Bbcode.php:
<?php
/*
===========================
boastMachine v3.1
Released : Sunday, June 5th 2005 ( 06/05/2005 )
http://boastology.com
Developed by Kailash Nadh
Email : mail <at> kailashnadh <dot> name
Website : http://kailashnadh.name, http://bnsoft.net
Blog : http://boastology.com/blog
boastMachine is a free software licensed under GPL (General public license)
===========================
*/
// BB CODE PARSER FOR boastMachine
// bbCode parsing engine adopted from phpBB2 (http://phpbb.com)
//error_reporting(E_ALL ^ E_NOTICE);
function bmc_bbCode($text=null) {
if(!$text) return false;
// Initialize random numbers
mt_srand( (double) microtime() * 1000000);
// Unique ID for this text..
$uid = md5(mt_rand());
$uid = substr($uid, 0, 10);
return bbencode_second_pass(bbencode_first_pass($text, $uid ), $uid);
}
// global that holds loaded-and-prepared bbcode templates, so we only have to do
// that stuff once.
$bbcode_tpl = null;
/**
* Loads bbcode templates from the bbcode.tpl file of the current template set.
* Creates an array, keys are bbcode names like "b_open" or "url", values
* are the associated template.
* Probably pukes all over the place if there's something really screwed
* with the bbcode.tpl file.
*
* Nathan Codding, Sept 26 2001.
*/
function load_bbcode_template()
{
global $template;
$tpl = fread(fopen(CFG_ROOT."/inc/users/bbcode.inc", "r"), filesize(CFG_ROOT."/inc/users/bbcode.inc"));
// replace \ with \\ and then ' with \'.
$tpl = str_replace('\\', '\\\\', $tpl);
$tpl = str_replace('\'', '\\\'', $tpl);
// strip newlines.
$tpl = str_replace("\n", '', $tpl);
// Turn template blocks into PHP assignment statements for the values of $bbcode_tpls..
$tpl = preg_replace('#<!-- BEGIN (.*?) -->(.*?)<!-- END (.*?) -->#', "\n" . '$bbcode_tpls[\'\\1\'] = \'\\2\';', $tpl);
$bbcode_tpls = array();
eval($tpl);
return $bbcode_tpls;
}
/**
* Prepares the loaded bbcode templates for insertion into preg_replace()
* or str_replace() calls in the bbencode_second_pass functions. This
* means replacing template placeholders with the appropriate preg backrefs
* or with language vars. NOTE: If you change how the regexps work in
* bbencode_second_pass(), you MUST change this function.
*
* Nathan Codding, Sept 26 2001
*
*/
function prepare_bbcode_template($bbcode_tpl)
{
global $lang;
$bbcode_tpl['olist_open'] = str_replace('{LIST_TYPE}', '\\1', $bbcode_tpl['olist_open']);
$bbcode_tpl['color_open'] = str_replace('{COLOR}', '\\1', $bbcode_tpl['color_open']);
$bbcode_tpl['font_open'] = str_replace('{FONT}', '\\1', $bbcode_tpl['font_open']); // boastMachine 3.1
$bbcode_tpl['size_open'] = str_replace('{SIZE}', '\\1', $bbcode_tpl['size_open']);
$bbcode_tpl['quote_open'] = str_replace('{L_QUOTE}', "QUOTE:", $bbcode_tpl['quote_open']);
$bbcode_tpl['code_open'] = str_replace('{L_CODE}', "CODE:", $bbcode_tpl['code_open']);
$bbcode_tpl['img'] = str_replace('{URL}', '\\1', $bbcode_tpl['img']);
// We do URLs in several different ways..
