*/
function cleanup_url($url)
{
- $obj_url = new Url($url);
- return $obj_url->cleanup();
+ $obj_url = new \Shaarli\Http\Url($url);
+ return $obj_url->cleanup();
}
/**
*/
function get_url_scheme($url)
{
- $obj_url = new Url($url);
- return $obj_url->getScheme();
+ $obj_url = new \Shaarli\Http\Url($url);
+ return $obj_url->getScheme();
}
/**
- * URL representation and cleanup utilities
+ * Adds a trailing slash at the end of URL if necessary.
*
- * Form
- * scheme://[username:password@]host[:port][/path][?query][#fragment]
+ * @param string $url URL to check/edit.
*
- * Examples
- * http://username:password@hostname:9090/path?arg1=value1&arg2=value2#anchor
- * https://host.name.tld
- * https://h2.g2/faq/?vendor=hitchhiker&item=guide&dest=galaxy#answer
- *
- * @see http://www.faqs.org/rfcs/rfc3986.html
+ * @return string $url URL with a end trailing slash.
*/
-class Url
+function add_trailing_slash($url)
{
- private static $annoyingQueryParams = array(
- // Facebook
- 'action_object_map=',
- 'action_ref_map=',
- 'action_type_map=',
- 'fb_',
- 'fb=',
-
- // Scoop.it
- '__scoop',
-
- // Google Analytics & FeedProxy
- 'utm_',
-
- // ATInternet
- 'xtor='
- );
-
- private static $annoyingFragments = array(
- // ATInternet
- 'xtor=RSS-',
-
- // Misc.
- 'tk.rss_all'
- );
-
- /*
- * URL parts represented as an array
- *
- * @see http://php.net/parse_url
- */
- protected $parts;
-
- /**
- * Parses a string containing a URL
- *
- * @param string $url a string containing a URL
- */
- public function __construct($url)
- {
- $this->parts = parse_url($url);
-
- if (!empty($url) && empty($this->parts['scheme'])) {
- $this->parts['scheme'] = 'http';
- }
- }
-
- /**
- * Returns a string representation of this URL
- */
- public function toString()
- {
- return unparse_url($this->parts);
- }
-
- /**
- * Removes undesired query parameters
- */
- protected function cleanupQuery()
- {
- if (! isset($this->parts['query'])) {
- return;
- }
-
- $queryParams = explode('&', $this->parts['query']);
-
- foreach (self::$annoyingQueryParams as $annoying) {
- foreach ($queryParams as $param) {
- if (startsWith($param, $annoying)) {
- $queryParams = array_diff($queryParams, array($param));
- continue;
- }
- }
- }
-
- if (count($queryParams) == 0) {
- unset($this->parts['query']);
- return;
- }
-
- $this->parts['query'] = implode('&', $queryParams);
- }
-
- /**
- * Removes undesired fragments
- */
- protected function cleanupFragment()
- {
- if (! isset($this->parts['fragment'])) {
- return;
- }
-
- foreach (self::$annoyingFragments as $annoying) {
- if (startsWith($this->parts['fragment'], $annoying)) {
- unset($this->parts['fragment']);
- break;
- }
- }
- }
+ return $url . (!endsWith($url, '/') ? '/' : '');
+}
- /**
- * Removes undesired query parameters and fragments
- *
- * @return string the string representation of this URL after cleanup
- */
- public function cleanup()
- {
- $this->cleanupQuery();
- $this->cleanupFragment();
- return $this->toString();
+/**
+ * Replace not whitelisted protocols by 'http://' from given URL.
+ *
+ * @param string $url URL to clean
+ * @param array $protocols List of allowed protocols (aside from http(s)).
+ *
+ * @return string URL with allowed protocol
+ */
+function whitelist_protocols($url, $protocols)
+{
+ if (startsWith($url, '?') || startsWith($url, '/')) {
+ return $url;
}
-
- /**
- * Get URL scheme.
- *
- * @return string the URL scheme or false if none is provided.
- */
- public function getScheme() {
- if (!isset($this->parts['scheme'])) {
- return false;
- }
- return $this->parts['scheme'];
+ $protocols = array_merge(['http', 'https'], $protocols);
+ $protocol = preg_match('#^(\w+):/?/?#', $url, $match);
+ // Protocol not allowed: we remove it and replace it with http
+ if ($protocol === 1 && ! in_array($match[1], $protocols)) {
+ $url = str_replace($match[0], 'http://', $url);
+ } elseif ($protocol !== 1) {
+ $url = 'http://' . $url;
}
+ return $url;
}