function encodeEntities($text)
{
/* Convert high and low ascii to entities. If multibyte string functions are
available (on by default in php 4.3+), we convert using unicode mapping as
defined in the function encode_high(). If not, we use php's nasty
built-in htmlentities() */
return function_exists('mb_encode_numericentity') ? encode_high($text) : htmlentities($text, ENT_NOQUOTES, "utf-8");
}
function textile($text)
{
$text = stripslashes($text);
$text = preg_replace("/&(?![#a-zA-Z0-9]+;)/", "x%x%", $text);
if (function_exists('mb_encode_numericentity')) {
$text = encode_high($text);
} else {
$text = htmlentities($text, ENT_NOQUOTES, "utf-8");
}
$text = str_replace(array(">", "<", "&"), array(">", "<", "&"), $text);
$text = str_replace("\r\n", "\n", $text);
$text = str_replace("\t", "", $text);
$text = preg_split("/\n/", $text);
foreach ($text as $line) {
$line = rtrim($line);
$lineout[] = $line;
}
$text = implode("\n", $lineout);
$text = preg_replace('/(^|\\s)==(.*)==(\\s|$)?/msU', '$1<notextile>$2</notextile>$3', $text);
$text = preg_replace('/!([^\\s\\(=]+)\\s?(?:\\(([^\\)]+)\\))?!(\\s)?/mU', '<img src="$1" alt="$2" border="0" />$3', $text);
$text = preg_replace('/(<img.+ \\/>):(\\S+)(\\s)/U', '<a href="$2">$1</a>$3', $text);
$text = preg_replace('/
([\\s[{(]|[[:punct:]])? # 1 optional space or brackets before
" # starting "
([^"\\(]+) # 2 text of link
\\s? # opt space
(?:\\(([^\\(]*)\\))? # 3 opt title attribute in parenths
": # dividing ":
(\\S+\\b) # 4 suppose this is the url
(\\/)? # 5 opt trailing slash
([^[:alnum:]\\/;]*) # 6 opt punctuation after the url
(\\s|$) # 7 either white space or end of string
/x', '$1<a href="$4$5" title="$3">$2</a>$6$7', $text);
$qtags = array('\\*\\*' => 'b', '\\*' => 'strong', '\\?\\?' => 'cite', '-' => 'del', '\\+' => 'ins', '~' => 'sub', '@' => 'code');
foreach ($qtags as $f => $r) {
$text = preg_replace('/(^|\\s|>)' . $f . '\\b(.+)\\b([[:punct:]]*)' . $f . '([[:punct:]]{0,2})(\\s|$)?/mU', '$1<' . $r . '>$2$3</' . $r . '>$4$5', $text);
}
$text = preg_replace('/(^|\\s)__(.*)__([[:punct:]]{0,2})(\\s|$)?/mU', '$1<i>$2</i>$3$4', $text);
$text = preg_replace('/(^|\\s)_(.*)_([[:punct:]]{0,2})(\\s|$)?/mU', '$1<em>$2</em>$3$4', $text);
$text = preg_replace('/\\^(.*)\\^/mU', '<sup>$1</sup>', $text);
$text = preg_replace('/"$/', "\" ", $text);
$glyph_search = array('/([^\\s[{(>])?\'(?(1)|(?=\\s|s\\b))/', '/\'/', '/([^\\s[{(])?"(?(1)|(?=\\s))/', '/"/', '/\\b( )?\\.{3}/', '/\\b([A-Z][A-Z0-9]{2,})\\b(?:[(]([^)]*)[)])/', '/(^|[^"][>\\s])([A-Z][A-Z0-9 ]{2,})([^<a-z0-9]|$)/', '/\\s?--\\s?/', '/\\s-\\s/', '/(\\d+) ?x ?(\\d+)/', '/\\b ?[([]TM[])]/i', '/\\b ?[([]R[])]/i', '/\\b ?[([]C[])]/i');
# copyright
$glyph_replace = array('$1’$2', '‘', '$1”', '“', '$1…', '<acronym title="$2">$1</acronym>', '$1<span class="caps">$2</span>$3', '—', ' – ', '$1×$2', '™', '®', '©');
# copyright
$codepre = false;
if (!preg_match("/<.*>/", $text)) {
$text = preg_replace($glyph_search, $glyph_replace, $text);
} else {
$text = preg_split("/(<.*>)/U", $text, -1, PREG_SPLIT_DELIM_CAPTURE);
