class UrlHelper
Same name in other branches
- 9 core/lib/Drupal/Component/Utility/UrlHelper.php \Drupal\Component\Utility\UrlHelper
- 8.9.x core/lib/Drupal/Component/Utility/UrlHelper.php \Drupal\Component\Utility\UrlHelper
- 10 core/lib/Drupal/Component/Utility/UrlHelper.php \Drupal\Component\Utility\UrlHelper
Helper class URL based methods.
Hierarchy
- class \Drupal\Component\Utility\UrlHelper
Expanded class hierarchy of UrlHelper
Related topics
82 files declare their use of UrlHelper
- AjaxPageState.php in core/
lib/ Drupal/ Core/ StackMiddleware/ AjaxPageState.php - AjaxPageStateTest.php in core/
modules/ system/ tests/ src/ Functional/ Render/ AjaxPageStateTest.php - AjaxTest.php in core/
tests/ Drupal/ FunctionalJavascriptTests/ Ajax/ AjaxTest.php - AssetControllerBase.php in core/
modules/ system/ src/ Controller/ AssetControllerBase.php - AssetOptimizationTest.php in core/
tests/ Drupal/ FunctionalTests/ Asset/ AssetOptimizationTest.php
File
-
core/
lib/ Drupal/ Component/ Utility/ UrlHelper.php, line 10
Namespace
Drupal\Component\UtilityView source
class UrlHelper {
/**
* The list of allowed protocols.
*
* @var array
*/
protected static $allowedProtocols = [
'http',
'https',
];
/**
* Parses an array into a valid query string encoded with rawurlencode().
*
* Function rawurlencode() is RFC3986 compliant, and as a consequence RFC3987
* compliant. The latter defines the required format of "URLs" in HTML5.
* urlencode() is almost the same as rawurlencode(), except that it encodes
* spaces as "+" instead of "%20". This makes its result non compliant to
* RFC3986 and as a consequence non compliant to RFC3987 and as a consequence
* not valid as a "URL" in HTML5.
*
* @param array $query
* The query parameter array to be processed; for instance,
* \Drupal::request()->query->all().
* @param string $parent
* (optional) Internal use only. Used to build the $query array key for
* nested items. Defaults to an empty string.
*
* @return string
* A string encoded with rawurlencode() which can be used as or appended to
* the URL query string.
*
* @ingroup php_wrappers
*/
public static function buildQuery(array $query, $parent = '') {
$params = [];
foreach ($query as $key => $value) {
$key = $parent ? $parent . rawurlencode('[' . $key . ']') : rawurlencode($key);
// Recurse into children.
if (is_array($value)) {
$params[] = static::buildQuery($value, $key);
}
elseif (!isset($value)) {
$params[] = $key;
}
else {
// For better readability of paths in query strings, we decode slashes.
$params[] = $key . '=' . str_replace('%2F', '/', rawurlencode($value));
}
}
return implode('&', $params);
}
/**
* Compresses a string for use in a query parameter.
*
* While RFC 1738 doesn't specify a maximum length for query strings,
* browsers or server configurations may restrict URLs and/or query strings to
* a certain length, often 1000 or 2000 characters. This method can be used to
* compress a string into a URL-safe query parameter which will be shorter
* than if it was used directly.
*
* @see \Drupal\Component\Utility\UrlHelper::uncompressQueryParameter()
*
* @param string $data
* The data to compress.
*
* @return string
* The data compressed into a URL-safe string.
*/
public static function compressQueryParameter(string $data) : string {
// Use 'base64url' encoding. Note that the '=' sign is only used for padding
// on the right of the string, and is otherwise not part of the data.
// @see https://datatracker.ietf.org/doc/html/rfc4648#section-5
// @see https://www.php.net/manual/en/function.base64-encode.php#123098
return str_replace([
'+',
'/',
'=',
], [
'-',
'_',
'',
], base64_encode(gzcompress($data)));
}
/**
* Takes a compressed parameter and converts it back to the original.
*
* @see \Drupal\Component\Utility\UrlHelper::compressQueryParameter()
*
* @param string $compressed
* A string as compressed by
* \Drupal\Component\Utility\UrlHelper::compressQueryParameter().
*
* @return string
* The uncompressed data, or the original string if it cannot be
* uncompressed.
*/
public static function uncompressQueryParameter(string $compressed) : string {
// Because this comes from user data, suppress the PHP warning that
// gzcompress() throws if the base64-encoded string is invalid.
