{

return preg_replace_callback('/ # Rev:20110220_1200 github.com\/jmrware\/LinkifyURL

/imx', '_linkify_callback', $text);

{ // Only linkify valid urls.

$url = $m[2] . $m[5] . $m[8] . $m[11] . $m[14];

if (is_array($u = url_valid($url))) {

if (preg_match('%\.(?:jpe?g|gif|png)$%Si', $u['path_abempty'])) {

return $m[1].$m[4].$m[7].$m[10].$m[13] .'[img]'. $u['url'] .'[/img]'. $m[3].$m[6].$m[9].$m[12];

} else {

return $m[1].$m[4].$m[7].$m[10].$m[13] .'[url]'. $u['url'] .'[/url]'. $m[3].$m[6].$m[9].$m[12];

}

} else {

return $m[1].$m[4].$m[7].$m[10].$m[13]. $url. $m[3].$m[6].$m[9].$m[12];

}

{

global $lang_common, $errors, $pd;

// Get Slim current session

$feather = \Slim\Slim::getInstance();

// Initialize some local variables. Use reference variables where possible.

$tagname =& $matches[1]; // BBCode tag name.

$contents =& $matches[6]; // BBCode tag contents.

$tag =& $pd['bbcd'][$tagname]; // alias to this tags array element of the BBCD database

$parent = end($pd['tag_stack']); // Name of parent tag. ("_ROOT_" is base parent tag).

/* $new_errors = array(); */ // BBCode tag error messages. (Create on error.)

// First things first.

$tag['depth']++; // Increment tag-specific nesting level depth.

$tagname = strtolower($tagname); // Force lowercase tags name.

array_push($pd['tag_stack'], $tagname); // Push this tags name onto the tag stack.

// ---------------------------------------------------------------------------

// Recursively parse any nested BBCode tag markup (unless tag type is hidden):

// ---------------------------------------------------------------------------

if ($tag['tag_type'] !== 'hidden' && strpos($contents, '[') !== false) {

$contents = preg_replace_callback($pd['re_bbcode'], '_preparse_bbcode_callback', $contents);

if ($contents === null) {

// On error, preg_replace_callback returns NULL.

// Error #1: '(%s) Message is too long or too complex. Please shorten.'

$new_errors[] = sprintf($lang_common['BBerr pcre'], preg_error());

$contents = ''; // Zero out the contents.

}

}

// ---------------------------------------------------------------------------------------

// Process optional $attribute. Set $fmt_open, $fmt_close and $handler based on attribute.

// ---------------------------------------------------------------------------------------

$fmt_close = '[/'. $tagname .']'; // BBCode closing tag format specifier string.

if ($matches[2]) { // Check if attribute specified?

// Attribute specified. Pick value from one of the three possible quote delimitations.

if ($matches[3]) { // Non-empty single-quoted value.

$attribute =& $matches[3]; // Set attribute to quoted content.

$fmt_open = '['. $tagname .'=\'%a_str%\']'; // Set 'single-quoted' opening format.

} elseif ($matches[4]) { // Non-empty double-quoted value.

$attribute =& $matches[4]; // Set attribute to quoted content.

$fmt_open = '['. $tagname .'="%a_str%"]'; // Set "double-quoted" opening format.

} elseif ($matches[5]) { // Non-empty un-or-any-quoted value.

$attribute =& $matches[5]; // Set attribute to unquoted content.

$fmt_open = '['. $tagname .'=%a_str%]'; // Set un-'or'-"any"-quoted opening format.

} else { // Otherwise must be empty.

$attribute = ''; // Set empty attribute.

$fmt_open = '['. $tagname .'=%a_str%]'; // Set empty-attribute opening format.

}

// Consolidate consecutive attribute whitespace to a single space. Trim start and end.

$attribute = preg_replace(array('/\s++/S', '/^ /', '/ $/'), array(' ', '', ''), $attribute);

// Determine attribute handler: fixed or variable or none.

if (isset($tag['handlers'][$attribute])) { // If attribute matches handler key

$handler =& $tag['handlers'][$attribute]; // use the fixed-attribute handler.

} elseif (isset($tag['handlers']['ATTRIB'])) { // Else if we have one, use this tags

$handler =& $tag['handlers']['ATTRIB']; // variable attribute handler. Otherwise...

} elseif (isset($tag['handlers']['NO_ATTRIB']) && // Otherwise we have an erroneous attribute

count($tag['handlers']) === 1) {

// which is either unexpected or unrecognized.

// Error #2: 'Unexpected attribute: "%1$s". (No attribute allowed for [%2$s].'.

$handler =& $pd['bbcd']['_ROOT_']['handlers']['NO_ATTRIB'];

$new_errors[] = sprintf($lang_common['BBerr unexpected attribute'], $attribute, $tagname);

} else { // Error #3: 'Unrecognized attribute: "%1$s", is not valid for [%2$s].'