foreach ($text as $line) {
# matches are off if we're between <code>, <pre> etc.
if (preg_match('/<(code|pre|kbd|notextile)>/i', $line)) {
$codepre = true;
}
if (preg_match('/<\\/(code|pre|kbd|notextile)>/i', $line)) {
$codepre = false;
}
if (!preg_match("/<.*>/", $line) && $codepre == false) {
$line = preg_replace($glyph_search, $glyph_replace, $line);
}
# convert htmlspecial if between <code>
if ($codepre == true) {
$line = htmlspecialchars($line, ENT_NOQUOTES, "UTF-8");
$line = str_replace("<pre>", "<pre>", $line);
$line = str_replace("<code>", "<code>", $line);
$line = str_replace("<notextile>", "<notextile>", $line);
$line = str_replace("<kbd>", "<kbd>", $line);
}
# each line gets pushed to a new array
$glyph_out[] = $line;
}
# $text is now the new array, cast to a string
$text = implode('', $glyph_out);
}
$text = preg_replace("/(\\S)(_*)([[:punct:]]*) *\n([^#*\\s])/", "\$1\$2\$3<br />\$4", $text);
$text = str_replace("l><br />", "l>\n", $text);
$text = preg_split("/\n/", $text);
array_push($text, " ");
$list = '';
$pre = false;
$block_find = array('/^\\s?\\*\\s(.*)/', '/^\\s?#\\s(.*)/', '/^bq\\. (.*)/', '/^h(\\d)\\(([[:alnum:]]+)\\)\\.\\s(.*)/', '/^h(\\d)\\. (.*)/', '/^p\\(([[:alnum:]]+)\\)\\.\\s(.*)/', '/^p\\. (.*)/i', '/^([^\\t ]+.*)/i');
$block_replace = array("\t\t<li>\$1</li>", "\t\t\t<li>\$1</li>", "\t<blockquote>\$1</blockquote>", "\t<h\$1 class=\"\$2\">\$3</h\$1>\$4", "\t<h\$1>\$2</h\$1>\$3", "\t<p class=\"\$1\">\$2</p>\$3", "\t<p>\$1</p>", "\t<p>\$1</p>\$2");
foreach ($text as $line) {
if (preg_match('/<pre>/i', $line)) {
$pre = true;
}
if ($pre == false) {
$line = preg_replace($block_find, $block_replace, $line);
}
if ($pre == true) {
$line = str_replace("<br />", "\n", $line);
}
if (preg_match('/<\\/pre>/i', $line)) {
$pre = false;
}
if ($list == '' && preg_match('/^\\t\\t<li>/', $line)) {
$list = "ul";
$line = preg_replace('/^(\\t\\t<li>.*)/', "\t<ul>\n\$1", $line);
} else {
if ($list == '' && preg_match('/^\\t\\t\\t<li>/', $line)) {
$list = "ol";
$line = preg_replace('/^\\t(\\t\\t<li>.*)/', "\t<ol>\n\$1", $line);
# at the end of a ul
} else {
if ($list == 'ul' && !preg_match('/^\\t\\t<li>/', $line)) {
$list = '';
$line = preg_replace('/^(.*)$/', "\t</ul>\n\$1", $line);
# at the end of a ol
} else {
if ($list == 'ol' && !preg_match('/^\\t\\t\\t<li>/', $line)) {
$list = '';
$line = preg_replace('/^(.*)$/', "\t</ol>\n\$1", $line);
}
}
}
}
$block_out[] = $line;
}
$text = implode("\n", $block_out);
$text = preg_replace('/<\\/?notextile>/', "", $text);
$text = str_replace("x%x%", "&", $text);
$text = str_replace("<br />", "<br />\n", $text);
return $text;
}
function textile($text)
{
### Basic global changes
$text = stripslashes($text);
# turn any incoming ampersands into a dummy character for now.
# This uses a negative lookahead for alphanumerics followed by a semicolon,
# implying an incoming html entity, to be skipped
$text = preg_replace("/&(?![#a-zA-Z0-9]+;)/", "x%x%", $text);
# entify everything
if (function_exists('mb_encode_numericentity')) {
$text = encode_high($text);
} else {
$text = htmlentities($text, ENT_NOQUOTES, "utf-8");
}
# unentify angle brackets and ampersands
$text = str_replace(array(">", "<", "&"), array(">", "<", "&"), $text);
# zap carriage returns
$text = str_replace("\r\n", "\n", $text);
# zap tabs
$text = str_replace("\t", "", $text);
$text = preg_split("/\n/", $text);
foreach ($text as $line) {
$line = trim($line);
$lineout[] = $line;
}
$text = implode("\n", $lineout);
### Find and replace quick tags
# double equal signs means <notextile>
$text = preg_replace('/(^|\\s)==(.*)==(\\s|$)?/msU', '$1<notextile>$2</notextile>$3', $text);
# image qtag
$text = preg_replace('/!([^\\s\\(=]+)\\s?(?:\\(([^\\)]+)\\))?!(\\s)?/mU', '<img src="$1" alt="$2" border="0" />$3', $text);
# image with hyperlink
$text = preg_replace('/(<img.+ \\/>):(\\S+)(\\s)/U', '<a href="$2">$1</a>$3', $text);
# hyperlink qtag
$text = preg_replace('/
([\\s[{(]|[[:punct:]])? # 1 optional space or brackets before
" # starting "
([^"\\(]+) # 2 text of link
\\s? # opt space
(?:\\(([^\\(]*)\\))? # 3 opt title attribute in parenths
": # dividing ":
(\\S+\\b) # 4 suppose this is the url
(\\/)? # 5 opt trailing slash
([^[:alnum:]\\/;]*) # 6 opt punctuation after the url
(\\s|$) # 7 either white space or end of string
/x', '$1<a href="$4$5" title="$3">$2</a>$6$7', $text);
# arrange qtag delineators and replacements in an array
$qtags = array('\\*\\*' => 'b', '\\*' => 'strong', '\\?\\?' => 'cite', '-' => 'del', '\\+' => 'ins', '~' => 'sub', '@' => 'code');
# loop through the array, replacing qtags with html
foreach ($qtags as $f => $r) {
$text = preg_replace('/(^|\\s|>)' . $f . '\\b(.+)\\b([[:punct:]]*)' . $f . '([[:punct:]]{0,2})(\\s|$)?/mU', '$1<' . $r . '>$2$3</' . $r . '>$4$5', $text);
}
# some weird bs with underscores and \b word boundaries,
# so we'll do those on their own
$text = preg_replace('/(^|\\s)__(.*)__([[:punct:]]{0,2})(\\s|$)?/mU', '$1<i>$2</i>$3$4', $text);
$text = preg_replace('/(^|\\s)_(.*)_([[:punct:]]{0,2})(\\s|$)?/mU', '$1<em>$2</em>$3$4', $text);
$text = preg_replace('/\\^(.*)\\^/mU', '<sup>$1</sup>', $text);
### Find and replace typographic chars and special tags
# small problem with double quotes at the end of a string
$text = preg_replace('/"$/', "\" ", $text);
# NB: all these will wreak havoc inside <html> tags
$glyph_search = array('/([^\\s[{(>])?\'(?(1)|(?=\\s|s\\b))/', '/\'/', '/([^\\s[{(])?"(?(1)|(?=\\s))/', '/"/', '/\\b( )?\\.{3}/', '/\\b([A-Z][A-Z0-9]{2,})\\b(?:[(]([^)]*)[)])/', '/(^|[^"][>\\s])([A-Z][A-Z0-9 ]{2,})([^<a-z0-9]|$)/', '/\\s?--\\s?/', '/\\s-\\s/', '/(\\d+) ?x ?(\\d+)/', '/\\b ?[([]TM[])]/i', '/\\b ?[([]R[])]/i', '/\\b ?[([]C[])]/i');
# copyright
$glyph_replace = array('$1’$2', '‘', '$1”', '“', '$1…', '<acronym title="$2">$1</acronym>', '$1<span class="caps">$2</span>$3', '—', ' – ', '$1×$2', '™', '®', '©');
# copyright
# set toggle for turning off replacements between <code> or <pre>
$codepre = false;
# if there is no html, do a simple search and replace
if (!preg_match("/<.*>/", $text)) {
$text = preg_replace($glyph_search, $glyph_replace, $text);
} else {
# else split the text into an array at <.*>
$text = preg_split("/(<.*>)/U", $text, -1, PREG_SPLIT_DELIM_CAPTURE);
