max number of attributes
private $ValidTags = array('b'=>0, 'u'=>0, 'i'=>0, 's'=>0, '*'=>0, 'artist'=>0, 'user'=>0, 'n'=>0, 'inlineurl'=>0, 'inlinesize'=>1, 'align'=>1, 'color'=>1, 'colour'=>1, 'size'=>1, 'url'=>1, 'img'=>1, 'quote'=>1, 'pre'=>1, 'code'=>1, 'tex'=>0, 'hide'=>1, 'plain'=>0
);
private $Smileys = array(
':angry:' => 'angry.gif',
':-D' => 'biggrin.gif',
':D' => 'biggrin.gif',
':|' => 'blank.gif',
':-|' => 'blank.gif',
':blush:' => 'blush.gif',
':cool:' => 'cool.gif',
':'(' => 'crying.gif',
':crying:' => 'crying.gif',
'>.>' => 'eyesright.gif',
':frown:' => 'frown.gif',
'<3' => 'heart.gif',
':unsure:' => 'hmm.gif',
':\\' => 'hmm.gif',
':whatlove:' => 'ilu.gif',
':lol:' => 'laughing.gif',
':loveflac:' => 'loveflac.gif',
':ninja:' => 'ninja.gif',
':no:' => 'no.gif',
':nod:' => 'nod.gif',
':ohno:' => 'ohnoes.gif',
':ohnoes:' => 'ohnoes.gif',
':omg:' => 'omg.gif',
':o' => 'ohshit.gif',
':O' => 'ohshit.gif',
':paddle:' => 'paddle.gif',
':(' => 'sad.gif',
':-(' => 'sad.gif',
':shifty:' => 'shifty.gif',
':sick:' => 'sick.gif',
':)' => 'smile.gif',
':-)' => 'smile.gif',
':sorry:' => 'sorry.gif',
':thanks:' => 'thanks.gif',
':P' => 'tongue.gif',
':-P' => 'tongue.gif',
':-p' => 'tongue.gif',
':wave:' => 'wave.gif',
';-)' => 'wink.gif',
':wink:' => 'wink.gif',
':creepy:' => 'creepy.gif',
':worried:' => 'worried.gif',
':wtf:' => 'wtf.gif',
':wub:' => 'wub.gif',
);
private $NoImg = 0; // If images should be turned into URLs
private $Levels = 0; // If images should be turned into URLs
function __construct() {
foreach($this->Smileys as $Key=>$Val) {
$this->Smileys[$Key] = '';
}
reset($this->Smileys);
}
function full_format($Str) {
$Str = display_str($Str);
//Inline links
$URLPrefix = '(\[url\]|\[url\=|\[img\=|\[img\])';
$Str = preg_replace('/'.$URLPrefix.'\s+http/i', '$1http', $Str);
$Str = preg_replace('/(?'?http', '=[inlineurl]http'=>'=http'));
$callback = create_function('$matches', 'return str_replace("[inlineurl]","",$matches[0]);');
$Str = preg_replace_callback('/(?<=\[inlineurl\]|'.$URLPrefix.')(\S*\[inlineurl\]\S*)/m', $callback, $Str);
$Str = preg_replace('/\=\=\=\=([^=].*)\=\=\=\=/i', '[inlinesize=3]$1[/inlinesize]', $Str);
$Str = preg_replace('/\=\=\=([^=].*)\=\=\=/i', '[inlinesize=5]$1[/inlinesize]', $Str);
$Str = preg_replace('/\=\=([^=].*)\=\=/i', '[inlinesize=7]$1[/inlinesize]', $Str);
$Str = $this->parse($Str);
$HTML = $this->to_html($Str);
$HTML = nl2br($HTML);
return $HTML;
}
function strip_bbcode($Str) {
$Str = display_str($Str);
//Inline links
$Str = preg_replace('/(?parse($Str);
$Str = $this->raw_text($Str);
$Str = nl2br($Str);
return $Str;
}
function valid_url($Str, $Extension = '', $Inline = false) {
$Regex = '/^';
$Regex .= '(https?|ftps?|irc):\/\/'; // protocol
$Regex .= '(\w+(:\w+)?@)?'; // user:pass@
$Regex .= '(';
$Regex .= '(([0-9]{1,3}\.){3}[0-9]{1,3})|'; // IP or...
$Regex .= '(([a-z0-9\-\_]+\.)+\w{2,6})'; // sub.sub.sub.host.com
$Regex .= ')';
$Regex .= '(:[0-9]{1,5})?'; // port
$Regex .= '\/?'; // slash?
