BBCode Parser
I've been working on building a custom forum software from the ground up the past few months. When I got to parsing messages, I wrote a TextParser class to handle bbcode, smilies and the like. It worked well until I realized that embedded bbcodes of the same type weren't being parsed.
I found some discussions on the topic and they pointed to using preg_replace_callback to recursively replace the bbcodes. I had some trouble implementing it and got some strange results. In search for more answers I came across a blog post where someone said to avoid regular expressions to parse bbcode structures. How would I not use regular expressions to parse bbcode? Are there any other alternatives?
For the curious, here's my class. My attempt at recursion is commented out and the original attempt isn't.
PHP Code:<?php
class Penelope_TextParser {
private $BBCodes = array();
private $smilies = array();
private $currentReplacement;
private $currentBBCode;
public function __construct() {
$this->loadCodes();
$this->loadSmilies();
}
private function loadCodes() {
// this function loads all the bbcodes in the database into the object for use later when parsing is needed.
// get the database connection
$db = Penelope_DB::getInstance();
$getCodes = $db->prepare("SELECT * FROM bbcodes");
$getCodes->execute();
$this->BBCodes = $getCodes->fetchAll(PDO::FETCH_ASSOC);
if($getCodes->errorCode() == "0000") {
return true;
}
else {
return false;
}
}
private function loadSmilies() {
// this function loads all the smilies in the database into the object for use later when parsing
$db = Penelope_DB::getInstance();
$getSmilies = $db->prepare("SELECT * FROM smilies");
$getSmilies->execute();
$this->smilies = $getSmilies->fetchAll(PDO::FETCH_ASSOC);
if($getSmilies->errorCode() == "0000")
{
return true;
}
else {
return false;
}
}
private function prepareReplacement($replacement, $use_option) {
if($use_option) {
$replacement = preg_replace('/{option}/', '\$1', $replacement);
$replacement = preg_replace('/{param}/', '\$2', $replacement);
}
else {
$replacement = preg_replace('/{param}/', '\$1', $replacement);
}
return $replacement;
}
public function parse($text) {
$text = htmlentities($text);
$text = $this->parseBBCode($text);
$text = $this->parseSmilies($text);
$text = nl2br($text);
return $text;
}
protected function recursiveReplaceSimple($input) {
//$arr = $this->currentBBCode;
if(is_array($input)) {
$replacement = $this->currentReplacement;
$input = str_replace('{param}', $input[1], $replacement);
}
return preg_replace_callback('/\['.$arr['tag'].'\](.+)\[\/'.$arr['tag'].'\]/is', array(&$this, 'recursiveReplaceSimple'), $input, 5);
}
protected function recursiveReplaceAdvanced($input) {
$arr = $this->currentBBCode;
if(is_array($input)) {
$replacement = $this->currentReplacement;
$option = $input[1];
$param = $input[2];
$input = str_replace('{param}', $param, $replacement);
$input = str_replace('{option}', $option, $input);
}
return preg_replace_callback('/\['.$arr['tag'].'\](.+)\[\/'.$arr['tag'].'\]/is', array(&$this, 'recursiveReplaceAdvanced'), $input, 5);
}
public function parseBBCode($text) {
foreach($this->BBCodes as $key => $arr)
{
if(!$arr['use_option']) {
/*
$this->currentReplacement = $arr['replacement'];
$this->currentBBcode = $arr;
// RECURSIVE REPLACEMENT
$text = preg_replace_callback('/\['.$arr['tag'].'\](.+)\[\/'.$arr['tag'].'\]/is', array(&$this, 'recursiveReplaceSimple'), $text, 5);
*/
//ORIGINAL REPLACEMENT:
$replacement = $this->prepareReplacement($arr['replacement'], 0);
$text = preg_replace('/\['.$arr['tag'].'\](.+)\[\/'.$arr['tag'].'\]/isU', $replacement, $text);
}
else {
/*
$this->currentReplacement = $arr['replacement'];
$this->currentBBCode = $arr;
// RECURSIVE REPLACEMENT
$text = preg_replace_callback('/\['.$arr['tag'].'=([^\]]+)\](.+)\[\/'.$arr['tag'].'\]/is', array(&$this, 'recursiveReplaceAdvanced'), $text, 5);
*/
//ORIGINAL REPLACEMENT:
$replacement = $this->prepareReplacement($arr['replacement'], 1);
$text = preg_replace('/\['.$arr['tag'].'=([^\]]+)\](.+)\[\/'.$arr['tag'].'\]/isU', $replacement, $text);
}
}
return $text;
}
public function parseSmilies($text) {
$config = Penelope_Config::getInstance();
$website_url = $config->getProp("website_url");
foreach($this->smilies as $key => $arr)
{
$text = str_replace($arr['search'], '<img src="'.$website_url.'/imgs/smilies/'.$arr['filepath'].'" alt="'.$arr['name'].'" />', $text);
}
return $text;
}
public function addDefaultBBCode($text, $font, $color, $size) {
$tmp = "";
if($font) {
$tmp .= "[font=".$font."]";
}
if($color) {
$tmp .= "[color=".$color."]";
}
if($size) {
$tmp .= "[size=".$size."]";
}
$tmp .= $text;
if($size) {
$tmp .= "[/size]";
}
if($color) {
$tmp .= "[/color]";
}
if($font) {
$tmp .= "[/font]";
}
return $tmp;
}
}
?>
CSS Codes
Reply With Quote
Bookmarks