* Link: http://code.shishnet.org/shimmie2/
* License: GPLv2
* Description: Turns BBCode into HTML
* Documentation:
* Supported tags:
*
* - [img]url[/img]
*
- [url]http://code.shishnet.org/[/url]
*
- [email]webmaster@shishnet.org[/email]
*
- [b]bold[/b]
*
- [i]italic[/i]
*
- [u]underline[/u]
*
- [s]
strikethrough[/s]
* - [sup]superscript[/sup]
*
- [sub]subscript[/sub]
*
- [[wiki article]]
*
- [[wiki article|with some text]]
*
- [quote]text[/quote]
*
- [quote=Username]text[/quote]
*
- >>123 (link to image #123)
*
*/
class BBCode extends FormatterExtension
{
public function format(string $text): string
{
$text = $this->extract_code($text);
foreach ([
"b", "i", "u", "s", "sup", "sub", "h1", "h2", "h3", "h4",
] as $el) {
$text = preg_replace("!\[$el\](.*?)\[/$el\]!s", "<$el>$1", $text);
}
$text = preg_replace('!^>>([^\d].+)!', '$1
', $text);
$text = preg_replace('!>>(\d+)(#c?\d+)?!s', '>>$1$2', $text);
$text = preg_replace('!\[anchor=(.*?)\](.*?)\[/anchor\]!s', '$2 ΒΆ ', $text); // add "bb-" to avoid clashing with eg #top
$text = preg_replace('!\[url=site://(.*?)(#c\d+)?\](.*?)\[/url\]!s', '$3', $text);
$text = preg_replace('!\[url\]site://(.*?)(#c\d+)?\[/url\]!s', '$1$2', $text);
$text = preg_replace('!\[url=((?:https?|ftp|irc|mailto)://.*?)\](.*?)\[/url\]!s', '$2', $text);
$text = preg_replace('!\[url\]((?:https?|ftp|irc|mailto)://.*?)\[/url\]!s', '$1', $text);
$text = preg_replace('!\[email\](.*?)\[/email\]!s', '$1', $text);
$text = preg_replace('!\[img\](https?:\/\/.*?)\[/img\]!s', '
', $text);
$text = preg_replace('!\[\[([^\|\]]+)\|([^\]]+)\]\]!s', '$2', $text);
$text = preg_replace('!\[\[([^\]]+)\]\]!s', '$1', $text);
$text = preg_replace("!\n\s*\n!", "\n\n", $text);
$text = str_replace("\n", "\n
", $text);
$text = preg_replace("/\[quote\](.*?)\[\/quote\]/s", "\\1
", $text);
$text = preg_replace("/\[quote=(.*?)\](.*?)\[\/quote\]/s", "\\1 said:
\\2
", $text);
while (preg_match("/\[list\](.*?)\[\/list\]/s", $text)) {
$text = preg_replace("/\[list\](.*?)\[\/list\]/s", "", $text);
}
while (preg_match("/\[ul\](.*?)\[\/ul\]/s", $text)) {
$text = preg_replace("/\[ul\](.*?)\[\/ul\]/s", "", $text);
}
while (preg_match("/\[ol\](.*?)\[\/ol\]/s", $text)) {
$text = preg_replace("/\[ol\](.*?)\[\/ol\]/s", "\\1
", $text);
}
$text = preg_replace("/\[li\](.*?)\[\/li\]/s", "\\1", $text);
$text = preg_replace("#\[\*\]#s", "", $text);
$text = preg_replace("#
<(li|ul|ol|/ul|/ol)>#s", "<\\1>", $text);
$text = preg_replace("#\[align=(left|center|right)\](.*?)\[\/align\]#s", "\\2
", $text);
$text = $this->filter_spoiler($text);
$text = $this->insert_code($text);
return $text;
}
public function strip(string $text): string
{
foreach ([
"b", "i", "u", "s", "sup", "sub", "h1", "h2", "h3", "h4",
"code", "url", "email", "li",
] as $el) {
$text = preg_replace("!\[$el\](.*?)\[/$el\]!s", '$1', $text);
}
$text = preg_replace("!\[anchor=(.*?)\](.*?)\[/anchor\]!s", '$2', $text);
$text = preg_replace("!\[url=(.*?)\](.*?)\[/url\]!s", '$2', $text);
$text = preg_replace("!\[img\](.*?)\[/img\]!s", "", $text);
$text = preg_replace("!\[\[([^\|\]]+)\|([^\]]+)\]\]!s", '$2', $text);
$text = preg_replace("!\[\[([^\]]+)\]\]!s", '$1', $text);
$text = preg_replace("!\[quote\](.*?)\[/quote\]!s", "", $text);
$text = preg_replace("!\[quote=(.*?)\](.*?)\[/quote\]!s", "", $text);
$text = preg_replace("!\[/?(list|ul|ol)\]!", "", $text);
$text = preg_replace("!\[\*\](.*?)!s", '$1', $text);
$text = $this->strip_spoiler($text);
return $text;
}
private function filter_spoiler(string $text): string
{
return str_replace(
["[spoiler]","[/spoiler]"],
["",""],
$text
);
}
private function strip_spoiler(string $text): string
{
$l1 = strlen("[spoiler]");
$l2 = strlen("[/spoiler]");
while (true) {
$start = strpos($text, "[spoiler]");
if ($start === false) {
break;
}
$end = strpos($text, "[/spoiler]");
if ($end === false) {
break;
}
if ($end < $start) {
break;
}
$beginning = substr($text, 0, $start);
$middle = str_rot13(substr($text, $start+$l1, ($end-$start-$l1)));
$ending = substr($text, $end + $l2, (strlen($text)-$end+$l2));
$text = $beginning . $middle . $ending;
}
return $text;
}
private function extract_code(string $text): string
{
# at the end of this function, the only code! blocks should be
# the ones we've added -- others may contain malicious content,
# which would only appear after decoding
$text = str_replace("[code!]", "[code]", $text);
$text = str_replace("[/code!]", "[/code]", $text);
$l1 = strlen("[code]");
$l2 = strlen("[/code]");
while (true) {
$start = strpos($text, "[code]");
if ($start === false) {
break;
}
$end = strpos($text, "[/code]", $start);
if ($end === false) {
break;
}
if ($end < $start) {
break;
}
$beginning = substr($text, 0, $start);
$middle = base64_encode(substr($text, $start+$l1, ($end-$start-$l1)));
$ending = substr($text, $end + $l2, (strlen($text)-$end+$l2));
$text = $beginning . "[code!]" . $middle . "[/code!]" . $ending;
}
return $text;
}
private function insert_code(string $text): string
{
$l1 = strlen("[code!]");
$l2 = strlen("[/code!]");
while (true) {
$start = strpos($text, "[code!]");
if ($start === false) {
break;
}
$end = strpos($text, "[/code!]");
if ($end === false) {
break;
}
$beginning = substr($text, 0, $start);
$middle = base64_decode(substr($text, $start+$l1, ($end-$start-$l1)));
$ending = substr($text, $end + $l2, (strlen($text)-$end+$l2));
$text = $beginning . "" . $middle . "
" . $ending;
}
return $text;
}
}