.meta	core/src/context/context.ts:113
near	protected	number	undefined	-	Context3D.near	3d/src/core/context.ts:194
parent?	public	EventBus<ContextEventMap>	undefined	The parent event bus that emitted events bubble up to, if any.	Context3D.parent	core/src/core/event-bus.ts:107
projectionMatrix	public	Matrix4	undefined	The projection matrix transforming view space into clip space.	Context3D.projectionMatrix	3d/src/core/context.ts:179
renderDepth	protected	number	0	-	Context3D.renderDepth	core/src/context/context.ts:105
renderedElements	public	RenderElement[]	undefined	Elements rendered during the current pass, used for hit testing.	Context3D.renderedElements	core/src/context/context.ts:132
renderElement?	public	RenderElement	undefined	The element currently being rendered, if any.	Context3D.renderElement	core/src/context/context.ts:130
root	readonly	HTMLElement	undefined	The host DOM element that the context's rendering surface is mounted into.	Context3D.root	dom/src/context.ts:48
saveDepth	protected	number	0	-	Context3D.saveDepth	core/src/context/context.ts:106
scaleX	public	Scale<number, number>	undefined	Scale mapping logical x coordinates onto this backend's surface x coordinates. Drives drawing, not a conversion seam for callers — use Context.toSurfacePoint.	Context3D.scaleX	core/src/context/context.ts:126
scaleY	public	Scale<number, number>	undefined	Scale mapping logical y coordinates onto this backend's surface y coordinates. Drives drawing, not a conversion seam for callers — use Context.toSurfacePoint.	Context3D.scaleY	core/src/context/context.ts:128
states	protected	BaseState[]	undefined	-	Context3D.states	core/src/context/context.ts:103
type	readonly	string	undefined	The context type identifier (e.g. canvas, svg).	Context3D.type	core/src/context/context.ts:109
viewMatrix	public	Matrix4	undefined	The view matrix transforming world space into camera (view) space.	Context3D.viewMatrix	3d/src/core/context.ts:177
viewProjectionMatrix	public	Matrix4	undefined	The combined view-projection matrix, transforming world space directly into clip space.	Context3D.viewProjectionMatrix	3d/src/core/context.ts:181
width	public	number	undefined	Current width of the rendering surface, in logical pixels — CSS pixels, unaffected by the device pixel ratio.	Context3D.width	core/src/context/context.ts:122
defaultKey	readonly	typeof defaultKey	undefined	The key under which resources are retained when no explicit key is provided.	Context3D.defaultKey	core/src/core/disposer.ts:11

Accessors

$events

Get Signature

get $events(): keyof ContextEventMap[]

Defined in: core/src/context/context.ts:135

The event types a context can emit. See EventBus.$events.

Returns

keyof ContextEventMap[]

Inherited from

Context3D.$events

---

currentRenderElement

Get Signature

get currentRenderElement(): RenderElement | undefined

Defined in: core/src/context/context.ts:153

The element currently being rendered; setting a non-abstract element also records it in Context.renderedElements.

Returns

RenderElement | undefined

Set Signature

set currentRenderElement(element): void

Defined in: core/src/context/context.ts:157

Parameters

Parameter	Type
element	RenderElement | undefined

Returns

void

Inherited from

Context3D.currentRenderElement

---

direction

Get Signature

get direction(): Direction

Defined in: core/src/context/context.ts:184

Directionality used when rendering text.

Returns

Direction

Set Signature

set direction(value): void

Defined in: core/src/context/context.ts:188

Directionality used when rendering text.

Parameters

Parameter	Type
value	Direction

Returns

void

Directionality used when rendering text.

Inherited from

Context3D.direction

---

fill

Get Signature

get fill(): string

Defined in: core/src/context/context.ts:166

Fill style (CSS color, gradient, or pattern) used to paint filled regions.

Returns

string

Set Signature

set fill(value): void

Defined in: core/src/context/context.ts:170

Fill style (CSS color, gradient, or pattern) used to paint filled regions.

Parameters

Parameter	Type
value	string

Returns

void

Fill style (CSS color, gradient, or pattern) used to paint filled regions.

Inherited from

Context3D.fill

---

filter

Get Signature

get filter(): string

Defined in: core/src/context/context.ts:175

CSS filter string applied to subsequent drawing operations (e.g. blur(4px)).

Returns

string

Set Signature

set filter(value): void

Defined in: core/src/context/context.ts:179

CSS filter string applied to subsequent drawing operations (e.g. blur(4px)).

Parameters

Parameter	Type
value	string

Returns

void

CSS filter string applied to subsequent drawing operations (e.g. blur(4px)).

Inherited from

Context3D.filter

---

fog

Get Signature

get fog(): Fog | null

Defined in: 3d/src/core/context.ts:212

Atmospheric haze blending distant geometry towards a colour, or null for none.

Both backends resolve it identically. Assign a new object to change it — mutating the existing one in place will not repaint.

Returns

Fog | null

Set Signature

set fog(value): void

Defined in: 3d/src/core/context.ts:216

Parameters

Parameter	Type
value	Fog | null

Returns

void

Inherited from

Context3D.fog

---

font

Get Signature

get font(): string

Defined in: core/src/context/context.ts:193

CSS font shorthand used when rendering text.

