/**
* Method to test an external or internal url for all valid parts.
*
* @param SimpleXMLElement $element The SimpleXMLElement object representing the `<field>` tag for the form field object.
* @param mixed $value The form field value to validate.
* @param string $group The field name group control value. This acts as as an array container for the field.
* For example if the field has name="foo" and the group value is set to "bar" then the
* full field name would end up being "bar[foo]".
* @param Registry $input An optional Registry object with the entire data set to validate against the entire form.
* @param JForm $form The form object for which the field is being tested.
*
* @return boolean True if the value is valid, false otherwise.
*
* @since 11.1
* @link http://www.w3.org/Addressing/URL/url-spec.txt
* @see JString
*/
public function test(SimpleXMLElement $element, $value, $group = null, Registry $input = null, JForm $form = null)
{
// If the field is empty and not required, the field is valid.
$required = (string) $element['required'] == 'true' || (string) $element['required'] == 'required';
if (!$required && empty($value)) {
return true;
}
$urlParts = UriHelper::parse_url($value);
// See http://www.w3.org/Addressing/URL/url-spec.txt
// Use the full list or optionally specify a list of permitted schemes.
if ($element['schemes'] == '') {
$scheme = array('http', 'https', 'ftp', 'ftps', 'gopher', 'mailto', 'news', 'prospero', 'telnet', 'rlogin', 'sftp', 'tn3270', 'wais', 'url', 'mid', 'cid', 'nntp', 'tel', 'urn', 'ldap', 'file', 'fax', 'modem', 'git');
} else {
$scheme = explode(',', $element['schemes']);
}
/*
* Note that parse_url() does not always parse accurately without a scheme,
* but at least the path should be set always. Note also that parse_url()
* returns False for seriously malformed URLs instead of an associative array.
* @see https://secure.php.net/manual/en/function.parse-url.php
*/
if ($urlParts === false or !array_key_exists('scheme', $urlParts)) {
/*
* The function parse_url() returned false (seriously malformed URL) or no scheme
* was found and the relative option is not set: in both cases the field is not valid.
*/
if ($urlParts === false or !$element['relative']) {
return false;
}
// The best we can do for the rest is make sure that the path exists and is valid UTF-8.
if (!array_key_exists('path', $urlParts) || !StringHelper::valid((string) $urlParts['path'])) {
return false;
}
// The internal URL seems to be good.
return true;
}
// Scheme found, check all parts found.
$urlScheme = (string) $urlParts['scheme'];
$urlScheme = strtolower($urlScheme);
if (in_array($urlScheme, $scheme) == false) {
return false;
}
// For some schemes here must be two slashes.
$scheme = array('http', 'https', 'ftp', 'ftps', 'gopher', 'wais', 'prospero', 'sftp', 'telnet', 'git');
if (in_array($urlScheme, $scheme) && substr($value, strlen($urlScheme), 3) !== '://') {
return false;
}
// The best we can do for the rest is make sure that the strings are valid UTF-8
// and the port is an integer.
if (array_key_exists('host', $urlParts) && !StringHelper::valid((string) $urlParts['host'])) {
return false;
}
if (array_key_exists('port', $urlParts) && !is_int((int) $urlParts['port'])) {
return false;
}
if (array_key_exists('path', $urlParts) && !StringHelper::valid((string) $urlParts['path'])) {
return false;
}
return true;
}
/**
* Method to test an external url for a valid parts.
*
* @param SimpleXMLElement $element The SimpleXMLElement object representing the <field /> tag for the form field object.
* @param mixed $value The form field value to validate.
* @param string $group The field name group control value. This acts as as an array container for the field.
* For example if the field has name="foo" and the group value is set to "bar" then the
* full field name would end up being "bar[foo]".
* @param Registry $input An optional Registry object with the entire data set to validate against the entire form.
* @param Form $form The form object for which the field is being tested.
*
* @return boolean True if the value is valid, false otherwise.