$return = @gzuncompress(base64_decode(str_replace([
'-',
'_',
], [
'+',
'/',
], $compressed)));
// If we failed to uncompress the query parameter, it may be a stale link
// from before compression was implemented with the URL parameter
// uncompressed already, or it may be an incorrectly formatted URL.
// In either case, pass back the original string to the caller.
return $return === FALSE ? $compressed : $return;
}
/**
* Filters a URL query parameter array to remove unwanted elements.
*
* @param array $query
* An array to be processed.
* @param array $exclude
* (optional) A list of $query array keys to remove. Use "parent[child]" to
* exclude nested items.
* @param string $parent
* Internal use only. Used to build the $query array key for nested items.
*
* @return array
* An array containing query parameters.
*/
public static function filterQueryParameters(array $query, array $exclude = [], $parent = '') {
// If $exclude is empty, there is nothing to filter.
if (empty($exclude)) {
return $query;
}
elseif (!$parent) {
$exclude = array_flip($exclude);
}
$params = [];
foreach ($query as $key => $value) {
$string_key = $parent ? $parent . '[' . $key . ']' : $key;
if (isset($exclude[$string_key])) {
continue;
}
if (is_array($value)) {
$params[$key] = static::filterQueryParameters($value, $exclude, $string_key);
}
else {
$params[$key] = $value;
}
}
return $params;
}
/**
* Parses a URL string into its path, query, and fragment components.
*
* This function splits both internal paths like @code node?b=c#d @endcode and
* external URLs like @code https://example.com/a?b=c#d @endcode into their
* component parts. See
* @link http://tools.ietf.org/html/rfc3986#section-3 RFC 3986 @endlink for an
* explanation of what the component parts are.
*
* Note that, unlike the RFC, when passed an external URL, this function
* groups the scheme, authority, and path together into the path component.
*
* @param string $url
* The internal path or external URL string to parse.
*
* @return array
* An associative array containing:
* - path: The path component of $url. If $url is an external URL, this
* includes the scheme, authority, and path.
* - query: An array of query parameters from $url, if they exist.
* - fragment: The fragment component from $url, if it exists.
*
* @see \Drupal\Core\Utility\LinkGenerator
* @see http://tools.ietf.org/html/rfc3986
*
* @ingroup php_wrappers
*/
public static function parse($url) {
$options = [
'path' => NULL,
'query' => [],
'fragment' => '',
];
// External URLs: not using parse_url() here, so we do not have to rebuild
// the scheme, host, and path without having any use for it.
// The URL is considered external if it contains the '://' delimiter. Since
// a URL can also be passed as a query argument, we check if this delimiter
// appears in front of the '?' query argument delimiter.
$scheme_delimiter_position = strpos($url, '://');
$query_delimiter_position = strpos($url, '?');
if ($scheme_delimiter_position !== FALSE && ($query_delimiter_position === FALSE || $scheme_delimiter_position < $query_delimiter_position)) {
// Split off the fragment, if any.
if (str_contains($url, '#')) {
[
$url,
$options['fragment'],
] = explode('#', $url, 2);
}
// Split off everything before the query string into 'path'.
$parts = explode('?', $url, 2);
// Don't support URLs without a path, like 'http://'.
[
,
$path,
] = explode('://', $parts[0], 2);
if ($path != '') {
$options['path'] = $parts[0];
}
// If there is a query string, transform it into keyed query parameters.
if (isset($parts[1])) {
parse_str($parts[1], $options['query']);
}
}
else {
// parse_url() does not support relative URLs, so make it absolute. For
// instance, the relative URL "foo/bar:1" isn't properly parsed.
$parts = parse_url('http://example.com/' . $url);
// Strip the leading slash that was just added.
$options['path'] = substr($parts['path'], 1);
if (isset($parts['query'])) {
parse_str($parts['query'], $options['query']);
}
if (isset($parts['fragment'])) {
$options['fragment'] = $parts['fragment'];
}
}
return $options;
}
/**
* Encodes a Drupal path for use in a URL.
*
* For aesthetic reasons slashes are not escaped.
*
* @param string $path
* The Drupal path to encode.
*
* @return string
* The encoded path.
*/
public static function encodePath($path) {
return str_replace('%2F', '/', rawurlencode($path));
}
/**
* Determines whether a path is external to Drupal.
*
* An example of an external path is http://example.com. If a path cannot be
* assessed by Drupal's menu handler, then we must treat it as potentially
* insecure.
*
* @param string $path
* The internal path or external URL being linked to, such as "node/34" or
* "http://example.com/foo".
*
* @return bool
* TRUE or FALSE, where TRUE indicates an external path.
*/
public static function isExternal($path) {
$colon_position = strpos($path, ':');
// Some browsers treat \ as / so normalize to forward slashes.
$path = str_replace('\\', '/', $path);
// If the path starts with 2 slashes then it is always considered an
// external URL without an explicit protocol part.
return str_starts_with($path, '//') || preg_match('/^\\p{C}/u', $path) !== 0 || $colon_position !== FALSE && !preg_match('![/?#]!', substr($path, 0, $colon_position)) && static::stripDangerousProtocols($path) == $path;
}
/**
* Determines if an external URL points to this installation.
*
* @param string $url
* A string containing an external URL, such as "http://example.com/foo".
* @param string $base_url
* The base URL string to check against, such as "http://example.com/"
*
* @return bool
* TRUE if the URL has the same domain and base path.
*
* @throws \InvalidArgumentException
* Exception thrown when either $url or $base_url are not fully qualified.
*/
public static function externalIsLocal($url, $base_url) {
// Some browsers treat \ as / so normalize to forward slashes.
$url = str_replace('\\', '/', $url);
// Leading control characters may be ignored or mishandled by browsers, so
// assume such a path may lead to a non-local location. The \p{C} character
// class matches all UTF-8 control, unassigned, and private characters.
if (preg_match('/^\\p{C}/u', $url) !== 0) {
return FALSE;
}
$url_parts = parse_url($url);
$base_parts = parse_url($base_url);
if (empty($base_parts['host']) || empty($url_parts['host'])) {
throw new \InvalidArgumentException('A path was passed when a fully qualified domain was expected.');
}
if (!isset($url_parts['path']) || !isset($base_parts['path'])) {
return (!isset($base_parts['path']) || $base_parts['path'] == '/') && $url_parts['host'] == $base_parts['host'];
}
else {
// When comparing base paths, we need a trailing slash to make sure a
// partial URL match isn't occurring. Since base_path() always returns
// with a trailing slash, we don't need to add the trailing slash here.
return $url_parts['host'] == $base_parts['host'] && stripos($url_parts['path'], $base_parts['path']) === 0;
}
}
/**
* Processes an HTML attribute value and strips dangerous protocols from URLs.
*
* @param string $string
* The string with the attribute value.
*
* @return string
* Cleaned up and HTML-escaped version of $string.
*/
public static function filterBadProtocol($string) {
// Get the plain text representation of the attribute value (i.e. its
// meaning).
$string = Html::decodeEntities($string);
return Html::escape(static::stripDangerousProtocols($string));
}
/**
* Gets the allowed protocols.
*
* @return array
* An array of protocols, for example http, https and irc.
*/
public static function getAllowedProtocols() {
return static::$allowedProtocols;
}
/**
* Sets the allowed protocols.
*
* @param array $protocols
* An array of protocols, for example http, https and irc.
*/
public static function setAllowedProtocols(array $protocols = []) {
static::$allowedProtocols = $protocols;
}
/**
* Strips dangerous protocols (for example, 'javascript:') from a URI.
*
* This function must be called for all URIs within user-entered input prior
* to being output to an HTML attribute value. It is often called as part of
* \Drupal\Component\Utility\UrlHelper::filterBadProtocol() or
* \Drupal\Component\Utility\Xss::filter(), but those functions return an
* HTML-encoded string, so this function can be called independently when the
* output needs to be a plain-text string for passing to functions that will
* call Html::escape() separately. The exact behavior depends on the value:
* - If the value is a well-formed (per RFC 3986) relative URL or
* absolute URL that does not use a dangerous protocol (like
* "javascript:"), then the URL remains unchanged. This includes all
* URLs generated via Url::toString().
* - If the value is a well-formed absolute URL with a dangerous protocol,
* the protocol is stripped. This process is repeated on the remaining URL
* until it is stripped down to a safe protocol.
* - If the value is not a well-formed URL, the same sanitization behavior as
* for well-formed URLs will be invoked, which strips most substrings that
* precede a ":". The result can be used in URL attributes such as "href"
* or "src" (only after calling Html::escape() separately), but this may not
* produce valid HTML (for example, malformed URLs within "href" attributes
* fail HTML validation). This can be avoided by using
* Url::fromUri($possibly_not_a_url)->toString(), which either throws an
* exception or returns a well-formed URL.
*
* @param string $uri
* A plain-text URI that might contain dangerous protocols.
*
* @return string
* A plain-text URI stripped of dangerous protocols. As with all plain-text
* strings, this return value must not be output to an HTML page without
* being sanitized first. However, it can be passed to functions
* expecting plain-text strings.
