Your IP : 216.73.216.182
<?php
/**
* @package Komento
* @copyright Copyright (C) Stack Ideas Sdn Bhd. All rights reserved.
* @license GNU/GPL, see LICENSE.php
* Komento is free software. This version may have been modified pursuant
* to the GNU General Public License, and as distributed it includes or
* is derivative of works licensed under the GNU General Public License or
* other free or open source software licenses.
* See COPYRIGHT.php for copyright notices and details.
*/
defined('_JEXEC') or die('Unauthorized Access');
use Foundry\Libraries\VideoParser;
class KomentoParser
{
protected $emojiPath = [
'path' => '',
'originalPath' => ''
];
protected $emoticons = [];
protected $emojiMap = [];
public $config = null;
public function __construct()
{
$this->config = KT::config();
}
/**
* Remove the BBcodes which has no value with empty string
*
* @since 4.0
* @access public
*/
public function removeEmptyBBcodes($comment)
{
$bbcodes = [
'/\[b\]\[\/b\]/ms',
'/\[i\]\[\/i\]/ms',
'/\[u\]\[\/u\]/ms',
'/\[img\]\[\/img\]/ms',
'/\[quote\]\[\/quote\]/ms',
'/\[spoiler\]\[\/spoiler\]/ms',
'/\[code type=\"(.*?)\"\]\[\/code\]/ms'
];
$comment = preg_replace($bbcodes, '', $comment);
return $comment;
}
public function parseComment($text)
{
// Remove the empty bbcodes first
$text = $this->removeEmptyBBcodes($text);
// word censoring
$text = $this->parseCensor($text);
// parseBBcode to HTML
if ($this->config->get('enable_bbcode')) {
$text = $this->parseBBCode($text);
}
return $text;
}
public function parseBBCode($text)
{
$maxdimension = ' style="max-width:100%"';
if (!$this->config->get('enable_media_max_width') && ($this->config->get('max_image_width') || $this->config->get('max_image_height'))) {
$maxdimension = ' style="';
if ($this->config->get('max_image_width')) {
$maxdimension .= 'max-width:' . $this->config->get('max_image_width') . 'px;';
}
if ($this->config->get('max_image_height')) {
$maxdimension .= 'max-height:' . $this->config->get('max_image_height') . 'px;';
}
$maxdimension .= '"';
}
// Converts all html entities properly
$text = htmlspecialchars($text , ENT_NOQUOTES);
$text = trim($text);
// Replace [code] blocks
if ($this->config->get('bbcode_code')) {
$text = $this->replaceCodes($text);
}
if ($this->config->get('bbcode_giphy')) {
$text = $this->replaceGiphy($text);
}
if ($this->config->get('bbcode_quote')) {
$text = $this->replaceQuote($text);
}
$text = preg_replace_callback('/\[code(type="(.*?)")?\](.*?)\[\/code\]/ms', ['KomentoParser', 'escape'], $text);
// avoid smileys in pre tag gets replaced
$text = $this->encodePre($text);
// BBCode to find...
$in = [
'/\[b\](.*?)\[\/b\]/ms',
'/\[i\](.*?)\[\/i\]/ms',
'/\[u\](.*?)\[\/u\]/ms',
'/\[img\]((http|https):\/\/([a-z0-9\%._\s\*_\/+-?]+)\.(jpg|JPG|jpeg|JPEG|png|PNG|gif|GIF).*?)\[\/img]/ims',
'/\[s\](.*?)\[\/s\]/ims'
];
// And replace them by...
$out = [
'<b>\1</b>',
'<em>\1</em>',
'<u>\1</u>',
'<img src="\1" alt="\1"' . $maxdimension . ' />',
'<del>\1</del>'
];
// strip out bbcode data first
$tmp = preg_replace($in, '', $text);
// strip out bbcode url data
$urlin = '/\[url\="?(.*?)"?\](.*?)\[\/url\]/ms';
$tmp = preg_replace($urlin, '', $tmp);
// strip out video links too
$tmp = VideoParser::strip($tmp);
// replace video links
if ($this->config->get('bbcode_video')) {
$text = VideoParser::replace($text, $this->config->get('bbcode_video_width'), $this->config->get('bbcode_video_height'), (bool) $this->config->get('enable_media_max_width'));
}
// replace bbcode with html
$text = preg_replace($in, $out, $text);
// Replace url bbcode with html
$text = $this->replaceBBUrl($text);
// manual fix for unwrapped li issue
$text = $this->replaceBBList($text);
// Check if the content still contain [img] tag
// if yes, dont allow to replace to hyperlink for prevent XSS attack
// Always let system to process bbcode replacement first before process this replace URL part
if ((strpos($tmp, '[img]') === false || strpos($tmp, '[/img]') === false) && $this->config->get('auto_hyperlink')) {
$text = $this->replaceURL($tmp, $text);
}
// change new line to br (without affecting pre)
$text = nl2br($text);
// done parsing emoticons and bbcode, decode pre text back
$text = $this->decodePre($text);
// paragraphs
$text = str_replace("\r", "", $text);
// $text = "<p>".preg_replace("/(\n){2,}/", "</p><p>", $text)."</p>";
$text = preg_replace_callback('/<ul>(.*?)<\/ul>/ms', ['KomentoParser','removeBr'], $text);
// fix [list] within [*] causing dom errors
$text = preg_replace('/<li>(.*?)<ul>(.*?)<\/ul>(.*?)<\/li>/ms', "\\1<ul>\\2</ul>\\3", $text);
$text = preg_replace('/<p><ul>(.*?)<\/ul><\/p>/ms', "<ul>\\1</ul>", $text);
// Replace [spoiler] block
if ($this->config->get('bbcode_spoiler')) {
$text = $this->replaceSpoiler($text);
}
// Replace mentions
if ($this->config->get('enable_mention')) {
$text = $this->replaceMention($text);
}
return $text;
}
/**
* Replace mentions
*
* @since 4.0
* @access public
*/
public function replaceMention($text)
{
$users = KT::string()->detectNames($text);
if (!$users) {
return $text;
}
foreach ($users as $user) {
$link = $user->getPermalink();
$name = htmlentities($user->getName());
$search = ['@' . $name . ''];
$replace = '<a href="' . $link . '">' . $name . '</a>';
if ($this->config->get('layout_avatar_integration') === 'easysocial' && $this->config->get('easysocial_profile_popbox')) {
$replace = '<a href="' . $link . '" data-popbox="module://easysocial/profile/popbox" data-popbox-position="top-left" data-user-id="' . $user->id . '" class="mentions-user">' . $name . '</a>';
}
$text = str_ireplace($search, $replace, $text);
}
return $text;
}
/**
* Callback to process smileys
*
* @since 3.1.4
* @access public
*/
protected function emoticonCallback($matches) {
$smiley = trim($matches[0]);
$baseUrl = rtrim(JURI::root(), '/');
if (count($matches) === 1 && isset($this->emojiMap[$smiley])) {
$image = '<img class="emoji" src="' . $baseUrl . $this->emojiPath['path'] . $this->emojiMap[$smiley] . '.png" alt="" width="20" height="20">';
return $image;
}
if (count($matches) === 1 || !isset($this->emojiMap[$smiley])) {
return $matches[0];
}
$l = isset($matches[1]) ? $matches[1] : '';
$r = isset($matches[2]) ? $matches[2] : '';
$smileyPath = $baseUrl . $this->emojiPath['path'] . $this->emojiMap[$smiley] . '.png';
// Check if the emoji is exists if override folder exists.
if ($this->emojiPath['path'] != $this->emojiPath['originalPath']) {
$overrideFile = JPATH_ROOT . $smileyPath;
// Override file is not exists
if (!JFile::exists($overrideFile)) {
$smileyPath = $baseUrl . $this->emojiPath['originalPath'] . $this->emojiMap[$smiley] . '.png';
}
}
$image = '<img class="emoji" src="' . $smileyPath . '" alt="" width="20" height="20">';
return $l . $image . $r;
}
public function parseCensor($text)
{
if (!$this->config->get('filter_word')) {
return $text;
}
// We need to determine whether we should use nl2br or not during the process #633
// By default BBcode content will not use nl2br
$useBrTag = false;
$decoda = KT::decoda($text);
$decoda->initHook('CensorHook');
$decoda->setEscaping(false);
$decoda->setNl2br($useBrTag);
$result = $decoda->parse();
return $result;
}
public function encodePre($text)
{
$pattern = '/<pre.*?>(.*?)<\/pre>/s';
preg_match_all($pattern , $text , $matches);
if (isset($matches[0]) && is_array($matches[0])) {
foreach ($matches[1] as $match) {
$text = str_ireplace($match , base64_encode($match), $text);
}
}
return $text;
}
public function decodePre($text)
{
$pattern = '/<pre.*?>(.*?)<\/pre>/s';
preg_match_all($pattern , $text , $matches);
if (isset($matches[0]) && is_array($matches[0])) {
foreach ($matches[1] as $match) {
$text = str_ireplace($match , base64_decode($match) , $text);
}
}
return $text;
}
/**
* Replace the [quote] with Foundry's quote.item helper
*
* @since 4.0.0
* @access public
*/
public function replaceQuote($text)
{
$fd = KT::themes()->fd;
$search = [
'/\[quote]([^\[\/quote\]].*?)\[\/quote\]/ims',
'/\[quote](.*?)\[\/quote\]/ims',
'/\[quote name="(.*?)"\](.*?)\[\/quote\]/ims'
];
$quote = trim(preg_replace('/\s\s+/', '', $fd->html('quote.item', '\1')));
$quoteWithAuthor = trim(preg_replace('/\s\s+/', '', $fd->html('quote.item', '\2', ['author' => '\1'])));
$replace = [
$quote,
$quote,
$quoteWithAuthor
];
$text = preg_replace($search, $replace, $text);
// Parse again if there is still recursive quote tags
if (preg_match('/\[quote](.*?)\[\/quote\]/ims', $text) || preg_match('/\[quote name="(.*?)"\](.*?)\[\/quote\]/ims', $text)) {
$text = $this->replaceQuote($text);
}
return $text;
}
/**
* Replace the [giphy] with its embed
*
* @since 4.0.0
* @access public
*/
public function replaceGiphy($text)
{
$giphy = KT::giphy();
if (!KT::giphy()->isEnabled()) {
return $text;
}
preg_match_all('/\[giphy\](.*?)\[\/giphy\]/ims', $text, $matches);
if (empty($matches) || !isset($matches[0]) || empty($matches[0])) {
return $text;
}
$codes = $matches[0];
$urls = $matches[1];
$i = 0;
foreach ($urls as $url) {
if (!$giphy->isValidUrl($url)) {
continue;
}
$item = $giphy->getItem($url);
$text = FCJString::str_ireplace($codes[$i], $item, $text);
$i++;
}
return $text;
}
/**
* Replace code blocks with prism.js compatible codes
*
* @since 3.0
* @access public
*/
public function replaceCodes($text, $replace = null)
{
// [code type="*"]*[/code]
$codesPattern = '/\[code( type="(.*?)")?\](.*?)\[\/code\]/ms';
$text = preg_replace_callback($codesPattern, ['KomentoParser', 'processCodeBlocks'], $text);
// Replace [code type="*"]*[/code]
$codesPattern = '/\[code( type="(.*?)")?\](.*?)\[\/code\]/ms';
$text = preg_replace_callback($codesPattern, ['KomentoParser', 'processCodeBlocks'], $text);
return $text;
}
/**
* Replace [code] blocks with prism.js compatibility
*
* @since 3.0
* @access public
*/
public function processCodeBlocks($blocks)
{
$code = $blocks[3];
// Remove break tags
$code = str_ireplace("<br />", "", $code);
$code = str_replace("[", "[", $code);
$code = str_replace("]", "]", $code);
// Determine the language type
$language = isset($blocks[2]) && !empty($blocks[2]) ? $blocks[2] : 'markup';
// Fix legacy code blocks
if ($language == 'xml' || $language == 'html') {
$language = 'markup';
}
// Because the text / contents are already escaped, we need to revert back to the original html codes only
// for the codes.
$code = html_entity_decode($code);
// Fix html codes not displaying correctly
$code = htmlspecialchars($code, ENT_NOQUOTES);
return '<pre class="line-numbers"><code class="language-' . $language . '">'.$code.'</code></pre>';
}
public function removeBr($s)
{
// return str_replace("<br />", "", $s[0]);
return str_replace("hello", "*****", $s[0]);
}
public function escape($s)
{
$code = $s[3];
$code = str_replace("[", "[", $code);
$code = str_replace("]", "]", $code);
$brush = isset($s[2]) && !empty($s[2]) ? $s[2] : 'xml';
$code = html_entity_decode($code);
$code = FH::escape($code);
if ($brush != '') {
$result = '<pre><code class="language-' . htmlspecialchars($brush) . '">' . $code . '</code></pre>';
} else {
$result = '<pre><code>' . $code . '</code></pre>';
}
return $result;
}
public function replaceBBList($content)
{
// BBCode to find... e.g.
