/**
* Tests {@link idna_encode()}
*/
function test_idna_encode()
{
global $evo_charset;
$old_evo_charset = $evo_charset;
$evo_charset = 'utf-8';
// this file
$this->assertEqual(idna_encode('läu.de'), 'xn--lu-via.de');
$evo_charset = $old_evo_charset;
}
/**
* Get the base domain (without protocol and any subdomain) of an URL.
*
* Gets a max of 3 domain parts (x.y.tld)
*
* @param string URL
* @return string the base domain (may become empty, if found invalid)
*/
function get_base_domain($url)
{
global $evo_charset;
//echo '<p>'.$url;
// Chop away the http part and the path:
$domain = preg_replace('~^([a-z]+://)?([^:/#]+)(.*)$~i', '\\2', $url);
if (empty($domain) || preg_match('~^(\\d+\\.)+\\d+$~', $domain)) {
// Empty or All numeric = IP address, don't try to cut it any further
return $domain;
}
//echo '<br>'.$domain;
// Get the base domain up to 3 levels (x.y.tld):
// NOTE: "_" is not really valid, but for Windows it is..
// NOTE: \w includes "_"
// convert URL to IDN:
$domain = idna_encode($domain);
$domain_pattern = '~ ( \\w (\\w|-|_)* \\. ){0,2} \\w (\\w|-|_)* $~ix';
if (!preg_match($domain_pattern, $domain, $match)) {
return '';
}
$base_domain = convert_charset(idna_decode($match[0]), $evo_charset, 'UTF-8');
// Remove any www. prefix:
$base_domain = preg_replace('~^www\\.~i', '', $base_domain);
//echo '<br>'.$base_domain.'</p>';
return $base_domain;
}
/**
* Check the validity of a given URL
*
* Checks allowed URI schemes and URL ban list.
* URL can be empty.
*
* Note: We have a problem when trying to "antispam" a keyword which is already blacklisted
* If that keyword appears in the URL... then the next page has a bad referer! :/
*
* {@internal This function gets tested in misc.funcs.simpletest.php.}}
*
* @param string Url to validate
* @param string Context ("posting", "commenting", "download_src", "http-https")
* @param boolean also do an antispam check on the url
* @return mixed false (which means OK) or error message
*/
function validate_url($url, $context = 'posting', $antispam_check = true)
{
global $Debuglog, $debug;
if (empty($url)) {
// Empty URL, no problem
return false;
}
// Do not give verbose info for comments, unless debug is enabled.
$verbose = $debug || $context != 'commenting';
$allowed_uri_schemes = get_allowed_uri_schemes($context);
// Validate URL structure
if ($url[0] == '$') {
// This is a 'special replace code' URL (used in footers)
if (!preg_match('~\\$([a-z_]+)\\$~', $url)) {
return T_('Invalid URL $code$ format');
}
} elseif (preg_match('~^\\w+:~', $url)) {
// there's a scheme and therefor an absolute URL:
if (substr($url, 0, 7) == 'mailto:') {
// mailto:link
if (!in_array('mailto', $allowed_uri_schemes)) {
// Scheme not allowed
$scheme = 'mailto:';
$Debuglog->add('URI scheme «' . $scheme . '» not allowed!', 'error');
return $verbose ? sprintf(T_('URI scheme "%s" not allowed.'), htmlspecialchars($scheme)) : T_('URI scheme not allowed.');
}
preg_match('~^(mailto):(.*?)(\\?.*)?$~', $url, $match);
if (!$match) {
return $verbose ? sprintf(T_('Invalid email link: %s.'), htmlspecialchars($url)) : T_('Invalid email link.');
} elseif (!is_email($match[2])) {
return $verbose ? sprintf(T_('Supplied email address (%s) is invalid.'), htmlspecialchars($match[2])) : T_('Invalid email address.');
}
} elseif (substr($url, 0, 6) == 'clsid:') {
// clsid:link
if (!in_array('clsid', $allowed_uri_schemes)) {
// Scheme not allowed
$scheme = 'clsid:';
$Debuglog->add('URI scheme «' . $scheme . '» not allowed!', 'error');
return $verbose ? sprintf(T_('URI scheme "%s" not allowed.'), htmlspecialchars($scheme)) : T_('URI scheme not allowed.');
}
if (!preg_match('~^(clsid):([a-fA-F0-9\\-]+)$~', $url, $match)) {
return T_('Invalid class ID format');
}
} elseif (substr($url, 0, 11) == 'javascript:') {
// javascript:
// Basically there could be anything here
if (!in_array('javascript', $allowed_uri_schemes)) {
// Scheme not allowed
$scheme = 'javascript:';
$Debuglog->add('URI scheme «' . $scheme . '» not allowed!', 'error');
return $verbose ? sprintf(T_('URI scheme "%s" not allowed.'), htmlspecialchars($scheme)) : T_('URI scheme not allowed.');
}
preg_match('~^(javascript):~', $url, $match);
} else {
// convert URL to IDN:
$url = idna_encode($url);
if (!preg_match('~^ # start
([a-z][a-z0-9+.\\-]*) # scheme
:// # authorize absolute URLs only ( // not present in clsid: -- problem? ; mailto: handled above)
(\\w+(:\\w+)?@)? # username or username and password (optional)
( localhost |
[a-z0-9]([a-z0-9\\-])* # Don t allow anything too funky like entities
\\. # require at least 1 dot
[a-z0-9]([a-z0-9.\\-])+ # Don t allow anything too funky like entities
)
(:[0-9]+)? # optional port specification
.* # allow anything in the path (including spaces - used in FileManager - but no newlines).
$~ix', $url, $match)) {
// Cannot validate URL structure
$Debuglog->add('URL «' . $url . '» does not match url pattern!', 'error');
return $verbose ? sprintf(T_('Invalid URL format (%s).'), htmlspecialchars($url)) : T_('Invalid URL format.');
}
$scheme = strtolower($match[1]);
if (!in_array($scheme, $allowed_uri_schemes)) {
// Scheme not allowed
$Debuglog->add('URI scheme «' . $scheme . '» not allowed!', 'error');
return $verbose ? sprintf(T_('URI scheme "%s" not allowed.'), htmlspecialchars($scheme)) : T_('URI scheme not allowed.');
}
}
} else {
// URL is relative..
if ($context == 'commenting' || $context == 'download_src' || $context == 'http-https') {
// We do not allow relative URLs in comments and download urls
return $verbose ? sprintf(T_('URL "%s" must be absolute.'), htmlspecialchars($url)) : T_('URL must be absolute.');
}
$char = substr($url, 0, 1);
if ($char != '/' && $char != '#') {
// must start with a slash or hash (for HTML anchors to the same page)
return $verbose ? sprintf(T_('URL "%s" must be a full path starting with "/" or an anchor starting with "#".'), htmlspecialchars($url)) : T_('URL must be a full path starting with "/" or an anchor starting with "#".');
}
}
if ($antispam_check) {
// Search for blocked keywords:
if ($block = antispam_check($url)) {
return $verbose ? sprintf(T_('URL "%s" not allowed: blacklisted word "%s".'), htmlspecialchars($url), $block) : T_('URL not allowed');
}
}
return false;
// OK
}