$Regex .= '(\/?[0-9a-z\-_.,&=@~%\/:;()+|!#]+)*'; // /file
if(!empty($Extension)) {
$Regex.=$Extension;
}
// query string
if ($Inline) {
$Regex .= '(\?([0-9a-z\-_.,%\/\@~&=:;()+*\^$!#|]|\[\d*\])*)?';
} else {
$Regex .= '(\?[0-9a-z\-_.,%\/\@[\]~&=:;()+*\^$!#|]*)?';
}
$Regex .= '(#[a-z0-9\-_.,%\/\@[\]~&=:;()+*\^$!]*)?'; // #anchor
$Regex .= '$/i';
return preg_match($Regex, $Str, $Matches);
}
function local_url($Str) {
$URLInfo = parse_url($Str);
if(!$URLInfo) { return false; }
$Host = $URLInfo['host'];
// If for some reason your site does not require subdomains or contains a directory in the SITE_URL, revert to the line below.
//if($Host == NONSSL_SITE_URL || $Host == SSL_SITE_URL || $Host == 'www.'.NONSSL_SITE_URL) {
if(preg_match('/(\S+\.)*'.NONSSL_SITE_URL.'/', $Host)) {
$URL = $URLInfo['path'];
if(!empty($URLInfo['query'])) {
$URL.='?'.$URLInfo['query'];
}
if(!empty($URLInfo['fragment'])) {
$URL.='#'.$URLInfo['fragment'];
}
return $URL;
} else {
return false;
}
}
/* How parsing works
Parsing takes $Str, breaks it into blocks, and builds it into $Array.
Blocks start at the beginning of $Str, when the parser encounters a [, and after a tag has been closed.
This is all done in a loop.
EXPLANATION OF PARSER LOGIC
1) Find and parse the next tag (regex)
1a) See if it's a [[wiki-link]] or an ordinary tag, and get the tag name
1b) If it's not a wiki link:
1bi) check it against the $this->ValidTags array to see if it's actually a tag and not [bullshit]
1bii) Get the attribute, if it exists [name=attribute]
If it's [not a tag] or has an invalid attribute, just leave it as plaintext and move on
2) If there aren't any tags left, write everything remaining to a block and return (done parsing)
3) If the next tag isn't where the pointer is, write everything up to there to a text block.
4) Move the pointer past the end of the tag
5) Find out where the tag closes (beginning of [/tag])
5a) Different for different types of tag. Some tags don't close, others are weird like [*]
5b) If it's a normal tag, it may have versions of itself nested inside - eg:
[quote=bob]*
[quote=joe]I am a redneck!**[/quote]
Me too!
***[/quote]
If we're at the position *, the first [/quote] tag is denoted by **.
However, our quote tag doesn't actually close there. We must perform
a loop which checks the number of opening [quote] tags, and make sure
they are all closed before we find our final [/quote] tag (***).
5c) Get the contents between [open] and [/close] and call it the block.
In many cases, this will be parsed itself later on, in a new parse() call.
5d) Move the pointer past the end of the [/close] tag.
6) Depending on what type of tag we're dealing with, create an array with the attribute and block.
In many cases, the block may be parsed here itself. Stick them in the $Array.
7) Increment array pointer, start again (past the end of the [/close] tag)
*/
function parse($Str) {
$i = 0; // Pointer to keep track of where we are in $Str
$Len = strlen($Str);
$Array = array();
$ArrayPos = 0;
while($i<$Len) {
$Block = '';
// 1) Find and parse the next tag (regex)
// [name=|[name]|[[wiki-link]]
$IsTag = preg_match("/((\[[a-zA-Z*]{1,100})([=\]]))|(\[\[[^\n\"'\[\]]+\]\])/", $Str, $Tag, PREG_OFFSET_CAPTURE, $i);
$TagPos = $Tag[0][1];
// At this point we still don't know whether the tag code is valid
// $IsTag will be updated with new information
// 1a) See if it's a [[wiki-link]] or an ordinary tag, and get the tag name
if ($IsTag && !empty($Tag[4][0])) { // Wiki-link
$WikiLink = true;
$TagName = substr($Tag[4][0], 2, -2);
$Attrib = '';
$ExtraTagLen = 0;
} elseif ($IsTag) { // 1b) If it's not a wiki link:
$WikiLink = false;
$TagName = strtolower(substr($Tag[2][0], 1));
$MaxAttribs = $this->ValidTags[$TagName];
// 1bi) check it against the $this->ValidTags array to see if it's actually a tag and not [bullshit]
if(!isset($this->ValidTags[$TagName])) {
$IsTag = 0;
}
// 1bii) Get the attribute, if it exists [name=attribute]
$HasAttrib = ($Tag[3][0] == '=');
if ($IsTag && $HasAttrib) {
$AttribFrom = $TagPos + strlen($Tag[0][0]);
$Attrib = substr($Str, $AttribFrom, strpos($Str, ']', $AttribFrom) - $AttribFrom);
$ExtraTagLen = strlen($Attrib) + 1;
// check $Attrib is valid, i.e. the tag is in a valid format
// disallowed characters
if ((strpos($Attrib, "\n") !== false)
|| (strpos($Attrib, "'") !== false)
|| (strpos($Attrib, "\"") !== false)) {
$IsTag = 0;
} elseif (strpos($Attrib, "[") !== false) {
// attributes aren't allowed to contain any tag (or something which looks like it might be one) inside them
foreach ($this->ValidTags as $PossibleTag=>$_) {
if (strpos($Attrib, "[".$PossibleTag) !== false) {
$IsTag = 0;
}
}
}
} else {
$Attrib = '';
$ExtraTagLen = 0;
}
if (!$IsTag) {
// we jump past the beginning of the "tag", but not the attribute value, as this may contain real tags
$Array[$ArrayPos] = substr($Str, $i, ($TagPos-$i)+strlen($Tag[0][0]));
$i=$TagPos+strlen($Tag[0][0]);
++$ArrayPos;
continue;
}
}
// 2) If there aren't any tags left, write everything remaining to a block
if(!$IsTag) {
// No more tags
$Array[$ArrayPos] = substr($Str, $i);
break;
}
// 3) If the next tag isn't where the pointer is, write everything up to there to a text block.