$bbcode_tpl['url1'] = str_replace('{URL}', '\\1', $bbcode_tpl['url']);
$bbcode_tpl['url1'] = str_replace('{DESCRIPTION}', '\\1', $bbcode_tpl['url1']);
$bbcode_tpl['url2'] = str_replace('{URL}', 'http://\\1', $bbcode_tpl['url']);
$bbcode_tpl['url2'] = str_replace('{DESCRIPTION}', '\\1', $bbcode_tpl['url2']);
$bbcode_tpl['url3'] = str_replace('{URL}', '\\1', $bbcode_tpl['url']);
$bbcode_tpl['url3'] = str_replace('{DESCRIPTION}', '\\2', $bbcode_tpl['url3']);
$bbcode_tpl['url4'] = str_replace('{URL}', 'http://\\1', $bbcode_tpl['url']);
$bbcode_tpl['url4'] = str_replace('{DESCRIPTION}', '\\3', $bbcode_tpl['url4']);
$bbcode_tpl['email'] = str_replace('{EMAIL}', '\\1', $bbcode_tpl['email']);
define("BBCODE_TPL_READY", true);
return $bbcode_tpl;
}
/**
* Does second-pass bbencoding. This should be used before displaying the message in
* a thread. Assumes the message is already first-pass encoded, and we are given the
* correct UID as used in first-pass encoding.
*/
function bbencode_second_pass($text, $uid)
{
global $lang, $bbcode_tpl;
// pad it with a space so we can distinguish between FALSE and matching the 1st char (index 0).
// This is important; bbencode_quote(), bbencode_list(), and bbencode_code() all depend on it.
$text = " " . $text;
// First: If there isn't a "[" and a "]" in the message, don't bother.
if (! (strpos($text, "[") && strpos($text, "]")) )
{
// Remove padding, return.
$text = substr($text, 1);
return $text;
}
// Only load the templates ONCE..
if (!defined("BBCODE_TPL_READY"))
{
// load templates from file into array.
$bbcode_tpl = load_bbcode_template();
// prepare array for use in regexps.
$bbcode_tpl = prepare_bbcode_template($bbcode_tpl);
}
// [CODE] and [/CODE] for posting code (HTML, PHP, C etc etc) in your posts.
$text = bbencode_second_pass_code($text, $uid, $bbcode_tpl);
// [QUOTE] and [/QUOTE] for posting replies with quote, or just for quoting stuff.
$text = str_replace("[quote:$uid]", $bbcode_tpl['quote_open'], $text);
$text = str_replace("[/quote:$uid]", $bbcode_tpl['quote_close'], $text);
// [list] and [list=x] for (un)ordered lists.
// unordered lists
$text = str_replace("[list:$uid]", $bbcode_tpl['ulist_open'], $text);
// li tags
$text = str_replace("[*:$uid]", $bbcode_tpl['listitem'], $text);
// ending tags
$text = str_replace("[/list:u:$uid]", $bbcode_tpl['ulist_close'], $text);
$text = str_replace("[/list:o:$uid]", $bbcode_tpl['olist_close'], $text);
// Ordered lists
$text = preg_replace("/\[list=([a1]):$uid\]/si", $bbcode_tpl['olist_open'], $text);
// colours
$text = preg_replace("/\[color=(\#[0-9A-F]{6}|[a-z]+):$uid\]/si", $bbcode_tpl['color_open'], $text);
$text = str_replace("[/color:$uid]", $bbcode_tpl['color_close'], $text);
// fonts (boastMachine 3.1)
$text = preg_replace("/\[font=([a-z\s]+):$uid\]/si", $bbcode_tpl['font_open'], $text);
$text = str_replace("[/font:$uid]", $bbcode_tpl['font_close'], $text);
// size
$text = preg_replace("/\[size=([1-2]?[0-9]):$uid\]/si", $bbcode_tpl['size_open'], $text);
$text = str_replace("[/size:$uid]", $bbcode_tpl['size_close'], $text);
// [b] and [/b] for bolding text.
$text = str_replace("[b:$uid]", $bbcode_tpl['b_open'], $text);
$text = str_replace("[/b:$uid]", $bbcode_tpl['b_close'], $text);
// [s] and [/s] for striking text. (boastMachine 3.1)
$text = str_replace("[s:$uid]", $bbcode_tpl['s_open'], $text);
$text = str_replace("[/s:$uid]", $bbcode_tpl['s_close'], $text);
// [u] and [/u] for underlining text.