$handler =& $pd['bbcd']['_ROOT_']['handlers']['NO_ATTRIB'];

$new_errors[] = sprintf($lang_common['BBerr unrecognized attribute'], $attribute, $tagname);

}

// Make sure attribute does nor contain a valid BBcode tag.

if (preg_match($pd['re_bbtag'], $attribute)) { // Error #4: 'Attribute may NOT contain open or close bbcode tags'

$handler =& $pd['bbcd']['_ROOT_']['handlers']['NO_ATTRIB'];

$new_errors[] = $lang_common['BBerr bbcode attribute'];

}

// Validate and filter tag's attribute value if and according to custom attribute regex.

if (isset($handler['a_regex'])) { // Check if this tag has an attribute regex? (very rare)

if (preg_match($handler['a_regex'], $attribute, $m)) { // Yes. Check if regex matches attribute?

$attribute = $m[1];

} else { // Error #4b: 'Invalid attribute, [%s] requires specific attribute.'

$new_errors[] = sprintf($lang_common['BBerr invalid attrib'], $tagname);

}

}

} else { // Attribute not specified. Use the NO_ATTRIB handler if it exixts else error.

$attribute = ''; // No attribute? Make it so.

$fmt_open = '['. $tagname .']'; // Set no-attribute fmt_open string.

if (isset($tag['handlers']['NO_ATTRIB'])) { // If we have one, use this tags

$handler =& $tag['handlers']['NO_ATTRIB']; // no-attribute handler. Otherwise...

} else { // Error #5: '[%1$s] is missing a required attribute.'.

$handler =& $pd['bbcd']['_ROOT_']['handlers']['NO_ATTRIB'];

$new_errors[] = sprintf($lang_common['BBerr missing attribute'], $tagname);

}

}

// -------------------------------------------------------

// Do some validation checks. Fix problems where possible:

// -------------------------------------------------------

// Handle tag nesting depth overflow.

if ($tag['depth'] > $tag['depth_max']) { // Allowable tag nesting level exceeded?

switch ($tag['nest_type']) { // Overflow. Handle based upon tag's "nest_type"

case 'clip': // Silently strip overly nested tags and content.

$contents = '';

break;

case 'fix': // Silently strip overly-nested tags (keep contents).

$fmt_open = $fmt_close = '';

break;

case 'err': // Error #6: '[%1$s] tag nesting depth: %2$d exceeds allowable limit: %3$d.'.

$new_errors[] = sprintf($lang_common['BBerr nesting overflow'],

$tagname, $tag['depth'], $tag['depth_max']);

break;

default:

}

}

// Verify this tag is not in its parent's excluded tags list.

if (isset($pd['bbcd'][$parent]['tags_excluded'][$tagname])) {

// Are we illegitimate?

// Yes. Pick between error #6 and #7.

if ($parent === $tagname) { // Error #7: '[%s] was opened within itself, this is not allowed.'

$new_errors[] = sprintf($lang_common['BBerr self-nesting'], $tagname);

} else { // Error #8: '[%1$s] was opened within [%2$s], this is not allowed.'

$new_errors[] = sprintf($lang_common['BBerr invalid nesting'], $tagname, $parent);

}

}

// Verfify our parent tag is in our 'parents' allowable array if it exists.

if (isset($tag['parents']) && !isset($tag['parents'][$parent])) { // Error #9: '[%1$s] cannot be within: [%2$s]. Allowable parent tags: %3$s.'.

$new_errors[] = sprintf($lang_common['BBerr invalid parent'],

$tagname, $parent, '('. implode('), (', array_keys($tag['parents'])) .')');

}

// -----------------------------------------

// Perform content-type-specific processing:

// -----------------------------------------

switch ($handler['c_type']) {

case 'width_height':

if (preg_match('/\b(\d++)[Xx](\d++)\b/S', $contents, $m)) {

$width = (int)$m[1];

$height = (int)$m[2];

}

if (preg_match('/\bw(?:idth)?+\s*+=\s*+[\'"]?+(\d++)\b/Si', $contents, $m)) {

$width = (int)$m[1];

}

if (preg_match('/\bh(?:eight)?+\s*+=\s*+[\'"]?+(\d++)\b/Si', $contents, $m)) {

$height = (int)$m[1];

}

if (isset($height, $tag['x_padding'], $tag['y_padding'])) {

$height -= $tag['y_padding'] - $tag['x_padding']; // Adjust for height of embedded controller.

}

break;

case 'url':

// Sanitize contents which is (hopefully) a url link. Trim spaces.

$contents = preg_replace(array('/^\s+/', '/\s+$/S'), '', $contents);

// Handle special case link to a

if ($feather->user->g_post_links != '1') {

$new_errors[] = $lang_common['BBerr cannot post URLs'];

}

else if (($m = url_valid($contents))) {

$contents = $m['url']; // Fetch possibly more complete url address.

} else { // Error #10a: 'Invalid URL name: %s'.

$new_errors[] = sprintf($lang_common['BBerr Invalid URL name'], $contents);

}

break;

case 'email':

if (filter_var($contents, FILTER_VALIDATE_EMAIL)) { // Error #10c: 'Invalid email address: %s'.

$new_errors[] = sprintf($lang_common['BBerr Invalid email address'], $contents);

}

break;

default:

} // End c_type switch().

// -------------------------------------------

// Perform attribute-type-specific processing:

// -------------------------------------------

switch ($handler['a_type']) {

case 'width_height':

if ($attribute) {

if (preg_match('/\b(\d++)[Xx](\d++)\b/', $attribute, $m)) { // Check for a "123x456" WxH spec?

$width = (int)$m[1]; // Yes. Set both dimensions.

$height = (int)$m[2];

}

if (preg_match('/\bw(?:idth)?+\s*+=\s*+[\'"]?+(\d++)\b/i', $attribute, $m)) {

$width = (int)$m[1];

}

if (preg_match('/\bh(?:eight)?+\s*+=\s*+[\'"]?+(\d++)\b/i', $attribute, $m)) {

$height = (int)$m[1];

}

$attribute = preg_replace('/[;\s]?+\b(?:(?:w(?:idth)?+|h(?:eight)?+)\s*+=\s*+|\d++[Xx])\d++\b/Si',

'', $attribute);

}

break;

case 'url':

if (($m = url_valid($attribute))) {

$attribute = $m['url']; // Fetch possibly more complete url address.

} else { // Error #10b: 'Invalid URL name: %s'.

$new_errors[] = sprintf($lang_common['BBerr Invalid URL name'], $attribute);

}

break;

case 'color':

if (!preg_match($pd['re_color'], $attribute)) { // Error #11: 'Invalid color attribute: %s'.

$new_errors[] = sprintf($lang_common['BBerr Invalid color'], $attribute);

}

break;

case 'email':

// TODO: improve this quick-n-dirty email check.

if (!preg_match('/^[A-Z0-9._%+-]+@[A-Z0-9.-]+\.[A-Z]{2,6}$/i', $attribute)) { // Error #10c: 'Invalid email address: %s'.

$new_errors[] = sprintf($lang_common['BBerr Invalid email address'], $attribute);

}

break;

default:

} // End a_type switch().

// ----------------------------------------------------------

// Perform tag-specific processing of attribute and contents:

// ----------------------------------------------------------

switch ($tagname) {

case 'img': // Handle bad image url, file too big, then scale-to-fit within forum defaults if too large.

if ($tag['depth'] === 1) { // Check if not overly nested?

if (($pd['ipass'] === 2) && $pd['config']['valid_imgs'] && url_valid($contents)) { // Valid URI?

// Yes. Fetch file headers containing file type and size ("Content-Type" and "Content-Length").

if (($http = @get_headers($contents)) !== false && is_array($http)) {

if (preg_match('/\b200\s++OK\s*+$/i', $http[0])) { // Good response header?

for ($i = 1, $len = count($http); $i < $len; ++$i) { // Yes. Loop through HTTP response headers.

if (preg_match('/^\s*+Content-Length\s*+:\s*+(\d++)\s*+$/i', $http[$i], $m)) {

$size = (int)$m[1];

} // File size found.

if (preg_match('/^\s*+Content-Type\s*+:\s*+image\/(.++)$/i', $http[$i], $m)) {

$type = $m[1];

} // Image file type found.

}

// Verify Content-Type is an image.

if (isset($type)) {

// Verify remote file size is not too big. (If too big, handle error.)

if (isset($size)) {

if ($size <= $pd['config']['max_size']) {

// Filesize is ok. Do nothing.

if (($info = @getimagesize($contents)) && is_array($info)) {

// Fetch width & height.

// Now we know the filesize, width and height of remote image.

if (($iwidth = (int)$info[0]) && ($iheight = (int)$info[1])) {

// To resize or not resize, that is the question.

// If bigger than default, scale down. Otherwise dont touch.

// Scale image to fit within forum default width/height box dimensions.

$ar = ((float)$iwidth) / ((float)$iheight);

// Otherwise, for images that naturally fit inside the box,

// leave the attribute clean (or unset).

if (!isset($width) && !isset($height) &&

($iwidth > $pd['config']['def_width'] ||

$iheight > $pd['config']['def_height'])

) { // Remote file dimensions are too big to fit within default box.

// Explicitly scale a new width and height in IMG attribute.

$width = $pd['config']['def_width'];

$height = (int)((((float)$width) / $ar) + 0.5);

if ($height > $pd['config']['def_height']) {

$height = $pd['config']['def_height'];

$width = (int)((((float)$height) * $ar) + 0.5);

}

} // Else remote image fits. Do nothing special with width and height.

}

} else { // Error #13: 'Unable to retrieve image data from remote url: %s'.

$new_errors[] = sprintf($lang_common['BBerr bad meta data'], $contents);

} // NOTE: cannot generate this error.

} else { // Filesize of remote image is too big. Silently convert to link if possible.

if (isset($pd['bbcd']['url']) && $pd['bbcd']['url']['depth'] === 0) {

$fmt_open = '{[url='. $contents .']';

$fmt_close = '[/url]}';

$contents = $lang_common['BBmsg big image'];

} else { // Image within a url cannot be linkified. Just display url name.

$contents = '{'. $contents .'}';

$fmt_open = '';

$fmt_close = '';

}

}

} else {

// $size not set.

// Error #14: 'Unable to determine remote file size.'.

$new_errors[] = $lang_common['BBerr no file size'];

} // NOTE: cannot generate this error.

} else { // Error #15: 'Remote url does not have Content-Type: "image".'

$new_errors[] = $lang_common['BBerr non image'];

}

} else { // Error #16: 'Bad HTTP response header: "%s"'.

$new_errors[] = sprintf($lang_common['BBerr bad http response'], $http[0]);

}

} else { // Error #17: 'Unable to read remote image http headers.'.

$new_errors[] = $lang_common['BBerr bad headers'];

}

} // Image validation turned off. Do nothing.

} else { // Non-Error: IMG tag self nesting. Handle by silently stripping tags with no error.

$fmt_open = $fmt_close = '';

}

break;

case 'list': // Fixup lists within lists. In lists, everything must be in a [*] tag.

// Check if LIST contents well-formed.

if ($pd['ipass'] === 2 && !preg_match('% # Rev:20110220_1200

%x', $contents)) { // Not well formed. Do fixup to ensure list contents are only * tags.

// First regex wraps invalid characters at start of LIST in a [*]...[/*] tag.

$contents = preg_replace($pd['re_fixlist_1'], '[*]$1[/*]', $contents);

// Second regex wraps invalid characters between [/*] and [*] (or [/list]).

$contents = preg_replace($pd['re_fixlist_2'], '$1[/*]', $contents);

} // Well-formed LIST contents!

if ($parent === 'list') {

$fmt_open = '[*]'. $fmt_open;

$fmt_close .= '[/*]';

}

break;

default:

break;

} // End switch statement.

// -------------------------------------------

// Process width and height values if present.

// -------------------------------------------

if (isset($width) || isset($height)) { // Check if dimension specified in attrib or contents?

// Yes. Clip both $width and/or $height to their respective config maximums.

if (isset($width)) { // Clip to max. Set to default if zero.

if ($width > $pd['config']['max_width']) {

$width = $pd['config']['max_width'];

} elseif ($width === 0) {

$width = $pd['config']['def_width'];

}

}

if (isset($height)) { // Clip to max. Set to default if zero.

if ($height > $pd['config']['max_height']) {

$height = $pd['config']['max_height'];

} elseif ($height === 0) {

$height = $pd['config']['def_height'];

}

}

if (isset($ar)) { // If the real image dimensions are known ($ar), then adjust to fit in box and maintain $ar.

if (isset($width) && isset($height)) { // Check if both dimensions set?

if ($ar > (((float)$width) / ((float)$height))) { // Yes. Check if $width more precise than $height?

$height = (int)((((float)$width) / $ar) + 0.5); // Yes. Compute height from width and AR.

if ($height > $pd['config']['max_height']) {

$height = $pd['config']['max_height'];

$width = (int)((((float)$height) * $ar) + 0.5);

}

} else {

$width = (int)((((float)$height) * $ar) + 0.5); // Compute width from height and AR.

if ($width > $pd['config']['max_width']) {

$width = $pd['config']['max_width'];

$height = (int)((((float)$width) / $ar) + 0.5);

}

}

} elseif (isset($width)) {

$height = (int)((((float)$width) / $ar) + 0.5); // Compute height from width and AR.

if ($height > $pd['config']['max_height']) {

$height = $pd['config']['max_height'];

$width = (int)((((float)$height) * $ar) + 0.5);

}

} else {

$width = (int)((((float)$height) * $ar) + 0.5); // Compute width from height and AR.

if ($width > $pd['config']['max_width']) {

$width = $pd['config']['max_width'];

$height = (int)((((float)$width) / $ar) + 0.5);

}

}

}

// Unconditionally write width and/or height data back into attribute.

if ($width === 0) {

$width = 1;

}

if ($height === 0) {

$height = 1;

}

if ($attribute) {

$attribute .= ';';

} // Add delimiter for non-empty attrib.

if (isset($width) && isset($height)) {

$attribute .= $width .'x'. $height;

} elseif (isset($width)) {

$attribute .= 'w='. $width;

} else {

$attribute .= 'h='. $height;

}

$fmt_open = '['. $tagname .'=%a_str%]'; // Set open tag format to receive attribute.