// [list]
// [*]hello world
// [*]marihome
// [/list]
$bbcodeListItemsSearch = '#\[list.*?\](.*?)\[\/list\]#ims';
// BBCode to find... e.g.
// [*]hello world
$bbcodeLISearch = [
'/\[\*\]\s?(.*?)\n/ims',
'/\[\*\]\s?(.*?)/ims'
];
// And replace them by...
$bbcodeLIReplace = [
'<li>\1</li>',
'<li>\1</li>'
];
// And replace them by...
$bbcodeLIReplaceString = [
'\1',
'\1'
];
// BBCode to find...
$bbcodeListPattern = [
'/\[list\=(.*?)\]/ims',
'/\[list\]/ims',
'/\[\/list\]/ims'
];
// And replace them by...
$bbcodeULReplaceString = ['\2', '\1'];
preg_match_all($bbcodeListItemsSearch, $content, $matches);
if (!$matches || !$matches[0]) {
return $content;
}
$lists = [];
$oldListIndicators = [
'a' => 'lower-alpha',
'A' => 'upper-alpha',
'i' => 'lower-roman',
'I' => 'upper-roman'
];
// Fix any unclosed list tags
foreach ($matches[0] as &$contents) {
$original = $contents;
$listStylePattern = ['\[list\=(.*?)(\stype\=(.*?))?\]', '\[list\]'];
$listStylePattern = implode('|', $listStylePattern);
preg_match('/' . $listStylePattern . '/ims', $contents, $listStyleMatches);
// The match of lists within this block should always be the first and last. Anything within the "list" should be considered as unclosed.
$contents = preg_replace($bbcodeListPattern, '', $contents);
// this match of list have to follow back what original list code pass in
$contents = $listStyleMatches[0] . $contents . '[/list]';
$hasStyleType = isset($listStyleMatches[3]);
$item = new stdClass();
$item->original = $original;
$item->contents = $contents;
$item->indicator = isset($listStyleMatches[1]) ? $listStyleMatches[1] : null;
$item->style = $hasStyleType ? $listStyleMatches[3] : 'decimal';
$isOldIndicator = $item->indicator && in_array($item->indicator, array_keys($oldListIndicators));
// Backward compatibility
if ($isOldIndicator) {
$item->style = $oldListIndicators[$item->indicator];
}
$styleTypeRegex = $hasStyleType ? '\stype\=' . $item->style : '';
// Regex to replace
$item->regex = $item->indicator ? '/\[list\=' . $item->indicator . $styleTypeRegex . '\](.*?)\[\/list\]/ims' : '/\[list\](.*?)\[\/list\]/ims';
// The html output
$item->html = $item->indicator ? '<ol start=' . $item->indicator . ' style="list-style-type: ' . $item->style . '">\1</ol>' : '<ul>\1</ul>';
$lists[] = $item;
}
foreach ($lists as $list) {
// Check if this list contains any list items "[*]"
$containsListItems = FCJString::strpos($list->contents, '[*]') !== false;
if ($containsListItems) {
$text = preg_replace($list->regex, $list->html, $list->contents);
$text = preg_replace($bbcodeLISearch, $bbcodeLIReplace, $text);
}
if (!$containsListItems) {
$text = preg_replace($bbcodeULSearch , $bbcodeULReplaceString, $list->contents);
}
// Update the content
$content = str_replace($list->original, $text, $content);
}
return $content;
}
/**
* Replaces the bbcode url tag
*
* @since 2.0
* @access public
*/
public function replaceBBUrl($text)
{
$nofollow = $this->config->get('links_nofollow') ? ' rel="nofollow"' : '';
preg_match_all('/\[url\=(.*?)\](.*?)\[\/url\]/ims', $text, $matches);
if (!empty($matches)) {
$sources = array_shift($matches);
$urls = $matches[0];
$txts = $matches[1];
for ($i = 0; $i < count($sources); $i++) {
$source = $sources[$i];
if (!empty($source)) {
$url = $urls[$i];
// prevent user add javascript in the url element
$segments = explode(' ', $url);
$url = $segments[0];
// Ensure that the url doesn't contain " or ' or "
$url = str_ireplace(['"', "'", '"'], '', $url);
$txt = $txts[$i];
if (stripos($url, 'http://') !== 0 && stripos($url, 'https://') !== 0 && stripos($url, 'ftp://') !== 0) {
$url = 'http://' . $url;
}
$replace = '<a target="_blank" href="' . $url . '"' . $nofollow . '>' . $txt . '</a>';
$text = str_ireplace($source, $replace, $text);
}
}
}
return $text;
}
public function replaceURL($tmp, $text)
{
$pattern = '@(?i)\b((?:https?://|www\d{0,3}[.]|[a-z0-9.\-]+[.][a-z]{2,4}/)(?:[^\s()<>]+|\(([^\s()<>]+|(\([^\s()<>]+\)))*\))+(?:\(([^\s()<>]+|(\([^\s()<>]+\)))*\)|[^\s`!()\[\]{};:\'".,<>?«»“”‘’]))@';
preg_match_all($pattern, $tmp, $matches);
$targetBlank = ' target="_blank"';
$noFollow = $this->config->get('links_nofollow') ? ' rel="nofollow"' : '';
// Do not proceed if there are no links to process
if (!isset($matches[0]) || !is_array($matches[0]) || empty($matches[0])) {
return $text;
}
$tmplinks = $matches[0];
$linksWithProtocols = [];
$linksWithoutProtocols = [];
// We need to separate the link with and without protocols to avoid conflict when there are similar url present in the content.
