- // 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:
- 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 .
- ')?@' . $Host_RegExp_Match;
-
- function parseEmail (&$body) {
- global $color, $Email_RegExp_Match;
- $Size = strlen($body);
-
- /*
- This is here in case we ever decide to use highlighting of searched
- text. this does it for email addresses
-
- if ($what && ($where == "BODY" || $where == "TEXT")) {
- eregi ($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 {
- $body = eregi_replace ($Email_RegExp_Match, "<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))
- return 1;
- return 0;
- }
-
-
- // We don't want to re-initialize this stuff for every line. Save work
- // and just do it once here.
- global $url_parser_url_tokens;
- $url_parser_url_tokens = array(
- 'http://',
- 'https://',
- 'ftp://',
- 'telnet:', // Special case -- doesn't need the slashes
- 'gopher://',
- 'news://');
-
- global $url_parser_poss_ends;
- $url_parser_poss_ends = array(' ', "\n", "\r", '<', '>', ".\r", ".\n",
- '. ', ' ', ')', '(', '"', '<', '>', '.<',
- ']', '[', '{', '}', "\240", ', ', '. ', ",\n", ",\r");
-
-
- function parseUrl (&$body)
- {
- global $url_parser_poss_ends, $url_parser_url_tokens;;
- $start = 0;
- $target_pos = strlen($body);
-
- while ($start != $target_pos)
- {
+/* Having this defined in just one spot could help when changes need
+ * to be made to the pattern
+ * Make sure that the expression is evaluated case insensitively
+ *
+ * RFC2822 (and RFC822) defines the left side of an email address as (roughly):
+ * 1*atext *("." 1*atext)
+ * where atext is: a-zA-Z0-9!#$%&'*+-/=?^_`{|}~
+ *
+ * Here's pretty sophisticated IP matching:
+ * $IPMatch = '(2[0-5][0-9]|1?[0-9]{1,2})';
+ * $IPMatch = '\[?' . $IPMatch . '(\.' . $IPMatch . '){3}\]?';
+ */
+/* Here's enough: */
+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 = '(' . $IP_RegExp_Match .
+ '|[0-9a-z]([-.]?[0-9a-z])*\\.[a-z][a-z]+)';
+$atext = '([a-z0-9!#$&%*+/=?^_`{|}~-]|&)';
+$dot_atom = $atext . '+(\.' . $atext . '+)*';
+$Email_RegExp_Match = $dot_atom . '(%' . $Host_RegExp_Match . ')?@' .
+ $Host_RegExp_Match;
+
+/**
+ * Parses a body and converts all found email addresses to clickable links.
+ *
+ * @param string body the body to process, by ref
+ * @return int the number of unique addresses found
+ */
+function parseEmail (&$body) {
+ 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]] = 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 $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);
+}
+
+
+/* We don't want to re-initialize this stuff for every line. Save work
+ * and just do it once here.
+ */
+global $url_parser_url_tokens;
+$url_parser_url_tokens = array(
+ 'http://',
+ 'https://',
+ 'ftp://',
+ 'telnet:', // Special case -- doesn't need the slashes
+ 'mailto:', // Special case -- doesn't use the slashes
+ 'gopher://',
+ 'news://');
+
+global $url_parser_poss_ends;
+$url_parser_poss_ends = array(' ', "\n", "\r", '<', '>', ".\r", ".\n",
+ '. ', ' ', ')', '(', '"', '<', '>', '.<',
+ ']', '[', '{', '}', "\240", ', ', '. ', ",\n", ",\r");
+
+
+/**
+ * rfc 2368 (mailto URL) preg_match() regexp
+ * @link http://www.ietf.org/rfc/rfc2368.txt
+ * @global string MailTo_PReg_Match the encapsulated regexp for preg_match()
+ */
+global $MailTo_PReg_Match;
+$Mailto_Email_RegExp = '[0-9a-z%]([-_.+%]?[0-9a-z])*(%' . $Host_RegExp_Match . ')?@' . $Host_RegExp_Match;
+$MailTo_PReg_Match = '/((?:' . $Mailto_Email_RegExp . ')*)((?:\?(?:to|cc|bcc|subject|body)=[^\s\?&=,()]+)?(?:&(?:to|cc|bcc|subject|body)=[^\s\?&=,()]+)*)/i';
+
+/**
+ * Parses a body and converts all found URLs to clickable links.
+ *
+ * @param string body the body to process, by ref
+ * @return void
+ */
+function parseUrl (&$body) {
+ global $url_parser_poss_ends, $url_parser_url_tokens;
+ $start = 0;
+ $blength = strlen($body);
+
+ while ($start < $blength) {