Cleaned up validation code and added support for mailto: format links. Issue #128225.

link.module

@@ -6,6 +6,11 @@
  * Defines simple link field types.
  */
 
+define('LINK_EXTERNAL', 'external');
+define('LINK_INTERNAL', 'internal');
+define('LINK_FRONT', 'front');
+define('LINK_EMAIL', 'email');
+
 /**
  * Implementation of hook_help().
  */
@@ -313,12 +318,15 @@ function link_field_formatter($field, $item, $formatter, $node) {
     }
   }
   
+  $type = link_validate_url($item['url']);
   $url = link_cleanup_url($item['url']);
-      
+  
+  // Seperate out the anchor if any
   if (strpos($url, '#') !== FALSE) {
     $fragment = substr($url, strpos($url, '#') + 1);
     $url = substr($url, 0, strpos($url, '#'));
   }
+  // Seperate out the query string if any
   if (strpos($url, '?') !== FALSE) {
     $query = substr($url, strpos($url, '?') + 1);
     $url = substr($url, 0, strpos($url, '?'));
@@ -332,10 +340,11 @@ function link_field_formatter($field, $item, $formatter, $node) {
   elseif (strlen(trim($item['title']))) {
     $output = l($item['title'], $url, $attributes, $query, $fragment);
   }
-  // Build the link with the URL as the title (max 80 characters)
+  // Build the link with the URL or email address as the title (max 80 characters)
   else {
-    $display_url = url($url, $query, $fragment, TRUE);
-    $output = l(strlen($display_url) > 80 ? substr($display_url,0,80)."..." : $display_url, $url, $attributes, $query, $fragment);
+    $display_url = $type == LINK_EMAIL ? str_replace('mailto:', '', $url) : url($url, $query, $fragment, TRUE);
+    $display_url = strlen($display_url) > 80 ? substr($display_url, 0, 80) . "..." : $display_url;
+    $output = l($display_url, $url, $attributes, $query, $fragment);
   }
   return $output;
 }
@@ -350,12 +359,12 @@ function link_field_formatter($field, $item, $formatter, $node) {
 function link_cleanup_url($url, $protocol = "http") {
   $url = trim($url);
   $type = link_validate_url($url);
-  if ($type == 1) {
+  if ($type == LINK_EXTERNAL) {
     // Check if there is no protocol specified
     $protocol_match = preg_match("/^([a-z0-9][a-z0-9\.\-_]*:\/\/)/i",$url);
     if (empty($protocol_match)) {
-      // But should it be? Add an automatic http:// if it starts with a domain name
-      $domain_match = preg_match('/^(([a-z0-9]([a-z0-9\-_]*\.)+)(aero|arpa|biz|com|coop|edu|gov|info|int|jobs|mil|museum|name|nato|net|org|pro|travel|[a-z]{2}))/i',$url);
+      // But should there be? Add an automatic http:// if it starts with a domain name
+      $domain_match = preg_match('/^(([a-z0-9]([a-z0-9\-_]*\.)+)(aero|arpa|biz|com|cat|coop|edu|gov|info|int|jobs|mil|museum|name|nato|net|org|pro|travel|mobi|[a-z]{2}))/i',$url);
       if (!empty($domain_match)) {
         $url = $protocol."://".$url;
       }
@@ -373,43 +382,39 @@ function link_cleanup_url($url, $protocol = "http") {
  * the following attributes: protocol, hostname, ip, and port.
  */
 function link_validate_url($text) {
-  static $allowed_protocols;
-  if (!isset($allowed_protocols)) {
-    $allowed_protocols = implode("|",variable_get('filter_allowed_protocols', array('http', 'https', 'ftp', 'news', 'nntp', 'telnet', 'mailto', 'irc', 'ssh', 'sftp', 'webcal')));
-  } 
   
-  $external_pattern = 
-  // protocol
-  '/^((' . $allowed_protocols . '):\/\/)?'.
-  '('.
-    // domains
-    '(([a-z0-9]([a-z0-9\-_]*\.)+)(aero|arpa|biz|com|cat|coop|edu|gov|info|int|jobs|mil|museum|name|nato|net|org|pro|travel|mobi|[a-z]{2}))'.
-    // OR ip addresses
-    '|(\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})'.
-  ')'.
-  // port number
-  '(:([0-9]{1,4}))?';
+  $allowed_protocols = variable_get('filter_allowed_protocols', array('http', 'https', 'ftp', 'news', 'nntp', 'telnet', 'mailto', 'irc', 'ssh', 'sftp', 'webcal'));
+  
+  $protocol = '((' . implode("|", $allowed_protocols) . '):\/\/)';
+  $domain = '(([a-z0-9]([a-z0-9\-_]*\.)+)(aero|arpa|biz|com|cat|coop|edu|gov|info|int|jobs|mil|museum|name|nato|net|org|pro|travel|mobi|[a-z]{2}))';
+  $ipv4 = '([0-9]{1,3}(\.[0-9]{1,3}){3})'; 
+  $ipv6 = '([0-9a-fA-F]{1,4}(\:[0-9a-fA-F]{1,4}){7})'; 
+  $port = '(:([0-9]{1,4}))';
   
-  // Starting path
+  // Pattern specific to eternal links
+  $external_pattern = '/^' . $protocol . '?'. '(' . $domain . '|' . $ipv4 . '|' . $ipv6 . ')' . $port . '?';
+  
+  // Pattern specific to internal links
   $internal_pattern = "/^([a-z0-9_\-+]+)";
   
-  // the rest of the path
-  $end = "(\/[a-z0-9_\-\.~+%=&,$'():;*@]*)*".
-  // query string
-  "(\/?\?[a-z0-9+_\-\.\/%=&,$'():;*@]*)?".
-  // anchors
-  "(#[a-z0-9_\-\.~+%=&,$'():;*@]*)?".
-  // end of the expression, case insensitive
-  '$/i';
+  $directories = "(\/[a-z0-9_\-\.~+%=&,$'():;*@]*)*";
+  $query = "(\/?\?[a-z0-9+_\-\.\/%=&,$'():;*@]*)";
+  $anchor = "(#[a-z0-9_\-\.~+%=&,$'():;*@]*)";
+  
+  // the rest of the path for a standard URL
+  $end = $directories . '?' . $query . '?' .  $anchor . '?' . '$/i';
   
-  if (preg_match($external_pattern . $end, $text, $m)) {
-    return 1;
+  if (preg_match($external_pattern . $end, $text)) {
+    return LINK_EXTERNAL;
+  }
+  elseif (preg_match($internal_pattern . $end, $text)) {
+    return LINK_INTERNAL;
   }
-  elseif (preg_match($internal_pattern . $end, $text, $m)) {
-    return 2;
+  elseif (in_array('mailto', $allowed_protocols) && ($address = preg_replace('/^mailto:/', '', $text)) && valid_email_address($address)) {
+    return LINK_EMAIL;
   }
   elseif (strpos($text, '<front>') === 0) {
-    return 3;
+    return LINK_FRONT;
   }
   return FALSE;
 }