130

GitHub - jae-jae/QueryList: QueryList是基于phpQuery的无比强大的PHP采集工具

 6 years ago
source link: https://github.com/jae-jae/QueryList
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.

QueryList

QueryList is a simple, elegant, extensible PHP Web Scraper (crawler/spider) ,based on phpQuery.

API Documentation

中文文档

Features

  • Have the same CSS3 DOM selector as jQuery
  • Have the same DOM manipulation API as jQuery
  • Have a generic list crawling program
  • Have a strong HTTP request suite, easy to achieve such as: simulated landing, forged browser, HTTP proxy and other complex network requests
  • Have a messy code solution
  • Have powerful content filtering, you can use the jQuey selector to filter content
  • Has a high degree of modular design, scalability and strong
  • Have an expressive API
  • Has a wealth of plug-ins

Through plug-ins you can easily implement things like:

  • Multithreaded crawl
  • Crawl JavaScript dynamic rendering page (PhantomJS/headless WebKit)
  • Image downloads to local
  • Simulate browser behavior such as submitting Form forms
  • Web crawler
  • .....

Requirements

  • PHP >= 7.1

Installation

By Composer installation:

composer require jaeger/querylist

Usage

DOM Traversal and Manipulation

  • Crawl「GitHub」all picture links
QueryList::get('https://github.com')->find('img')->attrs('src');
  • Crawl Google search results
$ql = QueryList::get('https://www.google.co.jp/search?q=QueryList');

$ql->find('title')->text(); //The page title
$ql->find('meta[name=keywords]')->content; //The page keywords

$ql->find('h3>a')->texts(); //Get a list of search results titles
$ql->find('h3>a')->attrs('href'); //Get a list of search results links

$ql->find('img')->src; //Gets the link address of the first image
$ql->find('img:eq(1)')->src; //Gets the link address of the second image
$ql->find('img')->eq(2)->src; //Gets the link address of the third image
// Loop all the images
$ql->find('img')->map(function($img){
	echo $img->alt;  //Print the alt attribute of the image
});
  • More usage
$ql->find('#head')->append('<div>Append content</div>')->find('div')->htmls();
$ql->find('.two')->children('img')->attrs('alt'); // Get the class is the "two" element under all img child nodes
// Loop class is the "two" element under all child nodes
$data = $ql->find('.two')->children()->map(function ($item){
    // Use "is" to determine the node type
    if($item->is('a')){
        return $item->text();
    }elseif($item->is('img'))
    {
        return $item->alt;
    }
});

$ql->find('a')->attr('href', 'newVal')->removeClass('className')->html('newHtml')->...
$ql->find('div > p')->add('div > ul')->filter(':has(a)')->find('p:first')->nextAll()->andSelf()->...
$ql->find('div.old')->replaceWith( $ql->find('div.new')->clone())->appendTo('.trash')->prepend('Deleted')->...

List crawl

Crawl the title and link of the Google search results list:

$data = QueryList::get('https://www.google.co.jp/search?q=QueryList')
	// Set the crawl rules
    ->rules([ 
	    'title'=>array('h3','text'),
	    'link'=>array('h3>a','href')
	])
	->query()->getData();

print_r($data->all());

Results:

Array
(
    [0] => Array
        (
            [title] => Angular - QueryList
            [link] => https://angular.io/api/core/QueryList
        )
    [1] => Array
        (
            [title] => QueryList | @angular/core - Angularリファレンス - Web Creative Park
            [link] => http://www.webcreativepark.net/angular/querylist/
        )
    [2] => Array
        (
            [title] => QueryListにQueryを追加したり、追加されたことを感知する | TIPS ...
            [link] => http://www.webcreativepark.net/angular/querylist_query_add_subscribe/
        )
        //...
)

Encode convert

// Out charset :UTF-8
// In charset :GB2312
QueryList::get('https://top.etao.com')->encoding('UTF-8','GB2312')->find('a')->texts();

// Out charset:UTF-8
// In charset:Automatic Identification
QueryList::get('https://top.etao.com')->encoding('UTF-8')->find('a')->texts();

HTTP Client (GuzzleHttp)