*
* @since 1.0
* @link http://www.w3.org/Addressing/URL/url-spec.txt
* @see Jstring
*/
public function test(SimpleXMLElement $element, $value, $group = null, Registry $input = null, Form $form = null)
{
// If the field is empty and not required, the field is valid.
$required = (string) $element['required'] == 'true' || (string) $element['required'] == 'required';
if (!$required && empty($value)) {
return true;
}
$urlParts = UriHelper::parse_url($value);
// See http://www.w3.org/Addressing/URL/url-spec.txt
// Use the full list or optionally specify a list of permitted schemes.
if ($element['schemes'] == '') {
$scheme = array('http', 'https', 'ftp', 'ftps', 'gopher', 'mailto', 'news', 'prospero', 'telnet', 'rlogin', 'tn3270', 'wais', 'url', 'mid', 'cid', 'nntp', 'tel', 'urn', 'ldap', 'file', 'fax', 'modem', 'git');
} else {
$scheme = explode(',', $element['schemes']);
}
/*
* This rule is only for full URLs with schemes because parse_url does not parse
* accurately without a scheme.
* @see http://php.net/manual/en/function.parse-url.php
*/
if ($urlParts && !array_key_exists('scheme', $urlParts)) {
return false;
}
$urlScheme = (string) $urlParts['scheme'];
$urlScheme = strtolower($urlScheme);
if (in_array($urlScheme, $scheme) == false) {
return false;
}
// For some schemes here must be two slashes.
if (($urlScheme == 'http' || $urlScheme == 'https' || $urlScheme == 'ftp' || $urlScheme == 'sftp' || $urlScheme == 'gopher' || $urlScheme == 'wais' || $urlScheme == 'gopher' || $urlScheme == 'prospero' || $urlScheme == 'telnet' || $urlScheme == 'git') && substr($value, strlen($urlScheme), 3) !== '://') {
return false;
}
// The best we can do for the rest is make sure that the strings are valid UTF-8
// and the port is an integer.
if (array_key_exists('host', $urlParts) && !String::valid((string) $urlParts['host'])) {
return false;
}
if (array_key_exists('port', $urlParts) && !is_int((int) $urlParts['port'])) {
return false;
}
if (array_key_exists('path', $urlParts) && !String::valid((string) $urlParts['path'])) {
return false;
}
return true;
}
/**
* Test the parse_url method.
*
* @return array
*
* @covers Joomla\Uri\UriHelper::parse_url
* @since 1.0
*/
public function testParse_Url()
{
$url = 'http://localhost/joomla_development/j16_trunk/administrator/index.php?option=com_contact&view=contact&layout=edit&id=5';
$expected = parse_url($url);
$actual = UriHelper::parse_url($url);
$this->assertEquals($expected, $actual, 'Line: ' . __LINE__ . ' Results should be equal');
// Test all parts of query
$url = 'https://*****:*****@www.google.com:80/folder/page.html#id?var=kay&var2=key&true';
$expected = parse_url($url);
$actual = UriHelper::parse_url($url);
$this->assertEquals($expected, $actual, 'Line: ' . __LINE__ . ' Results should be equal');
// Test special characters in URL
$url = 'http://joomla.org/mytestpath/È';
$expected = parse_url($url);
// Fix up path for UTF-8 characters
$expected['path'] = '/mytestpath/È';
$actual = UriHelper::parse_url($url);
$this->assertEquals($expected, $actual, 'Line: ' . __LINE__ . ' Results should be equal');
// Test special characters in URL
$url = 'http://mydomain.com/!*\'();:@&=+$,/?%#[]" \\';
$expected = parse_url($url);
$actual = UriHelper::parse_url($url);
$this->assertEquals($expected, $actual, 'Line: ' . __LINE__ . ' Results should be equal');
// Test url encoding in URL
$url = 'http://mydomain.com/%21%2A%27%28%29%3B%3A%40%26%3D%24%2C%2F%3F%25%23%5B%22%20%5C';
$expected = parse_url($url);
$actual = UriHelper::parse_url($url);
$this->assertEquals($expected, $actual, 'Line: ' . __LINE__ . ' Results should be equal');
// Test a mix of the above
$url = 'http://*****:*****@mydomain.com:80/%È21%25È3*%(';
$expected = parse_url($url);
// Fix up path for UTF-8 characters
$expected['path'] = '/%È21%25È3*%(';
$actual = UriHelper::parse_url($url);
$this->assertEquals($expected, $actual, 'Line: ' . __LINE__ . ' Results should be equal');
// Test invalild URL
$url = 'http:///mydomain.com';
$expected = parse_url($url);
$actual = UriHelper::parse_url($url);
$this->assertEquals($expected, $actual, 'Line: ' . __LINE__ . ' Results should be equal');
}
/**
* Does a UTF-8 safe version of PHP parse_url function
*
* @param string $url URL to parse
*
* @return mixed Associative array or false if badly formed URL.
