Output options
toFile
toFile(fileOut, [callback]) ⇒
Promise.<Object>
Write output image data to a file.
If an explicit output format is not selected, it will be inferred from the extension, with JPEG, PNG, WebP, AVIF, TIFF, GIF, DZI, and libvips’ V format supported. Note that raw pixel data is only supported for buffer output.
By default all metadata will be removed, which includes EXIF-based orientation. See withMetadata for control over this.
The caller is responsible for ensuring directory structures and permissions exist.
A Promise
is returned when callback
is not provided.
Returns: Promise.<Object>
- - when no callback is provided
Throws:
Error
Invalid parameters
Param | Type | Description |
---|---|---|
fileOut | string | the path to write the image data to. |
[callback] | function | called on completion with two arguments (err, info) . info contains the output image format , size (bytes), width , height , channels and premultiplied (indicating if premultiplication was used). When using a crop strategy also contains cropOffsetLeft and cropOffsetTop . When using the attention crop strategy also contains attentionX and attentionY , the focal point of the cropped region. Animated output will also contain pageHeight and pages . May also contain textAutofitDpi (dpi the font was rendered at) if image was created from text. |
Example
sharp(input) .toFile('output.png', (err, info) => { ... });
Example
sharp(input) .toFile('output.png') .then(info => { ... }) .catch(err => { ... });
toBuffer
toBuffer([options], [callback]) ⇒
Promise.<Buffer>
Write output to a Buffer. JPEG, PNG, WebP, AVIF, TIFF, GIF and raw pixel data output are supported.
Use toFormat or one of the format-specific functions such as jpeg, png etc. to set the output format.
If no explicit format is set, the output format will match the input image, except SVG input which becomes PNG output.
By default all metadata will be removed, which includes EXIF-based orientation. See withMetadata for control over this.
callback
, if present, gets three arguments (err, data, info)
where:
err
is an error, if any.data
is the output image data.info
contains the output imageformat
,size
(bytes),width
,height
,channels
andpremultiplied
(indicating if premultiplication was used). When using a crop strategy also containscropOffsetLeft
andcropOffsetTop
. Animated output will also containpageHeight
andpages
. May also containtextAutofitDpi
(dpi the font was rendered at) if image was created from text.
A Promise
is returned when callback
is not provided.
Returns: Promise.<Buffer>
- - when no callback is provided
Param | Type | Description |
---|---|---|
[options] | Object | |
[options.resolveWithObject] | boolean | Resolve the Promise with an Object containing data and info properties instead of resolving only with data . |
[callback] | function |
Example
sharp(input) .toBuffer((err, data, info) => { ... });
Example
sharp(input) .toBuffer() .then(data => { ... }) .catch(err => { ... });
Example
sharp(input) .png() .toBuffer({ resolveWithObject: true }) .then(({ data, info }) => { ... }) .catch(err => { ... });
Example
const { data, info } = await sharp('my-image.jpg') // output the raw pixels .raw() .toBuffer({ resolveWithObject: true });
// create a more type safe way to work with the raw pixel data// this will not copy the data, instead it will change `data`s underlying ArrayBuffer// so `data` and `pixelArray` point to the same memory locationconst pixelArray = new Uint8ClampedArray(data.buffer);
// When you are done changing the pixelArray, sharp takes the `pixelArray` as an inputconst { width, height, channels } = info;await sharp(pixelArray, { raw: { width, height, channels } }) .toFile('my-changed-image.jpg');
keepExif
keepExif() ⇒
Sharp
Keep all EXIF metadata from the input image in the output image.
EXIF metadata is unsupported for TIFF output.
Since: 0.33.0
Example
const outputWithExif = await sharp(inputWithExif) .keepExif() .toBuffer();
withExif
withExif(exif) ⇒
Sharp
Set EXIF metadata in the output image, ignoring any EXIF in the input image.
