# NPM npm install ogl react-ogl # Yarn yarn add ogl react-ogl # PNPM pnpm add ogl react-ogl
react-ogl itself is super minimal, but you can use the familiar @react-three/fiber API with some helpers targeted for different platforms:
This example uses create-react-app
for the sake of simplicity, but you can use your own environment or create a codesandbox.
# Create app npx create-react-app my-app cd my-app # Install dependencies npm install ogl react-ogl # Start npm run start
The following creates a re-usable component that has its own state, reacts to events and participates a shared render-loop.
</details>import * as React from 'react' import { useFrame, Canvas } from 'react-ogl' import { createRoot } from 'react-dom/client' function Box(props) { // This reference will give us direct access to the mesh const mesh = React.useRef() // Set up state for the hovered and active state const [hovered, setHover] = React.useState(false) const [active, setActive] = React.useState(false) // Subscribe this component to the render-loop, rotate the mesh every frame useFrame(() => (mesh.current.rotation.x += 0.01)) // Return view, these are regular OGL elements expressed in JSX return ( <mesh {...props} ref={mesh} scale={active ? 1.5 : 1} onClick={() => setActive((value) => !value)} onPointerOver={() => setHover(true)} onPointerOut={() => setHover(false)} > <box /> <program vertex={` attribute vec3 position; attribute vec3 normal; uniform mat4 modelViewMatrix; uniform mat4 projectionMatrix; uniform mat3 normalMatrix; varying vec3 vNormal; void main() { vNormal = normalize(normalMatrix * normal); gl_Position = projectionMatrix * modelViewMatrix * vec4(position, 1.0); } `} fragment={` precision highp float; uniform vec3 uColor; varying vec3 vNormal; void main() { vec3 normal = normalize(vNormal); float lighting = dot(normal, normalize(vec3(10))); gl_FragColor.rgb = uColor + lighting * 0.1; gl_FragColor.a = 1.0; } `} uniforms={{ uColor: hovered ? 'hotpink' : 'orange' }} /> </mesh> ) } createRoot(document.getElementById('root')).render( <Canvas camera={{ position: [0, 0, 8] }}> <Box position={[-1.2, 0, 0]} /> <Box position={[1.2, 0, 0]} /> </Canvas>, )
This example uses expo-cli
but you can create a bare app with react-native
CLI as well.
# Create app and cd into it npx expo init my-app # or npx react-native init my-app cd my-app # Automatically install & link expo modules npx install-expo-modules@latest expo install expo-gl # Install NPM dependencies npm install ogl react-ogl # Start npm run start
We'll also need to configure metro.config.js
to look for the mjs file extension that OGL uses.
module.exports = { resolver: { resolverMainFields: ['browser', 'exports', 'main'], // https://github.com/facebook/metro/issues/670 sourceExts: ['json', 'js', 'jsx', 'ts', 'tsx', 'cjs', 'mjs'], assetExts: ['glb', 'gltf', 'png', 'jpg'], }, }
Inside of our app, you can use the same API as web while running on native OpenGL ES — no webview needed.
</details>import * as React from 'react' import { useFrame, Canvas } from 'react-ogl' function Box(props) { // This reference will give us direct access to the mesh const mesh = React.useRef() // Set up state for the hovered and active state const [hovered, setHover] = React.useState(false) const [active, setActive] = React.useState(false) // Subscribe this component to the render-loop, rotate the mesh every frame useFrame(() => (mesh.current.rotation.x += 0.01)) // Return view, these are regular OGL elements expressed in JSX return ( <mesh {...props} ref={mesh} scale={active ? 1.5 : 1} onClick={() => setActive((value) => !value)} onPointerOver={() => setHover(true)} onPointerOut={() => setHover(false)} > <box /> <program vertex={` attribute vec3 position; attribute vec3 normal; uniform mat4 modelViewMatrix; uniform mat4 projectionMatrix; uniform mat3 normalMatrix; varying vec3 vNormal; void main() { vNormal = normalize(normalMatrix * normal); gl_Position = projectionMatrix * modelViewMatrix * vec4(position, 1.0); } `} fragment={` precision highp float; uniform vec3 uColor; varying vec3 vNormal; void main() { vec3 normal = normalize(vNormal); float lighting = dot(normal, normalize(vec3(10))); gl_FragColor.rgb = uColor + lighting * 0.1; gl_FragColor.a = 1.0; } `} uniforms={{ uColor: hovered ? 'hotpink' : 'orange' }} /> </mesh> ) } export default () => ( <Canvas camera={{ position: [0, 0, 8] }}> <Box position={[-1.2, 0, 0]} /> <Box position={[1.2, 0, 0]} /> </Canvas> )
react-ogl provides an x-platform <Canvas />
component for web and native that serves as the entrypoint for your OGL scenes. It is a real DOM canvas or native view that accepts OGL elements as children (see creating elements).
