/**
* Make a string lower case
*
* @param string $str String
*
* @return string
*/
public static function strtolower($str)
{
return self::$instance->strtolower($str);
}
/**
* Filter some text cutting out all non words and stop words
*
* @param string $input text to filter
*
* @return string
*/
public function clean($input)
{
$words = explode(' ', ereg_replace('[[:punct:][:space:]]+', ' ', $input));
$stop_words = $this->getByLang(null, true);
$retval = array();
foreach ($words as $word) {
$word = PMF_String::strtolower($word);
if (!is_numeric($word) && 1 < PMF_String::strlen($word) && !in_array($word, $stop_words) && !in_array($word, $retval)) {
$retval[] = $word;
}
}
return $retval;
}
/**
* This function checks the content against a bad word list if the banned
* word spam protection has been activated from the general phpMyFAQ
* configuration.
*
* @param string $content
*
* @return bool
*/
public function checkBannedWord($content)
{
// Sanity checks
$content = PMF_String::strtolower(trim($content));
if ('' === $content || !$this->_config->get('spam.checkBannedWords')) {
return true;
}
// Check if we check more than one word
$checkWords = explode(' ', $content);
if (1 === count($checkWords)) {
$checkWords = array($content);
}
$bannedWords = $this->getBannedWords();
// We just search a match of, at least, one banned word into $content
if (is_array($bannedWords)) {
foreach ($bannedWords as $bannedWord) {
foreach ($checkWords as $word) {
if (PMF_String::strtolower($word) === PMF_String::strtolower($bannedWord)) {
return false;
}
}
}
}
return true;
}
/**
* This function checks the content against a dab word list
* if the banned word spam protection has been activated from the general PMF configuration.
*
* @param string $content
* @return bool
* @access public
* @author Katherine A. Bouton
* @author Thorsten Rinne <*****@*****.**>
* @author Matteo Scaramuccia <*****@*****.**>
* @author Peter Beauvain <*****@*****.**>
*/
function checkBannedWord($content)
{
// Sanity checks
$content = trim($content);
if ('' == $content && !PMF_Configuration::getInstance()->get('spam.checkBannedWords')) {
return true;
}
$bannedWords = getBannedWords();
// We just search a match of, at least, one banned word into $content
$content = PMF_String::strtolower($content);
if (is_array($bannedWords)) {
foreach ($bannedWords as $bannedWord) {
if (PMF_String::strpos($content, PMF_String::strtolower($bannedWord)) !== false) {
return false;
}
}
}
return true;
}
/**
* Returns a search engine optimized title
*
* @param string $title
*
* @return string
*/
public function getSEOItemTitle($title = '')
{
if ('' === $title) {
$title = $this->itemTitle;
}
$itemTitle = trim($title);
// Lower the case (aesthetic)
$itemTitle = PMF_String::strtolower($itemTitle);
// Use '_' for some other characters for:
// 1. avoiding regexp match break;
// 2. improving the reading.
$itemTitle = str_replace(array('-', "'", '/', '''), '_', $itemTitle);
// 1. Remove any CR LF sequence
// 2. Use a '-' for the words separation
$itemTitle = PMF_String::preg_replace('/\\s/m', '-', $itemTitle);
// Hack: remove some chars for having a better readable title
$itemTitle = str_replace(array('+', ',', ';', ':', '.', '?', '!', '"', '(', ')', '[', ']', '{', '}', '<', '>'), '', $itemTitle);
// Hack: move some chars to "similar" but plain ASCII chars
$itemTitle = str_replace(array('à', 'è', 'é', 'ì', 'ò', 'ù', 'ä', 'ö', 'ü', 'ß', 'Ä', 'Ö', 'Ü', 'č', 'ę', 'ė', 'į', 'š', 'ų', 'ū', 'ž'), array('a', 'e', 'e', 'i', 'o', 'u', 'ae', 'oe', 'ue', 'ss', 'Ae', 'Oe', 'Ue', 'c', 'e', 'e', 'i', 's', 'u', 'u', 'z'), $itemTitle);
// Clean up
$itemTitle = PMF_String::preg_replace('/-[\\-]+/m', '-', $itemTitle);
return rawurlencode($itemTitle);
}
/**
* Saves all tags from a FAQ record
*
* @param integer $record_id Record ID
* @param array $tags Array of tags
*/
public function saveTags($record_id, $tags)
{
if (!is_array($tags)) {
return false;
}
$current_tags = $this->getAllTags();
// Delete all tag references for the faq record
if (count($tags) > 0) {
$this->deleteTagsFromRecordId($record_id);
}
// Store tags and references for the faq record
foreach ($tags as $tagging_name) {
$tagging_name = trim($tagging_name);
if (PMF_String::strlen($tagging_name) > 0) {
