Last active
December 22, 2015 03:59
-
-
Save julp/6414365 to your computer and use it in GitHub Desktop.
[PHP] Truncate HTML text to a given length
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| <?php | |
| var_dump( | |
| html_cut('<b>12<a href="#">34<i>56</i></a></b><u>78</u>', 3), | |
| // html_cut('<b>1é<a href="#">34<i>56</i></a></b><u>78</u>', 3), | |
| html_cut('<b>1<![CDATA[é]]><a href="#">34<i>56</i></a></b><u>78</u>', 3), | |
| NULL | |
| ); | |
| function html_cut(/* UTF-8 */ $string, $length) | |
| { | |
| $dom = new DomDocument; | |
| // Create fake root which will wrap our HTML code | |
| // $dom->loadHTML('<html><body/></html>', LIBXML_NOBLANKS | LIBXML_NOENT); // with s/$dom->documentElement/$dom->getElementsByTagName('body')->item(0)/ | |
| $dom->appendChild($dom->createElement('span')); | |
| $dom->documentElement->setAttribute('class', 'root'); | |
| $dom->preserveWhiteSpace = FALSE; // no effect ? | |
| // Import HTML into document | |
| $fragment = $dom->createDocumentFragment(); | |
| $fragment->appendXML($string); # html_entity_decode($string, ENT_NOQUOTES/* | ENT_HTML5*/, 'UTF-8'), ENT_NOQUOTES, 'UTF-8') => UNSAFE (XSS) !!! | |
| $dom->documentElement->appendChild($fragment); | |
| $stack = array(); | |
| $t = function ($node) use (&$stack, &$t) { | |
| if ($node->hasChildNodes()) { | |
| foreach ($node->childNodes as $child) { | |
| $t($child); | |
| } | |
| } | |
| $stack[] = $node; | |
| }; | |
| $t($dom->documentElement); | |
| $count = 0; | |
| $done = FALSE; | |
| $parents = array($dom->documentElement); | |
| foreach ($stack as $node) { | |
| if ($done) { | |
| $skip = FALSE; | |
| foreach ($parents as $n) { | |
| if ($n->isSameNode($node)) { | |
| $skip = TRUE; | |
| break; | |
| } | |
| } | |
| if (!$skip) { | |
| try { | |
| $node->parentNode->removeChild($node); | |
| } catch (Exception $e) { | |
| /* NOP */ | |
| } | |
| } | |
| } else { | |
| if (in_array($node->nodeType, array(XML_TEXT_NODE, XML_CDATA_SECTION_NODE))) { | |
| if ($count + mb_strlen($node->nodeValue, 'UTF-8') >= $length) { | |
| $done = TRUE; | |
| $node->nodeValue = mb_substr($node->nodeValue, 0, $length - $count); | |
| $n = $node->parentNode; | |
| while (!$dom->documentElement->isSameNode($n)) { | |
| $parents[] = $n; | |
| $n = $n->parentNode; | |
| } | |
| } else { | |
| $count += mb_strlen($node->nodeValue, 'UTF-8'); | |
| } | |
| } | |
| } | |
| } | |
| // Unwrap | |
| $output = ''; | |
| foreach ($dom->documentElement->childNodes as $child) { | |
| $output .= $dom->saveXML($child); | |
| } | |
| return $output; | |
| } |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment