Extracts info from a Cheerio-based DOM using a json object.

README.md

Usage

Code:

var DomExtractor = require('./index.js');
var html = '<h3 data-universe="marvel">List of Superheros:</h3><ul class="heros"><li data-id="1"><span>Spiderman<span></li><li data-id="2"><span>Captain America<span></li><ul>';
var query = {
      universe: 'h3[data-universe]',
      heros: {
        _: 'ul.heros li',
        name: 'span',
        id: '@[data-id]'
      }
    };
var results = (new DomExtractor(html)).extract(query);
console.log(results);

Output:

{ universe: 'marvel',
  heros:
   [ { name: 'Spiderman', id: '1' },
     { name: 'Captain America', id: '2' } ] }

index.js

var cheerio = require('cheerio');

var DomExtractor = module.exports = function(html) {
  var $ = cheerio.load(html);
  var extract = this.extract = function(q, root) {
    var results = {};

    for (var name in q) {
      var selector = q[name];

      if (typeof selector === 'string') {
        if (name !== '_') {

          var attr = (selector.match(/\[([^\]]+)\]$/) || [])[1],
            useRoot = /^@/.test(selector),
            node = useRoot ? $(root) : $(selector, root);

          if (!node.length) {
            node = $(root);
          }

          results[name] = attr ? node.attr(attr) : node.text();
        }
      } else {
        results[name] = $(selector._, root).map(function(i, el) {
          return extract(selector, el);
        }).get();
      }


    }
    return results;
  }
  
  this.find = function(selector) {
    return $(selector);
  };
}

package.json

{
  "name": "DomExtractor",
  "version": "1.0.0",
  "description": "Extracts info from a Cheerio-based DOM using a json object.",
  "author": "Matt Powell",
  "dependencies": {
    "cheerio": "0.18.0"
  }
}