Throws:
Error
Invalid parameters
Since: 0.33.0
Param | Type | Description |
---|---|---|
exif | Object.<string, Object.<string, string>> | Object keyed by IFD0, IFD1 etc. of key/value string pairs to write as EXIF data. |
Example
const dataWithExif = await sharp(input) .withExif({ IFD0: { Copyright: 'The National Gallery' }, IFD3: { GPSLatitudeRef: 'N', GPSLatitude: '51/1 30/1 3230/100', GPSLongitudeRef: 'W', GPSLongitude: '0/1 7/1 4366/100' } }) .toBuffer();
withExifMerge
withExifMerge(exif) ⇒
Sharp
Update EXIF metadata from the input image in the output image.
Throws:
Error
Invalid parameters
Since: 0.33.0
Param | Type | Description |
---|---|---|
exif | Object.<string, Object.<string, string>> | Object keyed by IFD0, IFD1 etc. of key/value string pairs to write as EXIF data. |
Example
const dataWithMergedExif = await sharp(inputWithExif) .withExifMerge({ IFD0: { Copyright: 'The National Gallery' } }) .toBuffer();
keepIccProfile
keepIccProfile() ⇒
Sharp
Keep ICC profile from the input image in the output image.
Where necessary, will attempt to convert the output colour space to match the profile.
Since: 0.33.0
Example
const outputWithIccProfile = await sharp(inputWithIccProfile) .keepIccProfile() .toBuffer();
withIccProfile
withIccProfile(icc, [options]) ⇒
Sharp
Transform using an ICC profile and attach to the output image.
This can either be an absolute filesystem path or
built-in profile name (srgb
, p3
, cmyk
).
Throws:
Error
Invalid parameters
Since: 0.33.0
Param | Type | Default | Description |
---|---|---|---|
icc | string | Absolute filesystem path to output ICC profile or built-in profile name (srgb, p3, cmyk). | |
[options] | Object | ||
[options.attach] | number | true | Should the ICC profile be included in the output image metadata? |
Example
const outputWithP3 = await sharp(input) .withIccProfile('p3') .toBuffer();
keepMetadata
keepMetadata() ⇒
Sharp
Keep all metadata (EXIF, ICC, XMP, IPTC) from the input image in the output image.
The default behaviour, when keepMetadata
is not used, is to convert to the device-independent
sRGB colour space and strip all metadata, including the removal of any ICC profile.
Since: 0.33.0
Example
const outputWithMetadata = await sharp(inputWithMetadata) .keepMetadata() .toBuffer();
withMetadata
withMetadata([options]) ⇒
Sharp
Keep most metadata (EXIF, XMP, IPTC) from the input image in the output image.
This will also convert to and add a web-friendly sRGB ICC profile if appropriate.
Allows orientation and density to be set or updated.
Throws:
Error
Invalid parameters
Param | Type | Description |
---|---|---|
[options] | Object | |
[options.orientation] | number | Used to update the EXIF Orientation tag, integer between 1 and 8. |
[options.density] | number | Number of pixels per inch (DPI). |
Example
const outputSrgbWithMetadata = await sharp(inputRgbWithMetadata) .withMetadata() .toBuffer();
Example
// Set output metadata to 96 DPIconst data = await sharp(input) .withMetadata({ density: 96 }) .toBuffer();
toFormat
toFormat(format, options) ⇒
Sharp
Force output to a given format.
Throws:
Error
unsupported format or options
Param | Type | Description |
---|---|---|
format | string | Object | as a string or an Object with an ‘id’ attribute |
options | Object | output options |
Example
// Convert any input to PNG outputconst data = await sharp(input) .toFormat('png') .toBuffer();
jpeg
jpeg([options]) ⇒
Sharp
Use these JPEG options for output image.
