/**
* Turn a string or array into valid, standards-compliant (x)HTML
*
* Uses configuraton options in tidy.conf - which should minimally have show-body-only set to yes
*
* @param mixed $text The data to be tidied up
* @return mixed $result Tidied data
*/
function tidy($text)
{
static $tidy_funcs;
static $tidy_conf;
if (!isset($tidy_conf)) {
$tidy_conf = SETTINGS_INC . 'tidy.conf';
}
if (is_array($text)) {
$result = array();
foreach (array_keys($text) as $key) {
$result[$key] = tidy($text[$key]);
}
return $result;
}
// determine what tidy libraries are available
if (empty($tidy_funcs)) {
$tidy_funcs = get_extension_funcs('tidy');
}
$tidy_1_lib_available = !empty($tidy_funcs) && array_search('tidy_setopt', $tidy_funcs) !== false;
$tidy_2_lib_available = !empty($tidy_funcs) && array_search('tidy_setopt', $tidy_funcs) === false;
$tidy_command_line_available = TIDY_EXE ? file_exists(TIDY_EXE) : false;
$text = protect_string_from_tidy($text);
$text = '<html><body>' . $text . '</body></html>';
if ($tidy_2_lib_available) {
$tidy = new tidy();
$tidy->parseString($text, $tidy_conf, 'utf8');
$tidy->cleanRepair();
$result = $tidy;
} elseif ($tidy_1_lib_available) {
tidy_load_config($tidy_conf);
tidy_set_encoding('utf8');
tidy_parse_string($text);
tidy_clean_repair();
$result = tidy_get_output();
} elseif ($tidy_command_line_available) {
$arg = escapeshellarg($text);
// escape the bad stuff in the text
$cmd = 'echo ' . $arg . ' | ' . TIDY_EXE . ' -q -config ' . $tidy_conf . ' 2> /dev/null';
// the actual command - pipes the input to tidy which diverts its output to the random file
$result = shell_exec($cmd);
// execute the command
} else {
trigger_error('tidy does not appear to be available within php or at the command line - no tidying is taking place.');
$result = $text;
}
return trim($result);
}
/**
* Use the HTML tidy PECL extension to use the tidy library in-process,
* saving the overhead of spawning a new process. Currently written to
* the PHP 4.3.x version of the extension, may not work on PHP 5.
*
* 'pear install tidy' should be able to compile the extension module.
*/
private static function internal($text)
{
global $wgTidyConf;
$fname = 'Parser::internalTidy';
wfProfileIn($fname);
tidy_load_config($wgTidyConf);
tidy_set_encoding('utf8');
tidy_parse_string($text);
tidy_clean_repair();
if (tidy_get_status() == 2) {
// 2 is magic number for fatal error
// http://www.php.net/manual/en/function.tidy-get-status.php
$cleansource = null;
} else {
$cleansource = tidy_get_output();
}
wfProfileOut($fname);
return $cleansource;
}
function return_parsed_bbcode($message, $nowrap = false)
{
// never strip_tags here, see Page.Talks for details
$message = str_replace("[b]", "<b>", $message);
$message = str_replace("[/b]", "</b>", $message);
$message = str_replace("[i]", "<i>", $message);
$message = str_replace("[/i]", "</i>", $message);
$message = str_replace("[u]", "<u>", $message);
$message = str_replace("[/u]", "</u>", $message);
$message = str_replace("[center]", "<div align=\"center\">", $message);
$message = str_replace("[/center]", "</div>", $message);
$message = str_replace("[left]", "<div align=\"left\">", $message);
$message = str_replace("[/left]", "</div>", $message);
$message = str_replace("[right]", "<div align=\"right\">", $message);
$message = str_replace("[/right]", "</div>", $message);
$message = str_replace("[ol]", "<ol>", $message);
$message = str_replace("[ul]", "<ul>", $message);
$message = str_replace("[li]", "<li>", $message);