}

// Validate and filter tag's contents if and according to optional contents regex.

if (isset($handler['c_regex'])) { // Check if this tag has a contents regex? (youtube, vimeo, etc.)

// Yes. Check if regex matches contents?

if (preg_match($handler['c_regex'], $contents, $m)) {

$contents = $m[1];

} else { // Error #12: 'Invalid content, [%s] requires specific content.'

$new_errors[] = sprintf($lang_common['BBerr invalid content'], $tagname);

}

}

// Silently strip empty or all-white tags:

if (preg_match('/^\s*+$/', $contents)) {

$contents = '';

}

// Unconditionally hide all opening square brackets within hidden CODE contents.

// This is necessary otherwise the LIST fixup code would process "[*]" within CODE tags.

// These \3 byte markers are subsequently removed by preparse_bbcode().

if ($tag['tag_type'] === 'hidden') {

$contents = str_replace('[', "\3", $contents);

}

// On first pass, fix inline tags which span paragraphs by closting then re-opening.

if ($pd['ipass'] === 1 && $tag['html_type'] === 'inline' &&

$tag['tag_type'] !== 'hidden' && strpos($contents, "\n") !== false) {

$contents = preg_replace('/\n\s*?\n\s*/',

"\1\2". $fmt_close ."\1".'$0'."\1\2". str_replace('%a_str%', $attribute, $fmt_open) ."\1", $contents);

}

// ***********************************************************************************

// Handle errors. Wrap this tags open and close BBCode tag each in a valid [err] tag.

// ***********************************************************************************

if (isset($new_errors) && $fmt_open) {

// check if we detected any errors?

// Yes, we have detected one or more new error conditions.

foreach ($new_errors as $errmsg) { // Push all new errors on g errors array.

$pd['new_errors'][] = htmlspecialchars($errmsg);

}

// Wrap offending BBCode open and close tags each in its own valid error tag (last err only).

$fmt_open = '[err='. $errmsg .']'. $fmt_open . '[/err]'; // Wrap tags in the last error message.

$fmt_close = '[err='. $errmsg .']'. $fmt_close . '[/err]';

}

// -----------------------------------------------------------------------------

// All done processing. Substitute $attribute and $contents into format strings:

// -----------------------------------------------------------------------------

if ($contents) {

if ($pd['ipass'] === 1) { // Add byte markers on first pass.

if ($tag['tag_type'] === 'hidden' || $handler['c_type'] == 'url') {

$text = "\1\2". $fmt_open .'%c_str%'. $fmt_close ."\1";

} else {

$text = "\1\2". $fmt_open ."\1%c_str%\1\2". $fmt_close ."\1";

}

} else {

$text = $fmt_open .'%c_str%'. $fmt_close;

} // Pass 2, dont bother with byte markers.

$text = str_replace('%a_str%', $attribute, $text);

$text = str_replace('%c_str%', $contents, $text);

} else {

$text = '';

}

array_pop($pd['tag_stack']); // Were done. Pop this tag off the stack.

$tag['depth']--; // Restore pre-call tag specific depth.

return $text;

{

global $lang_common, $feather_config, $pd;

$pd['new_errors'] = array(); // Reset the parser error message stack.

$pd['in_signature'] = ($is_signature) ? true : false;

$pd['ipass'] = 1;

$newtext = preg_replace_callback($pd['re_bbcode'], '_preparse_bbcode_callback', $text);

if ($newtext === null) {

// On error, preg_replace_callback returns NULL.

// Error #1: '(%s) Message is too long or too complex. Please shorten.'

$errors[] = sprintf($lang_common['BBerr pcre'], preg_error());

return $text;

}

$newtext = str_replace("\3", '[', $newtext); // Fixup CODE sections.

$parts = explode("\1", $newtext); // Hidden chunks pre-marked like so: "\1\2<code.../code>\1"

for ($i = 0, $len = count($parts); $i < $len; ++$i) { // Loop through hidden and non-hidden text chunks.

$part =& $parts[$i]; // Use shortcut alias

if (empty($part)) {

continue;

} // Skip empty string chunks.

if ($part[0] !== "\2") { // If not hidden, process this normal text content.

// Mark erroneous orphan tags.

$part = preg_replace_callback($pd['re_bbtag'], '_orphan_callback', $part);

// Process do-clickeys if enabled.

if ($feather_config['o_make_links']) {

$part = linkify($part);

}

// Process textile syntax tag shortcuts.

if ($pd['config']['textile']) {

// Do phrase replacements.

$part = preg_replace_callback($pd['re_textile'],

'_textile_phrase_callback', $part);

// Do lists.