foreach ($text as $line) {
# matches are off if we're between <code>, <pre> etc.
if (preg_match('/<(code|pre|kbd|notextile)>/i', $line)) {
$codepre = true;
}
if (preg_match('/<\\/(code|pre|kbd|notextile)>/i', $line)) {
$codepre = false;
}
if (!preg_match("/<.*>/", $line) && $codepre == false) {
$line = preg_replace($glyph_search, $glyph_replace, $line);
}
# convert htmlspecial if between <code>
if ($codepre == true) {
$line = htmlspecialchars($line, ENT_NOQUOTES, "UTF-8");
$line = str_replace("<pre>", "<pre>", $line);
$line = str_replace("<code>", "<code>", $line);
$line = str_replace("<notextile>", "<notextile>", $line);
$line = str_replace("<kbd>", "<kbd>", $line);
}
# each line gets pushed to a new array
$glyph_out[] = $line;
}
# $text is now the new array, cast to a string
$text = implode('', $glyph_out);
}
### Block level formatting
# deal with forced breaks; this is going to be a problem between
# <pre> tags, but we'll clean them later
$text = preg_replace("/(\\S)(_*)([[:punct:]]*) *\n([^#*\\s])/", "\$1\$2\$3<br />\$4", $text);
# might be a problem with lists
$text = str_replace("l><br />", "l>\n", $text);
# clear out multiple newlines for now
# $text = preg_replace("/\n+/","\n",$text);
# split the text into an array by newlines
$text = preg_split("/\n/", $text);
array_push($text, " ");
$list = '';
$pre = false;
$block_find = array('/^\\s?\\*\\s(.*)/', '/^\\s?#\\s(.*)/', '/^bq\\. (.*)/', '/^h(\\d)\\(([[:alnum:]]+)\\)\\.\\s(.*)/', '/^h(\\d)\\. (.*)/', '/^p\\(([[:alnum:]]+)\\)\\.\\s(.*)/', '/^p\\. (.*)/i', '/^([^\\t ]+.*)/i');
$block_replace = array("\t\t<li>\$1</li>", "\t\t\t<li>\$1</li>", "\t<blockquote>\$1</blockquote>", "\t<h\$1 class=\"\$2\">\$3</h\$1>\$4", "\t<h\$1>\$2</h\$1>\$3", "\t<p class=\"\$1\">\$2</p>\$3", "\t<p>\$1</p>", "\t<p>\$1</p>\$2");
# loop through lines
foreach ($text as $line) {
# matches are off if we're between <pre> or <code> tags
if (preg_match('/<pre>/i', $line)) {
$pre = true;
}
# deal with block replacements first, then see if we're in a list
if ($pre == false) {
$line = preg_replace($block_find, $block_replace, $line);
}
# kill any br tags that slipped in earlier
if ($pre == true) {
$line = str_replace("<br />", "\n", $line);
}
# matches back on after </pre>
if (preg_match('/<\\/pre>/i', $line)) {
$pre = false;
}
# on entry to a list, $list switches to a value
# two tabs means unordered list
if ($list == '' && preg_match('/^\\t\\t<li>/', $line)) {
$list = "ul";
$line = preg_replace('/^(\\t\\t<li>.*)/', "\t<ul>\n\$1", $line);
} else {
if ($list == '' && preg_match('/^\\t\\t\\t<li>/', $line)) {
$list = "ol";
$line = preg_replace('/^\\t(\\t\\t<li>.*)/', "\t<ol>\n\$1", $line);
# at the end of a ul
} else {
if ($list == 'ul' && !preg_match('/^\\t\\t<li>/', $line)) {
$list = '';
$line = preg_replace('/^(.*)$/', "\t</ul>\n\$1", $line);
# at the end of a ol
} else {
if ($list == 'ol' && !preg_match('/^\\t\\t\\t<li>/', $line)) {
$list = '';
$line = preg_replace('/^(.*)$/', "\t</ol>\n\$1", $line);
}
}
}
}
# push each line to a new array once processed
$block_out[] = $line;
}
$text = implode("\n", $block_out);
#clean up <notextile>
$text = preg_replace('/<\\/?notextile>/', "", $text);
# turn the temp char back to an ampersand entity
$text = str_replace("x%x%", "&", $text);
# Newline linebreaks, just for markup tidiness
$text = str_replace("<br />", "<br />\n", $text);
return $text;
}