*
* @see \Drupal\Component\Utility\Html::escape()
* @see \Drupal\Core\Url::toString()
* @see \Drupal\Core\Url::fromUri()
*/
public static function stripDangerousProtocols($uri) {
$allowed_protocols = array_flip(static::$allowedProtocols);
// Iteratively remove any invalid protocol found.
do {
$before = $uri;
$colon_position = strpos($uri, ':');
if ($colon_position > 0) {
// We found a colon, possibly a protocol. Verify.
$protocol = substr($uri, 0, $colon_position);
// If a colon is preceded by a slash, question mark or hash, it cannot
// possibly be part of the URL scheme. This must be a relative URL, which
// inherits the (safe) protocol of the base document.
if (preg_match('![/?#]!', $protocol)) {
break;
}
// Check if this is a disallowed protocol. Per RFC2616, section 3.2.3
// (URI Comparison) scheme comparison must be case-insensitive.
if (!isset($allowed_protocols[strtolower($protocol)])) {
$uri = substr($uri, $colon_position + 1);
}
}
} while ($before != $uri);
return $uri;
}
/**
* Verifies the syntax of the given URL.
*
* This function should only be used on actual URLs. It should not be used for
* Drupal menu paths, which can contain arbitrary characters.
* Valid values per RFC 3986.
*
* @param string $url
* The URL to verify.
* @param bool $absolute
* Whether the URL is absolute (beginning with a scheme such as "http:").
*
* @return bool
* TRUE if the URL is in a valid format, FALSE otherwise.
*/
public static function isValid($url, $absolute = FALSE) {
if ($absolute) {
return (bool) preg_match("\n /^ # Start at the beginning of the text\n (?:ftp|https?|feed):\\/\\/ # Look for ftp, http, https or feed schemes\n (?: # Userinfo (optional) which is typically\n (?:(?:[\\w\\.\\-\\+!\$&'\\(\\)*\\+,;=]|%[0-9a-f]{2})+:)* # a username or a username and password\n (?:[\\w\\.\\-\\+%!\$&'\\(\\)*\\+,;=]|%[0-9a-f]{2})+@ # combination\n )?\n (?:\n (?:[a-z0-9\\-\\.]|%[0-9a-f]{2})+ # A domain name or a IPv4 address\n |(?:\\[(?:[0-9a-f]{0,4}:)*(?:[0-9a-f]{0,4})\\]) # or a well formed IPv6 address\n )\n (?::[0-9]+)? # Server port number (optional)\n (?:[\\/|\\?]\n (?:[\\w#!:\\.\\?\\+=&@\$'~*,;\\/\\(\\)\\[\\]\\-]|%[0-9a-f]{2}) # The path and query (optional)\n *)?\n \$/xi", $url);
}
else {
return (bool) preg_match("/^(?:[\\w#!:\\.\\?\\+=&@\$'~*,;\\/\\(\\)\\[\\]\\-]|%[0-9a-f]{2})+\$/i", $url);
}
}
}
Members
Title Sort descending | Modifiers | Object type | Summary |
---|---|---|---|
UrlHelper::$allowedProtocols | protected static | property | The list of allowed protocols. |
UrlHelper::buildQuery | public static | function | Parses an array into a valid query string encoded with rawurlencode(). |
UrlHelper::compressQueryParameter | public static | function | Compresses a string for use in a query parameter. |
UrlHelper::encodePath | public static | function | Encodes a Drupal path for use in a URL. |
UrlHelper::externalIsLocal | public static | function | Determines if an external URL points to this installation. |
UrlHelper::filterBadProtocol | public static | function | Processes an HTML attribute value and strips dangerous protocols from URLs. |
UrlHelper::filterQueryParameters | public static | function | Filters a URL query parameter array to remove unwanted elements. |
UrlHelper::getAllowedProtocols | public static | function | Gets the allowed protocols. |
UrlHelper::isExternal | public static | function | Determines whether a path is external to Drupal. |
UrlHelper::isValid | public static | function | Verifies the syntax of the given URL. |
UrlHelper::parse | public static | function | Parses a URL string into its path, query, and fragment components. |
UrlHelper::setAllowedProtocols | public static | function | Sets the allowed protocols. |
UrlHelper::stripDangerousProtocols | public static | function | Strips dangerous protocols (for example, 'javascript:') from a URI. |
UrlHelper::uncompressQueryParameter | public static | function | Takes a compressed parameter and converts it back to the original. |
Buggy or inaccurate documentation? Please file an issue. Need support? Need help programming? Connect with the Drupal community.