使用casperjs和phantomjs抓取多个页面 [英] Using casperjs and phantomjs to scrape multiple pages

问题描述

我正在尝试抓取许多具有标准格式的页面.我已经能够使用Phantomjs成功地刮取单个页面，但是当我尝试遍历多个页面时，异步处理会使事情挂起.告诉Casper/Phantom等待的正确方法是什么?

I'm trying to scrape a number of pages that have a standard format. I've been able to use Phantomjs to successfully scrape a single page, but when I try to iterate over multiple ones, the asynchronous processing makes things hang up. What's the proper way to tell Casper/Phantom to wait?

var page = require('webpage').create();
var fs = require('fs');

page.onConsoleMessage = function(msg) {
    phantom.outputEncoding = "utf-8";
    console.log(msg);
};


// this overwrites the previous output file

f = fs.open("lat_long.txt", "w");
f.write("--");
f.close();


   // this is the unique identifier for the locations. For now, I just have three datapoints
  var EPAID = ["KYD980501076","ME8170022018", "MEN000103584"]; 

 /// this code will be used to loop through the different locations. For now, set to look at only one.  
 for (q= 0;  q < 1; q++)  {
    var processing = false;



   //we construct the target url
   var url  = "http://iaspub.epa.gov/enviro/efsystemquery.cerclis?fac_search=site_epa_id&fac_value=" + EPAID[0]  + "&fac_search_type=Beginning+With&postal_code=&location_address=&add_search_type=Beginning+With&city_name=&county_name=&state_code=&program_search=1&report=2&page_no=1&output_sql_switch=TRUE&database_type=CERCLIS" ;


   page.open(url);
   page.onLoadFinished = function(status) {
   if ( status === "success" ) {
       page.includeJs("http://ajax.googleapis.com/ajax/libs/jquery/1.6.1/jquery.min.js", function() {
           var str = page.evaluate(function() {                   
               $value = [];
               $Object = $(".result tr");                
               for (i =0 ; i < 10; i++) { 
             $value.push($Object.find('td').html(),$Object.find('td').next().next().html() );          
             $Object = $Object.next();
            } 

            $string = "{ EPAID: "+  $value[0] +  ", " + 
                     "Name: "+  $value[1] +  ", " +                
                     "City: "+  $value[4] +  ", " +
                     "State: "+  $value[6] +  ", " +
                     "ZipCode: "+  $value[8] +  ", " +  
                     "Latitude: "+  $value[14] +  ", " +
                     "Longitude: "+  $value[16] +  " }" ;          
            return $string;
        });

        f = fs.open("lat_long.txt", "a");
        f.write(str);
        f.close();
        processing = true;
        console.log("writing to file");
       phantom.exit();    

    });
 }


 // right here it should delay until the previous page is completed        
 //  while (!processing)  {    
 //       setTimeout(function(){ console.log("waiting....");},1000);
 //    }


};

}

console.log("finished all pages");

推荐答案

以下是最终有效的代码(使用超时方法，因为我无法使成功回调更好地工作).

Here is the code that ultimately works (using the timeout approach since I wasn't able to get the success callback to work better).

在安装了casperjs的情况下，我将此文件命名为"process.js"，并能够在命令行中将其作为"casperjs process.js"

With casperjs installed, I named this file "process.js" and was able to run it from the command line as "casperjs process.js"

var page = require('webpage').create();
var fs = require('fs');

page.onConsoleMessage = function(msg) {
    phantom.outputEncoding = "utf-8";
    console.log(msg);
};


// this overwrites the previous output f
 // this is the unique identifier for the locations. 
    var EPAID = ["NED981713837",... , "FLD049985302", "NJD986643153"]; 


f = fs.open("lat_long.txt", "w");
f.write("-<>-");
f.close();


var count = 0;
var target = 1400;
var written = [];

function yourFunction(){

   if (count < target) {

      process(count);
      count++;
      setTimeout(yourFunction, 5000);

   } else {
       console.log("exiting");
       phantom.exit();    
       return;
   }    
}




function process(counter){    

    var processing = false;

         console.log("Beginning record #" + counter); 

    //we construct the target url
    var url  = "http://iaspub.epa.gov/enviro/efsystemquery.cerclis?fac_search=site_epa_id&fac_value=" + EPAID[counter]  + "&fac_search_type=Beginning+With&postal_code=&location_address=&add_search_type=Beginning+With&city_name=&county_name=&state_code=&program_search=1&report=2&page_no=1&output_sql_switch=TRUE&database_type=CERCLIS" ;


    page.open(url);
    page.onLoadFinished = function(status) {
    if ( status === "success" ) {
        page.includeJs("http://ajax.googleapis.com/ajax/libs/jquery/1.6.1/jquery.min.js", function() {
            var str = page.evaluate(function() {                   
                $value = [];
                $Object = $(".result tr");                
              for (i =0 ; i < 10; i++) { 
                 $value.push($Object.find('td').html(),$Object.find('td').next().next().html() );          
                 $Object = $Object.next();
              } 

                $string = "{ \"EPAID\": \""+  $value[0] +  "\", " + 
                         "\"Name\": \""+  $value[1] +  "\", " +                
                         "\"City\": \""+  $value[4] +  "\", " +
                         "\"State\": \""+  $value[6] +  "\", " +
                         "\"ZipCode\": \""+  $value[8] +  "\", " +  
                         "\"Latitude\": "+  $value[14] +  ", " +
                         "\"Longitude\": "+  $value[16] +  " }," ;          
                return $string;
            });


           if (written[counter] === undefined) { 

             f = fs.open("lat_long.txt", "a");
             f.write(str);
             f.close();
             written[counter] = true;
             console.log("Writing to file #"+  counter);
           }  

        });
    }

    };
}

 console.log("Start...");

yourFunction();

这篇关于使用casperjs和phantomjs抓取多个页面的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！