if (!in_array(PMF_String::strtolower($tagging_name), array_map(array('PMF_String', 'strtolower'), $current_tags))) {
// Create the new tag
$new_tagging_id = $this->db->nextID(SQLPREFIX . 'faqtags', 'tagging_id');
$query = sprintf("\n INSERT INTO\n %sfaqtags\n (tagging_id, tagging_name)\n VALUES\n (%d, '%s')", SQLPREFIX, $new_tagging_id, $tagging_name);
$this->db->query($query);
// Add the tag reference for the faq record
$query = sprintf("\n INSERT INTO\n %sfaqdata_tags\n (record_id, tagging_id)\n VALUES\n (%d, %d)", SQLPREFIX, $record_id, $new_tagging_id);
$this->db->query($query);
} else {
// Add the tag reference for the faq record
$query = sprintf("\n INSERT INTO\n %sfaqdata_tags\n (record_id, tagging_id)\n VALUES\n (%d, %d)", SQLPREFIX, $record_id, array_search(PMF_String::strtolower($tagging_name), array_map(array('PMF_String', 'strtolower'), $current_tags)));
$this->db->query($query);
}
}
}
return true;
}
/**
* Decode a given ACE domain name
* @param string Domain name (ACE string)
* [@param string Desired output encoding, see {@link set_parameter}]
* @return string Decoded Domain name (UTF-8 or UCS-4)
*/
public function decode($input, $one_time_encoding = false)
{
// Optionally set
if ($one_time_encoding) {
switch ($one_time_encoding) {
case 'utf8':
case 'ucs4_string':
case 'ucs4_array':
break;
default:
$this->_error('Unknown encoding ' . $one_time_encoding);
return false;
}
}
// Make sure to drop any newline characters around
$input = trim($input);
// Negotiate input and try to determine, whether it is a plain string,
// an email address or something like a complete URL
if (PMF_String::strpos($input, '@')) {
// Maybe it is an email address
// No no in strict mode
if ($this->_strict_mode) {
$this->_error('Only simple domain name parts can be handled in strict mode');
return false;
}
list($email_pref, $input) = explode('@', $input, 2);
$arr = explode('.', $input);
foreach ($arr as $k => $v) {
if (PMF_String::preg_match('!^' . preg_quote($this->_punycode_prefix, '!') . '!', $v)) {
$conv = $this->_decode($v);
if ($conv) {
$arr[$k] = $conv;
}
}
}
$input = join('.', $arr);
$arr = explode('.', $email_pref);
foreach ($arr as $k => $v) {
if (PMF_String::preg_match('!^' . preg_quote($this->_punycode_prefix, '!') . '!', $v)) {
$conv = $this->_decode($v);
if ($conv) {
$arr[$k] = $conv;
}
}
}
$email_pref = join('.', $arr);
$return = $email_pref . '@' . $input;
} elseif (PMF_String::preg_match('![:\\./]!', $input)) {
// Or a complete domain name (with or without paths / parameters)
// No no in strict mode
if ($this->_strict_mode) {
$this->_error('Only simple domain name parts can be handled in strict mode');
return false;
}
$parsed = parse_url($input);
if (isset($parsed['host'])) {
$arr = explode('.', $parsed['host']);
foreach ($arr as $k => $v) {
$conv = $this->_decode($v);
if ($conv) {
$arr[$k] = $conv;
}
}
$parsed['host'] = join('.', $arr);
$return = (empty($parsed['scheme']) ? '' : $parsed['scheme'] . (PMF_String::strtolower($parsed['scheme']) == 'mailto' ? ':' : '://')) . (empty($parsed['user']) ? '' : $parsed['user'] . (empty($parsed['pass']) ? '' : ':' . $parsed['pass']) . '@') . $parsed['host'] . (empty($parsed['port']) ? '' : ':' . $parsed['port']) . (empty($parsed['path']) ? '' : $parsed['path']) . (empty($parsed['query']) ? '' : '?' . $parsed['query']) . (empty($parsed['fragment']) ? '' : '#' . $parsed['fragment']);
} else {
// parse_url seems to have failed, try without it
$arr = explode('.', $input);
foreach ($arr as $k => $v) {
$conv = $this->_decode($v);
$arr[$k] = $conv ? $conv : $v;
}
$return = join('.', $arr);
}
} else {
// Otherwise we consider it being a pure domain name string
$return = $this->_decode($input);
if (!$return) {
$return = $input;
}
}
// The output is UTF-8 by default, other output formats need conversion here
// If one time encoding is given, use this, else the objects property
switch ($one_time_encoding ? $one_time_encoding : $this->_api_encoding) {
case 'utf8':
return $return;
break;
case 'ucs4_string':
return $this->_ucs4_to_ucs4_string($this->_utf8_to_ucs4($return));
break;
case 'ucs4_array':
return $this->_utf8_to_ucs4($return);
break;
default:
$this->_error('Unsupported output format');
return false;
}
}