Throws:
Error
Invalid options
Param | Type | Default | Description |
---|---|---|---|
[options] | Object | output options | |
[options.quality] | number | 80 | quality, integer 1-100 |
[options.progressive] | boolean | false | use progressive (interlace) scan |
[options.chromaSubsampling] | string | “‘4:2:0‘“ | set to ‘4:4:4’ to prevent chroma subsampling otherwise defaults to ‘4:2:0’ chroma subsampling |
[options.optimiseCoding] | boolean | true | optimise Huffman coding tables |
[options.optimizeCoding] | boolean | true | alternative spelling of optimiseCoding |
[options.mozjpeg] | boolean | false | use mozjpeg defaults, equivalent to { trellisQuantisation: true, overshootDeringing: true, optimiseScans: true, quantisationTable: 3 } |
[options.trellisQuantisation] | boolean | false | apply trellis quantisation |
[options.overshootDeringing] | boolean | false | apply overshoot deringing |
[options.optimiseScans] | boolean | false | optimise progressive scans, forces progressive |
[options.optimizeScans] | boolean | false | alternative spelling of optimiseScans |
[options.quantisationTable] | number | 0 | quantization table to use, integer 0-8 |
[options.quantizationTable] | number | 0 | alternative spelling of quantisationTable |
[options.force] | boolean | true | force JPEG output, otherwise attempt to use input format |
Example
// Convert any input to very high quality JPEG outputconst data = await sharp(input) .jpeg({ quality: 100, chromaSubsampling: '4:4:4' }) .toBuffer();
Example
// Use mozjpeg to reduce output JPEG file size (slower)const data = await sharp(input) .jpeg({ mozjpeg: true }) .toBuffer();
png
png([options]) ⇒
Sharp
Use these PNG options for output image.
By default, PNG output is full colour at 8 bits per pixel.
Indexed PNG input at 1, 2 or 4 bits per pixel is converted to 8 bits per pixel.
Set palette
to true
for slower, indexed PNG output.
For 16 bits per pixel output, convert to rgb16
via
toColourspace.
Throws:
Error
Invalid options
Param | Type | Default | Description |
---|---|---|---|
[options] | Object | ||
[options.progressive] | boolean | false | use progressive (interlace) scan |
[options.compressionLevel] | number | 6 | zlib compression level, 0 (fastest, largest) to 9 (slowest, smallest) |
[options.adaptiveFiltering] | boolean | false | use adaptive row filtering |
[options.palette] | boolean | false | quantise to a palette-based image with alpha transparency support |
[options.quality] | number | 100 | use the lowest number of colours needed to achieve given quality, sets palette to true |
[options.effort] | number | 7 | CPU effort, between 1 (fastest) and 10 (slowest), sets palette to true |
[options.colours] | number | 256 | maximum number of palette entries, sets palette to true |
[options.colors] | number | 256 | alternative spelling of options.colours , sets palette to true |
[options.dither] | number | 1.0 | level of Floyd-Steinberg error diffusion, sets palette to true |
[options.force] | boolean | true | force PNG output, otherwise attempt to use input format |
Example
// Convert any input to full colour PNG outputconst data = await sharp(input) .png() .toBuffer();
Example
// Convert any input to indexed PNG output (slower)const data = await sharp(input) .png({ palette: true }) .toBuffer();
Example
// Output 16 bits per pixel RGB(A)const data = await sharp(input) .toColourspace('rgb16') .png() .toBuffer();
webp
webp([options]) ⇒
Sharp
Use these WebP options for output image.
Throws:
Error
Invalid options
Param | Type | Default | Description |
---|---|---|---|
[options] | Object | output options | |
[options.quality] | number | 80 | quality, integer 1-100 |
[options.alphaQuality] | number | 100 | quality of alpha layer, integer 0-100 |
[options.lossless] | boolean | false | use lossless compression mode |
[options.nearLossless] | boolean | false | use near_lossless compression mode |
[options.smartSubsample] | boolean | false | use high quality chroma subsampling |
[options.smartDeblock] | boolean | false | auto-adjust the deblocking filter, can improve low contrast edges (slow) |
[options.preset] | string | “‘default‘“ | named preset for preprocessing/filtering, one of: default, photo, picture, drawing, icon, text |
[options.effort] | number | 4 | CPU effort, between 0 (fastest) and 6 (slowest) |
[options.loop] | number | 0 | number of animation iterations, use 0 for infinite animation |
[options.delay] | number | Array.<number> | delay(s) between animation frames (in milliseconds) | |
[options.minSize] | boolean | false | prevent use of animation key frames to minimise file size (slow) |
[options.mixed] | boolean | false | allow mixture of lossy and lossless animation frames (slow) |
[options.force] | boolean | true | force WebP output, otherwise attempt to use input format |
Example
// Convert any input to lossless WebP outputconst data = await sharp(input) .webp({ lossless: true }) .toBuffer();
Example
// Optimise the file size of an animated WebPconst outputWebp = await sharp(inputWebp, { animated: true }) .webp({ effort: 6 }) .toBuffer();
gif
gif([options]) ⇒
Sharp
Use these GIF options for the output image.