Returns

string

Set Signature

set font(value): void

Defined in: core/src/context/context.ts:197

CSS font shorthand used when rendering text.

Parameters

Parameter	Type
value	string

Returns

void

CSS font shorthand used when rendering text.

Inherited from

Context3D.font

---

fontKerning

Get Signature

get fontKerning(): FontKerning

Defined in: core/src/context/context.ts:202

Whether font kerning is applied when rendering text.

Returns

FontKerning

Set Signature

set fontKerning(value): void

Defined in: core/src/context/context.ts:206

Whether font kerning is applied when rendering text.

Parameters

Parameter	Type
value	FontKerning

Returns

void

Whether font kerning is applied when rendering text.

Inherited from

Context3D.fontKerning

---

globalCompositeOperation

Get Signature

get globalCompositeOperation(): unknown

Defined in: core/src/context/context.ts:220

Compositing operation controlling how new drawing is blended with existing content.

Returns

unknown

Set Signature

set globalCompositeOperation(value): void

Defined in: core/src/context/context.ts:224

Compositing operation controlling how new drawing is blended with existing content.

Parameters

Parameter	Type
value	unknown

Returns

void

Compositing operation controlling how new drawing is blended with existing content.

Inherited from

Context3D.globalCompositeOperation

---

lightDirection

Get Signature

get lightDirection(): Vector3

Defined in: 3d/src/core/context.ts:244

The direction of the default rig's directional light.

A convenience over reaching into lights for the single-light case. Replacing the rig through Context3DOptions.lights, or removing that light, leaves this inert — set the light's own direction instead.

Returns

Vector3

Set Signature

set lightDirection(value): void

Defined in: 3d/src/core/context.ts:248

Parameters

Parameter	Type
value	Vector3

Returns

void

Inherited from

Context3D.lightDirection

---

lightMode

Get Signature

get lightMode(): LightMode

Defined in: 3d/src/core/context.ts:261

Whether the default rig's directional light is fixed in world space or follows the camera.

Inert once the default rig is replaced, exactly as lightDirection is.

Returns

LightMode

Set Signature

set lightMode(value): void

Defined in: 3d/src/core/context.ts:265

Parameters

Parameter	Type
value	LightMode

Returns

void

Inherited from

Context3D.lightMode

---

lineCap

Get Signature

get lineCap(): LineCap

Defined in: core/src/context/context.ts:229

Cap style drawn at the endpoints of stroked lines.

Returns

LineCap

Set Signature

set lineCap(value): void

Defined in: core/src/context/context.ts:233

Cap style drawn at the endpoints of stroked lines.

Parameters

Parameter	Type
value	LineCap

Returns

void

Cap style drawn at the endpoints of stroked lines.

Inherited from

Context3D.lineCap

---

lineDash

Get Signature

get lineDash(): number[]

Defined in: core/src/context/context.ts:238

Dash pattern as alternating stroke and gap lengths; empty for a solid line.

Returns

number[]

Set Signature

set lineDash(value): void

Defined in: core/src/context/context.ts:242

Dash pattern as alternating stroke and gap lengths; empty for a solid line.

Parameters

Parameter	Type
value	number[]

Returns

void

Dash pattern as alternating stroke and gap lengths; empty for a solid line.

Inherited from

Context3D.lineDash

---

lineDashOffset

Get Signature

get lineDashOffset(): number

Defined in: core/src/context/context.ts:247

Distance into the line dash pattern at which dashing begins.

Returns

number

Set Signature

set lineDashOffset(value): void

Defined in: core/src/context/context.ts:251

Distance into the line dash pattern at which dashing begins.

Parameters

Parameter	Type
value	number

Returns

void

Distance into the line dash pattern at which dashing begins.

Inherited from

Context3D.lineDashOffset

---

lineJoin

Get Signature

get lineJoin(): LineJoin

Defined in: core/src/context/context.ts:256

Join style drawn where two stroked segments meet.

Returns

LineJoin

Set Signature

set lineJoin(value): void

Defined in: core/src/context/context.ts:260

Join style drawn where two stroked segments meet.

Parameters

Parameter	Type
value	LineJoin

Returns

void

Join style drawn where two stroked segments meet.

Inherited from

Context3D.lineJoin

---

lineWidth

Get Signature

get lineWidth(): number

Defined in: core/src/context/context.ts:265

Width, in pixels, of stroked lines.

Returns

number

Set Signature

set lineWidth(value): void

Defined in: core/src/context/context.ts:269

Width, in pixels, of stroked lines.

Parameters

Parameter	Type
value	number

Returns

void

Width, in pixels, of stroked lines.

Inherited from

Context3D.lineWidth

---

miterLimit

Get Signature

get miterLimit(): number

Defined in: core/src/context/context.ts:274

Miter length limit ratio applied to miter line joins.

Returns

number

Set Signature

set miterLimit(value): void

Defined in: core/src/context/context.ts:278

Miter length limit ratio applied to miter line joins.

