* Improved speed of URL parser

* Made useful regexps global so other plugins (like abook_take) could use them
* Made host regexp match any TLD because of what New.net and ICANN are
  squabbling about -- might as well be safe and match a little too much than
  not match a valid host.

functions/url_parser.php

@@ -4,48 +4,48 @@
 
    $url_parser_php = true;
 
-   function replaceBlock ($in, $replace, $start, $end) {
+   function replaceBlock (&$in, $replace, $start, $end) {
       $begin = substr($in,0,$start);
       $end   = substr($in,$end,strlen($in)-$end);
-      $ret   = $begin.$replace.$end;
-      return $ret;
+      $in    = $begin.$replace.$end;
    }
 
+   // 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:
+   $IP_RegExp_Match = '\\[?[0-9]{1,3}(\\.[0-9]{1,3}){3}\\]?';
+   $Host_RegExp_Match = '(' . $IP_RegExp_Match . 
+       '|[0-9a-z]([-.]?[0-9a-z])*\\.[a-z][a-z]+)';
+   $Email_RegExp_Match = '[0-9a-z]([-_.]?[0-9a-z])*(%' . $Host_RegExp_Match . 
+       ')?@' . $Host_RegExp_Match;
+      
    function parseEmail (&$body) {
-      global $color;
+      global $color, $Email_RegExp_Match;
       $Size = strlen($body);
       
-      // 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](fo|g|l|m|me|o|op|pa|ro|seum|t|u|v|z)?)';
-      $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" || $where == "TEXT")) {
-         eregi ($Expression, $body, $regs);
+         eregi ($Email_RegExp_Match, $body, $regs);
          $oldaddr = $regs[0];
          if ($oldaddr) {
             $newaddr = eregi_replace ($what, "<b><font color=\"$color[2]\">$what</font></font></b>", $oldaddr);
             $body = str_replace ($oldaddr, "<a href=\"../src/compose.php?send_to=$oldaddr\">$newaddr</a>", $body); 
          }
       } else { 
-         $body = eregi_replace ($Expression, "<a href=\"../src/compose.php?send_to=\\0\">\\0</a>", $body);
+         $body = eregi_replace ($Email_RegExp_Match, "<a href=\"../src/compose.php?send_to=\\0\">\\0</a>", $body);
       }
       */
       
-      $body = eregi_replace ($Expression, "<a href=\"../src/compose.php?send_to=\\0\">\\0</a>", $body); 
+      $body = eregi_replace ($Email_RegExp_Match, "<a href=\"../src/compose.php?send_to=\\0\">\\0</a>", $body); 
       
       // If there are any changes, it'll just get bigger.
       if ($Size != strlen($body))
@@ -54,20 +54,24 @@
    }
 
 
-   function parseUrl (&$body)
-   {
-      $url_tokens = array(
-         'http://',
-         'https://',
-         'ftp://',
-         'telnet:',  // Special case -- doesn't need the slashes
-         'gopher://',
-         'news://');
+   // We don't want to re-initialize this stuff for every line.  Save work
+   // and just do it once here.
+   $url_parser_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");
+   $url_parser_poss_ends = array(' ', "\n", "\r", '<', '>', ".\r", ".\n", 
+       '.&nbsp;', '&nbsp;', ')', '(', '&quot;', '&lt;', '&gt;', '.<', 
+       ']', '[', '{', '}', "\240");
 
+
+   function parseUrl (&$body)
+   {
+      global $url_parser_poss_ends, $url_parser_url_tokens;;
       $start = 0;
       $target_pos = strlen($body);
       
@@ -76,7 +80,7 @@
         $target_token = '';
         
         // Find the first token to replace
-        foreach ($url_tokens as $the_token)
+        foreach ($url_parser_url_tokens as $the_token)
         {
           $pos = strpos(strtolower($body), $the_token, $start);
           if (is_int($pos) && $pos < $target_pos)
@@ -91,7 +95,7 @@
        
         if (parseEmail($check_str))
         {
-          $body = replaceBlock($body, $check_str, $start, $target_pos);
+          replaceBlock($body, $check_str, $start, $target_pos);
           $target_pos = strlen($check_str) + $start;
         }
 
@@ -100,7 +104,7 @@
         {
           // Find the end of the URL
           $end=strlen($body); 
-          foreach ($poss_ends as $key => $val)
+          foreach ($url_parser_poss_ends as $key => $val)
           {
             $enda = strpos($body,$val,$target_pos);
             if (is_int($enda) && $enda < $end) 
@@ -114,7 +118,7 @@
           if ($url != '' && $url != $target_token) 
           {
             $url_str = "<a href=\"$url\" target=\"_blank\">$url</a>";
-            $body = replaceBlock($body,$url_str,$target_pos,$end);
+            replaceBlock($body,$url_str,$target_pos,$end);
             $target_pos += strlen($url_str);
           } 
           else