$message = str_replace("[/ol]", "</ol>", $message);
$message = str_replace("[/ul]", "</ul>", $message);
$message = str_replace("[br]", "<br>", $message);
$message = eregi_replace("\\[img\\]([^\\[]*)\\[/img\\]", "<img src=\"\\1\" border=\"0\">", $message);
$message = eregi_replace("\\[url\\](https?://[^\\[]*)\\[/url\\]", "<a href=\"\\1\">\\1</a>", $message);
if (function_exists("tidy_get_output")) {
if (!$nowrap) {
$config = array('indent' => FALSE, 'output-xhtml' => TRUE, 'show-body-only' => TRUE, 'wrap' => 80);
} else {
$config = array('indent' => FALSE, 'output-xhtml' => TRUE, 'show-body-only' => TRUE);
}
tidy_set_encoding('UTF8');
foreach ($config as $key => $value) {
tidy_setopt($key, $value);
}
tidy_parse_string($message);
tidy_clean_repair();
$message = tidy_get_output();
}
return $message;
}
function TidyClean()
{
if (!class_exists('tidy')) {
if (function_exists('tidy_parse_string')) {
//use procedural style for compatibility with PHP 4.3
tidy_set_encoding($this->Encoding);
foreach ($this->TidyConfig as $key => $value) {
tidy_setopt($key, $value);
}
tidy_parse_string($this->html);
tidy_clean_repair();
$this->html = tidy_get_output();
} else {
print "<b>No tidy support. Please enable it in your php.ini.\r\nOnly basic cleaning is beeing applied\r\n</b>";
}
} else {
//PHP 5 only !!!
$tidy = new tidy();
$tidy->parseString($this->html, $this->TidyConfig, $this->Encoding);
$tidy->cleanRepair();
$this->html = $tidy;
}
}
/**
* Generates HTML from an array of tokens.
* @param $tokens Array of HTMLPurifier_Token
* @param $config HTMLPurifier_Config object
* @return Generated HTML
*/
function generateFromTokens($tokens, $config, &$context)
{
$html = '';
if (!$config) {
$config = HTMLPurifier_Config::createDefault();
}
$this->_scriptFix = $config->get('Output', 'CommentScriptContents');
$this->_def = $config->getHTMLDefinition();
$this->_xhtml = $this->_def->doctype->xml;
if (!$tokens) {
return '';
}
for ($i = 0, $size = count($tokens); $i < $size; $i++) {
if ($this->_scriptFix && $tokens[$i]->name === 'script' && $i + 2 < $size && $tokens[$i + 2]->type == 'end') {
// script special case
// the contents of the script block must be ONE token
// for this to work
$html .= $this->generateFromToken($tokens[$i++]);
$html .= $this->generateScriptFromToken($tokens[$i++]);
// We're not going to do this: it wouldn't be valid anyway
//while ($tokens[$i]->name != 'script') {
// $html .= $this->generateScriptFromToken($tokens[$i++]);
//}
}
$html .= $this->generateFromToken($tokens[$i]);
}
if ($config->get('Output', 'TidyFormat') && extension_loaded('tidy')) {
$tidy_options = array('indent' => true, 'output-xhtml' => $this->_xhtml, 'show-body-only' => true, 'indent-spaces' => 2, 'wrap' => 68);
if (version_compare(PHP_VERSION, '5', '<')) {
tidy_set_encoding('utf8');
foreach ($tidy_options as $key => $value) {
tidy_setopt($key, $value);
}
tidy_parse_string($html);
tidy_clean_repair();
$html = tidy_get_output();
} else {
$tidy = new Tidy();
$tidy->parseString($html, $tidy_options, 'utf8');
$tidy->cleanRepair();
$html = (string) $tidy;
}
}
// normalize newlines to system
$nl = $config->get('Output', 'Newline');
if ($nl === null) {
$nl = PHP_EOL;
}
$html = str_replace("\n", $nl, $html);
return $html;
}
/**
* BlockenController()
*/
function BlockenController()
{
if (BLOCKEN_DEBUG_MODE && (BLOCKEN_ALWAYS_DEBUG || BlockenCommon::isMember())) {
$this->_fStart = BlockenCommon::getMicroTime();
}
if (function_exists('tidy_set_encoding')) {
tidy_set_encoding('UTF8');
tidy_setopt('input-xml', true);
}
}