In addition to its platform props, <Canvas />
accepts a set of RenderProps
to configure react-ogl and its rendering behavior.
<Canvas // Configures the react rendering mode. Defaults to `blocking` mode={"legacy" | "blocking" | "concurrent"} // Creates, sets, or configures the default renderer. // Accepts a callback, an external renderer, or renderer constructor params/properties. // Defaults to `new OGL.Renderer({ alpha: true, antialias: true, powerPreference: 'high-performance' }) renderer={(canvas: HTMLCanvasElement) => new Renderer(canvas) | renderer | { ...params, ...props }} // Sets the renderer pixel ratio from a clamped range or value. Default is `[1, 2]` dpr={[min, max] | value} // Sets or configures the default camera. // Accepts an external camera, or camera constructor params/properties. // Defaults to `new OGL.Camera(gl, { fov: 75, near: 1, far: 1000 })` with position-z `5` camera={camera | { ...params, ...props }} // Enables orthographic projection when using OGL's built-in camera. Default is `false` orthographic={true | false} // Defaults to `always` frameloop={'always' | 'never'} // An optional callback invoked after canvas creation and before commit. onCreated={(state: RootState) => void} // Optionally configures custom events. Defaults to built-in events exported as `events` events={EventManager | undefined} > {/* Accepts OGL elements as children */} <transform /> </Canvas> // e.g. <Canvas renderer={{ alpha: true }} camera={{ fov: 45, position: [0, 1.3, 3] }} onCreated={(state) => void state.gl.clearColor(1, 1, 1, 0)} > <transform /> </Canvas>
A react 18 style createRoot
API creates an imperative Root
with the same options as <Canvas />
, but you're responsible for updating it and configuring things like events (see events). This root attaches to an HTMLCanvasElement
and renders OGL elements into a scene. Useful for creating an entrypoint with react-ogl and for headless contexts like a server or testing (see testing).
import { createRoot, events } from 'react-ogl' const canvas = document.querySelector('canvas') const root = createRoot(canvas, { events }) root.render( <mesh> <box /> <normalProgram /> </mesh>, ) root.unmount()
createRoot
can also be used to create a custom <Canvas />
. The following constructs a custom canvas that renders its children into react-ogl.
import * as React from 'react' import { createRoot, events } from 'react-ogl' function CustomCanvas({ children }) { // Init root from canvas const [canvas, setCanvas] = React.useState() const root = React.useMemo(() => canvas && createRoot(canvas, { events }), [canvas]) // Render children as a render-effect root?.render(children) // Cleanup on unmount React.useEffect(() => () => root?.unmount(), [root]) // Use callback-style ref to access canvas in render return <canvas ref={setCanvas} /> }
react-ogl renders React components into an OGL scene-graph, and can be used on top of other renderers like react-dom and react-native that render for web and native, respectively. react-ogl components are defined by primitives or lower-case elements native to the OGL namespace (for custom elements, see extend).
function Component(props) { return ( <mesh {...props}> <box /> <normalProgram /> </mesh> ) } ;<transform> <Component position={[1, 2, 3]} /> </transform>
These elements are not exported or implemented internally, but merely expressed as JSX — <mesh />
becomes new OGL.Mesh()
. This happens dynamically; there's no wrapper involved.
react-ogl elements can be modified with JSX attributes or props. These are native to their underlying OGL objects.
<transform // Set non-atomic properties with literals // transform.visible = false visible={false} // Copy atomic properties with a stable reference (e.g. useMemo) // transform.rotation.copy(rotation) rotation={rotation} // Set atomic properties with declarative array syntax // transform.position.set(1, 2, 3) position={[1, 2, 3]} // Set scalars with shorthand for vector properties // transform.scale.set(1, 1, 1) scale={1} // Set CSS names or hex values as shorthand for color properties // transform.color.set('red') color="red" // Set sub properties with prop piercing or dash-case // transform.rotation.x = Math.PI / 2 rotation-x={Math.PI / 2} />
args
An array of constructor arguments (args
) can be passed to instantiate elements' underlying OGL objects. Changing args
will reconstruct the object and update any associated refs.
// new OGL.Text({ font, text: 'Text' }) <text args={[{ font, text: 'Text' }]} />
Built-in elements that require a gl
context such as <mesh />
, <geometry />
, or <program />
are marked as effectful (see extend) and do not require an OGLRenderingContext
to be passed via args
. They can be constructed mutably and manipulated via props:
<mesh> <box /> <normalProgram /> </mesh>
<geometry />
and <program />
also accept attributes and shader sources as props, which are passed to their respective constructors. This does not affect other properties like drawRange
or uniforms
.
