note 103925 deleted from function.substr by danbrown

Note Submitter: frederic dot leroy at adeqweb dot com ---- Maybe this can help a few people around, this is a slight evolution of the mb_substrws function posted hereafter. Many thanks to kueb, erez and benny /** * Function to return $length first characters from html buffer. HTML tags are not taken into account in the char count and preserved * The function expects W3C well formed html, especially regarding self closing tags that MUST have a space before the closing slash : * For example, a '<br/>' tag or '<img src=""/>' tag will not be correctly interpreted, function expects proper '<br />' and '<img src="" />' (with spaces) * * @param string $html: The html to truncate * @param integer $length : The number of characters to keep in summary (default 250) * @return string The trunctaed HTML */ public static function html_summary($html, $length = 250) { if ((mb_strlen($html) > $length)) { $explodedText = preg_split('/(<[^>]*>| )/', $html, -1, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY); $html = ''; list($index, $word) = each($explodedText); while ($word && $length > 0) { //add this word $html.=$word; //only count chars if not a tag if (!preg_match('/<[^>]*>/', $word)) { $length-=mb_strlen($word); } //get next word list($index, $word) = each($explodedText); } // close unclosed html tags if (preg_match_all("/(<([\w]+) [^>]*[^\/]>|<([a-zA-Z]+)>)/", $html, $aBuffer)) { $openingTags = array(); foreach ($aBuffer[2] as $index => $tagWithOptions) { if (!$tagWithOptions) { $openingTags[] = $aBuffer[3][$index]; } else { $openingTags[] = $tagWithOptions; } } if (!empty($aBuffer[1])) { preg_match_all("|</([a-zA-Z]+)>|", $html, $aBuffer2); $closingTags = $aBuffer2[1]; if (count($openingTags) != count($closingTags)) { $tagsToClose = array_diff($openingTags, $closingTags); $tagsToClose = array_reverse($tagsToClose); foreach ($tagsToClose as $tag) { $html .= '</' . $tag . '>'; } } } } } return $html; }