Proof of concept demonstration for generating custom infographics, built in javascript using node.js, angular.js, d3.js and GraphicsMagick.

• http://infographic.jamesmcguigan.com - live demo with unminified sources
• https://production.infographic.jamesmcguigan.com - live SSL demo with minified js/css

sudo mkdir /var/log/node/;
chmod a+rw /var/log/node/;
sudo apt install nodejs ruby-dev ruby-rubygems graphicsmagick 
sudo gem install compass

git clone [email protected]:JamesMcGuigan/infographic-generator.git
cd infographic-generator
npm install  # will also download bower dependencies and compile the client side browserify.js file
npm start    # runs nodemon and compass for development

# For production deployment
npm run production  # compiles minified js/css into ./production/
node InfographicGenerator.js NODE_ENV=production PORT_HTTP=3012 PORT_HTTPS=3013

Then open up the following localhost url http://localhost:4000/

See package.json for a list of other project npm scripts

Demo server deployment is managed via puppet using the following project:
https://github.com/JamesMcGuigan/puppet-config

• /data/ - Example configuration files
• /data/sources/ - Example data sources
• /app/config/config.js - Node configuration file
• /app/controllers/ - Node API logic
• /app/routes/ - Node URL routing
• /app/views/ - Mustache templates for generating initial HTML page
• /app/public/scss-src/ - SASS source files
• /app/public/scss/ - Compiled SASS -> CSS
• /app/public/html/ - Angualar HTML Snippits
• /app/public/angular/ - Angualar.js Application
• /app/public/angular/directive/directive.infographic.js - The main logic for the SVG generator

Sample Configuration Files:

• https://github.com/JamesMcGuigan/infographic-generator/blob/master/data/chart1.json
• https://github.com/JamesMcGuigan/infographic-generator/blob/master/data/chart2.json

Sample Defaults Files:

• http://infographic.jamesmcguigan.com/data/sources/defaults.json

Sample Data Files:

• http://infographic.jamesmcguigan.com/data/sources/mexico-brazil-shots.json
• http://infographic.jamesmcguigan.com/data/sources/player-passes.json

The JSON configuration file starts with a root "type":"svg" element, followed by an array of "children" elements. The code will auto-convert between arrays and objects as needed.

Any property name can be suffixed with URL to load data from a remote url, ie "defaults" -> "defaultsURL"

{

• "defaults": this hash is organized by type, and inherits down the tree. It provides defaults options that can be overridden in children elements.
• "title": this property on the "svg" root node is shown in the file listings dialog
• "label": this property is just a human readable string for easily identifying elements
• "type": "svg"|"image"|"rect"|"text"|"chart" - creates a SVG element of the same name, "chart" will render a custom d3.js SVG object
• "attr": this property exposes the raw SVG attributes for each element - either read the book "SVG Essentials" or see the the full technical SVG spec: http://www.w3.org/TR/SVG11/
• "attr": { x: 0, y: 0, height: "100px", width: "100px", "xlink:href": , "font-size": "24px", "line-height": "24px", "fill": "black", "fill-opacity": 0.5 } - these are the most common and useful attributes
• "align": "Top|Middle|Bottom" + "Left|Center|Right" -
• "margin": custom property which adds some additional xy spacing when used with "align"
• "content": for "type":"text" objects only, provides the content for the string
• "children" [{},{},{},...] of child nodes to be rendered relative to the parent xy and height|width

Chart Specific Options

• "chart": currently only "BarChart"|"DonutChart"|"DotChart" - which type of chart to draw
• "data"|"dataURL": provides configuration data for the chart, typically: { "values": { "field": value, ... }, "colors": { "field": color, ... } }
• "options": "DonutChart" only { "borderRadius": 20, "highlightRadius": 10, "highlight": { "outside": "white" } }

Text Interpolation

• "content" strings can be include "#{values.outside} / #{stats.total}" using dot separated notation
• "#{values.field}" reads values from "data.values"
• "#{colors.field}" reads values from "data.colors"
• "#{stats.length}" number of entries in data.values
• "#{stats.max}" maximum number in data.values
• "#{stats.min}" minimum number in data.values
• "#{stats.total}" sum of numbers in data.values
• "#{stats.percent.field}" percentage string (ie "85%") calculated using: values.field / stats.total
• "#{stats.factor.value}" number of times each individual string/number value was encountered
• #{stats.colors.factor.white} number of "white" entries in data.colors

}

The JSON configuration file needs to be syntactically valid, which mostly means quoting "key": "value" (except raw numbers) and making sure commas are in the correct places (after every item in a list, except the last)

• JSON configuration files are currently saved only to the filesystem, this made life easier for development. It would be relatively easy to add in MongoDB support.

• There is currently no way of editing the /data/sources/*.json via the web interface, but you can always switch from "dataURL" to "data" for inline editing

• In theory a custom AJAX interface could be created for editing this configuration file could be created, but learning JSON and SVG attributes is fairly easy

• BUG: the positioning of "children" inside "type":"chart" entries is currently buggy