$text = str_replace("[u:$uid]", $bbcode_tpl['u_open'], $text);
$text = str_replace("[/u:$uid]", $bbcode_tpl['u_close'], $text);
// [i] and [/i] for italicizing text.
$text = str_replace("[i:$uid]", $bbcode_tpl['i_open'], $text);
$text = str_replace("[/i:$uid]", $bbcode_tpl['i_close'], $text);
// Patterns and replacements for URL and email tags..
$patterns = array();
$replacements = array();
// [img]image_url_here[/img] code..
// This one gets first-passed..
$patterns[] = "#\[img:$uid\](.*?)\[/img:$uid\]#si";
$replacements[] = $bbcode_tpl['img'];
// matches a [url]xxxx://www.site.com[/url] code..
$patterns[] = "#\[url\]([\w]+?://[^ \"\n\r\t<]*?)\[/url\]#is";
$replacements[] = $bbcode_tpl['url1'];
// [url]www.site.com[/url] code.. (no xxxx:// prefix).
$patterns[] = "#\[url\]((www|ftp)\.[^ \"\n\r\t<]*?)\[/url\]#is";
$replacements[] = $bbcode_tpl['url2'];
// [url=xxxx://www.site.com]Site[/url] code..
$patterns[] = "#\[url=([\w]+?://[^ \"\n\r\t<]*?)\](.*?)\[/url\]#is";
$replacements[] = $bbcode_tpl['url3'];
// [url=www.site.com]Site[/url] code.. (no xxxx:// prefix).
$patterns[] = "#\[url=((www|ftp)\.[^ \"\n\r\t<]*?)\](.*?)\[/url\]#is";
$replacements[] = $bbcode_tpl['url4'];
// [email]user <at> domain <dot> tld[/email] code..
$patterns[] = "#\[email\]([a-z0-9&\-_.]+?@[\w\-]+\.([\w\-\.]+\.)?[\w]+)\[/email\]#si";
$replacements[] = $bbcode_tpl['email'];
$text = preg_replace($patterns, $replacements, $text);
// Remove our padding from the string..
$text = substr($text, 1);
return $text;
} // bbencode_second_pass()
function bbencode_first_pass($text, $uid)
{
// pad it with a space so we can distinguish between FALSE and matching the 1st char (index 0).
// This is important; bbencode_quote(), bbencode_list(), and bbencode_code() all depend on it.
$text = " " . $text;
// [CODE] and [/CODE] for posting code (HTML, PHP, C etc etc) in your posts.
$text = bbencode_first_pass_pda($text, $uid, '[code]', '[/code]', '', true, '');
// [QUOTE] and [/QUOTE] for posting replies with quote, or just for quoting stuff.
$text = bbencode_first_pass_pda($text, $uid, '[quote]', '[/quote]', '', false, '');
$text = bbencode_first_pass_pda($text, $uid, '/\[quote=(\\\".*?\\\")\]/is', '[/quote]', '', false, '', "[quote:$uid=\\1]");
// [list] and [list=x] for (un)ordered lists.
$open_tag = array();
$open_tag[0] = "[list]";
// unordered..
$text = bbencode_first_pass_pda($text, $uid, $open_tag, "[/list]", "[/list:u]", false, 'replace_listitems');
$open_tag[0] = "[list=1]";
$open_tag[1] = "[list=a]";
// ordered.
$text = bbencode_first_pass_pda($text, $uid, $open_tag, "[/list]", "[/list:o]", false, 'replace_listitems');
// [color] and [/color] for setting text color
$text = preg_replace("#\[color=(\#[0-9A-F]{6}|[a-z\-]+)\](.*?)\[/color\]#si", "[color=\\1:$uid]\\2[/color:$uid]", $text);
// [font] and [/font] for setting font (boastMachine 3.1)
$text = preg_replace("#\[font=([a-z\s]+)\](.*?)\[/font\]#si", "[font=\\1:$uid]\\2[/font:$uid]", $text);
// [size] and [/size] for setting text size
$text = preg_replace("#\[size=([1-2]?[0-9])\](.*?)\[/size\]#si", "[size=\\1:$uid]\\2[/size:$uid]", $text);
// [b] and [/b] for bolding text.