Parameters

Parameter	Type
value	number

Returns

void

Miter length limit ratio applied to miter line joins.

Inherited from

Context3D.miterLimit

---

opacity

Get Signature

get opacity(): number

Defined in: core/src/context/context.ts:211

Global alpha applied to everything drawn, from 0 to 1. Assignment replaces the current alpha; element and group opacity instead composite multiplicatively through CONTEXT_OPERATIONS.

Returns

number

Set Signature

set opacity(value): void

Defined in: core/src/context/context.ts:215

Global alpha applied to everything drawn, from 0 to 1.

Parameters

Parameter	Type
value	number

Returns

void

Global alpha applied to everything drawn, from 0 to 1.

Inherited from

Context3D.opacity

---

projectionMode

Get Signature

get projectionMode(): "perspective" | "orthographic"

Defined in: 3d/src/core/context.ts:228

The projection currently in effect, which a resize preserves.

Returns

"perspective" | "orthographic"

Inherited from

Context3D.projectionMode

---

renderStrategy

Get Signature

get renderStrategy(): RenderStrategy

Defined in: 3d/src/core/context.ts:233

The active rendering strategy (cpu or gpu) for this context.

Returns

RenderStrategy

Inherited from

Context3D.renderStrategy

---

resolvedFog

Get Signature

get resolvedFog(): ResolvedFog | null

Defined in: 3d/src/core/context.ts:223

The resolved fog both backends shade against, or null when there is none.

Returns

ResolvedFog | null

Inherited from

Context3D.resolvedFog

---

rotation

Get Signature

get rotation(): Rotation

Defined in: core/src/context/context.ts:391

Rotation applied to the element's transform, as radians or a deg/rad string.

Returns

Rotation

Set Signature

set rotation(value): void

Defined in: core/src/context/context.ts:395

Rotation applied to the element's transform, as radians or a deg/rad string.

Parameters

Parameter	Type
value	Rotation

Returns

void

Rotation applied to the element's transform, as radians or a deg/rad string.

Inherited from

Context3D.rotation

---

shadowBlur

Get Signature

get shadowBlur(): number

Defined in: core/src/context/context.ts:283

Gaussian blur radius applied to drawn shadows.

Returns

number

Set Signature

set shadowBlur(value): void

Defined in: core/src/context/context.ts:287

Gaussian blur radius applied to drawn shadows.

Parameters

Parameter	Type
value	number

Returns

void

Gaussian blur radius applied to drawn shadows.

Inherited from

Context3D.shadowBlur

---

shadowColor

Get Signature

get shadowColor(): string

Defined in: core/src/context/context.ts:292

Color of drawn shadows.

Returns

string

Set Signature

set shadowColor(value): void

Defined in: core/src/context/context.ts:296

Color of drawn shadows.

Parameters

Parameter	Type
value	string

Returns

void

Color of drawn shadows.

Inherited from

Context3D.shadowColor

---

shadowOffsetX

Get Signature

get shadowOffsetX(): number

Defined in: core/src/context/context.ts:301

Horizontal offset, in pixels, of drawn shadows.

Returns

number

Set Signature

set shadowOffsetX(value): void

Defined in: core/src/context/context.ts:305

Horizontal offset, in pixels, of drawn shadows.

Parameters

Parameter	Type
value	number

Returns

void

Horizontal offset, in pixels, of drawn shadows.

Inherited from

Context3D.shadowOffsetX

---

shadowOffsetY

Get Signature

get shadowOffsetY(): number

Defined in: core/src/context/context.ts:310

Vertical offset, in pixels, of drawn shadows.

Returns

number

Set Signature

set shadowOffsetY(value): void

Defined in: core/src/context/context.ts:314

Vertical offset, in pixels, of drawn shadows.

Parameters

Parameter	Type
value	number

Returns

void

Vertical offset, in pixels, of drawn shadows.

Inherited from

Context3D.shadowOffsetY

---

stroke

Get Signature

get stroke(): string

Defined in: core/src/context/context.ts:319

Stroke style (CSS color, gradient, or pattern) used to paint outlines.

Returns

string

Set Signature

set stroke(value): void

Defined in: core/src/context/context.ts:323

Stroke style (CSS color, gradient, or pattern) used to paint outlines.

Parameters

Parameter	Type
value	string

Returns

void

Stroke style (CSS color, gradient, or pattern) used to paint outlines.

Inherited from

Context3D.stroke

---

supportsPathCaching

Get Signature

get supportsPathCaching(): boolean

Defined in: core/src/context/context.ts:733

Whether cached ContextPaths may be safely reused across render cycles. false by default; backends whose Context.createPath is free of per-frame side effects (e.g. canvas) override this to true. SVG, for example, returns false because createPath registers the path into a per-frame virtual DOM tree.

Returns

boolean

Inherited from

Context3D.supportsPathCaching

---

textAlign

Get Signature

get textAlign(): TextAlignment

Defined in: core/src/context/context.ts:328

Horizontal alignment of text relative to the drawing position.

Returns

TextAlignment

Set Signature

set textAlign(value): void

Defined in: core/src/context/context.ts:332

Horizontal alignment of text relative to the drawing position.