if($TagPos>$i) {
$Array[$ArrayPos] = substr($Str, $i, $TagPos-$i);
++$ArrayPos;
$i=$TagPos;
}
// 4) Move the pointer past the end of the tag
$i=$TagPos+strlen($Tag[0][0])+$ExtraTagLen;
// 5) Find out where the tag closes (beginning of [/tag])
// Unfortunately, BBCode doesn't have nice standards like xhtml
// [*], [img=...], and http:// follow different formats
// Thus, we have to handle these before we handle the majority of tags
//5a) Different for different types of tag. Some tags don't close, others are weird like [*]
if($TagName == 'img' && $HasAttrib) { //[img=...]
$Block = ''; // Nothing inside this tag
// Don't need to touch $i
} elseif($TagName == 'inlineurl') { // We did a big replace early on to turn http:// into [inlineurl]http://
// Let's say the block can stop at a newline or a space
$CloseTag = strcspn($Str, " \n\r", $i);
if($CloseTag === false) { // block finishes with URL
$CloseTag = $Len;
}
if(preg_match('/[!;,.?:]+$/',substr($Str, $i, $CloseTag), $Match)) {
$CloseTag -= strlen($Match[0]);
}
$URL = substr($Str, $i, $CloseTag);
if(substr($URL, -1) == ')' && substr_count($URL, '(') < substr_count($URL, ')')) {
$CloseTag--;
$URL = substr($URL, 0, -1);
}
$Block = $URL; // Get the URL
// strcspn returns the number of characters after the offset $i, not after the beginning of the string
// Therefore, we use += instead of the = everywhere else
$i += $CloseTag; // 5d) Move the pointer past the end of the [/close] tag.
} elseif($WikiLink == true || $TagName == 'n') {
// Don't need to do anything - empty tag with no closing
} elseif($TagName == '*') {
// We're in a list. Find where it ends
$NewLine = $i;
do { // Look for \n[*]
$NewLine = strpos($Str, "\n", $NewLine+1);
} while($NewLine!== false && substr($Str, $NewLine+1, 3) == '[*]');
$CloseTag = $NewLine;
if($CloseTag === false) { // block finishes with list
$CloseTag = $Len;
}
$Block = substr($Str, $i, $CloseTag-$i); // Get the list
$i = $CloseTag; // 5d) Move the pointer past the end of the [/close] tag.
} else {
//5b) If it's a normal tag, it may have versions of itself nested inside
$CloseTag = $i-1;
$InTagPos = $i-1;
$NumInOpens = 0;
$NumInCloses = -1;
$InOpenRegex = '/\[('.$TagName.')';
if($MaxAttribs>0) {
$InOpenRegex.="(=[^\n'\"\[\]]+)?";
}
$InOpenRegex.='\]/i';
// Every time we find an internal open tag of the same type, search for the next close tag
// (as the first close tag won't do - it's been opened again)
do {
$CloseTag = stripos($Str, '[/'.$TagName.']', $CloseTag+1);
if($CloseTag === false) {
$CloseTag = $Len;
break;
} else {
$NumInCloses++; // Majority of cases
}
// Is there another open tag inside this one?
$OpenTag = preg_match($InOpenRegex, $Str, $InTag, PREG_OFFSET_CAPTURE, $InTagPos+1);
if(!$OpenTag || $InTag[0][1]>$CloseTag) {
break;
} else {
$InTagPos = $InTag[0][1];
$NumInOpens++;
}
} while($NumInOpens>$NumInCloses);
// Find the internal block inside the tag
$Block = substr($Str, $i, $CloseTag-$i); // 5c) Get the contents between [open] and [/close] and call it the block.
