Call the file using node with a keyword to search. `node search-results.js "nodejs rocks"`

search-results.js

var start = new Date().getTime();
var jsdom = require('jsdom').jsdom;
var request = require('request');
var argv = require('optimist')
    .usage('numresults <word>')
    .check(function(argv) { if (argv._.length == 0) throw '' })
    .argv;
var url = 'http://www.google.com/search?hl=en&q=' + encodeURIComponent(argv._[0]);
console.log(url);
request({uri: url}, function(error, response, body) {
  var window = jsdom(body.toString(), undefined, {
    ProcessExternalResources: false,
    FetchExternalResources: false,
  }).createWindow();
  var $ = require('jquery').create(window);
  var numResults = $('#resultStats').html().split(' ')[1];
  console.log(numResults, 'results for', url);
  var items = [];
  var c = 0;
  $('#ires ol li .l').each(function() {
    if (c < 10) {
      items.push( $(this).text() );
      c++;
    } else {
      return;
    }
  });
  console.log('Top 10 results');
  items.forEach(function(item, i) {
    console.log('%d %s', i+1, item);
  })
  var end = new Date().getTime();
  console.log('Total Time: %d s', (end - start) / 1000);
});

https://github.com/chriso/node.io - so much nicer for scraping.

+1 chris, I've been looking for something that's a simple package to do scraping with, this was just a wee experiment to test the flexibility of using nodejs + jQuery to do screen scraping. I'll check your package out today