$part = preg_replace_callback('/^([*#]) .*+(?:\n\1 .*+)++$/Sm',

'_textile_list_callback', $part);

}

$part = preg_replace('/^[ \t]++$/m', '', $part); // Clear "white" lines of spaces and tabs.

} else {

$part = substr($part, 1);

} // For hidden chunks, strip \2 marker byte.

}

$text = implode("", $parts); // Put hidden and non-hidden chunks back together.

$pd['ipass'] = 2; // Run a second pass through parser to clean changed content.

$text = preg_replace_callback($pd['re_bbcode'], '_preparse_bbcode_callback', $text);

$text = str_replace("\3", '[', $text); // Fixup CODE sections.

if (!empty($pd['new_errors'])) {

foreach ($pd['new_errors'] as $errmsg) {

$errors[] = $errmsg;

} // Push all new errors on global array.

}

return $text;

{

global $pd, $lang_common;

if ($matches[0][1] === '/') { // Error #18: 'Orphan close tag: [/%s] is missing its open tag.'

$errmsg = sprintf($lang_common['BBerr orphan close'], $matches[1]);

} else { // Error #19: 'Orphan open tag: [%s] is missing its close tag.'

$errmsg = sprintf($lang_common['BBerr orphan open'], $matches[1]);

}

$pd['new_errors'][] = $errmsg; // Append to array of errors so far.

return '[err='. $errmsg .']'. $matches[0] .'[/err]';

{

global $pd;

if (!isset($pd['bbcd']['list'])) {

return $matches[0];

}

$parts = preg_split('/(?:^|\n)\\'. $matches[1] .' /S', $matches[0], -1, PREG_SPLIT_NO_EMPTY);

switch ($matches[1]) {

case '*': return "[list]\n[*]". implode("[/*]\n[*]", $parts) ."[/*]\n[/list]";

case '#': return "[list=1]\n[*]". implode("[/*]\n[*]", $parts) ."[/*]\n[/list]";

}

{

global $pd;

$matches[2] = preg_replace_callback($pd['re_textile'], '_textile_phrase_callback', $matches[2]);

switch ($matches[1]) {

case '_': return (isset($pd['bbcd']['i'])) ? '[i]'. $matches[2] .'[/i]' : $matches[0];

case '*': return (isset($pd['bbcd']['b'])) ? '[b]'. $matches[2] .'[/b]' : $matches[0];

case '@': return (isset($pd['bbcd']['tt'])) ? '[tt]'. $matches[2] .'[/tt]' : $matches[0];

case '^': return (isset($pd['bbcd']['sup'])) ? '[sup]'. $matches[2] .'[/sup]' : $matches[0];

case '~': return (isset($pd['bbcd']['sub'])) ? '[sub]'. $matches[2] .'[/sub]' : $matches[0];

case '-': return (isset($pd['bbcd']['del'])) ? '[del]'. $matches[2] .'[/del]' : $matches[0];

case '+': return (isset($pd['bbcd']['ins'])) ? '[ins]'. $matches[2] .'[/ins]' : $matches[0];

}

{

global $pd;

$errmsg = ''; // assume no error has occured. return empty string

if ($pd['newer_php_version']) { // this function requires PHP 5.2.0 or higher

switch (preg_last_error()) { // this returns the last error condition as a number

case PREG_NO_ERROR:

break; // no error? return empty string === FALSE

case PREG_INTERNAL_ERROR:

$errmsg = 'PREG_INTERNAL_ERROR';

break;

case PREG_BACKTRACK_LIMIT_ERROR:

$errmsg = 'PREG_BACKTRACK_LIMIT_ERROR';

break;

case PREG_RECURSION_LIMIT_ERROR:

$errmsg = 'PREG_RECURSION_LIMIT_ERROR';

break;

case PREG_BAD_UTF8_ERROR:

$errmsg = 'PREG_BAD_UTF8_ERROR';

break;

case PREG_BAD_UTF8_OFFSET_ERROR:

$errmsg = 'PREG_BAD_UTF8_OFFSET_ERROR';

break;

default:

$errmsg = 'Unrecognized PREG error';

break;

}

}

return $errmsg;

{

global $pd, $lang_common;

$tagname =& $matches[1]; // TAGNAME we are currently servicing.

$contents =& $matches[6]; // Shortcut to contents.

$tag =& $pd['bbcd'][$tagname]; // Shortcut to bbcd array entry.

$tag['depth']++; // update tag-specific nesting level

// Set local parse enable flag based upon message type and global flags.

$enabled = (!$pd['in_signature'] && $tag['in_post'] || // If in a post and post-enabled, or

$pd['in_signature'] && $tag['in_sig']) ? true : false; // in a sig and sig-enabled, then enabled.

