X-Git-Url: https://vcs.fsf.org/?a=blobdiff_plain;f=functions%2Furl_parser.php;h=23eb8c9acdeb806013bd77cf6fcd81d25800835a;hb=96862638ef65bfd5438320fc30adf6edf5f1429d;hp=82da7973e48c92451ba89861ac4eed6ab74ac095;hpb=6f1450f4c0f726d0592ff8942723fdad6f14a573;p=squirrelmail.git

diff --git a/functions/url_parser.php b/functions/url_parser.php
index 82da7973..23eb8c9a 100644
--- a/functions/url_parser.php
+++ b/functions/url_parser.php
@@ -26,6 +26,10 @@ function replaceBlock (&$in, $replace, $start, $end) {
  * to be made to the pattern
  * Make sure that the expression is evaluated case insensitively
  *
+ * RFC2822 (and RFC822) defines the left side of an email address as (roughly):
+ *  1*atext *("." 1*atext)
+ * where atext is: a-zA-Z0-9!#$%&'*+-/=?^_`{|}~
+ *
  * Here's pretty sophisticated IP matching:
  * $IPMatch = '(2[0-5][0-9]|1?[0-9]{1,2})';
  * $IPMatch = '\[?' . $IPMatch . '(\.' . $IPMatch . '){3}\]?';
@@ -35,8 +39,10 @@ global $IP_RegExp_Match, $Host_RegExp_Match, $Email_RegExp_Match;
 $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 .
-    ')?@' . $Host_RegExp_Match;
+$atext = '([a-z0-9!#$&%*+/=?^_`{|}~-]|&)';
+$dot_atom = $atext . '+(\.' . $atext . '+)*';
+$Email_RegExp_Match = $dot_atom . '(%' . $Host_RegExp_Match . ')?@' .
+                      $Host_RegExp_Match;
 
 /**
  * Parses a body and converts all found email addresses to clickable links.
@@ -45,21 +51,23 @@ $Email_RegExp_Match = '[0-9a-z]([-_.+]?[0-9a-z])*(%' . $Host_RegExp_Match .
  * @return int the number of unique addresses found
  */
 function parseEmail (&$body) {
-    global $color, $Email_RegExp_Match;
+    global $Email_RegExp_Match;
     $sbody     = $body;
     $addresses = array();
 
     /* Find all the email addresses in the body */
     while(eregi($Email_RegExp_Match, $sbody, $regs)) {
-        $addresses[$regs[0]] = $regs[0];
+        $addresses[$regs[0]] = strtr($regs[0], array('&' => '&'));
         $start = strpos($sbody, $regs[0]) + strlen($regs[0]);
         $sbody = substr($sbody, $start);
     }
+
     /* Replace each email address with a compose URL */
-    foreach ($addresses as $email) {
-        $comp_uri = makeComposeLink('src/compose.php?send_to='.urlencode($email), $email);
-        $body = str_replace($email, $comp_uri, $body);
+    foreach ($addresses as $text => $email) {
+        $comp_uri = makeComposeLink('src/compose.php?send_to='.urlencode($email), $text);
+        $body = str_replace($text, $comp_uri, $body);
     }
+
     /* Return number of unique addresses found */
     return count($addresses);
 }
@@ -79,8 +87,8 @@ $url_parser_url_tokens = array(
     'news://');
 
 global $url_parser_poss_ends;
-$url_parser_poss_ends = array(' ', "\n", "\r", '<', '>', ".\r", ".\n", 
-    '.&nbsp;', '&nbsp;', ')', '(', '&quot;', '&lt;', '&gt;', '.<', 
+$url_parser_poss_ends = array(' ', "\n", "\r", '<', '>', ".\r", ".\n",
+    '.&nbsp;', '&nbsp;', ')', '(', '&quot;', '&lt;', '&gt;', '.<',
     ']', '[', '{', '}', "\240", ', ', '. ', ",\n", ",\r");
 
 
@@ -127,8 +135,8 @@ function parseUrl (&$body) {
         }
 
         /* If there was a token to replace, replace it */
-        if ($target_token == 'mailto:') {	// rfc 2368 (mailto URL)
-            $target_pos += 7;	//skip mailto:
+        if ($target_token == 'mailto:') {    // rfc 2368 (mailto URL)
+            $target_pos += 7;    //skip mailto:
             $end = $blength;
 
             $mailto = substr($body, $target_pos, $end-$target_pos);
@@ -138,12 +146,12 @@ function parseUrl (&$body) {
                 //sm_print_r($regs);
                 $mailto_before = $target_token . $regs[0];
                 $mailto_params = $regs[10];
-                if ($regs[1]) {	//if there is an email addr before '?', we need to merge it with the params
+                if ($regs[1]) {    //if there is an email addr before '?', we need to merge it with the params
                     $to = 'to=' . $regs[1];
-                    if (strpos($mailto_params, 'to=') > -1)	//already a 'to='
+                    if (strpos($mailto_params, 'to=') > -1)    //already a 'to='
                         $mailto_params = str_replace('to=', $to . '%2C%20', $mailto_params);
                     else {
-                        if ($mailto_params)	//already some params, append to them
+                        if ($mailto_params)    //already some params, append to them
                             $mailto_params .= '&amp;' . $to;
                         else
                             $mailto_params .= '?' . $to;
@@ -191,5 +199,28 @@ function parseUrl (&$body) {
         $start   = $target_pos;
         $blength = strlen($body);
     }
-} 
-?>
+}
+
+/**
+ * Parses a string and returns the first e-mail address found.
+ *
+ * @param string string the string to process
+ * @return string the first e-mail address found
+ */
+function getEmail($string) {
+    global $Email_RegExp_Match;
+    $addresses = array();
+
+    /* Find all the email addresses in the body */
+    while (eregi($Email_RegExp_Match, $string, $regs)) {
+        $addresses[$regs[0]] = strtr($regs[0], array('&amp;' => '&'));
+        $start = strpos($string, $regs[0]) + strlen($regs[0]);
+        $string = substr($string, $start);
+    }
+
+    /* Return the first address, or an empty string if no address was found */
+    $addresses = array_values($addresses);
+    return (array_key_exists(0, $addresses) ? $addresses[0] : '');
+}
+
+?>
\ No newline at end of file