Adapters
8 years ago
Providers
8 years ago
Autoload.php
8 years ago
Embera.php
8 years ago
FakeResponse.php
8 years ago
Formatter.php
8 years ago
HtmlProcessor.php
8 years ago
HttpRequest.php
8 years ago
Oembed.php
8 years ago
Providers.php
8 years ago
Url.php
8 years ago
HtmlProcessor.php
136 lines
| 1 | <?php |
| 2 | /** |
| 3 | * HtmlProcessor.php |
| 4 | * |
| 5 | * @package Embera |
| 6 | * @author Michael Pratt <pratt@hablarmierda.net> |
| 7 | * @link http://www.michael-pratt.com/ |
| 8 | * |
| 9 | * For the full copyright and license information, please view the LICENSE |
| 10 | * file that was distributed with this source code. |
| 11 | * |
| 12 | */ |
| 13 | namespace Embera; |
| 14 | |
| 15 | /** |
| 16 | * This Class is in charge of replacing content, ignoring specific |
| 17 | * html tags. |
| 18 | */ |
| 19 | class HtmlProcessor |
| 20 | { |
| 21 | /** @var array with tags that should be ignored */ |
| 22 | protected $tags = array(); |
| 23 | |
| 24 | /** @var array With url to html conversion code */ |
| 25 | protected $replacementTable = array(); |
| 26 | |
| 27 | /** @var array with placeholders to be reserved/restored */ |
| 28 | protected $holder = array(); |
| 29 | |
| 30 | /** |
| 31 | * Constructor |
| 32 | * |
| 33 | * @param array $tags The tags that should be ignored |
| 34 | * @param array $replacementTable Array with the conversions that should be made |
| 35 | * Normally in the url -> html mode. |
| 36 | * @return void |
| 37 | */ |
| 38 | public function __construct(array $tags, array $replacementTable) |
| 39 | { |
| 40 | $this->replacementTable = $replacementTable; |
| 41 | $this->tags = $tags; |
| 42 | } |
| 43 | |
| 44 | /** |
| 45 | * Process the content and does the replacements |
| 46 | * |
| 47 | * @param string $content |
| 48 | * @return string |
| 49 | */ |
| 50 | public function process($content) |
| 51 | { |
| 52 | foreach ($this->tags as $t) { |
| 53 | $content = $this->reserveTag($content, $t); |
| 54 | } |
| 55 | |
| 56 | $content = strtr($content, $this->replacementTable); |
| 57 | return $this->restoreHolders($content); |
| 58 | } |
| 59 | |
| 60 | /** |
| 61 | * Reserves a tag, which in this case, all it does is replace the html tag |
| 62 | * with a placeholder. |
| 63 | * |
| 64 | * @param string $content |
| 65 | * @param string $tag |
| 66 | * @return string |
| 67 | */ |
| 68 | protected function reserveTag($content, $tag) |
| 69 | { |
| 70 | if (stripos($content, '<' . $tag) !== false) { |
| 71 | |
| 72 | if (in_array(strtolower($tag), array('img'))) { |
| 73 | $pattern = '~(<' . preg_quote($tag, '~') . '[^>]+\>)~i'; |
| 74 | } else { |
| 75 | $pattern = '~(<' . preg_quote($tag, '~') . '\\b[^>]*?>[\\s\\S]*?<\\/' . preg_quote($tag, '~') . '>)~i'; |
| 76 | } |
| 77 | |
| 78 | return preg_replace_callback($pattern, array($this, 'placeHolder'), $content); |
| 79 | } |
| 80 | |
| 81 | return $content; |
| 82 | } |
| 83 | |
| 84 | /** |
| 85 | * Assigns a placeholder to the matched content. Its used by preg_replace_callback |
| 86 | * |
| 87 | * @param array $matches |
| 88 | * @return string |
| 89 | */ |
| 90 | protected function placeHolder(array $matches) |
| 91 | { |
| 92 | $placeholder = '%{{' . md5(time() . count($this->holder) . mt_rand(0, 500)) . '}}%'; |
| 93 | $this->holder[$placeholder] = $matches['1']; |
| 94 | |
| 95 | return $placeholder; |
| 96 | } |
| 97 | |
| 98 | /** |
| 99 | * Restores all the dummy placeholders with the original content |
| 100 | * |
| 101 | * @param string $content |
| 102 | * @return string |
| 103 | */ |
| 104 | protected function restoreHolders($content) |
| 105 | { |
| 106 | if (!empty($this->holder)) { |
| 107 | |
| 108 | /** |
| 109 | * Now the question that you might be asking yourself is, why use a 'for' loop instead |
| 110 | * of the plain 'str_replace' function once? |
| 111 | * |
| 112 | * Well, since html can be nested, there might be cases were a placeholder was placed inside |
| 113 | * another placeholder and if we do just one 'str_replace', we might not get the full original |
| 114 | * content. |
| 115 | * |
| 116 | * In order to ensure that no data is lost, we need to run 'str_replace' for each time a tag |
| 117 | * was registered in the constructor. That is the worst case scenario, most of the times, the loop |
| 118 | * is just going to run once and repeat the cycle only if there are signs that another placeholder is in |
| 119 | * the content. |
| 120 | */ |
| 121 | $count = count($this->tags) + 1; |
| 122 | for ($i=0; $i < $count; $i++) { |
| 123 | $content = str_replace(array_keys($this->holder), array_values($this->holder), $content); |
| 124 | if (strpos($content, '%{{') === false) { |
| 125 | break; |
| 126 | } |
| 127 | } |
| 128 | |
| 129 | } |
| 130 | |
| 131 | return $content; |
| 132 | } |
| 133 | } |
| 134 | |
| 135 | ?> |
| 136 |