Parameters

Parameter	Type
value	TextAlignment

Returns

void

Horizontal alignment of text relative to the drawing position.

Inherited from

Context3D.textAlign

---

textBaseline

Get Signature

get textBaseline(): TextBaseline

Defined in: core/src/context/context.ts:337

Vertical baseline used when positioning text.

Returns

TextBaseline

Set Signature

set textBaseline(value): void

Defined in: core/src/context/context.ts:341

Vertical baseline used when positioning text.

Parameters

Parameter	Type
value	TextBaseline

Returns

void

Vertical baseline used when positioning text.

Inherited from

Context3D.textBaseline

---

transformOriginX

Get Signature

get transformOriginX(): TransformOrigin

Defined in: core/src/context/context.ts:400

Horizontal origin about which rotation and scaling are applied.

Returns

TransformOrigin

Set Signature

set transformOriginX(value): void

Defined in: core/src/context/context.ts:404

Horizontal origin about which rotation and scaling are applied.

Parameters

Parameter	Type
value	TransformOrigin

Returns

void

Horizontal origin about which rotation and scaling are applied.

Inherited from

Context3D.transformOriginX

---

transformOriginY

Get Signature

get transformOriginY(): TransformOrigin

Defined in: core/src/context/context.ts:409

Vertical origin about which rotation and scaling are applied.

Returns

TransformOrigin

Set Signature

set transformOriginY(value): void

Defined in: core/src/context/context.ts:413

Vertical origin about which rotation and scaling are applied.

Parameters

Parameter	Type
value	TransformOrigin

Returns

void

Vertical origin about which rotation and scaling are applied.

Inherited from

Context3D.transformOriginY

---

transformScaleX

Get Signature

get transformScaleX(): number

Defined in: core/src/context/context.ts:373

Horizontal scale factor applied to the element's transform.

Returns

number

Set Signature

set transformScaleX(value): void

Defined in: core/src/context/context.ts:377

Horizontal scale factor applied to the element's transform.

Parameters

Parameter	Type
value	number

Returns

void

Horizontal scale factor applied to the element's transform.

Inherited from

Context3D.transformScaleX

---

transformScaleY

Get Signature

get transformScaleY(): number

Defined in: core/src/context/context.ts:382

Vertical scale factor applied to the element's transform.

Returns

number

Set Signature

set transformScaleY(value): void

Defined in: core/src/context/context.ts:386

Vertical scale factor applied to the element's transform.

Parameters

Parameter	Type
value	number

Returns

void

Vertical scale factor applied to the element's transform.

Inherited from

Context3D.transformScaleY

---

translateX

Get Signature

get translateX(): number

Defined in: core/src/context/context.ts:355

Horizontal translation, in pixels, applied to the element's transform.

Returns

number

Set Signature

set translateX(value): void

Defined in: core/src/context/context.ts:359

Horizontal translation, in pixels, applied to the element's transform.

Parameters

Parameter	Type
value	number

Returns

void

Horizontal translation, in pixels, applied to the element's transform.

Inherited from

Context3D.translateX

---

translateY

Get Signature

get translateY(): number

Defined in: core/src/context/context.ts:364

Vertical translation, in pixels, applied to the element's transform.

Returns

number

Set Signature

set translateY(value): void

Defined in: core/src/context/context.ts:368

Vertical translation, in pixels, applied to the element's transform.

Parameters

Parameter	Type
value	number

Returns

void

Vertical translation, in pixels, applied to the element's transform.

Inherited from

Context3D.translateY

---

zIndex

Get Signature

get zIndex(): number

Defined in: core/src/context/context.ts:346

Stacking order used to sort elements during rendering; higher values draw on top.

Returns

number

Set Signature

set zIndex(value): void

Defined in: core/src/context/context.ts:350

Stacking order used to sort elements during rendering; higher values draw on top.

Parameters

Parameter	Type
value	number

Returns

void

Stacking order used to sort elements during rendering; higher values draw on top.

Inherited from

Context3D.zIndex

Methods

applyClip()

applyClip(): void

Defined in: webgpu/src/context.ts:283

No-op; the WebGPU pipeline rasterises submitted meshes only. Warns once.

Returns

void

Overrides

Context3D.applyClip

---

applyFill()

applyFill(): void

Defined in: webgpu/src/context.ts:273

No-op; the WebGPU pipeline rasterises submitted meshes only. Warns once.

Returns

void

Overrides

Context3D.applyFill

---

applyGroupPaint()

protected applyGroupPaint(group): void

Defined in: core/src/context/context.ts:603

Applies a group's own inherited paint (fill, stroke, opacity, font, line, shadow, text) to the context so descendants pick it up from the copied state. Transforms are applied separately, so they are skipped here.

Two boundary semantics are fixed here and every backend must honour them:

• Opacity composites multiplicatively. CONTEXT_OPERATIONS multiplies rather than assigns, so nested groups compound and a leaf's own alpha stacks under its ancestors' instead of replacing it.
• A group's gradient or pattern resolves against the group's own box. The group is the Context.currentRenderElement for the duration of the boundary, so a paint that bakes geometry at set time (canvas) and one that resolves it per leaf (SVG) agree on the same reference box: group.getBoundingBox(true), never a sibling's and never the surface.

