Node canvas is a Cairo backed Canvas implementation for NodeJS.
Latest commit fd17135 Oct 16, 2018


Build Status NPM version

node-canvas is a Cairo-backed Canvas implementation for Node.js.


$ npm install canvas

By default, binaries for macOS, Linux and Windows will be downloaded. If you want to build from source, use npm install --build-from-source and see the Compiling section below.

The minimum version of Node.js required is 6.0.0.


If you don't have a supported OS or processor architecture, or you use --build-from-source, the module will be compiled on your system. This requires several dependencies, including Cairo and Pango.

For detailed installation information, see the wiki. One-line installation instructions for common OSes are below. Note that libgif/giflib, librsvg and libjpeg are optional and only required if you need GIF, SVG and JPEG support, respectively.

OS Command
OS X Using Homebrew:
brew install pkg-config cairo pango libpng jpeg giflib librsvg
Ubuntu sudo apt-get install build-essential libcairo2-dev libpango1.0-dev libjpeg-dev libgif-dev librsvg2-dev
Fedora sudo yum install gcc-c++ cairo-devel pango-devel libjpeg-turbo-devel giflib-devel
Solaris pkgin install cairo pango pkg-config xproto renderproto kbproto xextproto
OpenBSD doas pkg_add cairo pango png jpeg giflib
Windows See the wiki
Others See the wiki

Mac OS X v10.11+: If you have recently updated to Mac OS X v10.11+ and are experiencing trouble when compiling, run the following command: xcode-select --install. Read more about the problem on Stack Overflow.

Quick Example

const { createCanvas, loadImage } = require('canvas')
const canvas = createCanvas(200, 200)
const ctx = canvas.getContext('2d')

// Write "Awesome!"
ctx.font = '30px Impact'
ctx.fillText('Awesome!', 50, 100)

// Draw line under text
var text = ctx.measureText('Awesome!')
ctx.strokeStyle = 'rgba(0,0,0,0.5)'
ctx.lineTo(50, 102)
ctx.lineTo(50 + text.width, 102)

// Draw cat with lime helmet
loadImage('examples/images/lime-cat.jpg').then((image) => {
  ctx.drawImage(image, 50, 0, 70, 70)

  console.log('<img src="' + canvas.toDataURL() + '" />')

Upgrading from 2.x

See the changelog for a guide to upgrading from 1.x to 2.x.

For version 1.x documentation, see the v1.x branch.


This project is an implementation of the Web Canvas API and implements that API as closely as possible. For API documentation, please visit Mozilla Web Canvas API. (See Compatibility Status for the current API compliance.) All utility methods and non-standard APIs are documented below.

Utility methods

Non-standard APIs


createCanvas(width: number, height: number, type?: 'PDF'|'SVG') => Canvas

Creates a Canvas instance. This method works in both Node.js and Web browsers, where there is no Canvas constructor. (See browser.js for the implementation that runs in browsers.)

const { createCanvas } = require('canvas')
const mycanvas = createCanvas(200, 200)
const myPDFcanvas = createCanvas(600, 800, 'pdf') // see "PDF Support" section


createImageData(width: number, height: number) => ImageData
createImageData(data: Uint8ClampedArray, width: number, height?: number) => ImageData
// for alternative pixel formats:
createImageData(data: Uint16Array, width: number, height?: number) => ImageData

Creates an ImageData instance. This method works in both Node.js and Web browsers.

const { createImageData } = require('canvas')
const width = 20, height = 20
const arraySize = width * height * 4
const mydata = createImageData(new Uint8ClampedArray(arraySize), width)


loadImage() => Promise<Image>

Convenience method for loading images. This method works in both Node.js and Web browsers.

const { loadImage } = require('canvas')
const myimg = loadImage('')

myimg.then(() => {
  // do something with image
}).catch(err => {
  console.log('oh no!', err)

// or with async/await:
const myimg = await loadImage('')
// do something with image


registerFont(path: string, { family: string, weight?: string, style?: string }) => void

To use a font file that is not installed as a system font, use registerFont() to register the font with Canvas. This must be done before the Canvas is created.

const { registerFont, createCanvas } = require('canvas')
registerFont('comicsans.ttf', { family: 'Comic Sans' })

const canvas = createCanvas(500, 500)
const ctx = canvas.getContext('2d')

ctx.font = '12px "Comic Sans"'
ctx.fillText('Everyone hates this font :(', 250, 10)

The second argument is an object with properties that resemble the CSS properties that are specified in @font-face rules. You must specify at least family. weight, and style are optional and default to 'normal'.


img.src: string|Buffer

As in browsers, img.src can be set to a data: URI or a remote URL. In addition, node-canvas allows setting src to a local file path or Buffer instance.

const { Image } = require('canvas')

// From a buffer:
fs.readFile('images/squid.png', (err, squid) => {
  if (err) throw err
  const img = new Image()
  img.onload = () => ctx.drawImage(img, 0, 0)
  img.onerror = err => { throw err }
  img.src = squid

// From a local file path:
const img = new Image()
img.onload = () => ctx.drawImage(img, 0, 0)
img.onerror = err => { throw err }
img.src = 'images/squid.png'

// From a remote URL:
img.src = ''
// ... as above

// From a `data:` URI:
// ... as above

Note: In some cases, img.src= is currently synchronous. However, you should always use img.onload and img.onerror, as we intend to make img.src= always asynchronous as it is in browsers. See


img.dataMode: number

Applies to JPEG images drawn to PDF canvases only.

Setting img.dataMode = Image.MODE_MIME or Image.MODE_MIME|Image.MODE_IMAGE enables MIME data tracking of images. When MIME data is tracked, PDF canvases can embed JPEGs directly into the output, rather than re-encoding into PNG. This can drastically reduce filesize and speed up rendering.

const { Image, createCanvas } = require('canvas')
const canvas = createCanvas(w, h, 'pdf')
const img = new Image()
img.dataMode = Image.MODE_IMAGE // Only image data tracked
img.dataMode = Image.MODE_MIME // Only mime data tracked
img.dataMode = Image.MODE_MIME | Image.MODE_IMAGE // Both are tracked

If working with a non-PDF canvas, image data must be tracked; otherwise the output will be junk.

Enabling mime data tracking has no benefits (only a slow down) unless you are generating a PDF.


canvas.toBuffer((err: Error|null, result: Buffer) => void, mimeType?: string, config?: any) => void
canvas.toBuffer(mimeType?: string, config?: any) => Buffer

Creates a Buffer object representing the image contained in the canvas.

  • callback If provided, the buffer will be provided in the callback instead of being returned by the function. Invoked with an error as the first argument if encoding failed, or the resulting buffer as the second argument if it succeeded. Not supported for mimeType raw or for PDF or SVG canvases.
  • mimeType A string indicating the image format. Valid options are image/png, image/jpeg (if node-canvas was built with JPEG support) and raw (unencoded ARGB32 data in native-endian byte order, top-to-bottom). Defaults to image/png. If the canvas is a PDF or SVG canvas, this argument is ignored and a PDF or SVG is returned always.
  • config
    • For image/jpeg an object specifying the quality (0 to 1), if progressive compression should be used and/or if chroma subsampling should be used: {quality: 0.75, progressive: false, chromaSubsampling: true}. All properties are optional.

    • For image/png, an object specifying the ZLIB compression level (between 0 and 9), the compression filter(s), the palette (indexed PNGs only), the the background palette index (indexed PNGs only) and/or the resolution (ppi): {compressionLevel: 6, filters: canvas.PNG_ALL_FILTERS, palette: undefined, backgroundIndex: 0, resolution: undefined}. All properties are optional.

      Note that the PNG format encodes the resolution in pixels per meter, so if you specify 96, the file will encode 3780 ppm (~96.01 ppi). The resolution is undefined by default to match common browser behavior.

Return value

If no callback is provided, a Buffer. If a callback is provided, none.


// Default: buf contains a PNG-encoded image
const buf = canvas.toBuffer()

// PNG-encoded, zlib compression level 3 for faster compression but bigger files, no filtering
const buf2 = canvas.toBuffer('image/png', { compressionLevel: 3, filters: canvas.PNG_FILTER_NONE })

// JPEG-encoded, 50% quality
const buf3 = canvas.toBuffer('image/jpeg', { quality: 0.5 })

// Asynchronous PNG
canvas.toBuffer((err, buf) => {
  if (err) throw err // encoding failed
  // buf is PNG-encoded image

canvas.toBuffer((err, buf) => {
  if (err) throw err // encoding failed
  // buf is JPEG-encoded image at 95% quality
}, 'image/jpeg', { quality: 0.95 })

// ARGB32 pixel values, native-endian
const buf4 = canvas.toBuffer('raw')
const { stride, width } = canvas
// In memory, this is `canvas.height * canvas.stride` bytes long.
// The top row of pixels, in ARGB order, left-to-right, is:
const topPixelsARGBLeftToRight = buf4.slice(0, width * 4)
// And the third row is:
const row3 = buf4.slice(2 * stride, 2 * stride + width * 4)

// SVG and PDF canvases ignore the mimeType argument
const myCanvas = createCanvas(w, h, 'pdf')
myCanvas.toBuffer() // returns a buffer containing a PDF-encoded canvas


canvas.createPNGStream(config?: any) => ReadableStream

Creates a ReadableStream that emits PNG-encoded data.

  • config An object specifying the ZLIB compression level (between 0 and 9), the compression filter(s), the palette (indexed PNGs only) and/or the background palette index (indexed PNGs only): {compressionLevel: 6, filters: canvas.PNG_ALL_FILTERS, palette: undefined, backgroundIndex: 0, resolution: undefined}. All properties are optional.


const fs = require('fs')
const out = fs.createWriteStream(__dirname + '/test.png')
const stream = canvas.createPNGStream()
out.on('finish', () =>  console.log('The PNG file was created.'))

To encode indexed PNGs from canvases with pixelFormat: 'A8' or 'A1', provide an options object:

const palette = new Uint8ClampedArray([
  //r    g    b    a
    0,  50,  50, 255, // index 1
   10,  90,  90, 255, // index 2
  127, 127, 255, 255
  // ...
  palette: palette,
  backgroundIndex: 0 // optional, defaults to 0


canvas.createJPEGStream(config?: any) => ReadableStream

Creates a ReadableStream that emits JPEG-encoded data.

Note: At the moment, createJPEGStream() is synchronous under the hood. That is, it runs in the main thread, not in the libuv threadpool.

  • config an object specifying the quality (0 to 1), if progressive compression should be used and/or if chroma subsampling should be used: {quality: 0.75, progressive: false, chromaSubsampling: true}. All properties are optional.


const fs = require('fs')
const out = fs.createWriteStream(__dirname + '/test.jpeg')
const stream = canvas.createJPEGStream()
out.on('finish', () =>  console.log('The JPEG file was created.'))

// Disable 2x2 chromaSubsampling for deeper colors and use a higher quality
const stream = canvas.createJPEGStream({
  quality: 95,
  chromaSubsampling: false


canvas.createPDFStream(config?: any) => ReadableStream

Applies to PDF canvases only. Creates a ReadableStream that emits the encoded PDF. canvas.toBuffer() also produces an encoded PDF, but createPDFStream() can be used to reduce memory usage.


This is a standard API, but several non-standard calls are supported. The full list of supported calls is:

dataUrl = canvas.toDataURL() // defaults to PNG
dataUrl = canvas.toDataURL('image/png')
canvas.toDataURL((err, png) => { }) // defaults to PNG
canvas.toDataURL('image/png', (err, png) => { })
canvas.toDataURL('image/jpeg', (err, jpeg) => { }) // sync JPEG is not supported
canvas.toDataURL('image/jpeg', {...opts}, (err, jpeg) => { }) // see Canvas#createJPEGStream for valid options
canvas.toDataURL('image/jpeg', quality, (err, jpeg) => { }) // spec-following; quality from 0 to 1


context.patternQuality: 'fast'|'good'|'best'|'nearest'|'bilinear'

Defaults to 'good'. Affects pattern (gradient, image, etc.) rendering quality.


context.quality: 'fast'|'good'|'best'|'nearest'|'bilinear'

Defaults to 'good'. Like patternQuality, but applies to transformations affecting more than just patterns.


context.textDrawingMode: 'path'|'glyph'

Defaults to 'path'. The effect depends on the canvas type:

  • Standard (image) glyph and path both result in rasterized text. Glyph mode is faster than path, but may result in lower-quality text, especially when rotated or translated.

  • PDF glyph will embed text instead of paths into the PDF. This is faster to encode, faster to open with PDF viewers, yields a smaller file size and makes the text selectable. The subset of the font needed to render the glyphs will be embedded in the PDF. This is usually the mode you want to use with PDF canvases.

  • SVG glyph does not cause <text> elements to be produced as one might expect (cairo bug). Rather, glyph will create a <defs> section with a <symbol> for each glyph, then those glyphs be reused via <use> elements. path mode creates a <path> element for each text string. glyph mode is faster and yields a smaller file size.

In glyph mode, ctx.strokeText() and ctx.fillText() behave the same (aside from using the stroke and fill style, respectively).

This property is tracked as part of the canvas state in save/restore.

CanvasRenderingContext2D#globalCompositeOperator = 'saturate'

In addition to all of the standard global composite operators defined by the Canvas specification, the 'saturate' operator is also available.


context.antialias: 'default'|'none'|'gray'|'subpixel'

Sets the anti-aliasing mode.