// Recursively parse any nested BBCode tag markup (unless tag type is hidden).

if ($tag['tag_type'] !== 'hidden' && strpos($contents, '[') !== false) {

$contents = preg_replace_callback($pd['re_bbcode'], '_parse_bbcode_callback', $contents);

}

// ------------------------------------------------------------------------------

// Determine $attribute and format conversion $handler to use based on attribute.

// ------------------------------------------------------------------------------

if (!$matches[2]) { // No attribute specified? Use the NO_ATTRIB handler.

$attribute = '';

if (isset($tag['handlers']['NO_ATTRIB'])) { // If we have one, use this tags

$handler =& $tag['handlers']['NO_ATTRIB'];

} // no-attribute handler. Otherwise...

else {

$handler =& $pd['bbcd']['_ROOT_']['handlers']['NO_ATTRIB'];

} // Missing attribute! Strip tag.

} else { // Attribute specified. Assign it from one of the three possible delimitations.

if ($matches[3]) { // Non-empty single-quoted value.

$attribute =& $matches[3]; // Strip out escape from escapes and single quotes.

$attribute = str_replace(array('\\\\', '\\\''), array('\\', '\''), $attribute);

} elseif ($matches[4]) { // Non-empty double-quoted value.

$attribute =& $matches[4]; // Strip out escape from escapes and double quotes.

$attribute = str_replace(array('\\\\', '\\"'), array('\\', '"'), $attribute);

} elseif ($matches[5]) {

$attribute =& $matches[5];

} // Non-empty un-or-any-quoted value.

else {

$attribute = '';

} // Otherwise must be empty.

// Determine which type of attribute handler: fixed or variable.

if (isset($tag['handlers'][$attribute])) { // If attribute matches handler key

$handler =& $tag['handlers'][$attribute]; // use the fixed-attribute handler.

} elseif (isset($tag['handlers']['ATTRIB'])) { // Else if we have one, use this tags

$handler =& $tag['handlers']['ATTRIB']; // variable attribute handler. Otherwise...

} else {

$handler =& $pd['bbcd']['_ROOT_']['handlers']['NO_ATTRIB'];

} // Missing attribute! Strip tag.

// htmlspecialchars() was already called (but not for quotes).

// Hide any double quotes in attribute now.

$attribute = str_replace('"', '&quot;', $attribute);

}

// Set default format handler

$format = $handler['format'];

// -------------------------------------------

// Perform attribute-type-specific processing:

// -------------------------------------------

switch ($handler['a_type']) {

case 'width_height': // This attribute type is used for video BBCodes (YouTube, Vimeo, ...).

if ($attribute) {

if (preg_match('/\b(\d++)[Xx](\d++)\b/S', $attribute, $m)) {

// Check for a "123x456" WxH spec?

$width = (int)$m[1]; // Yes. Set both dimensions.

$height = (int)$m[2];

}

if (preg_match('%\bw(?:idth)?+\s*+=\s*+[\'"]?+(\d++)\b%Si', $attribute, $m)) {

$width = (int)$m[1];

}

if (preg_match('%\bh(?:eight)?+\s*+=\s*+[\'"]?+(\d++)\b%Si', $attribute, $m)) {

$height = (int)$m[1];

}

if (isset($width) && isset($height)) {

$ar = (float)$width / (float)$height;

}

// Clean the attribute of any and all width/height specs.

$attribute = preg_replace('/[ ;#]?\b(?:(?:w(?:idth)?+|h(?:eight)?+)\s*+=\s*+([\'"])?+|\d++X)\d++\b(?(1)\1)?+/Si',

'', $attribute);

if (isset($width)) { // If set, clip to max allowed. Convert zero to global default.

if ($width > $pd['config']['max_width']) {

$width = $pd['config']['max_width'];

} elseif ($width === 0) {

$width = $pd['config']['def_width'];

} // If zero, set to default.

}

if (isset($height)) { // If set, clip to max allowed. Convert zero to global default.

if ($height > $pd['config']['max_height']) {

$height = $pd['config']['max_height'];

} elseif ($height === 0) {

$height = $pd['config']['def_height'];

} // If zero, set to default.

}

if (!isset($width)) {

$format = str_replace('width="%w_str%"', '', $format);

}

if (!isset($height)) {

$format = str_replace('height="%h_str%"', '', $format);

}

} else {

$width = $pd['config']['def_width'];

$height = $pd['config']['def_height'];

}

if (isset($tag['x_padding']) && $width) {

$width += $tag['x_padding'];

}

if (isset($tag['y_padding']) && $height) {

$height += $tag['y_padding'];

}

break;

default:

}

// ----------------------------------------------------------

// Perform tag-specific processing of attribute and contents:

// ----------------------------------------------------------

switch ($tagname) {

case 'img': // Handle disabled image, image inside of QUOTE, and width/height dimensions in attribute.

if (!$enabled || (!$pd['config']['quote_imgs'] && isset($pd['bbcd']['quote']) && $pd['bbcd']['quote']['depth'] > 0)) { // IMG not enabled in this context. Convert to a text URL link if possible and re-enable.

if (isset($pd['bbcd']['url']) && $pd['bbcd']['url']['depth'] > 0) {

$format = "{%c_str%}";

} else {

if ($attribute) {

$format = '{<a href="%c_str%" title="%a_str%">'. $lang_common['Image link'] .'</a>}';

} else {

$format = '{<a href="%c_str%" title="'. $lang_common['BBmsg images disabled'] .'">'.

