#desprite#

Split a sprite image by the CSS usage with a simple command-line tool, written in node.

desprite searches your CSS for qualified1 rules, and crops out the rectangles they reference to seperate2 image files. 1 - rules containing width, height, and background or background-position. 2 - files are named by the rule that referenced their position (usually a class name).

##Usage##

1. desprite runs on Node.js, so you need that.
2. You must also install GraphicsMagick - that's how we crop out the images.

• Windows: SourceForge downloads
• Linux: $ sudo apt-get install graphicsmagick
• Mac:
  • Homebrew: $ brew install graphicsmagick
  • Old fashioned: http://mac-dev-env.patrickbougie.com/graphicsmagick/

Make sure that after your installation, your cmd/terminal responds to "gm" with the GraphicsMagick CLI. 3. That's it. Just git clone (or download zip) and run (for example): ``` $ npm install

$ node desprite.js -i test-src\sprites.png -c test-src\sprites.css
```

##Runtime Options##

  -i, --image    Sprite image                                                  [required]
  -c, --css      CSS file                                                      [required]
  -o, --output   Output folder path (default: split/)
  -v, --verbose  Verbose progress messages
  -p, --parsed   Verbose progress messages shown for valid rules only
  -u, --unique   Include duplicate rules if their rule identifiers are unique
  -s, --spawn    Max threads spawned for split operation (default: 50)
  -r, --ratio    Scale ratio between source image size and the background-image-size (default: 1)

##Dependencies##

• gm (https://github.com/aheckmann/gm)
• GraphicsMagick (http://www.graphicsmagick.org/)
• css-parse (https://npmjs.org/package/css-parse)
• optimist (https://github.com/substack/node-optimist)
• ansi (https://github.com/TooTallNate/ansi.js)
• when (https://github.com/cujojs/when)
• sanitize-filename (https://npmjs.org/package/sanitize-filename)

##Acknowledgments##

desprite was authored minimally to solve a problem with our team's project, and probably does not take all wacky cases into consideration. Contributions or feedback about your case are welcome.