JSPM

  • ESM via JSPM
  • ES Module Entrypoint
  • Export Map
  • Keywords
  • License
  • Repository URL
  • TypeScript Types
  • README
  • Created
  • Published
  • Downloads 21119
  • Score
    100M100P100Q136293F

convert PNG images to pixel arrays without dependencies

Package Exports

  • pngparse

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (pngparse) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

pngparse

pngparse is a pure-JavaScript library for Node.JS for converting a PNG file into an array of pixel values. It came out of a need for reading PNG images in Node.JS for the Dark Sky API, but all existing libraries either had compilation issues or did not support enough of the PNG standard to be practical.

There's a reason nobody writes PNG-parsing libraries. This stuff is complicated and reinventing the wheel is dumb. But, if you're wondering whether you can trust it, it has a full unit test suite and we've been using it in production since Sep 2012, so there you go.

It's reasonably complete, covering most PNG color types, depths, and filters; notable omissions are lack of support for 16-bit images and interlacing.

Comments, bug fixes, feature improvements, etc. are all welcome. If you do write code, please ensure that you write tests for it!

Usage

To install:

npm install pngparse

To use:

var pngparse = require("pngparse")

...

pngparse.parse(buffer, function(err, data) {
  if(err)
    throw err

  /* do things! */
})

...

pngparse.parseFile("path/to/file.png", function(err, data) {
  if(err)
    throw err

  /* do things! */
})

The data object returned from the callback bears a striking resemblance to the HTML5 Canvas ImageData object. A notable distinction is that the object returns has a channels property which indicates how many color channels it uses (while an HTML5 ImageData object is always 4-channel). The possible color channel combinations are as follows:

1 channel : grayscale
2 channels: grayscale + alpha
3 channels: RGB
4 channels: RGBA

If you use the ImageData.prototype.getPixel method, this is handled for you; however, if you access the data array manually, then you will have to be aware of it.