<mesh> <geometry position={{ size: 3, data: new Float32Array([-0.5, 0.5, 0, -0.5, -0.5, 0, 0.5, 0.5, 0, 0.5, -0.5, 0]) }} uv={{ size: 2, data: new Float32Array([0, 1, 1, 1, 0, 0, 1, 0]) }} index={{ data: new Uint16Array([0, 1, 2, 1, 3, 2]) }} /> {/* prettier-ignore */} <program vertex={/* glsl */ `...`} fragment={/* glsl */ `...`} uniforms={{ uniform: value }} /> </mesh>
attach
Some elements do not follow the traditional scene-graph and need to be added by other means. For this, the attach
prop can describe where an element is added via a property or a callback to add & remove the element.
// Attaches into parent.property, parent.sub.property, and parent.array[0] <parent> <element attach="property" /> <element attach="sub-property" /> <element attach="array-0" /> </parent> // Attaches via parent#setProperty and parent#removeProperty <parent> <element attach={(parent, self) => { parent.setProperty(self) return () => parent.removeProperty(self) }} // lambda version
AI小说写作助手,一站式润色、改写、扩写
蛙蛙写作—国内先进的AI写作平台,涵盖小说、学术、社交媒体等多场景。提供续写、改写、润色等功能,助力创作者高效优化写作流程。界面简洁,功能全面,适合各类写作者提升内容品质和工作效率。
字节跳动发布的AI编程神器IDE
Trae是一种自适应的集成开发环境(IDE),通过自动化和多元协作改变开发流程。利用Trae,团队能够更快速、精确地编写和部署代码,从而提高编程效率和项目交付速度。Trae具备上下文感知和代码自动完成功能,是提升开发效率的理想工具。
全能AI智能助手,随时解答生活与工作的多样问题
问小白,由元石科技研发的AI智能助手,快速准确地解答各种生活和工作问题,包括但不限于搜索、规划和社交互动,帮助用户在日常生活中提高效率,轻松管理个人事务。
实时语音翻译/同声传译工具
Transly是一个多场景的AI大语言模型驱动的同声传译、专业翻译助手,它拥有超精准的音频识别翻译能力,几乎零延迟的使用体验和支持多国语言可以让你带它走遍全球,无论你是留学生、商务人士、韩剧美剧爱好者,还是出国游玩、多国会议、跨国追星等等,都可以满足你所有需要同传的场景需求,线上线下通用,扫除语言障碍,让全世界的语言交流不再有国界。
一键生成PPT和Word,让学习生活更轻松
讯飞智文是一个利用 AI 技术的项目,能够帮助用户生成 PPT 以及各类文档。无论是商业领域的市场分析报告、年度目标制定,还是学生群体的职业生涯规划、实习避坑指南,亦或是活动策划、旅游攻略等内容,它都能提供支持,帮助用户精准表达,轻松呈现各种信息。
深度推理能力全新升级,全面对标OpenAI o1
科大讯飞的星火大模型,支持语言理解、知识问答和文本创作等多功能,适用于多种文件和业务场景,提升办公和日常生活的效率。讯飞星火是一个提供丰富智能服务的平台,涵盖科技资讯、图像创作、写作辅助、编程解答、科研文献解读等功能,能为不同需求的用户提供便捷高效的帮助,助力用户轻松获取信息、解决问题,满足多样化使用场景。
一种基于大语言模型的高效单流解耦语音令牌文本到语音合成模型
Spark-TTS 是一个基于 PyTorch 的开源文本到语音合成项目,由多个知名机构联合参与。该项目提供了高效的 LLM(大语言模型)驱动的语音合成方案,支持语音克隆和语音创建功能,可通过命令行界面(CLI)和 Web UI 两种方式使用。用户可以根据需求调整语音的性别、音高、速度等参数,生成高质量的语音。该项目适用于多种场景,如有声读物制作、智能语音助手开发等。
AI助力,做PPT更简单!
咔片是一款轻量化在线演示设计工具,借助 AI 技术,实现从内容生成到智能设计的一站式 PPT 制作服务。支持多种文档格式导入生成 PPT,提供海量模板、智能美化、素材替换等功能,适用于销售、教师、学生等各类人群,能高效制作出高品质 PPT,满足不同场景演示需求。
选题、配图、成文,一站式创作,让内容运营更高效
讯飞绘文,一个AI集成平台,支持写作、选题、配图、排版和发布。高效生成适用于各类媒体的定制内容,加速品牌传播,提升内容营销效果。
专业的AI公文写作平台,公文写作神器
AI 材料星,专业的 AI 公文写作辅助平台,为体制内工作人员提供高效的公文写作解决方案。拥有海量公文文库、9 大核心 AI 功能,支持 30 + 文稿类型生成,助力快速完成领导讲话、工作总结、述职报告等材料,提升办公效率,是体制打工人的得力写作神器。