*
* @see http://us3.php.net/manual/en/function.parse-url.php
* @since 11.1
* @deprecated 4.0 (CMS) - Use {@link \Joomla\Uri\UriHelper::parse_url()} instead.
*/
public static function parse_url($url)
{
JLog::add('JString::parse_url has been deprecated. Use \\Joomla\\Uri\\UriHelper::parse_url.', JLog::WARNING, 'deprecated');
return \Joomla\Uri\UriHelper::parse_url($url);
}
/**
* Transforms a Punycode URL to a UTF-8 URL
*
* @param string $uri The Punycode URL to transform
*
* @return string The UTF-8 URL
*
* @since 3.1.2
*/
public static function urlToUTF8($uri)
{
if (empty($uri)) {
return;
}
$parsed = UriHelper::parse_url($uri);
if (!isset($parsed['host']) || $parsed['host'] == '') {
// If there is no host we do not need to convert it.
return $uri;
}
$host = $parsed['host'];
$hostExploded = explode('.', $host);
$newhost = '';
foreach ($hostExploded as $hostex) {
$hostex = self::fromPunycode($hostex);
$newhost .= $hostex . '.';
}
$newhost = substr($newhost, 0, -1);
$newuri = '';
if (!empty($parsed['scheme'])) {
// Assume :// is required although it is not always.
$newuri .= $parsed['scheme'] . '://';
}
if (!empty($newhost)) {
$newuri .= $newhost;
}
if (!empty($parsed['port'])) {
$newuri .= ':' . $parsed['port'];
}
if (!empty($parsed['path'])) {
$newuri .= $parsed['path'];
}
if (!empty($parsed['query'])) {
$newuri .= '?' . $parsed['query'];
}
if (!empty($parsed['fragment'])) {
$newuri .= '#' . $parsed['fragment'];
}
return $newuri;
}
/**
* Teste la validité d'une URL.
*
* @param string $value L'URL à tester.
*
* @return bool True si valide, false sinon.
*/
protected function testUrl($value)
{
$urlParts = UriHelper::parse_url($value);
// Protocoles autorisés.
$scheme = array('http', 'https');
/*
* This rule is only for full URLs with schemes because parse_url does not parse
* accurately without a scheme.
* @see http://php.net/manual/en/function.parse-url.php
*/
if ($urlParts && !array_key_exists('scheme', $urlParts)) {
return false;
}
$urlScheme = (string) $urlParts['scheme'];
$urlScheme = strtolower($urlScheme);
if (in_array($urlScheme, $scheme) == false) {
return false;
}
// For some schemes here must be two slashes.
if (($urlScheme == 'http' || $urlScheme == 'https') && substr($value, strlen($urlScheme), 3) !== '://') {
return false;
}
// The best we can do for the rest is make sure that the strings are valid UTF-8
// and the port is an integer.
if (array_key_exists('host', $urlParts) && !StringHelper::valid((string) $urlParts['host'])) {
return false;
}
if (array_key_exists('port', $urlParts) && !is_int((int) $urlParts['port'])) {
return false;
}
if (array_key_exists('path', $urlParts) && !StringHelper::valid((string) $urlParts['path'])) {
return false;
}
return true;
}