$text = preg_replace("#\[b\](.*?)\[/b\]#si", "[b:$uid]\\1[/b:$uid]", $text);
// [bs] and [/s] for striking text.(boastMachine 3.1)
$text = preg_replace("#\[s\](.*?)\[/s\]#si", "[s:$uid]\\1[/s:$uid]", $text);
// [u] and [/u] for underlining text.
$text = preg_replace("#\[u\](.*?)\[/u\]#si", "[u:$uid]\\1[/u:$uid]", $text);
// [i] and [/i] for italicizing text.
$text = preg_replace("#\[i\](.*?)\[/i\]#si", "[i:$uid]\\1[/i:$uid]", $text);
// [img]image_url_here[/img] code..
$text = preg_replace("#\[img\]((ht|f)tp://)([^ \?&=\"\n\r\t<]*?(\.(jpg|jpeg|gif|png)))\[/img\]#sie", "'[img:$uid]\\1' . str_replace(' ', '%20', '\\3') . '[/img:$uid]'", $text);
// Remove our padding from the string..
return substr($text, 1);;
} // bbencode_first_pass()
/**
* $text - The text to operate on.
* $uid - The UID to add to matching tags.
* $open_tag - The opening tag to match. Can be an array of opening tags.
* $close_tag - The closing tag to match.
* $close_tag_new - The closing tag to replace with.
* $mark_lowest_level - boolean - should we specially mark the tags that occur
* at the lowest level of nesting? (useful for [code], because
* we need to match these tags first and transform HTML tags
* in their contents..
* $func - This variable should contain a string that is the name of a function.
* That function will be called when a match is found, and passed 2
* parameters: ($text, $uid). The function should return a string.
* This is used when some transformation needs to be applied to the
* text INSIDE a pair of matching tags. If this variable is FALSE or the
* empty string, it will not be executed.
* If open_tag is an array, then the pda will try to match pairs consisting of
* any element of open_tag followed by close_tag. This allows us to match things
* like [list=A]...[/list] and [list=1]...[/list] in one pass of the PDA.
*
* NOTES: - this function assumes the first character of $text is a space.
* - every opening tag and closing tag must be of the [...] format.
*/
function bbencode_first_pass_pda($text, $uid, $open_tag, $close_tag, $close_tag_new, $mark_lowest_level, $func, $open_regexp_replace = false)
{
$open_tag_count = 0;
if (!$close_tag_new || ($close_tag_new == ''))
{
$close_tag_new = $close_tag;
}
$close_tag_length = strlen($close_tag);
$close_tag_new_length = strlen($close_tag_new);
$uid_length = strlen($uid);
$use_function_pointer = ($func && ($func != ''));
$stack = array();
if (is_array($open_tag))
{
if (0 == count($open_tag))
{
// No opening tags to match, so return.
return $text;
}
$open_tag_count = count($open_tag);
}
else
{
// only one opening tag. make it into a 1-element array.
$open_tag_temp = $open_tag;
$open_tag = array();
$open_tag[0] = $open_tag_temp;
$open_tag_count = 1;
}
$open_is_regexp = false;
if ($open_regexp_replace)
{
$open_is_regexp = true;
if (!is_array($open_regexp_replace))
{
$open_regexp_temp = $open_regexp_replace;
$open_regexp_replace = array();
$open_regexp_replace[0] = $open_regexp_temp;
}
}
if ($mark_lowest_level && $open_is_regexp)
{
message_die(GENERAL_ERROR, "Unsupported operation for bbcode_first_pass_pda().");
}
// Start at the 2nd char of the string, looking for opening tags.
