Simple Crawler with Node Phantom

crawler.js

/**
 * Simple Crawler with Node Phantom
 * @author: loretoparisi at gmail dot com
*/
(function() {

  var Crawler,phantom,cheerio;

  cheerio = require("cheerio"),
  phantom = require('phantom');

  Crawler = (function() {

    /**
     * Wait for event condition to verify
     */
    function waitFor(testFx, onReady, timeOutMillis) {
        var maxtimeOutMillis = timeOutMillis ? timeOutMillis : 5000, //< Default Max Timout is 5s
            start = new Date().getTime(),
            condition = false,
            interval = setInterval(function () {
                if ((new Date().getTime() - start < maxtimeOutMillis) && !condition) {
                    // If not time-out yet and condition not yet fulfilled
                    condition = (typeof (testFx) === "string" ? eval(testFx) : testFx()); //< defensive code
                } else {
                    if (!condition) {
                        // If condition still not fulfilled (timeout but condition is 'false')
                        //console.log("'waitFor()' timeout");
                        typeof (onReady) === "string" ? eval(onReady) : onReady();
                        clearInterval(interval);
                        //phantom.exit(1);
                    } else {
                        // Condition fulfilled (timeout and/or condition is 'true')
                        console.log("'waitFor()' finished in " + (new Date().getTime() - start) + "ms.");
                        typeof (onReady) === "string" ? eval(onReady) : onReady(); //< Do what it's supposed to do once the condition is fulfilled
                        clearInterval(interval); //< Stop this interval
                    }
                }
            }, 500); //< repeat check every 500ms
    };

    /**
     * Simple Crawler
     * @uses phantom module
     * @author Loreto Parisi (loretoparisi at musixmatch dot com)
     */
    function Crawler(options,logger) {
      this.logger=logger;
      /** Command line phantom options */
      this.options = {
        debug : false, // true to debug page events
        error : true, // true to log page errors
        info : true, // true to log page info
        commandline : ['--ignore-ssl-errors=yes', '--load-images=no', '--ssl-protocol=tlsv1']
      }
      for (var attrname in options) { this.options[attrname] = options[attrname]; }

    }

    /**
     * Fetch url
     * @param done Block success Block
     * @param fail Block error Block
    */
    Crawler.prototype.fetchUrl = function(url, done, fail) {
      var self=this;

      var sitepage
        ,pageContent
        ,pageStatus
        ,pageResponse
        ,phInstance
        ,pageContext;

      sitepage = phInstance = null;
      // context for the page
      pageContext = {
          options : self.options
      }
      phantom.create(self.options.commandline)
          .then((instance) => {
              phInstance = instance;
              return instance.createPage();
          })
          .then((page) => {

              pageResponse = phInstance.createOutObject();
              sitepage = page;

              // attach events: resources
              page.property('onResourceRequested', function(requestData, networkRequest,pageContext) {
                  if(pageContext.options.debug) console.log(requestData.url);
              },pageContext);
              page.property('onResourceReceived', function(response, networkRequest,pageContext) {
                var msg = '  id: ' + response.id + ', stage: "' + response.stage + '", response: ' + JSON.stringify(response);
                if(pageContext.options.debug) console.log(msg);
              },pageContext);
              page.property('onResourceError', function(resourceError, networkRequest,pageContext) {
                var errorString = resourceError.errorString;
                var errorPageUrl = resourceError.url;
                if(pageContext.options.error) console.log(resourceError,errorString,errorPageUrl);
              },pageContext);
              // attach events: load
              page.property('onLoadStarted', function(pageContext) {
                  if(pageContext.options.info) console.log('onLoadStarted');
              },pageContext);
              page.property('onLoadFinished', function(response,pageContext,pageResponse) {
                  if(pageContext.options.info) console.log('onLoadFinished');

              },pageContext,pageResponse);
              // events: navigation
              page.property('onNavigationRequested', function(url, type, willNavigate, main,pageContext) {
                var msg = '  destination_url: ' + url;
                msg += '  type (cause): ' + type;
                msg += '  will navigate: ' + willNavigate;
                msg += '  from page\'s main frame: ' + main;
                if(pageContext.options.debug) console.log(msg);
              },pageContext);
              // error
              page.property('onError', function(requestData, networkRequest,pageContext) {
                  if(pageContext.options.error) console.log(requestData.url);
              },pageContext);

              return page.open(url);
          })
          .then((status) => {
              console.log("Page status [%s]",status);
              pageStatus=status;
              return sitepage.property('content');
            })
          .then((content) => {

            pageContent = content;
            console.log("Content ready Page Status [%s]",pageStatus);
            //var $ = cheerio.load( pageContent );

            if (pageStatus !== 'success') {
              console.log("Page status error %s", pageStatus);
              if(fail) fail(error);
              sitepage.close();
              phInstance.exit();
            } else {

              sitepage.evaluate(function() {
                return document.body;
                //return document.querySelectorAll('.content > ul > li');
              }).then(function(html) {
                  pageContent = sitepage.content;
                  if(done) done( pageContent, html );
                  sitepage.close();
                  phInstance.exit();
              });


            }
          })
          .catch((error) => {
              self.logger.error(error);
              if(fail) fail(error);
              if(sitepage) sitepage.close();
              phInstance.exit();
          });
    } //fetchUrl

    return Crawler;

  })();

  module.exports = Crawler;

}).call(this);