$lang_common['Image link'] .'</a>}';

}

$enabled = true; // Re-enable to override defauslt disabled handling (i.e. dont delete.)

}

} else { // IMG is enabled in this context. Wrap image inside a clickable link if global option is set.

if ($pd['config']['click_imgs'] && isset($pd['bbcd']['url']) && $pd['bbcd']['url']['depth'] === 0) {

$format = preg_replace('/^\x01\x02([^\x01]*+)\x01$/', '<a href="%c_str%">$1</a>', $format);

}

}

$format = "\1\2". $format ."\1";

// If user provided a verbose attribute (in addition to WxH), place this in the title attribute,

// otherwise place the URL ($contents) in the ALT attribute and remove the title attribute from format.

if (preg_match('/^\s*$/', $attribute)) {

$attribute = $contents;

$format = str_replace(' title="%a_str%"', '', $format);

}

break;

case 'quote': // Quotes require language-specific "wrote:" following the posters name.

if ($attribute) { // Quote attribute specified. Convert optional #post_id to a post link.

if (preg_match('/#(\d++)$/', $attribute, $m)) {

// Check for optional embedded post id.

// Attribute has optional '#1234' quoted post ID number. Convert to link back to quoted post.

$attribute = preg_replace('/\s*#\d++$/S', '', $attribute); // Strip post id from attribute.

$attribute .= ' '. $lang_common['wrote']; // Append language-specific "wrote:".

if ($pd['config']['quote_links']) {

$attribute = ' <a href="'. get_link('post/'.$m[1].'/#p'.$m[1]) .'">'. $attribute .'</a>';

}

} else {

$attribute .= ' '. $lang_common['wrote'];

} // If no post id, just add "wrote:".

}

break;

case 'list': // Clean whitespace cruft bordering <li> tags.

case 'table': // Clean whitespace cruft bordering <tr> tags.

case 'tr': // Clean whitespace cruft bordering <td> tags.

$contents = preg_replace(array('/^\s++/', '/\s+$/S', '%\s+(?=\x01\x02<(?:li|ol|ul|tr|td)>)%S', '%(?<=</(?:li|ol|ul|tr|td)>\x01)\s++%S'), '', $contents);

break;

case 'code':

if ($attribute) {

if (!isset($tag['handlers'][$attribute])) {

// Check for no matching attribute handler.

// Yes, the attribute may be more complex. Extract the extra verbage to add to code header.

if (preg_match('/^([\w\-.:]++)\s*+(.*)$/', $attribute, $m)) {

$type = strtolower($m[1]);

if (isset($tag['handlers'][$type])) { // Check if we recognize this first word?

$handler =& $tag['handlers'][$type]; // Yes. Set new fixed-attribute handler.

if ($m[2]) { // Check if there are extra words.

// Yes. Use this extra part of the attribute to augment the code header.

$format = str_replace('</h4>', ' - "'. $m[2] .'"</h4>', $handler['format']);

}

$attribute = $type;

}

}

}

}

break;

case 'color':

case 'colour':

if (preg_match($pd['re_color'], $attribute, $m)) {

if (isset($m[2]) && $m[2]) {

$attribute = $m[1] .'; background-color: '. $m[2];

} else {

$attribute = $m[1];

}

}

default:

}

// The return value depends upon whether this tag is currently enabled.

if ($enabled) { // Tag is enabled. Add byte markers to hidden chunks to allow easy explode() later.

if (isset($width)) {

$format = str_replace('%w_str%', $width, $format);

}

if (isset($height)) {

$format = str_replace('%h_str%', $height, $format);

}

// Subtitute attribute into format string.

$format = str_replace('%a_str%', $attribute, $format); // Encode attribute value.

// Finally, subtitute the content into the message.

$text = str_replace('%c_str%', $contents, $format); // Encode contents.

} else { // Tag is not enabled. Strip HTML tags and contents according to tag_type.

switch ($tag['tag_type']) {

case 'atomic': // Strip everything.

$text = '';

break;

case 'normal': // Strip HTML open and close tags only.

$text =& $contents;

break;

case 'hidden': // Hidden and 'zombie' tags are displayed without.

case 'zombie': // stripping anything. All [BBCODES] are displayed.