  • Carry cookie login GitHub
//Crawl GitHub content
$ql = QueryList::get('https://github.com','param1=testvalue & params2=somevalue',[
  'headers' => [
      // Fill in the cookie from the browser
      'Cookie' => 'SINAGLOBAL=546064; wb_cmtLike_2112031=1; wvr=6;....'
  ]
]);
//echo $ql->getHtml();
$userName = $ql->find('.header-nav-current-user>.css-truncate-target')->text();
echo $userName;
  • Use the Http proxy
$urlParams = ['param1' => 'testvalue','params2' => 'somevalue'];
$opts = [
	// Set the http proxy
    'proxy' => 'http://222.141.11.17:8118',
    //Set the timeout time in seconds
    'timeout' => 30,
     // Fake HTTP headers
    'headers' => [
        'Referer' => 'https://querylist.cc/',
        'User-Agent' => 'testing/1.0',
        'Accept'     => 'application/json',
        'X-Foo'      => ['Bar', 'Baz'],
        'Cookie'    => 'abc=111;xxx=222'
    ]
];
$ql->get('http://httpbin.org/get',$urlParams,$opts);
// echo $ql->getHtml();
  • Analog login
// Post login
$ql = QueryList::post('http://xxxx.com/login',[
    'username' => 'admin',
    'password' => '123456'
])->get('http://xxx.com/admin');
// Crawl pages that need to be logged in to access
$ql->get('http://xxx.com/admin/page');
//echo $ql->getHtml();

Submit forms

Login GitHub

// Get the QueryList instance
$ql = QueryList::getInstance();
// Get the login form
$form = $ql->get('https://github.com/login')->find('form');

// Fill in the GitHub username and password
$form->find('input[name=login]')->val('your github username or email');
$form->find('input[name=password]')->val('your github password');

// Serialize the form data
$fromData = $form->serializeArray();
$postData = [];
foreach ($fromData as $item) {
    $postData[$item['name']] = $item['value'];
}

// Submit the login form
$actionUrl = 'https://github.com'.$form->attr('action');
$ql->post($actionUrl,$postData);
// To determine whether the login is successful
// echo $ql->getHtml();
$userName = $ql->find('.header-nav-current-user>.css-truncate-target')->text();
if($userName)
{
    echo 'Login successful ! Welcome:'.$userName;
}else{
    echo 'Login failed !';
}

Bind function extension

Customize the extension of a myHttp method:

$ql = QueryList::getInstance();

//Bind a `myHttp` method to the QueryList object
$ql->bind('myHttp',function ($url){
	// $this is the current QueryList object
    $html = file_get_contents($url);
    $this->setHtml($html);
    return $this;
});

// And then you can call by the name of the binding
$data = $ql->myHttp('https://toutiao.io')->find('h3 a')->texts();
print_r($data->all());

Or package to class, and then bind:

$ql->bind('myHttp',function ($url){
    return new MyHttp($this,$url);
});

Plugin used

  • Use the PhantomJS plugin to crawl JavaScript dynamically rendered pages:
// Set the PhantomJS binary file path during installation
$ql = QueryList::use(PhantomJs::class,'/usr/local/bin/phantomjs');

// Crawl「500px」all picture links
$data = $ql->browser('https://500px.com/editors')->find('img')->attrs('src');
print_r($data->all());

// Use the HTTP proxy
$ql->browser('https://500px.com/editors',false,[
	'--proxy' => '192.168.1.42:8080',
    '--proxy-type' => 'http'
])
  • Using the CURL multithreading plug-in, multi-threaded crawling GitHub trending :
$ql = QueryList::use(CurlMulti::class);
$ql->curlMulti([
    'https://github.com/trending/php',
    'https://github.com/trending/go',
    //.....more urls
])
 // Called if task is success
 ->success(function (QueryList $ql,CurlMulti $curl,$r){
    echo "Current url:{$r['info']['url']} \r\n";
    $data = $ql->find('h3 a')->texts();
    print_r($data->all());
})
 // Task fail callback
->error(function ($errorInfo,CurlMulti $curl){
    echo "Current url:{$errorInfo['info']['url']} \r\n";
    print_r($errorInfo['error']);
})
->start([
	// Maximum number of threads
    'maxThread' => 10,
    // Number of error retries
    'maxTry' => 3,
]);

Plugins

View more QueryList plugins and QueryList-based products: QueryList Community

Contributing

Welcome to contribute code for the QueryList。About Contributing Plugins can be viewed:QueryList Plugin Contributing Guide

Author

Jaeger [email protected]

If this library is useful for you, say thanks !

Lisence

QueryList is licensed under the license of MIT. See the LICENSE for more details.


About Joyk


Aggregate valuable and interesting links.
Joyk means Joy of geeK