Parameters

Parameter	Type
group	Element

Returns

void

Inherited from

Context3D.applyGroupPaint

---

applyStroke()

applyStroke(): void

Defined in: webgpu/src/context.ts:278

No-op; the WebGPU pipeline rasterises submitted meshes only. Warns once.

Returns

void

Overrides

Context3D.applyStroke

---

batch()

batch<TResult>(body): TResult

Defined in: core/src/context/context.ts:534

Clears the rendering surface and brackets the callback in markRenderStart/markRenderEnd, returning the callback's result. On exit the state stack is unwound to the depth captured on entry, giving the scene root the same guarantee Context.popGroup gives a group: a root-level clip: true shape deliberately skips its own restore() so the clip persists to later siblings, and with no enclosing group to absorb it that save would otherwise leak one state per frame, unbounded.

The surface is only cleared at render depth 0: a pass entered while an outer one is open continues it (matching Context.markRenderStart) rather than wiping what it drew.

Type Parameters

Type Parameter	Default type
TResult	void

Parameters

Parameter	Type
body	() => TResult

Returns

TResult

Inherited from

Context3D.batch

---

captureFaceState()

captureFaceState(transform): ProjectedFaceState3D

Defined in: 3d/src/core/context.ts:585

Snapshots the drawing state a face must be painted with.

The CPU painter buffers faces and flushes them at the end of the frame, by which point every element restore() and popGroup() has already unwound the state the face was projected under. Capturing it here is what lets CanvasContext3D re-apply it at draw time.

Parameters

Parameter	Type	Description
transform	Matrix | null	The element's composed 2D world transform, or null for the identity.

Returns

ProjectedFaceState3D

The resolved drawing state to store on each of the element's faces.

Inherited from

Context3D.captureFaceState

---

clear()

clear(): void

Defined in: webgpu/src/context.ts:245

No-op; the surface is cleared by the render pass via loadOp: 'clear'.

Returns

void

Overrides

Context3D.clear

---

createPath()

createPath(id?): CanvasPath

Defined in: webgpu/src/context.ts:268

Creates a CanvasPath used for CPU-side hit testing.

Parameters

Parameter	Type
id?	string

Returns

CanvasPath

Overrides

Context3D.createPath

---

createText()

createText(options): ContextText

Defined in: webgpu/src/context.ts:293

Creates a text element that this backend cannot paint. Warns once.

Parameters

Parameter	Type
options	TextOptions

Returns

ContextText

Overrides

Context3D.createText

---

destroy()

destroy(): void

Defined in: webgpu/src/context.ts:422

Destroys the WebGPU context, unconfigures the swap chain, and releases every GPU resource.

Returns

void

Overrides

Context3D.destroy

---

disableInteraction()

disableInteraction(): void

Defined in: dom/src/context.ts:388

Disables DOM interaction events, unwinding any hover with a final mouseleave per element.

Returns

void

Inherited from

Context3D.disableInteraction

---

dispose()

protected dispose(key?): void

Defined in: core/src/core/disposer.ts:25

Disposes all resources under the given key, or all resources if no key is provided.

Parameters

Parameter	Type
key?	PropertyKey

Returns

void

Inherited from

Context3D.dispose

---

drawImage()

drawImage(): void

Defined in: webgpu/src/context.ts:288

No-op; the WebGPU pipeline rasterises submitted meshes only. Warns once.

Returns

void

Overrides

Context3D.drawImage

---

emit()

Call Signature

emit<TEvent>(event): TEvent

Defined in: core/src/core/event-bus.ts:185

Emits an event, invoking all matching handlers and bubbling to the parent if applicable. Handlers for the event's own type run first, then EVENT_WILDCARD subscriptions.

Type Parameters

Type Parameter	Default type
TEvent extends Event<undefined>	Event<undefined>

Parameters

Parameter	Type
event	TEvent

Returns

TEvent

Inherited from

Context3D.emit

Call Signature

emit<TEvent>(type, data): Event<ContextEventMap[TEvent]>

Defined in: core/src/core/event-bus.ts:186

Emits an event, invoking all matching handlers and bubbling to the parent if applicable. Handlers for the event's own type run first, then EVENT_WILDCARD subscriptions.

Type Parameters

Type Parameter
TEvent extends keyof ContextEventMap

Parameters

Parameter	Type
type	TEvent
data	ContextEventMap[TEvent]

Returns

Event<ContextEventMap[TEvent]>

Inherited from

Context3D.emit

---

enableInteraction()

enableInteraction(): void

Defined in: dom/src/context.ts:344

Enables DOM interaction events (mouse enter, leave, move, down, up, click, drag) with element hit testing.

Returns

void

Inherited from

Context3D.enableInteraction

---

export()

export(): ContextExport

Defined in: 3d/src/core/context.ts:627

Exports the rendered canvas. Shared by CPU and GPU 3D contexts; call after a frame has rendered (GPU present textures are transient, so createCanvasExport snapshots immediately).

