node-webcrawler

node-webcrawler

node-webcrawler is originally a fork of node-crawler which aims to be the best crawling/scraping package for Node, sice the author won't mantain I will take care of it.

Features:

• server-side DOM & automatic jQuery insertion with Cheerio (default) or JSDOM
• Configurable pool size and retries
• Priority of requests
• forceUTF8 mode to let crawler deal for you with charset detection and conversion

Here is the CHANGELOG

Help & Forks welcomed!

How to install

$ npm install node-webcrawler

Crash course

var Crawler = require("node-webcrawler");
var url = require('url');
 
var c = new Crawler({
    maxConnections : 10,
    // This will be called for each crawled page
    callback : function (error, result, $) {
        // $ is Cheerio by default
        //a lean implementation of core jQuery designed specifically for the server
        if(error){
            console.log(error);
        }else{
            console.log($("title").text());
        }
    }
});
 
// Queue just one URL, with default callback
c.queue('http://www.amazon.com');
 
// Queue a list of URLs
c.queue(['http://www.google.com/','http://www.yahoo.com']);
 
// Queue URLs with custom callbacks & parameters
c.queue([{
    uri: 'http://parishackers.org/',
    jQuery: false,
 
    // The global callback won't be called
    callback: function (error, result) {
        if(error){
            console.log(error);
        }else{
            console.log('Grabbed', result.body.length, 'bytes');
        }
    }
}]);
 
// Queue some HTML code directly without grabbing (mostly for tests)
c.queue([{
    html: '<p>This is a <strong>test</strong></p>'
}]);

Work with bottleneck

Control rate limits for each connection, usually used with proxy.

var Crawler = require("node-webcrawler");
 
var c = new Crawler({
    maxConnections : 3,
    rateLimits:2000,
    callback : function (error, result, $) {
        if(error){
            console.error(error);
        }else{
            console.log($('title').text());
        }
    }
});
 
c.queue({
    uri:"http://www.google.com",
    limiter:"key1",// for connection of 'key1'
    proxy:"http://user:pass@127.0.0.1:8080"
});
 
c.queue({
    uri:"http://www.google.com",
    limiter:"key2", // for connection of 'key2'
    proxy:"http://user:pass@127.0.0.1:8082"
});
 
c.queue({
    uri:"http://www.google.com",
    limiter:"key3", // for connection of 'key3'
    proxy:"http://user:pass@127.0.0.1:8081"
});
 

Options reference

You can pass these options to the Crawler() constructor if you want them to be global or as items in the queue() calls if you want them to be specific to that item (overwriting global options)

This options list is a strict superset of mikeal's request options and will be directly passed to the request() method.

Basic request options:

• uri: String, the URL you want to crawl
• timeout : Number, in milliseconds (Default 15000)
• All mikeal's requests options are accepted

Callbacks:

• callback(error, result, $): A request was completed
• onDrain(pool): There is no more queued requests, call pool.destroyAllNow() if you wanna release resources in pool to, or if you have follow-up tasks to queue you can ignore.

Pool options:

• maxConnections: Number, Size of the worker pool (Default 10),
• priorityRange: Number, Range of acceptable priorities starting from 0 (Default 10),
• priority: Number, Priority of this request (Default 5),

Retry options:

• retries: Number of retries if the request fails (Default 3),
• retryTimeout: Number of milliseconds to wait before retrying (Default 10000),

Server-side DOM options:

• jQuery: true, false or ConfObject (Default true)

Charset encoding:

• forceUTF8: Boolean, if true will get charset from HTTP headers or meta tag in html and convert it to UTF8 if necessary. Never worry about encoding anymore! (Default false),
• incomingEncoding: String, with forceUTF8: true to set encoding manually (Default null) incomingEncoding : 'windows-1255' for example

Cache:

• cache: Boolean, if true stores requests' result in memory (Default false), not recommend if you are doing with huge amount of pages as the process will exhaust momery
• skipDuplicates: Boolean, if true skips URIs that were already crawled, without even calling callback() (Default false)

Other:

• rotateUA: Boolean, if true, userAgent should be an array, and rotate it (Default false)
• userAgent: String or Array, if rotateUA is false, but userAgent is array, will use first one.
• referer: String, if truthy sets the HTTP referer header
• rateLimits: Number of milliseconds to delay between each requests (Default 0)

Class:Crawler

Instance of Crawler

crawler.queue(uri|options)

• uri String, options is Options

crawler.queueSize

Size of queue, read-only

Working with Cheerio or JSDOM

Crawler by default use Cheerio instead of Jsdom. Jsdom is more robust but can be hard to install (espacially on windows) because of contextify. Which is why, if you want to use jsdom you will have to build it, and require('jsdom') in your own script before passing it to crawler. This is to avoid cheerio crawler user to build jsdom when installing crawler.

Working with Cheerio

jQuery: true //(default)
//OR
jQuery: 'cheerio'
//OR
jQuery: {
    name: 'cheerio',
    options: {
        normalizeWhitespace: true,
        xmlMode: true
    }
}

These parsing options are taken directly from htmlparser2, therefore any options that can be used in htmlparser2 are valid in cheerio as well. The default options are:

{
    normalizeWhitespace: false,
    xmlMode: false,
    decodeEntities: true
}

For a full list of options and their effects, see this and htmlparser2's options. source

Working with JSDOM

In order to work with JSDOM you will have to install it in your project folder npm install jsdom, deal with compiling C++ and pass it to crawler.

var jsdom = require('jsdom');
var Crawler = require('node-webcrawler');
 
var c = new Crawler({
    jQuery: jsdom
});

How to test

Install and run Httpbin

node-webcrawler use a local httpbin for testing purpose. You can install httpbin as a library from PyPI and run it as a WSGI app. For example, using Gunicorn:

$ pip install httpbin
// launch httpbin as a daemon with 6 worker on localhost
$ gunicorn httpbin:app -b 127.0.0.1:8000 -w 6 --daemon

// Finally
$ npm install && npm test

Alternative: Docker

After installing Docker, you can run:

// Builds the local test environment
$ docker build -t node-webcrawler .

// Runs tests
$ docker run node-webcrawler sh -c "gunicorn httpbin:app -b 127.0.0.1:8000 -w 6 --daemon && npm install && npm test"

// You can also ssh into the container for easier debugging
$ docker run -i -t node-webcrawler bash

Rough todolist

• Using bottleneck to deal with rate limits
• Introducing zombie to deal with page with complex ajax
• Refactoring the code to be more maintenable, it's spaghetti code in there !
• Proxy feature
• This issue: https://github.com/sylvinus/node-crawler/issues/118
• Make Sizzle tests pass (jsdom bug? https://github.com/tmpvar/jsdom/issues#issue/81)
• More crawling tests
• Document the API more (+ the result object)
• Option to wait for callback to finish before freeing the pool resource (via another callback like next())

ChangeLog

See https://github.com/bda-research/node-webcrawler/releases