无法从lenskart网站上刮掉任何东西

时间:2013-10-25 18:19:10

标签: javascript web-scraping phantomjs

我正在尝试使用phantomjs从网站http://www.lenskart.com/eyeglasses/frame-style/half-rim.html抓取一些产品信息。以下是我使用的脚本:

    var page = require('webpage').create();
page.onConsoleMessage = function(msg){
    console.log(msg);
};
page.open('http://www.lenskart.com/eyeglasses/frame-style/half-rim.html', function () {
console.log('1234');
  // Checks for bottom div and scrolls down from time to time
  window.setInterval(function() {
console.log('ddddddddddd');
      // Checks if there is a div with class=".has-more-items" 
      // (not sure if this is the best way of doing it)
      var count = page.evaluate(function() {
try{       
 return document.getElementsByClassName('top-to-bottom')[0].style.display;
}catch(e){return e.message;}
    });
//count = 1
console.log(count);
 console.log('1111 '+count);

      if(count === 'block' || count ==='inline' ) { // Didn't find
      console.log('count '+count);
//this.echo('wdc');      
  page.evaluate(function() {
          // Scrolls to the bottom of page
          window.document.body.scrollTop = document.body.scrollHeight;
        });
      }
      else

     { // Found
        //Do what you want
console.log('len123');

            try {
              var links = page.evaluate(function() {
        return [].map.call(document.querySelectorAll('div.product-view-region a.product-image'), function(link) {
            //console.log(link.getAttribute('href'));
            return link.getAttribute('href');
        });
                });
            } catch (e) {

           console.log(e.message); return [];
            }

//console.log(links.join(','));
//console.log(links);


        phantom.exit();
      }
  }, 1000); // Number o ms to wait between scrolls

});

但我不断得到这个错误:

TypeError: 'undefined' is not a function (evaluating 'this.each(function(value,i
ndex){results.push(iterator.call(context,value,index));})')



http://d37l6i9uhpr090.cloudfront.net/js/top_common.js:809 in collect
  phantomjs://webpage.evaluate():6
  phantomjs://webpage.evaluate():7
  phantomjs://webpage.evaluate():7
null

我还试图抓取其他一些信息,但得到了相同的结果。我不明白我哪里错了!

0 个答案:

没有答案