if ($tmplinks) {
foreach($tmplinks as $link) {
if (stristr( $link , 'http://' ) === false && stristr( $link , 'https://' ) === false && stristr( $link , 'ftp://' ) === false ) {
$linksWithoutProtocols[] = $link;
} else if (stristr( $link , 'http://' ) !== false || stristr( $link , 'https://' ) !== false || stristr( $link , 'ftp://' ) === false ) {
$linksWithProtocols[] = $link;
}
}
}
// the idea is the first convert the url to [EDWURLx] and [EDWOURLx] where x is the index. This is to prevent same url get overwritten with wrong value.
$linkArrays = [];
// global indexing.
$idx = 1;
// lets process the one with protocol
if ($linksWithProtocols) {
$linksWithProtocols = array_unique($linksWithProtocols);
foreach($linksWithProtocols as $link) {
$mypattern = '[EDWURL' . $idx . ']';
$text = str_ireplace($link, $mypattern, $text);
$obj = new stdClass();
$obj->index = $idx;
$obj->link = $link;
$obj->newlink = $link;
$obj->customcode = $mypattern;
$linkArrays[] = $obj;
$idx++;
}
}
// Now we process the one without protocol
if ($linksWithoutProtocols) {
$linksWithoutProtocols = array_unique($linksWithoutProtocols);
foreach($linksWithoutProtocols as $link) {
$mypattern = '[EDWOURL' . $idx . ']';
$text = str_ireplace($link, $mypattern, $text);
$obj = new stdClass();
$obj->index = $idx;
$obj->link = $link;
$obj->newlink = 'http://'. $link;
$obj->customcode = $mypattern;
$linkArrays[] = $obj;
$idx++;
}
}
// Let's replace back the link now with the proper format based on the index given.
foreach ($linkArrays as $link) {
$text = str_ireplace($link->customcode, $link->newlink, $text);
$patternReplace = '@(?<![.*">])\b(?:(?:https?))[-A-Z0-9+&#/%=~_|$()?!:;,.]*[A-Z0-9+&#/%=~_|$]@i';
// Use preg_replace to only replace if the URL doesn't has <a> tag
$text = preg_replace($patternReplace, '<a href="\0" ' . $targetBlank . $noFollow . '>\0</a>', $text);
}
// Not really sure why this is needed as it will cause some of the content to not rendered correctly.
// We will comment this out for now. References : #1878
// $text = FCJString::str_ireplace('"', '"', $text);
return $text;
}
/**
* Parse spoiler block
*
* @since 4.0
* @access public
*/
public function replaceSpoiler($block)
{
// Clean block to remove line break.
$text = preg_replace('/(\r\n|\n|\r)/ms', "", $block);
$codesPattern = '/\[spoiler\](.*)\[\/spoiler\]/U';
preg_match_all($codesPattern, $text, $matches);
foreach ($matches[1] as $key => $match) {
$search = $matches[0][$key];
$replace = '<span class="kt-spoiler-block">' . $match . '</span>';
if (preg_match('/\<div/i', $match)) {
$replace = '<div class="kt-spoiler-block">' . $match . '</div>';
}
$text = str_replace($search, $replace, $text);
}
return $text;
}
}