$curr_pos = 1;
while ($curr_pos && ($curr_pos < strlen($text)))
{
$curr_pos = strpos($text, "[", $curr_pos);
// If not found, $curr_pos will be 0, and the loop will end.
if ($curr_pos)
{
// We found a [. It starts at $curr_pos.
// check if it's a starting or ending tag.
$found_start = false;
$which_start_tag = "";
$start_tag_index = -1;
for ($i = 0; $i < $open_tag_count; $i++)
{
// Grab everything until the first "]"...
$possible_start = substr($text, $curr_pos, strpos($text, ']', $curr_pos + 1) - $curr_pos + 1);
//
// We're going to try and catch usernames with "[' characters.
//
if( preg_match('#\[quote=\\\"#si', $possible_start, $match) && !preg_match('#\[quote=\\\"(.*?)\\\"\]#si', $possible_start) )
{
// OK we are in a quote tag that probably contains a ] bracket.
// Grab a bit more of the string to hopefully get all of it..
if ($close_pos = strpos($text, '"]', $curr_pos + 9))
{
if (strpos(substr($text, $curr_pos + 9, $close_pos - ($curr_pos + 9)), '[quote') === false)
{
$possible_start = substr($text, $curr_pos, $close_pos - $curr_pos + 2);
}
}
}
// Now compare, either using regexp or not.
if ($open_is_regexp)
{
$match_result = array();
if (preg_match($open_tag[$i], $possible_start, $match_result))
{
$found_start = true;
$which_start_tag = $match_result[0];
$start_tag_index = $i;
break;
}
}
else
{
// straightforward string comparison.
if (0 == strcasecmp($open_tag[$i], $possible_start))
{
$found_start = true;
$which_start_tag = $open_tag[$i];
$start_tag_index = $i;
break;
}
}
}
if ($found_start)
{
// We have an opening tag.
// Push its position, the text we matched, and its index in the open_tag array on to the stack, and then keep going to the right.
$match = array("pos" => $curr_pos, "tag" => $which_start_tag, "index" => $start_tag_index);
bbcode_array_push($stack, $match);
//
// Rather than just increment $curr_pos
// Set it to the ending of the tag we just found
// Keeps error in nested tag from breaking out
// of table structure..
//
$curr_pos += strlen($possible_start);
}
else
{
// check for a closing tag..
$possible_end = substr($text, $curr_pos, $close_tag_length);
if (0 == strcasecmp($close_tag, $possible_end))
{
// We have an ending tag.
// Check if we've already found a matching starting tag.
if (sizeof($stack) > 0)
{
// There exists a starting tag.
$curr_nesting_depth = sizeof($stack);
// We need to do 2 replacements now.
$match = array_pop($stack);
$start_index = $match['pos'];
$start_tag = $match['tag'];
$start_length = strlen($start_tag);
$start_tag_index = $match['index'];
if ($open_is_regexp)
{
$start_tag = preg_replace($open_tag[$start_tag_index], $open_regexp_replace[$start_tag_index], $start_tag);
}
// everything before the opening tag.
$before_start_tag = substr($text, 0, $start_index);
// everything after the opening tag, but before the closing tag.
$between_tags = substr($text, $start_index + $start_length, $curr_pos - $start_index - $start_length);
// Run the given function on the text between the tags..
if ($use_function_pointer)
{
$between_tags = $func($between_tags, $uid);
}
// everything after the closing tag.
$after_end_tag = substr($text, $curr_pos + $close_tag_length);
// Mark the lowest nesting level if needed.