The first entry in the palette is reserved for transparency.
The palette of the input image will be re-used if possible.
Throws:
Error
Invalid options
Since: 0.30.0
Param | Type | Default | Description |
---|---|---|---|
[options] | Object | output options | |
[options.reuse] | boolean | true | re-use existing palette, otherwise generate new (slow) |
[options.progressive] | boolean | false | use progressive (interlace) scan |
[options.colours] | number | 256 | maximum number of palette entries, including transparency, between 2 and 256 |
[options.colors] | number | 256 | alternative spelling of options.colours |
[options.effort] | number | 7 | CPU effort, between 1 (fastest) and 10 (slowest) |
[options.dither] | number | 1.0 | level of Floyd-Steinberg error diffusion, between 0 (least) and 1 (most) |
[options.interFrameMaxError] | number | 0 | maximum inter-frame error for transparency, between 0 (lossless) and 32 |
[options.interPaletteMaxError] | number | 3 | maximum inter-palette error for palette reuse, between 0 and 256 |
[options.loop] | number | 0 | number of animation iterations, use 0 for infinite animation |
[options.delay] | number | Array.<number> | delay(s) between animation frames (in milliseconds) | |
[options.force] | boolean | true | force GIF output, otherwise attempt to use input format |
Example
// Convert PNG to GIFawait sharp(pngBuffer) .gif() .toBuffer();
Example
// Convert animated WebP to animated GIFawait sharp('animated.webp', { animated: true }) .toFile('animated.gif');
Example
// Create a 128x128, cropped, non-dithered, animated thumbnail of an animated GIFconst out = await sharp('in.gif', { animated: true }) .resize({ width: 128, height: 128 }) .gif({ dither: 0 }) .toBuffer();
Example
// Lossy file size reduction of animated GIFawait sharp('in.gif', { animated: true }) .gif({ interFrameMaxError: 8 }) .toFile('optim.gif');
jp2
jp2([options]) ⇒
Sharp
Use these JP2 options for output image.
Requires libvips compiled with support for OpenJPEG. The prebuilt binaries do not include this - see installing a custom libvips.
Throws:
Error
Invalid options
Since: 0.29.1
Param | Type | Default | Description |
---|---|---|---|
[options] | Object | output options | |
[options.quality] | number | 80 | quality, integer 1-100 |
[options.lossless] | boolean | false | use lossless compression mode |
[options.tileWidth] | number | 512 | horizontal tile size |
[options.tileHeight] | number | 512 | vertical tile size |
[options.chromaSubsampling] | string | “‘4:4:4‘“ | set to ‘4:2:0’ to use chroma subsampling |
Example
// Convert any input to lossless JP2 outputconst data = await sharp(input) .jp2({ lossless: true }) .toBuffer();
Example
// Convert any input to very high quality JP2 outputconst data = await sharp(input) .jp2({ quality: 100, chromaSubsampling: '4:4:4' }) .toBuffer();
tiff
tiff([options]) ⇒
Sharp
Use these TIFF options for output image.
The density
can be set in pixels/inch via withMetadata
instead of providing xres
and yres
in pixels/mm.
Throws:
Error
Invalid options
Param | Type | Default | Description |
---|---|---|---|
[options] | Object | output options | |
[options.quality] | number | 80 | quality, integer 1-100 |
[options.force] | boolean | true | force TIFF output, otherwise attempt to use input format |
[options.compression] | string | “‘jpeg‘“ | compression options: none, jpeg, deflate, packbits, ccittfax4, lzw, webp, zstd, jp2k |
[options.predictor] | string | “‘horizontal‘“ | compression predictor options: none, horizontal, float |
[options.pyramid] | boolean | false | write an image pyramid |
[options.tile] | boolean | false | write a tiled tiff |
[options.tileWidth] | number | 256 | horizontal tile size |
[options.tileHeight] | number | 256 | vertical tile size |
[options.xres] | number | 1.0 | horizontal resolution in pixels/mm |
[options.yres] | number | 1.0 | vertical resolution in pixels/mm |
[options.resolutionUnit] | string | “‘inch‘“ | resolution unit options: inch, cm |
[options.bitdepth] | number | 8 | reduce bitdepth to 1, 2 or 4 bit |
[options.miniswhite] | boolean | false | write 1-bit images as miniswhite |
Example
// Convert SVG input to LZW-compressed, 1 bit per pixel TIFF outputsharp('input.svg') .tiff({ compression: 'lzw', bitdepth: 1 }) .toFile('1-bpp-output.tiff') .then(info => { ... });
avif
avif([options]) ⇒
Sharp
Use these AVIF options for output image.
AVIF image sequences are not supported. Prebuilt binaries support a bitdepth of 8 only.
Throws:
Error
Invalid options
Since: 0.27.0
Param | Type | Default | Description |
---|---|---|---|
[options] | Object | output options | |
[options.quality] | number | 50 | quality, integer 1-100 |
[options.lossless] | boolean | false | use lossless compression |
[options.effort] | number | 4 | CPU effort, between 0 (fastest) and 9 (slowest) |
[options.chromaSubsampling] | string | “‘4:4:4‘“ | set to ‘4:2:0’ to use chroma subsampling |
[options.bitdepth] | number | 8 | set bitdepth to 8, 10 or 12 bit |
Example
const data = await sharp(input) .avif({ effort: 2 }) .toBuffer();
Example
const data = await sharp(input) .avif({ lossless: true }) .toBuffer();
heif
heif(options) ⇒
Sharp
Use these HEIF options for output image.
Support for patent-encumbered HEIC images using hevc
compression requires the use of a
globally-installed libvips compiled with support for libheif, libde265 and x265.
Throws:
Error
Invalid options
Since: 0.23.0
Param | Type | Default | Description |
---|---|---|---|
options | Object | output options | |
options.compression | string | compression format: av1, hevc | |
[options.quality] | number | 50 | quality, integer 1-100 |
[options.lossless] | boolean | false | use lossless compression |
[options.effort] | number | 4 | CPU effort, between 0 (fastest) and 9 (slowest) |
[options.chromaSubsampling] | string | “‘4:4:4‘“ | set to ‘4:2:0’ to use chroma subsampling |
[options.bitdepth] | number | 8 | set bitdepth to 8, 10 or 12 bit |
Example
const data = await sharp(input) .heif({ compression: 'hevc' }) .toBuffer();
jxl
jxl([options]) ⇒
Sharp
Use these JPEG-XL (JXL) options for output image.
This feature is experimental, please do not use in production systems.
Requires libvips compiled with support for libjxl. The prebuilt binaries do not include this - see installing a custom libvips.
Throws:
Error
Invalid options
Since: 0.31.3
Param | Type | Default | Description |
---|---|---|---|
[options] | Object | output options | |
[options.distance] | number | 1.0 | maximum encoding error, between 0 (highest quality) and 15 (lowest quality) |
[options.quality] | number | calculate distance based on JPEG-like quality, between 1 and 100, overrides distance if specified | |
[options.decodingTier] | number | 0 | target decode speed tier, between 0 (highest quality) and 4 (lowest quality) |
[options.lossless] | boolean | false | use lossless compression |
[options.effort] | number | 7 | CPU effort, between 1 (fastest) and 9 (slowest) |
[options.loop] | number | 0 | number of animation iterations, use 0 for infinite animation |
[options.delay] | number | Array.<number> | delay(s) between animation frames (in milliseconds) |
raw
raw([options]) ⇒
Sharp
Force output to be raw, uncompressed pixel data. Pixel ordering is left-to-right, top-to-bottom, without padding. Channel ordering will be RGB or RGBA for non-greyscale colourspaces.