Returns

ContextExport

Inherited from

Context3D.export

---

getDefaultState()

protected getDefaultState(): BaseState

Defined in: core/src/context/context.ts:446

Returns

BaseState

Inherited from

Context3D.getDefaultState

---

getLightDirectionForRender()

getLightDirectionForRender(): Vector3

Defined in: 3d/src/core/context.ts:414

Returns the effective light direction for the current render, accounting for the light mode.

Both consumers dot this against a world-space normal, so 'world' is the identity and 'camera' reads lightDirection as camera-relative and carries it into world space through the inverse of the view rotation.

Returns

Vector3

Inherited from

Context3D.getLightDirectionForRender

---

has()

has(type): boolean

Defined in: core/src/core/event-bus.ts:123

Returns whether there are any listeners registered for the given event type. Wildcard subscriptions (EVENT_WILDCARD) are not listeners for any concrete type, so a bus observed only through the wildcard still reports false here.

Parameters

Parameter	Type
type	keyof ContextEventMap

Returns

boolean

Inherited from

Context3D.has

---

hitTest()

protected hitTest(events, x, y): RenderElement[]

Defined in: 3d/src/core/context.ts:480

Tests which rendered elements the pointer is over, ranking 3D shapes by how near their geometry is rather than by paint order.

The base class ranks by paint order, which for 2D is exactly right — the last thing painted is the thing on top. A 3D scene does not paint in element order: CanvasContext3D.flushFaces buffers the faces of every shape and depth-sorts them globally, so what you see is ordered by depth while what the base class hits is ordered by the position an element happened to take in the render stream. For an assembly of interleaved parts those two orders routinely disagree.

2D elements keep their paint order and their positions in the list, so a 2D element painted over a 3D one still wins: it flushed the face buffer, and paint order genuinely does decide there.

Parameters

Parameter	Type
events	string[]
x	number
y	number

Returns

RenderElement[]

Inherited from

Context3D.hitTest

---

init()

protected init(): void

Defined in: dom/src/context.ts:85

Returns

void

Inherited from

Context3D.init

---

invalidateTrackedElements()

invalidateTrackedElements(event?): void

Defined in: core/src/context/context.ts:496

Clears the cached list of tracked elements for interaction, forcing a rebuild on the next hit test.

Parameters

Parameter	Type
event?	string

Returns

void

Inherited from

Context3D.invalidateTrackedElements

---

isPointInPath()

isPointInPath(path, x, y, fillRule?): boolean

Defined in: webgpu/src/context.ts:300

Tests whether the logical-space point (x, y) lies inside the given path's fill, using an offscreen 2D canvas.

Parameters

Parameter	Type
path	ContextPath
x	number
y	number
fillRule?	FillRule

Returns

boolean

Overrides

Context3D.isPointInPath

---

isPointInStroke()

isPointInStroke(path, x, y): boolean

Defined in: webgpu/src/context.ts:312

Tests whether the logical-space point (x, y) lies on the given path's stroke, using an offscreen 2D canvas.

Parameters

Parameter	Type
path	ContextPath
x	number
y	number

Returns

boolean

Overrides

Context3D.isPointInStroke

---

layer()

layer<TResult>(body): TResult

Defined in: core/src/context/context.ts:472

Executes a callback within a save/restore pair, returning the callback's result.

Type Parameters

Type Parameter	Default type
TResult	void

Parameters

Parameter	Type
body	() => TResult

Returns

TResult

Inherited from

Context3D.layer

---

markRenderEnd()

markRenderEnd(): void

Defined in: webgpu/src/context.ts:234

Ends the render pass and, at the outermost depth, executes the queued GPU draw commands.

Returns

void

Overrides

Context3D.markRenderEnd

---

markRenderStart()

markRenderStart(): void

Defined in: webgpu/src/context.ts:225

Begins a render pass, resetting per-frame geometry accumulation at the outermost depth.

Returns

void

Overrides

Context3D.markRenderStart

---

measureText()

measureText(text, font?): TextMetrics

Defined in: webgpu/src/context.ts:417

Measures text using the offscreen 2D canvas context.

Parameters

Parameter	Type
text	string
font?	string

Returns

TextMetrics

Overrides

Context3D.measureText

---

off()

off<TEvent>(type, handler): void

Defined in: core/src/core/event-bus.ts:144

Removes a previously registered handler for the given event type.

Type Parameters

Type Parameter
TEvent extends keyof ContextEventMap

Parameters

Parameter	Type
type	TEvent
handler	EventHandler<ContextEventMap[TEvent]>

Returns

void

Inherited from

Context3D.off

---

on()

on<TEvent>(type, handler, options?): Disposable

Defined in: core/src/core/event-bus.ts:131

Subscribes a handler to the given event type and returns a disposable for cleanup. Pass EVENT_WILDCARD to receive every event emitted on the bus regardless of type.

Type Parameters

Type Parameter
TEvent extends keyof ContextEventMap

Parameters

Parameter	Type
type	TEvent
handler	EventHandler<ContextEventMap[TEvent]>
options?	EventSubscriptionOptions

Returns

Disposable

Inherited from

Context3D.on

---

once()

once<TEvent>(type, handler, options?): Disposable

Defined in: core/src/core/event-bus.ts:159

Subscribes a handler that is automatically removed after it fires once.

Type Parameters

Type Parameter
TEvent extends keyof ContextEventMap

Parameters

Parameter	Type
type	TEvent
handler	EventHandler<ContextEventMap[TEvent]>
options?	EventSubscriptionOptions

Returns

Disposable

Inherited from

Context3D.once

---

popGroup()

popGroup(): void

Defined in: core/src/context/context.ts:620

Closes the most recently opened group boundary, unwinding the state stack back to the depth captured at Context.pushGroup and restoring the render element that was current before it. This absorbs any dangling save() left by a group-scoped clip (which deliberately skips its own restore so the clip persists to later siblings), confining the clip to the group rather than leaking it to the scene.

Returns

void

Inherited from

Context3D.popGroup

---

project()

project(point): ProjectedPoint

Defined in: 3d/src/core/context.ts:571

Projects a 3D world-space point to 2D logical coordinates — CSS pixels relative to the context's top-left — plus a depth for z-ordering.

Parameters

Parameter	Type
point	Vector3

Returns

ProjectedPoint

Inherited from

Context3D.project

---

pushGroup()

pushGroup(group): void

Defined in: core/src/context/context.ts:573

Opens a group boundary: saves the current drawing state and applies the group's own transform and inherited paint (see applyGroupPaint for the opacity and gradient-bounds semantics every backend must honour), so that descendant elements render within the group's coordinate system, inherit its paint through the copied state (the render-tree cascade), and composite under its opacity. Any group-scoped clip is confined to the group. Every Context.pushGroup must be balanced by a matching Context.popGroup. Backends that render hierarchy structurally (such as SVG) override this to nest descendants under a group node.

Parameters

Parameter	Type	Description
group	Element	The group element whose boundary is being entered.

Returns

void

Inherited from

Context3D.pushGroup

---

raycast()

raycast(x, y): Ray | null

Defined in: 3d/src/core/context.ts:429

Builds the world-space ray through a point on the surface.

Parameters

Parameter	Type	Description
x	number	Screen-space x, in logical CSS pixels relative to the context's top-left.
y	number	Screen-space y.

Returns

Ray | null

The ray, or null when the view-projection matrix is singular.

Inherited from

Context3D.raycast

---

raycastAll()

raycastAll(root, x, y, options?): Intersection3D[]

Defined in: 3d/src/core/context.ts:446

Casts a ray through a point on the surface and returns every 3D shape it meets.

Reports the full intersection — point, face, normal and texture coordinate — where hitTest answers only which elements the pointer is over. It also ignores pointerEvents and whether a shape has any handlers, so it sees the whole scene.

Parameters

Parameter	Type	Description
root	Element	The element to search, typically the scene.
x	number	Screen-space x, in logical CSS pixels relative to the context's top-left.
y	number	Screen-space y.
options?	Raycast3DOptions	Whether to accept back-facing hits.

Returns

Intersection3D[]

The intersections, nearest first.

Inherited from

Context3D.raycastAll

---

requestRender()

requestRender(): void

Defined in: core/src/context/context.ts:505

Requests that the bound scene repaint on the next frame, for context-level changes (e.g. a 3D camera move) that mutate no element and so would otherwise be skipped by the renderer's dirty check.

Returns

void

Inherited from

Context3D.requestRender

---

rescale()

protected rescale(width, height): void

Defined in: webgpu/src/context.ts:136

Parameters

Parameter	Type
width	number
height	number

Returns

void

Overrides

Context3D.rescale

---

reset()

reset(): void

Defined in: core/src/context/context.ts:488

Resets the context to its initial state: drops the saved-state stack, restores the default drawing state, and closes any open group boundary.

Returns

void

Inherited from

Context3D.reset

---

resolveLights()

resolveLights(): ResolvedLight[]

Defined in: 3d/src/core/context.ts:518

Resolves the scene's lights into the flat numeric form both backends shade against.

Cached until a light changes or the camera moves, because the CPU painter asks for it once per shape and camera-space lights depend on the view matrix.

Returns

ResolvedLight[]

The enabled lights, capped at MAX_LIGHTS.

Inherited from

Context3D.resolveLights

---

restore()

restore(): void

Defined in: core/src/context/context.ts:462

Restores the most recently saved state from the stack.

Returns

void

Inherited from

Context3D.restore

---

retain()

protected retain(value, key?): void

Defined in: core/src/core/disposer.ts:14

Registers a disposable resource under an optional key for later cleanup.

Parameters

Parameter	Type	Default value
value	Disposable	undefined
key	PropertyKey	Disposer.defaultKey

Returns

void

Inherited from

Context3D.retain

---

rotate()

rotate(angle): void

Defined in: core/src/context/context.ts:634

Applies a rotation transformation.

