php/src/gadgets/rewrite/ContentRewriter.php

   1 <?php
   2
   3 /**
   4  * Licensed to the Apache Software Foundation (ASF) under one
   5  * or more contributor license agreements.  See the NOTICE file
   6  * distributed with this work for additional information
   7  * regarding copyright ownership.  The ASF licenses this file
   8  * to you under the Apache License, Version 2.0 (the
   9  * "License"); you may not use this file except in compliance
  10  * with the License.  You may obtain a copy of the License at
  11  *
  12  *     http://www.apache.org/licenses/LICENSE-2.0
  13  *
  14  * Unless required by applicable law or agreed to in writing,
  15  * software distributed under the License is distributed on an
  16  * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
  17  * KIND, either express or implied.  See the License for the
  18  * specific language governing permissions and limitations
  19  * under the License.
  20  */
  21
  22 /**
  23  * Implements the Content-Rewrite feature which rewrites all image, css and script
  24  * links to their proxied versions, which can be quite a latency improvement, and
  25  * save the gadget dev's server from melting down
  26  *
  27  */
  28 class ContentRewriter extends DomRewriter {
  29   private $rewrite;
  30   private $baseUrl;
  31   private $defaultRewrite = array('include-url' => array('*'), 'exclude-url' => array(), 'refresh' => '86400');
  32
  33   public function __construct(GadgetContext $context, Gadget &$gadget) {
  34     parent::__construct($context, $gadget);
  35     // if no rewrite params are set in the gadget but rewrite_by_default is on, use our default rules (rewrite all)
  36     if (! isset($gadget->gadgetSpec->rewrite) && Config::get('rewrite_by_default')) {
  37       $this->rewrite = $this->defaultRewrite;
  38     } else {
  39       $this->rewrite = $gadget->gadgetSpec->rewrite;
  40     }
  41     // the base url of the gadget is used for relative paths
  42     $this->baseUrl = substr($this->context->getUrl(), 0, strrpos($this->context->getUrl(), '/') + 1);
  43   }
  44
  45   /**
  46    * Register our dom node observers
  47    *
  48    * @param GadgetRewriter $gadgetRewriter
  49    */
  50   public function register(GadgetRewriter &$gadgetRewriter) {
  51     $gadgetRewriter->addObserver('img', $this, 'rewriteImage');
  52     $gadgetRewriter->addObserver('style', $this, 'rewriteStyle');
  53     $gadgetRewriter->addObserver('script', $this, 'rewriteScript');
  54     $gadgetRewriter->addObserver('link', $this, 'rewriteStyleLink');
  55   }
  56
  57   /**
  58    * Produces the proxied version of a URL if it falls within the content-rewrite params and
  59    * will append a refresh param to the proxied url based on the expires param, and the gadget
  60    * url so that the proxy server knows to rewrite it's content or not
  61    *
  62    * @param string $url
  63    */
  64   private function getProxyUrl($url) {
  65     if (strpos(strtolower($url), 'http://') === false && strpos(strtolower($url), 'https://') === false) {
  66       $url = $this->baseUrl . $url;
  67     }
  68     $url = Config::get('web_prefix') . '/gadgets/proxy?url=' . urlencode($url);
  69     $url .= '&refresh=' . (isset($this->rewrite['expires']) && is_numeric($this->rewrite['expires']) ? $this->rewrite['expires'] : '3600');
  70     $url .= '&gadget=' . urlencode($this->context->getUrl());
  71     return $url;
  72   }
  73
  74   /**
  75    * Checks the URL against the include-url and exclude-url params
  76    *
  77    * @param string $url
  78    */
  79   private function includedUrl($url) {
  80     $included = $excluded = false;
  81     if (isset($this->rewrite['include-url'])) {
  82       foreach ($this->rewrite['include-url'] as $includeUrl) {
  83         if ($includeUrl == '*' || strpos($url, $includeUrl) !== false) {
  84           $included = true;
  85           break;
  86         }
  87       }
  88     }
  89     if (isset($this->rewrite['exclude-url'])) {
  90       foreach ($this->rewrite['exclude-url'] as $excludeUrl) {
  91         if ($excludeUrl == '*' || strpos($url, $excludeUrl) !== false) {
  92           $excluded = true;
  93           break;
  94         }
  95       }
  96     }
  97     return ($included && ! $excluded);
  98   }
  99
 100   /**
 101    * Rewrites the src attribute of an img tag
 102    *
 103    * @param DOMElement $node
 104    */
 105   public function rewriteImage(DOMElement &$node) {
 106     if (($src = $node->getAttribute('src')) != null && $this->includedUrl($src)) {
 107       $node->setAttribute('src', $this->getProxyUrl($src));
 108     }
 109   }
 110
 111   /**
 112    * Tries to find url(<url tag>) constructs and rewrite them to their
 113    * proxied counterparts
 114    *
 115    * @param DOMElement $node
 116    */
 117   public function rewriteStyle(DOMElement &$node) {
 118     $content = $node->nodeValue;
 119     $newVal = '';
 120     // loop through the url elements in the content
 121     while (($pos = strpos($content, 'url')) !== false) {
 122       // output everything before this url tag
 123       $newVal .= substr($content, 0, $pos + 3);
 124       $content = substr($content, $pos + 3);
 125       // low tech protection against miss-reading tags, if the open ( is to far away, this is probabbly a miss-read
 126       if (($beginTag = strpos($content, '(')) < 4) {
 127         $content = substr($content, $beginTag + 1);
 128         $endTag = strpos($content, ')');
 129         $tag = str_replace(array("'", "\""), '', trim(substr($content, 0, $endTag)));
 130         // at this point $tag should be the actual url aka: http://example.org/bar/foo.gif
 131         if ($this->includedUrl($tag)) {
 132           $newVal .= "('" . $this->getProxyUrl($tag) . "')";
 133         } else {
 134           $newVal .= "('$tag')";
 135         }
 136         $content = substr($content, $endTag + 1);
 137       }
 138     }
 139     // append what's left
 140     $newVal .= $content;
 141     $node->nodeValue = $newVal;
 142   }
 143
 144   public function rewriteScript(DOMElement &$node) {
 145     if (($src = $node->getAttribute('src')) != null && $this->includedUrl($src)) {
 146       // make sure not to rewrite our forcedJsLibs src tag, else things break
 147       if (strpos($src, '/gadgets/js') === false) {
 148         $node->setAttribute('src', $this->getProxyUrl($src));
 149       }
 150     }
 151   }
 152
 153   public function rewriteStyleLink(DOMElement &$node) {
 154     if (($src = $node->getAttribute('href')) != null && $this->includedUrl($src)) {
 155       $node->setAttribute('href', $this->getProxyUrl($src));
 156     }
 157   }
 158 }