Throws:
Error
Invalid options
Param | Type | Default | Description |
---|---|---|---|
[options] | Object | output options | |
[options.depth] | string | “‘uchar‘“ | bit depth, one of: char, uchar (default), short, ushort, int, uint, float, complex, double, dpcomplex |
Example
// Extract raw, unsigned 8-bit RGB pixel data from JPEG inputconst { data, info } = await sharp('input.jpg') .raw() .toBuffer({ resolveWithObject: true });
Example
// Extract alpha channel as raw, unsigned 16-bit pixel data from PNG inputconst data = await sharp('input.png') .ensureAlpha() .extractChannel(3) .toColourspace('b-w') .raw({ depth: 'ushort' }) .toBuffer();
tile
tile([options]) ⇒
Sharp
Use tile-based deep zoom (image pyramid) output.
Set the format and options for tile images via the toFormat
, jpeg
, png
or webp
functions.
Use a .zip
or .szi
file extension with toFile
to write to a compressed archive file format.
The container will be set to zip
when the output is a Buffer or Stream, otherwise it will default to fs
.
Throws:
Error
Invalid parameters
Param | Type | Default | Description |
---|---|---|---|
[options] | Object | ||
[options.size] | number | 256 | tile size in pixels, a value between 1 and 8192. |
[options.overlap] | number | 0 | tile overlap in pixels, a value between 0 and 8192. |
[options.angle] | number | 0 | tile angle of rotation, must be a multiple of 90. |
[options.background] | string | Object | ”{r: 255, g: 255, b: 255, alpha: 1}“ | background colour, parsed by the color module, defaults to white without transparency. |
[options.depth] | string | how deep to make the pyramid, possible values are onepixel , onetile or one , default based on layout. | |
[options.skipBlanks] | number | -1 | Threshold to skip tile generation. Range is 0-255 for 8-bit images, 0-65535 for 16-bit images. Default is 5 for google layout, -1 (no skip) otherwise. |
[options.container] | string | “‘fs‘“ | tile container, with value fs (filesystem) or zip (compressed file). |
[options.layout] | string | “‘dz‘“ | filesystem layout, possible values are dz , iiif , iiif3 , zoomify or google . |
[options.centre] | boolean | false | centre image in tile. |
[options.center] | boolean | false | alternative spelling of centre. |
[options.id] | string | “‘https://example.com/iiif'“ | when layout is iiif /iiif3 , sets the @id /id attribute of info.json |
[options.basename] | string | the name of the directory within the zip file when container is zip . |
Example
sharp('input.tiff') .png() .tile({ size: 512 }) .toFile('output.dz', function(err, info) { // output.dzi is the Deep Zoom XML definition // output_files contains 512x512 tiles grouped by zoom level });
Example
const zipFileWithTiles = await sharp(input) .tile({ basename: "tiles" }) .toBuffer();
Example
const iiififier = sharp().tile({ layout: "iiif" });readableStream .pipe(iiififier) .pipe(writeableStream);
timeout
timeout(options) ⇒
Sharp
Set a timeout for processing, in seconds. Use a value of zero to continue processing indefinitely, the default behaviour.
The clock starts when libvips opens an input image for processing. Time spent waiting for a libuv thread to become available is not included.
Since: 0.29.2
Param | Type | Description |
---|---|---|
options | Object | |
options.seconds | number | Number of seconds after which processing will be stopped |
Example
// Ensure processing takes no longer than 3 secondstry { const data = await sharp(input) .blur(1000) .timeout({ seconds: 3 }) .toBuffer();} catch (err) { if (err.message.includes('timeout')) { ... }}