Improved link detection in bodies of mail messages.

functions/url_parser.php

@@ -13,18 +13,24 @@
    function parseEmail ($body) {
       global $color;
       
-      // Changed the expression to the one in abook_take
-      // This works very well, especially it looks like you might have
-      // three instances of it below.  Having it defined in
-      // just one spot could help when you need to change it.
-      $Expression = "[0-9a-z]([-_.]?[0-9a-z])*@[0-9a-z]([-.]?[0-9a-z])*\\.[a-wyz][a-z](g|l|m|pa|t|u|v)?";
+      // Having this defined in just one spot could help when changes need
+      // to be made to the pattern
+      // Make sure that the expression is evaluated case insensitively
+      // 
+      // Here's pretty sophisticated IP matching:
+      // $IPMatch = '(2[0-5][0-9]|1?[0-9]{1,2})';
+      // $IPMatch = '\[?' . $IPMatch . '(\.' . $IPMatch . '){3}\]?';
+      //
+      // Here's enough:
+      $IPMatch = '\[?[0-9]{1,3}(\.[0-9]{1,3}){3}\]?';
+      $Host = '(' . $IPMatch . '|[0-9a-z]([-.]?[0-9a-z])*\.[a-wyz][a-z](g|l|m|pa|t|u|v)?)';
+      $Expression = '[0-9a-z]([-_.]?[0-9a-z])*(%' . $Host . ')?@' . $Host;
       
       /*
         This is here in case we ever decide to use highlighting of searched
         text.  this does it for email addresses
         
       if ($what && ($where == "BODY" || $where == "TEXT")) {
-         // Use the $Expression
          eregi ($Expression, $body, $regs);
          $oldaddr = $regs[0];
          if ($oldaddr) {
@@ -32,94 +38,92 @@
             $body = str_replace ($oldaddr, "<a href=\"../src/compose.php?send_to=$oldaddr\">$newaddr</a>", $body); 
          }
       } else { 
-         // Use the $Expression
          $body = eregi_replace ($Expression, "<a href=\"../src/compose.php?send_to=\\0\">\\0</a>", $body);
       }
       */
-      // Use the $Expression
+      
       $body = eregi_replace ($Expression, "<a href=\"../src/compose.php?send_to=\\0\">\\0</a>", $body); 
       return $body;
    }
-   
-   function parseUrl ($body) {
-      #Possible ways a URL could finish.
 
-      // Removed "--" since it could be part of a URL
-      $poss_ends=array(" ", "\n", "\r", "<", ">", ".\r", ".\n", ".&nbsp;", "&nbsp;", ")", "(", 
-                       "&quot;", "&lt;", "&gt;", ".<", "]", "[", "{", "}");
-      $done=False;
-      while (!$done) {
-         #Look for when a URL starts
-         // Added gopher, news.  Modified telnet.
-         $url_tokens = array(
-                         "http://",
-                         "https://",
-                         "ftp://",
-                         "telnet:",  // Special case -- doesn't need the slashes
-                         "gopher://",
-                         "news://");
-         for($i = 0; $i < sizeof($url_tokens); $i++) {
-           // Removed the use of "^^" -- it is unneeded
-           if(is_int($where = strpos(strtolower($body), $url_tokens[$i], $start)))
-             break;
-         }
-         // Look between $start and $where for email links
-         $check_str = substr($body, $start, $where);
-         $new_str = parseEmail($check_str);
-       
-         if ($check_str != $new_str)
-         {
-             $body = replaceBlock($body, $new_str, $start, $where);
-             $where = strlen($new_str) + $start;
-         }
-         
-         //$where = strpos(strtolower($body),"http://",$start);
-         // Fixed this to work with $i instead of $where
-         if ($i < sizeof($url_tokens)) {
-            // Removed the "^^" so I removed the next line
-            //$where = $where - 2;  // because we added the ^^ at the begining
-            # Find the end of that URL
-            reset($poss_ends); $end=0; 
-            while (list($key, $val) = each($poss_ends)) {
-               $enda = strpos($body,$val,$where);
-               if ($end == 0) $end = $enda;
-               if ($enda < $end and $enda != 0) $end = $enda;
-            } 
-            if (!$end) $end = strlen($body);
-            #Extract URL
-            $url = substr($body,$where,$end-$where);
-            #Replace URL with HyperLinked Url
-            // Now this code doesn't simply match on url_tokens
-            // It will need some more text.  This is good.
-            if ($url != "" && $url != $url_tokens[$i]) {
-               $url_str = "<a href=\"$url\" target=\"_blank\">$url</a>";
-               #    $body = str_replace($url,$url_str,$body); 
-               # echo "$where, $end<br>";
-               $body = replaceBlock($body,$url_str,$where,$end);
-               // Removed unnecessary strpos call.  Searching
-               // a string takes longer than just figuring out
-               // the length.
-               // $start = strpos($body,"</a>",$where);
-               $start = $where + strlen($url_str);
-            } else { 
-               // Proper length increment -- Don't just assume 7
-               $start = $where + strlen($url_tokens[$i]); 
-            } 
-         } else {
-            $done=true;
-         }
-      }
 