Parameters

Parameter	Type
angle	number

Returns

void

Inherited from

Context3D.rotate

---

save()

save(): void

Defined in: core/src/context/context.ts:455

Pushes the current state onto the stack and continues with a copy of it, so nested scopes inherit the enclosing drawing state (matching native canvas). This is what lets a group's paint, applied once at its boundary via Context.pushGroup, cascade to descendants.

Returns

void

Inherited from

Context3D.save

---

scale()

scale(x, y): void

Defined in: core/src/context/context.ts:639

Applies a scale transformation.

Parameters

Parameter	Type
x	number
y	number

Returns

void

Inherited from

Context3D.scale

---

setCamera()

setCamera(eye, target, up): void

Defined in: 3d/src/core/context.ts:367

Sets the view matrix from an eye position, look-at target, and up direction.

Parameters

Parameter	Type
eye	Vector3
target	Vector3
up	Vector3

Returns

void

Inherited from

Context3D.setCamera

---

setOrthographic()

setOrthographic(left, right, bottom, top, near, far): void

Defined in: 3d/src/core/context.ts:393

Sets an orthographic projection with explicit frustum bounds.

The bounds are retained and replayed verbatim whenever the projection is rebuilt, so a resize keeps the projection orthographic. It does not re-fit the frustum to the new aspect ratio — the caller owns that, and Camera recomputes it on its next flush.

Parameters

Parameter	Type
left	number
right	number
bottom	number
top	number
near	number
far	number

Returns

void

Inherited from

Context3D.setOrthographic

---

setPerspective()

setPerspective(fov, near, far): void

Defined in: 3d/src/core/context.ts:377

Updates the perspective projection with the given field of view, near, and far planes, replacing any orthographic projection.

Parameters

Parameter	Type
fov	number
near	number
far	number

Returns

void

Inherited from

Context3D.setPerspective

---

setTransform()

setTransform(a, b, c, d, e, f): void

Defined in: core/src/context/context.ts:663

Replaces the current transformation matrix with [a, b, c, d, e, f].

The matrix is in logical space, so the identity (1, 0, 0, 1, 0, 0) restores the context's own baseline rather than the backend's — on canvas that baseline is the device pixel ratio, which the backend composes back underneath. Translations are therefore in CSS pixels.

Parameters

Parameter	Type	Description
a	number	Horizontal scaling.
b	number	Vertical skewing.
c	number	Horizontal skewing.
d	number	Vertical scaling.
e	number	Horizontal translation, in logical pixels.
f	number	Vertical translation, in logical pixels.

Returns

void

Inherited from

Context3D.setTransform

---

submitMesh()

submitMesh(submission): void

Defined in: webgpu/src/context.ts:216

Submits a mesh for GPU rendering this frame.

Parameters

Parameter	Type
submission	MeshSubmission

Returns

void

Overrides

Context3D.submitMesh

---

toLogicalPoint()

toLogicalPoint(x, y): [number, number]

Defined in: core/src/context/context.ts:697

Maps a point from surface space — this backend's own drawing coordinates, as produced by Context.scaleX and Context.scaleY — into the logical space elements are authored in and every public coordinate is expressed in.

Backends disagree on what surface space is: canvas maps logical coordinates onto device pixels while SVG leaves them identity, so a backend has to invert what this context actually does rather than assume the device pixel ratio.

This is a seam for backend implementors, not for consumers — nothing crossing the public API is in surface space, so a caller holding a pointer coordinate already has a logical one.

Parameters

Parameter	Type	Description
x	number	X coordinate in surface space.
y	number	Y coordinate in surface space.

Returns

[number, number]

The [x, y] pair in logical space.

Inherited from

Context3D.toLogicalPoint

---

toSurfacePoint()

toSurfacePoint(x, y): [number, number]

Defined in: core/src/context/context.ts:710

Maps a point from the logical space elements are authored in into this backend's own drawing coordinates. The inverse of Context.toLogicalPoint, and likewise a seam for backend implementors rather than consumers.

Parameters

Parameter	Type	Description
x	number	X coordinate in logical space.
y	number	Y coordinate in logical space.

Returns

[number, number]

The [x, y] pair in surface space.

Inherited from

Context3D.toSurfacePoint

---

transform()

transform(a, b, c, d, e, f): void

Defined in: core/src/context/context.ts:677

Multiplies the current transformation matrix by [a, b, c, d, e, f].

Parameters

Parameter	Type	Description
a	number	Horizontal scaling.
b	number	Vertical skewing.
c	number	Horizontal skewing.
d	number	Vertical scaling.
e	number	Horizontal translation.
f	number	Vertical translation.

Returns

void

Inherited from

Context3D.transform

---

translate()

translate(x, y): void

Defined in: core/src/context/context.ts:644

Applies a translation transformation.

Parameters

Parameter	Type
x	number
y	number

Returns

void

Inherited from

Context3D.translate

---

updateProjectionMatrix()

protected updateProjectionMatrix(): void

Defined in: 3d/src/core/context.ts:346

Returns

void

Inherited from

Context3D.updateProjectionMatrix

---

updateViewProjectionMatrix()

protected updateViewProjectionMatrix(): void

Defined in: 3d/src/core/context.ts:342

Returns

void

Inherited from

Context3D.updateViewProjectionMatrix