src/Wallabag/CoreBundle/Helper/DownloadImages.php

   1 <?php
   2
   3 namespace Wallabag\CoreBundle\Helper;
   4
   5 use Psr\Log\LoggerInterface as Logger;
   6 use Symfony\Component\DomCrawler\Crawler;
   7
   8 define('REGENERATE_PICTURES_QUALITY', 75);
   9 define('HTTP_PORT', 80);
  10 define('SSL_PORT', 443);
  11 define('BASE_URL','');
  12
  13 class DownloadImages {
  14     private $folder;
  15     private $url;
  16     private $html;
  17     private $fileName;
  18     private $logger;
  19
  20     public function __construct($html, $url, Logger $logger) {
  21         $this->html = $html;
  22         $this->url = $url;
  23         $this->setFolder();
  24         $this->logger = $logger;
  25     }
  26
  27     public function setFolder($folder = "assets/images") {
  28         // if folder doesn't exist, attempt to create one and store the folder name in property $folder
  29         if(!file_exists($folder)) {
  30             mkdir($folder);
  31         }
  32         $this->folder = $folder;
  33     }
  34
  35     public function process() {
  36         //instantiate the symfony DomCrawler Component
  37         $crawler = new Crawler($this->html);
  38         // create an array of all scrapped image links
  39         $this->logger->log('debug', 'Finding images inside document');
  40         $result = $crawler
  41             ->filterXpath('//img')
  42             ->extract(array('src'));
  43
  44         // download and save the image to the folder
  45         foreach ($result as $image) {
  46             $file = file_get_contents($image);
  47
  48             // Checks
  49             $absolute_path = self::getAbsoluteLink($image, $this->url);
  50             $filename = basename(parse_url($absolute_path, PHP_URL_PATH));
  51             $fullpath = $this->folder."/".$filename;
  52             self::checks($file, $fullpath, $absolute_path);
  53             $this->html = str_replace($image, self::getPocheUrl() . '/' . $fullpath, $this->html);
  54         }
  55
  56         return $this->html;
  57     }
  58
  59     private function checks($rawdata, $fullpath, $absolute_path) {
  60         $fullpath = urldecode($fullpath);
  61
  62         if (file_exists($fullpath)) {
  63             unlink($fullpath);
  64         }
  65
  66         // check extension
  67         $this->logger->log('debug','Checking extension');
  68
  69         $file_ext = strrchr($fullpath, '.');
  70         $whitelist = array('.jpg', '.jpeg', '.gif', '.png');
  71         if (!(in_array($file_ext, $whitelist))) {
  72             $this->logger->log('debug','processed image with not allowed extension. Skipping '.$fullpath);
  73
  74             return false;
  75         }
  76
  77         // check headers
  78         $this->logger->log('debug','Checking headers');
  79         $imageinfo = getimagesize($absolute_path);
  80         if ($imageinfo['mime'] != 'image/gif' && $imageinfo['mime'] != 'image/jpeg' && $imageinfo['mime'] != 'image/jpg' && $imageinfo['mime'] != 'image/png') {
  81             $this->logger->log('debug','processed image with bad header. Skipping '.$fullpath);
  82
  83             return false;
  84         }
  85
  86         // regenerate image
  87         $this->logger->log('debug','regenerating image');
  88         $im = imagecreatefromstring($rawdata);
  89         if ($im === false) {
  90             $this->logger->log('error','error while regenerating image '.$fullpath);
  91
  92             return false;
  93         }
  94
  95         switch ($imageinfo['mime']) {
  96             case 'image/gif':
  97                 $result = imagegif($im, $fullpath);
  98                 $this->logger->log('debug','Re-creating gif');
  99                 break;
 100             case 'image/jpeg':
 101             case 'image/jpg':
 102                 $result = imagejpeg($im, $fullpath, REGENERATE_PICTURES_QUALITY);
 103                 $this->logger->log('debug','Re-creating jpg');
 104                 break;
 105             case 'image/png':
 106                 $this->logger->log('debug','Re-creating png');
 107                 $result = imagepng($im, $fullpath, ceil(REGENERATE_PICTURES_QUALITY / 100 * 9));
 108                 break;
 109         }
 110         imagedestroy($im);
 111
 112         return $result;
 113     }
 114
 115     private static function getAbsoluteLink($relativeLink, $url)
 116     {
 117         /* return if already absolute URL */
 118         if (parse_url($relativeLink, PHP_URL_SCHEME) != '') {
 119             return $relativeLink;
 120         }
 121
 122         /* queries and anchors */
 123         if ($relativeLink[0] == '#' || $relativeLink[0] == '?') {
 124             return $url.$relativeLink;
 125         }
 126
 127         /* parse base URL and convert to local variables:
 128            $scheme, $host, $path */
 129         extract(parse_url($url));
 130
 131         /* remove non-directory element from path */
 132         $path = preg_replace('#/[^/]*$#', '', $path);
 133
 134         /* destroy path if relative url points to root */
 135         if ($relativeLink[0] == '/') {
 136             $path = '';
 137         }
 138
 139         /* dirty absolute URL */
 140         $abs = $host.$path.'/'.$relativeLink;
 141
 142         /* replace '//' or '/./' or '/foo/../' with '/' */
 143         $re = array('#(/\.?/)#', '#/(?!\.\.)[^/]+/\.\./#');
 144         for ($n = 1; $n > 0; $abs = preg_replace($re, '/', $abs, -1, $n)) {
 145         }
 146
 147         /* absolute URL is ready! */
 148         return $scheme.'://'.$abs;
 149     }
 150
 151     public static function getPocheUrl()
 152     {
 153         $baseUrl = "";
 154         $https = (!empty($_SERVER['HTTPS'])
 155                     && (strtolower($_SERVER['HTTPS']) == 'on'))
 156             || (isset($_SERVER["SERVER_PORT"])
 157                     && $_SERVER["SERVER_PORT"] == '443') // HTTPS detection.
 158             || (isset($_SERVER["SERVER_PORT"]) //Custom HTTPS port detection
 159                     && $_SERVER["SERVER_PORT"] == SSL_PORT)
 160              || (isset($_SERVER['HTTP_X_FORWARDED_PROTO'])
 161                     && $_SERVER['HTTP_X_FORWARDED_PROTO'] == 'https');
 162         $serverport = (!isset($_SERVER["SERVER_PORT"])
 163             || $_SERVER["SERVER_PORT"] == '80'
 164             || $_SERVER["SERVER_PORT"] == HTTP_PORT
 165             || ($https && $_SERVER["SERVER_PORT"] == '443')
 166             || ($https && $_SERVER["SERVER_PORT"]==SSL_PORT) //Custom HTTPS port detection
 167             ? '' : ':' . $_SERVER["SERVER_PORT"]);
 168
 169         if (isset($_SERVER["HTTP_X_FORWARDED_PORT"])) {
 170             $serverport = ':' . $_SERVER["HTTP_X_FORWARDED_PORT"];
 171         }
 172         // $scriptname = str_replace('/index.php', '/', $_SERVER["SCRIPT_NAME"]);
 173         // if (!isset($_SERVER["HTTP_HOST"])) {
 174         //     return $scriptname;
 175         // }
 176         $host = (isset($_SERVER['HTTP_X_FORWARDED_HOST']) ? $_SERVER['HTTP_X_FORWARDED_HOST'] : (isset($_SERVER['HTTP_HOST']) ? $_SERVER['HTTP_HOST'] : $_SERVER['SERVER_NAME']));
 177         if (strpos($host, ':') !== false) {
 178             $serverport = '';
 179         }
 180         // check if BASE_URL is configured
 181         if(BASE_URL) {
 182             $baseUrl = BASE_URL;
 183         } else {
 184             $baseUrl = 'http' . ($https ? 's' : '') . '://' . $host . $serverport;
 185         }
 186     return $baseUrl;
 187
 188     }
 189 }