-      // Look after $start for more email links.
-      $check_str = substr($body, $start);
-      $new_str = parseEmail($check_str);
-       
-      if ($check_str != $new_str)
+   function parseUrl ($body)
+   {
+      $url_tokens = array(
+         'http://',
+         'https://',
+         'ftp://',
+         'telnet:',  // Special case -- doesn't need the slashes
+         'gopher://',
+         'news://');
+
+      $poss_ends = array(' ', '\n', '\r', '<', '>', '.\r', '.\n', '.&nbsp;', 
+         '&nbsp;', ')', '(', '&quot;', '&lt;', '&gt;', '.<', ']', '[', '{', 
+         '}', "\240");
+
+      $start = 0;
+      $target_pos = strlen($body);
+      
+      while ($start != $target_pos)
       {
-          $body = replaceBlock($body, $new_str, $start, strlen($body));
-      }
+        $target_token = '';
+        
+        // Find the first token to replace
+        foreach ($url_tokens as $the_token)
+        {
+          $pos = strpos(strtolower($body), $the_token, $start);
+          if (is_int($pos) && $pos < $target_pos)
+          {
+            $target_pos = $pos;
+            $target_token = $the_token;
+          }
+        }
+        
+        // Look for email addresses between $start and $target_pos
+        $check_str = substr($body, $start, $target_pos);
+        $new_str = parseEmail($check_str);
+       
+        if ($check_str != $new_str)
+        {
+          $body = replaceBlock($body, $new_str, $start, $target_pos);
+          $target_pos = strlen($new_str) + $start;
+        }
 
-      return $body;
+        // If there was a token to replace, replace it
+        if ($target_token != '')
+        {
+          // Find the end of the URL
+          $end=strlen($body); 
+          foreach ($poss_ends as $key => $val)
+          {
+            $enda = strpos($body,$val,$target_pos);
+            if (is_int($enda) && $enda < $end) 
+              $end = $enda;
+          }
+          
+          // Extract URL
+          $url = substr($body, $target_pos, $end-$target_pos);
+          
+          // Replace URL with HyperLinked Url, requires 1 char in link
+          if ($url != '' && $url != $target_token) 
+          {
+            $url_str = "<a href=\"$url\" target=\"_blank\">$url</a>";
+            $body = replaceBlock($body,$url_str,$target_pos,$end);
+            $target_pos += strlen($url_str);
+          } 
+          else 
+          { 
+             // Not quite a valid link, skip ahead to next chance
+             $target_pos += strlen($target_token);
+          }
+        }
+        
+        // Move forward
+        $start = $target_pos;
+        $target_pos = strlen($body);
+      }
+      
+     return $body;
    }
-
+   
 ?>