$i = $CloseTag+strlen($TagName)+3; // 5d) Move the pointer past the end of the [/close] tag.
}
// 6) Depending on what type of tag we're dealing with, create an array with the attribute and block.
switch($TagName) {
case 'inlineurl':
$Array[$ArrayPos] = array('Type'=>'inlineurl', 'Attr'=>$Block, 'Val'=>'');
break;
case 'url':
$Array[$ArrayPos] = array('Type'=>'img', 'Attr'=>$Attrib, 'Val'=>$Block);
if(empty($Attrib)) { // [url]http://...[/url] - always set URL to attribute
$Array[$ArrayPos] = array('Type'=>'url', 'Attr'=>$Block, 'Val'=>'');
} else {
$Array[$ArrayPos] = array('Type'=>'url', 'Attr'=>$Attrib, 'Val'=>$this->parse($Block));
}
break;
case 'quote':
$Array[$ArrayPos] = array('Type'=>'quote', 'Attr'=>$this->Parse($Attrib), 'Val'=>$this->parse($Block));
break;
case 'img':
case 'image':
if(empty($Block)) {
$Block = $Attrib;
}
$Array[$ArrayPos] = array('Type'=>'img', 'Val'=>$Block);
break;
case 'aud':
case 'mp3':
case 'audio':
if(empty($Block)) {
$Block = $Attrib;
}
$Array[$ArrayPos] = array('Type'=>'aud', 'Val'=>$Block);
break;
case 'user':
$Array[$ArrayPos] = array('Type'=>'user', 'Val'=>$Block);
break;
case 'artist':
$Array[$ArrayPos] = array('Type'=>'artist', 'Val'=>$Block);
break;
case 'tex':
$Array[$ArrayPos] = array('Type'=>'tex', 'Val'=>$Block);
break;
case 'pre':
case 'code':
case 'plain':
$Block = strtr($Block, array('[inlineurl]'=>''));
$Block = preg_replace('/\[inlinesize\=3\](.*?)\[\/inlinesize\]/i', '====$1====', $Block);
$Block = preg_replace('/\[inlinesize\=5\](.*?)\[\/inlinesize\]/i', '===$1===', $Block);
$Block = preg_replace('/\[inlinesize\=7\](.*?)\[\/inlinesize\]/i', '==$1==', $Block);
$Array[$ArrayPos] = array('Type'=>$TagName, 'Val'=>$Block);
break;
case 'hide':
$Array[$ArrayPos] = array('Type'=>'hide', 'Attr'=>$Attrib, 'Val'=>$this->parse($Block));
break;
case '*':
$Array[$ArrayPos] = array('Type'=>'list');
$Array[$ArrayPos]['Val'] = explode('[*]', $Block);
foreach($Array[$ArrayPos]['Val'] as $Key=>$Val) {
$Array[$ArrayPos]['Val'][$Key] = $this->parse(trim($Val));
}
break;
case 'n':
$ArrayPos--;
break; // n serves only to disrupt bbcode (backwards compatibility - use [pre])
default:
if($WikiLink == true) {
$Array[$ArrayPos] = array('Type'=>'wiki','Val'=>$TagName);
} else {
// Basic tags, like [b] or [size=5]
$Array[$ArrayPos] = array('Type'=>$TagName, 'Val'=>$this->parse($Block));
if(!empty($Attrib) && $MaxAttribs>0) {
$Array[$ArrayPos]['Attr'] = strtolower($Attrib);
}
}
}
$ArrayPos++; // 7) Increment array pointer, start again (past the end of the [/close] tag)
}
return $Array;
}
function to_html($Array) {
$this->Levels++;
if($this->Levels>10) { return $Block['Val']; } // Hax prevention
$Str = '';
foreach($Array as $Block) {
if(is_string($Block)) {
$Str.=$this->smileys($Block);
continue;
}
switch($Block['Type']) {
case 'b':
$Str.=''.$this->to_html($Block['Val']).'';
break;
case 'u':
$Str.=''.$this->to_html($Block['Val']).'';
break;
case 'i':
$Str.=''.$this->to_html($Block['Val'])."";
break;
case 's':
$Str.=''.$this->to_html($Block['Val']).'';
break;
case 'user':
$Str.=''.$Block['Val'].'';
break;
case 'artist':
$Str.=''.$Block['Val'].'';
break;
case 'wiki':
$Str.=''.$Block['Val'].'';
break;
case 'tex':
$Str.='';
break;
case 'plain':
$Str.=$Block['Val'];
break;
case 'pre':
$Str.='
'.$Block['Val'].'
';
break;
case 'code':
$Str.=''.$Block['Val'].'';
break;
case 'list':
$Str .= '
';
break;
case 'plain':
$Str.=$Block['Val'];
break;
case 'pre':
$Str.='