Changeset 545 for trunk/lib

Timestamp:

Aug 19, 2015 4:18:55 AM (9 years ago)

Author:

anonymous

Message:

Updated hyperlinkTxt function

File:

• trunk/lib/Utilities.inc.php (modified) (3 diffs)

trunk/lib/Utilities.inc.php

@@ -225 +225 @@
 function hyperlinkTxt($text, $strict=false, $length=null, $delim='
')
 {
+    // A list of schemes we allow at the beginning of a URL.
+    $schemes = 'mailto:|tel:|skype:|callto:|facetime:|bitcoin:|geo:|magnet:\?|sip:|sms:|xmpp:|view-source:(?:https?://)?|[\w-]{2,}://';
+
     // Capture the full URL into the first match and only the first X characters into the second match.
     // This will match URLs not preceeded by " ' or = (URLs inside an attribute) or ` (Markdown quoted) or double-scheme (http://http://www.asdf.com)
@@ -233 +236 @@
         (                               # Begin match 1
             (                           # Begin match 2
-                (?:[\w-]{2,}://%s)      # URL starts with SCHEME:// or www. (if strict = false)
+                (?:%s)                  # URL starts with known scheme or www. if strict = false
                 [^\s/$.?#]+             # Any domain-valid characters
-                \.                      # At least one point
                 [^\s"`<>]{1,%s}         # Match 2 is limited to a maximum of LENGTH valid URL characters
             )
             [^\s"`<>]*                  # Match 1 continues with any further valid URL characters
-            [^\P{Any}\s
<>«»"—–%s]      # Final character not a space or common end-of-sentence punctuation (.,:;?!, etc). Using double negation set, see http://stackoverflow.com/a/4786560/277303
+            ([^\P{Any}\s
<>«»"—–%s])    # Final character not a space or common end-of-sentence punctuation (.,:;?!, etc). Using double negation set, see http://stackoverflow.com/a/4786560/277303
         )
         @Suxi
     ';
     $regex = sprintf($regex,
-        ($strict ? '' : '|www\.'), // Strict=false allows URLs beginning with www.
-        $length,
-        ($strict ? '' : '?!.,:;)\'-') // Strict=false excludes these characters as a possible last character of URL.
+        ($strict ? $schemes : $schemes . '|www\.'), // Strict=false adds "www." to the list of allowed start-of-URL.
+        ($length ? $length : ''),
+        ($strict ? '' : '?!.,:;)\'-') // Strict=false excludes some "URL-valid" characters from the last character of URL. (Hyphen must remain last character in this class.)
     );
 
@@ -253 +255 @@
     return preg_replace_callback($regex, function ($m) use ($length, $delim) {
         $url = $m[1];
-        $truncated_url = $m[2];
+        $truncated_url = $m[2] . $m[3];
         $absolute_url = preg_replace('!^www\.!', 'http://www.', $url);
         if (is_null($length) || $url == $truncated_url) {
             // If not truncating, or URL was not truncated.
-            $display_url = preg_replace('!^[\w-]{2,}://!', '', $url);
+            // Remove http schemas, and any single trailing / to make the display URL.
+            $display_url = preg_replace(['!^https?://!', '!^([^/]+)/$!'], ['', '$1'], $url);
             return sprintf('<a href="%s">%s</a>', oTxt($absolute_url), $display_url);
         } else {
             // Truncated URL.
-            $display_url = preg_replace('!^[\w-]{2,}://!', '', trim($truncated_url));
+            // Remove http schemas, and any single trailing / to make the display URL.
+            $display_url = preg_replace(['!^https?://!', '!^([^/]+)/$!'], ['', '$1'], trim($truncated_url));
             return sprintf('<a href="%s">%s%s</a>', oTxt($absolute_url), $display_url, $delim);
         }