X-Git-Url: https://vcs.fsf.org/?a=blobdiff_plain;ds=sidebyside;f=functions%2Furl_parser.php;h=2a6da6670a431306baef4f3ef36cba60abbd322b;hb=1bc52de52ab111ec3f6f2b9c88ac6e52b96eef00;hp=eadebd37c9fbe2ba48c418d02fec63cf62baec47;hpb=8b096f0a2427cf0019f4dc4433a3e02b9f6f5951;p=squirrelmail.git diff --git a/functions/url_parser.php b/functions/url_parser.php index eadebd37..2a6da667 100644 --- a/functions/url_parser.php +++ b/functions/url_parser.php @@ -3,13 +3,13 @@ /** * url_parser.php * - * Copyright (c) 1999-2003 The SquirrelMail Project Team + * Copyright (c) 1999-2005 The SquirrelMail Project Team * Licensed under the GNU GPL. For full terms see the file COPYING. * * This code provides various string manipulation functions that are - * used by the rest of the Squirrelmail code. + * used by the rest of the SquirrelMail code. * - * $Id$ + * @version $Id$ * @package squirrelmail */ @@ -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); } @@ -74,15 +82,25 @@ $url_parser_url_tokens = array( '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", - '. ', ' ', ')', '(', '"', '<', '>', '.<', +$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. * @@ -90,18 +108,18 @@ $url_parser_poss_ends = array(' ', "\n", "\r", '<', '>', ".\r", ".\n", * @return void */ function parseUrl (&$body) { - global $url_parser_poss_ends, $url_parser_url_tokens;; + global $url_parser_poss_ends, $url_parser_url_tokens; $start = 0; $blength = strlen($body); - while ($start != $blength) { + while ($start < $blength) { $target_token = ''; $target_pos = $blength; /* Find the first token to replace */ foreach ($url_parser_url_tokens as $the_token) { $pos = strpos(strtolower($body), $the_token, $start); - if (is_int($pos) && $pos < $blength) { + if (is_int($pos) && $pos < $target_pos) { $target_pos = $pos; $target_token = $the_token; } @@ -117,6 +135,35 @@ function parseUrl (&$body) { } /* If there was a token to replace, replace it */ + if ($target_token == 'mailto:') { // rfc 2368 (mailto URL) + $target_pos += 7; //skip mailto: + $end = $blength; + + $mailto = substr($body, $target_pos, $end-$target_pos); + + global $MailTo_PReg_Match; + if ((preg_match($MailTo_PReg_Match, $mailto, $regs)) && ($regs[0] != '')) { + //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 + $to = 'to=' . $regs[1]; + if (strpos($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 + $mailto_params .= '&' . $to; + else + $mailto_params .= '?' . $to; + } + } + $url_str = preg_replace(array('/to=/i', '/(? '&')); + $start = strpos($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] : ''); +} + ?>