PDF Output Support

node-canvas can create PDF documents instead of images. The canvas type must be set when creating the canvas as follows:

const canvas = createCanvas(200, 500, 'pdf')

An additional method .addPage() is then available to create multiple page PDFs:

// On first page
ctx.font = '22px Helvetica'
ctx.fillText('Hello World', 50, 80)

// Now on second page
ctx.font = '22px Helvetica'
ctx.fillText('Hello World 2', 50, 80)

canvas.toBuffer() // returns a PDF file
canvas.createPDFStream() // returns a ReadableStream that emits a PDF

See also:

SVG Output Support

node-canvas can create SVG documents instead of images. The canva type must be set when creating the canvas as follows:

const canvas = createCanvas(200, 500, 'svg')
// Use the normal primitives.
fs.writeFileSync('out.svg', canvas.toBuffer())

SVG Image Support

If librsvg is available when node-canvas is installed, node-canvas can render SVG images to your canvas context. This currently works by rasterizing the SVG image (i.e. drawing an SVG image to an SVG canvas will not preserve the SVG data).

const img = new Image()
img.onload = () => ctx.drawImage(img, 0, 0)
img.onerror = err => { throw err }
img.src = './example.svg'

Image pixel formats (experimental)

node-canvas has experimental support for additional pixel formats, roughly following the Canvas color space proposal.

const canvas = createCanvas(200, 200)
const ctx = canvas.getContext('2d', { pixelFormat: 'A8' })

By default, canvases are created in the RGBA32 format, which corresponds to the native HTML Canvas behavior. Each pixel is 32 bits. The JavaScript APIs that involve pixel data (getImageData, putImageData) store the colors in the order {red, green, blue, alpha} without alpha pre-multiplication. (The C++ API stores the colors in the order {alpha, red, green, blue} in native-endian ordering, with alpha pre-multiplication.)

These additional pixel formats have experimental support:

  • RGB24 Like RGBA32, but the 8 alpha bits are always opaque. This format is always used if the alpha context attribute is set to false (i.e. canvas.getContext('2d', {alpha: false})). This format can be faster than RGBA32 because transparency does not need to be calculated.
  • A8 Each pixel is 8 bits. This format can either be used for creating grayscale images (treating each byte as an alpha value), or for creating indexed PNGs (treating each byte as a palette index) (see the example using alpha values with fillStyle and the example using imageData).
  • RGB16_565 Each pixel is 16 bits, with red in the upper 5 bits, green in the middle 6 bits, and blue in the lower 5 bits, in native platform endianness. Some hardware devices and frame buffers use this format. Note that PNG does not support this format; when creating a PNG, the image will be converted to 24-bit RGB. This format is thus suboptimal for generating PNGs. ImageData instances for this mode use a Uint16Array instead of a Uint8ClampedArray.
  • A1 Each pixel is 1 bit, and pixels are packed together into 32-bit quantities. The ordering of the bits matches the endianness of the platform: on a little-endian machine, the first pixel is the least-significant bit. This format can be used for creating single-color images. Support for this format is incomplete, see note below.
  • RGB30 Each pixel is 30 bits, with red in the upper 10, green in the middle 10, and blue in the lower 10. (Requires Cairo 1.12 or later.) Support for this format is incomplete, see note below.

Notes and caveats:

  • Using a non-default format can affect the behavior of APIs that involve pixel data:

    • context2d.createImageData The size of the array returned depends on the number of bit per pixel for the underlying image data format, per the above descriptions.
    • context2d.getImageData The format of the array returned depends on the underlying image mode, per the above descriptions. Be aware of platform endianness, which can be determined using node.js's os.endianness() function.
    • context2d.putImageData As above.
  • A1 and RGB30 do not yet support getImageData or putImageData. Have a use case and/or opinion on working with these formats? Open an issue and let us know! (See #935.)

  • A1, A8, RGB30 and RGB16_565 with shadow blurs may crash or not render properly.

  • The ImageData(width, height) and ImageData(Uint8ClampedArray, width) constructors assume 4 bytes per pixel. To create an ImageData instance with a different number of bytes per pixel, use new ImageData(new Uint8ClampedArray(size), width, height) or new ImageData(new Uint16ClampedArray(size), width, height).


Benchmarks live in the benchmarks directory.


Examples line in the examples directory. Most produce a png image of the same name, and others such as live-clock.js launch an HTTP server to be viewed in the browser.

Original Authors


(The MIT License)

Copyright (c) 2010 LearnBoost, and contributors <>

Copyright (c) 2014 Automattic, Inc and contributors <>

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the 'Software'), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.