README

Spider

Spider 是一个简单、优雅、可扩展的基于 phpQuery 的 PHP 网络爬虫。

功能

• 拥有与 jQuery 相同的 CSS3 DOM 选择器
• 拥有与 jQuery 相同的 DOM 操作 API
• 拥有通用的列表爬虫程序
• 拥有强大的 HTTP 请求套件，易于实现诸如模拟登录、伪造浏览器、HTTP 代理等复杂网络请求
• 拥有混乱的代码解决方案
• 拥有强大的内容过滤功能，你可以使用 jQuery 选择器过滤内容
• 具有高度模块化设计，可扩展性和强大
• 拥有表达式丰富的 API
• 拥有丰富的插件

通过插件，你可以轻松实现以下功能

• 多线程爬取
• 爬取 JavaScript 动态渲染页面（PhantomJS/无头 WebKit）
• 图片下载到本地
• 模拟浏览器行为，如提交表单
• 网络爬虫
• .....

要求

• PHP >= 7.1

安装

通过 Composer 安装

composer require laoqianjunzi/spider

使用

DOM 遍历和操作

• 爬取「GitHub」所有图片链接

Spider::get('https://github.com')->find('img')->attrs('src');

• 爬取 Google 搜索结果

$ql = Spider::get('https://www.google.co.jp/search?q=Spider');

$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
});

• 更多使用方法

$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')->...

列表爬取

爬取 Google 搜索结果列表的标题和链接

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

print_r($data->all());

结果

Array
(
    [0] => Array
        (
            [title] => Angular - Spider
            [link] => https://angular.io/api/core/Spider
        )
    [1] => Array
        (
            [title] => Spider | @angular/core - Angularリファレンス - Web Creative Park
            [link] => http://www.webcreativepark.net/angular/Spider/
        )
    [2] => Array
        (
            [title] => Spider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/Spider_query_add_subscribe/
        )
        //...
)

编码转换

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

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

HTTP 客户端（GuzzleHttp）

• 携带 Cookie 登录 GitHub

  //Crawl GitHub content
  $ql = Spider::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;
  

• 模拟登录

  // Post login
  $ql = Spider::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();
  

提交表单

登录 GitHub

// Get the Spider instance
$ql = Spider::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 !';
}

绑定函数扩展

自定义 myHttp 方法的扩展

$ql = Spider::getInstance();

//Bind a `myHttp` method to the Spider object
$ql->bind('myHttp',function ($url){
	// $this is the current Spider 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());

或打包成类，然后绑定

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

• 使用 CURL 多线程插件，多线程爬取 GitHub 趋势

$ql = Spider::use(CurlMulti::class);
$ql->curlMulti([
    'https://github.com/trending/php',
    'https://github.com/trending/go',
    //.....more urls
])
 // Called if task is success
 ->success(function (Spider $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,
]);