/**
* Returns returns person name convention for a given language.
* @param string $language The input language.
* @param string $type The type of the requested convention. It may be 'format' for name order convention or 'sort_by' for name sorting convention.
* @return mixed Depending of the requested type, the returned result may be string or boolean; null is returned on error;
*/
function _api_get_person_name_convention($language, $type)
{
global $app;
$conventions = $app['name_order_conventions'];
$language = api_purify_language_id($language);
switch ($type) {
case 'format':
return is_string($conventions[$language]['format']) ? $conventions[$language]['format'] : '%t %f %l';
case 'sort_by':
return is_bool($conventions[$language]['sort_by']) ? $conventions[$language]['sort_by'] : true;
}
return null;
}
/**
* Chooses the default MySQL-specific collation from given encoding and language.
* @param string $encoding A conventional encoding id, i.e. 'UTF-8'
* @param string $language (optional) A conventional for the system language id, i.e. 'bulgarian'. If it is empty, the chosen collation is the default server value corresponding to the given encoding.
* @return string Returns a suitable default collation, for example 'utf8_general_ci', or NULL if collation was not found.
* @author Ivan Tcholakov
*/
public static function to_db_collation($encoding, $language = null)
{
static $result = array();
if (!isset($result[$encoding][$language])) {
$result[$encoding][$language] = null;
if (self::is_encoding_supported($encoding)) {
$db_encoding = self::to_db_encoding($encoding);
if (!empty($language)) {
$lang = api_purify_language_id($language);
$res = self::check_db_collation($db_encoding, $lang);
if (empty($res)) {
$db_collation_map =& self::get_db_collation_map();
if (isset($db_collation_map[$lang])) {
$res = self::check_db_collation($db_encoding, $db_collation_map[$lang]);
}
}
if (empty($res)) {
$res = self::check_db_collation($db_encoding, null);
}
$result[$encoding][$language] = $res;
} else {
$result[$encoding][$language] = self::check_db_collation($db_encoding, null);
}
}
}
return $result[$encoding][$language];
}
/**
* Returns returns person name convention for a given language.
* @param string $language The input language.
* @param string $type The type of the requested convention.
* It may be 'format' for name order convention or 'sort_by' for name sorting convention.
* @return mixed Depending of the requested type,
* the returned result may be string or boolean; null is returned on error;
*/
function _api_get_person_name_convention($language, $type)
{
static $conventions;
$language = api_purify_language_id($language);
if (!isset($conventions)) {
$file = dirname(__FILE__) . '/internationalization_database/name_order_conventions.php';
if (file_exists($file)) {
$conventions = (include $file);
} else {
$conventions = array('english' => array('format' => 'title first_name last_name', 'sort_by' => 'first_name'));
}
// Overwrite classic conventions
$customConventions = api_get_configuration_value('name_order_conventions');
if (!empty($customConventions)) {
foreach ($customConventions as $key => $data) {
$conventions[$key] = $data;
}
}
$search1 = array('FIRST_NAME', 'LAST_NAME', 'TITLE');
$replacement1 = array('%F', '%L', '%T');
$search2 = array('first_name', 'last_name', 'title');
$replacement2 = array('%f', '%l', '%t');
foreach (array_keys($conventions) as $key) {
$conventions[$key]['format'] = str_replace($search1, $replacement1, $conventions[$key]['format']);
$conventions[$key]['format'] = _api_validate_person_name_format(_api_clean_person_name(str_replace('%', ' %', str_ireplace($search2, $replacement2, $conventions[$key]['format']))));
$conventions[$key]['sort_by'] = strtolower($conventions[$key]['sort_by']) != 'last_name' ? true : false;
}
}
switch ($type) {
case 'format':
return is_string($conventions[$language]['format']) ? $conventions[$language]['format'] : '%t %f %l';
case 'sort_by':
return is_bool($conventions[$language]['sort_by']) ? $conventions[$language]['sort_by'] : true;
}
return null;
}
/**
* Detects encoding of plain text.
* @param string $string The input text.
* @param string $language (optional) The language of the input text, provided if it is known.
* @return string Returns the detected encoding.
*/
function api_detect_encoding($string, $language = null)
{
// Testing against valid UTF-8 first.
if (api_is_valid_utf8($string)) {
return 'UTF-8';
}
$result = null;
$delta_points_min = LANGUAGE_DETECT_MAX_DELTA;
// Testing non-UTF-8 encodings.
$encodings = api_get_valid_encodings();
foreach ($encodings as &$encoding) {
if (api_is_encoding_supported($encoding) && !api_is_utf8($encoding)) {
$stringToParse = api_substr($string, 0, LANGUAGE_DETECT_MAX_LENGTH, $encoding);
$strintToParse2 = _api_generate_n_grams($stringToParse, $encoding);
$result_array = _api_compare_n_grams($strintToParse2, $encoding);
if (!empty($result_array)) {
list($key, $delta_points) = each($result_array);
if ($delta_points < $delta_points_min) {
$pos = strpos($key, ':');
$result_encoding = api_refine_encoding_id(substr($key, $pos + 1));
if (api_equal_encodings($encoding, $result_encoding)) {
if ($string == api_utf8_decode(api_utf8_encode($string, $encoding), $encoding)) {
$delta_points_min = $delta_points;
$result = $encoding;
}
}
}
}
}
}
// "Broken" UTF-8 texts are to be detected as UTF-8.
// This functionality is enabled when language of the text is known.
$language = api_purify_language_id((string) $language);
if (!empty($language)) {
$encoding = 'UTF-8';
$result_array =& _api_compare_n_grams(_api_generate_n_grams(api_substr($string, 0, LANGUAGE_DETECT_MAX_LENGTH, $encoding), $encoding), $encoding);
if (!empty($result_array)) {
list($key, $delta_points) = each($result_array);
if ($delta_points < $delta_points_min) {
$pos = strpos($key, ':');
$result_encoding = api_refine_encoding_id(substr($key, $pos + 1));
$result_language = substr($key, 0, $pos);
if ($language == $result_language && api_is_utf8($result_encoding)) {
$delta_points_min = $delta_points;
$result = $encoding;
}
}
}
}
return $result;
}