if ($mark_lowest_level && ($curr_nesting_depth == 1))
{
if ($open_tag[0] == '[code]')
{
$code_entities_match = array('#<#', '#>#', '#"#', '#:#', '#\[#', '#\]#', '#\(#', '#\)#', '#\{#', '#\}#');
$code_entities_replace = array('<', '>', '"', ':', '[', ']', '(', ')', '{', '}');
$between_tags = preg_replace($code_entities_match, $code_entities_replace, $between_tags);
}
$text = $before_start_tag . substr($start_tag, 0, $start_length - 1) . ":$curr_nesting_depth:$uid]";
$text .= $between_tags . substr($close_tag_new, 0, $close_tag_new_length - 1) . ":$curr_nesting_depth:$uid]";
}
else
{
if ($open_tag[0] == '[code]')
{
$text = $before_start_tag . '[code]';
$text .= $between_tags . '[/code]';
}
else
{
if ($open_is_regexp)
{
$text = $before_start_tag . $start_tag;
}
else
{
$text = $before_start_tag . substr($start_tag, 0, $start_length - 1) . ":$uid]";
}
$text .= $between_tags . substr($close_tag_new, 0, $close_tag_new_length - 1) . ":$uid]";
}
}
$text .= $after_end_tag;
// Now.. we've screwed up the indices by changing the length of the string.
// So, if there's anything in the stack, we want to resume searching just after it.
// otherwise, we go back to the start.
if (sizeof($stack) > 0)
{
$match = array_pop($stack);
$curr_pos = $match['pos'];
// bbcode_array_push($stack, $match);
// ++$curr_pos;
}
else
{
$curr_pos = 1;
}
}
else
{
// No matching start tag found. Increment pos, keep going.
++$curr_pos;
}
}
else
{
// No starting tag or ending tag.. Increment pos, keep looping.,
++$curr_pos;
}
}
}
} // while
return $text;
} // bbencode_first_pass_pda()
/**
* Does second-pass bbencoding of the [code] tags. This includes
* running htmlspecialchars() over the text contained between
* any pair of [code] tags that are at the first level of
* nesting. Tags at the first level of nesting are indicated
* by this format: [code:1:$uid] ... [/code:1:$uid]
* Other tags are in this format: [code:$uid] ... [/code:$uid]
*/
function bbencode_second_pass_code($text, $uid, $bbcode_tpl)
{
global $lang;
$code_start_html = $bbcode_tpl['code_open'];
$code_end_html = $bbcode_tpl['code_close'];
// First, do all the 1st-level matches. These need an htmlspecialchars() run,
// so they have to be handled differently.
$match_count = preg_match_all("#\[code:1:$uid\](.*?)\[/code:1:$uid\]#si", $text, $matches);
for ($i = 0; $i < $match_count; $i++)
{
$before_replace = $matches[1][$i];
$after_replace = $matches[1][$i];
// Replace 2 spaces with " " so non-tabbed code indents without making huge long lines.
$after_replace = str_replace(" ", " ", $after_replace);
// now Replace 2 spaces with " " to catch odd #s of spaces.
$after_replace = str_replace(" ", " ", $after_replace);
// Replace tabs with " " so tabbed code indents sorta right without making huge long lines.
$after_replace = str_replace("\t", " ", $after_replace);
// now Replace space occurring at the beginning of a line
$after_replace = preg_replace("/^ {1}/m", ' ', $after_replace);
$str_to_match = "[code:1:$uid]" . $before_replace . "[/code:1:$uid]";
$replacement = $code_start_html;
$replacement .= $after_replace;
$replacement .= $code_end_html;
$text = str_replace($str_to_match, $replacement, $text);
}
// Now, do all the non-first-level matches. These are simple.
$text = str_replace("[code:$uid]", $code_start_html, $text);
$text = str_replace("[/code:$uid]", $code_end_html, $text);
return $text;
} // bbencode_second_pass_code()
/**
* This is used to change a [*] tag into a [*:$uid] tag as part
* of the first-pass bbencoding of [list] tags. It fits the
* standard required in order to be passed as a variable
* function into bbencode_first_pass_pda().
*/
function replace_listitems($text, $uid)
{
$text = str_replace("[*]", "[*:$uid]", $text);
return $text;
}
/**
* This function does exactly what the PHP4 function array_push() does
*/
function bbcode_array_push(&$stack, $value)
{
$stack[] = $value;
return(sizeof($stack));
}
?>
Other Blog Scripts:
