Source for file url_parser.php
Documentation is available at url_parser.php
* This code provides various string manipulation functions that are
* used by the rest of the SquirrelMail code.
* @copyright © 1999-2006 The SquirrelMail Project Team
* @license http://opensource.org/licenses/gpl-license.php GNU Public License
* @version $Id: url_parser.php,v 1.64 2006/07/15 12:00:45 tokul Exp $
* Undocumented - complain, then patch.
$begin =
substr($in,0,$start);
$in =
$begin.
$replace.
$end;
/* Having this defined in just one spot could help when changes need
* to be made to the pattern
* Make sure that the expression is evaluated case insensitively
* RFC2822 (and RFC822) defines the left side of an email address as (roughly):
* where atext is: a-zA-Z0-9!#$%&'*+-/=?^_`{|}~
* Here's pretty sophisticated IP matching:
* $IPMatch = '(2[0-5][0-9]|1?[0-9]{1,2})';
* $IPMatch = '\[?' . $IPMatch . '(\.' . $IPMatch . '){3}\]?';
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 .
')?@' .
* 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
global $Email_RegExp_Match;
/* Find all the email addresses in the body */
while(eregi($Email_RegExp_Match, $sbody, $regs)) {
$addresses[$regs[0]] =
strtr($regs[0], array('&' =>
'&'));
$sbody =
substr($sbody, $start);
/* Replace each email address with a compose URL */
foreach ($addresses as $text =>
$email) {
/* Return number of unique addresses found */
return count($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(
'telnet:', // Special case -- doesn't need the slashes
'mailto:', // Special case -- doesn't use the slashes
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
global $url_parser_poss_ends, $url_parser_url_tokens;
while ($start <
$blength) {
/* Find the first token to replace */
foreach ($url_parser_url_tokens as $the_token) {
if (is_int($pos) &&
$pos <
$target_pos) {
$target_token =
$the_token;
/* Look for email addresses between $start and $target_pos */
$check_str =
substr($body, $start, $target_pos-
$start);
$target_pos =
strlen($check_str) +
$start;
/* If there was a token to replace, replace it */
if ($target_token ==
'mailto:') { // rfc 2368 (mailto URL)
$target_pos +=
7; //skip mailto:
$mailto =
substr($body, $target_pos, $end-
$target_pos);
global $MailTo_PReg_Match;
if ((preg_match($MailTo_PReg_Match, $mailto, $regs)) &&
($regs[0] !=
'')) {
$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 (strpos($mailto_params, 'to=') > -
1) //already a 'to='
$mailto_params =
str_replace('to=', $to .
'%2C%20', $mailto_params);
if ($mailto_params) //already some params, append to them
$mailto_params .=
'&' .
$to;
$mailto_params .=
'?' .
$to;
$url_str =
preg_replace(array('/to=/i', '/(?<!b)cc=/i', '/bcc=/i'), array('send_to=', 'send_to_cc=', 'send_to_bcc='), $mailto_params);
$target_pos +=
strlen($comp_uri) -
7;
if ($target_token !=
'') {
/* Find the end of the URL */
foreach ($url_parser_poss_ends as $val) {
$enda =
strpos($body, $val, $target_pos);
if (is_int($enda) &&
$enda <
$end) {
/* make sure that there are no 8bit chars between $target_pos and suspected end of URL */
$url =
substr($body, $target_pos, $end-
$target_pos);
/* Needed since lines are not passed with \n or \r */
while ( ereg("[,\.]$", $url) ) {
/* Replace URL with HyperLinked Url, requires 1 char in link */
if ($url !=
'' &&
$url !=
$target_token) {
$url_str =
"<a href=\"$url\" target=\"_blank\">$url</a>";
$target_pos +=
strlen($url_str);
// Not quite a valid link, skip ahead to next chance
$target_pos +=
strlen($target_token);
* 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
global $Email_RegExp_Match;
/* Find all the email addresses in the body */
while (eregi($Email_RegExp_Match, $string, $regs)) {
$addresses[$regs[0]] =
strtr($regs[0], array('&' =>
'&'));
$string =
substr($string, $start);
/* Return the first address, or an empty string if no address was found */
* Finds first occurrence of 8bit data in the string
* Function finds first 8bit symbol or html entity that represents 8bit character.
* Search start is defined by $offset argument. Search ends at $maxlength position.
* If $maxlength is not defined or bigger than provided string, search ends when
* Check returned data type in order to avoid confusion between bool(false)
* (not found) and int(0) (first char in the string).
* @param string $haystack
* @param integer $maxlength
* @return mixed integer with first 8bit character position or boolean false
if ($maxlength===
false ||
strlen($haystack) <
$maxlength) {
for($i=
$offset;$i<
$maxlength;$i++
) {
/* rh7-8 compatibility. don't use full 8bit range in regexp */
if (preg_match('/[\200-\237]|\240|[\241-\377]/',$haystack[$i])) {
/* we have 8bit char. stop here and return position */
} elseif ($haystack[$i]==
'&') {
$substring =
substr($haystack,$i);
* 1. look for "&#(decimal number);" where decimal_number is bigger than 127
* 2. look for "&x(hexadecimal number);", where hex number is bigger than x7f
* 3. look for any html character entity that is not 7bit html special char. Use
* own sq_get_html_translation_table() function with 'utf-8' character set in
* order to get all html entities.
if ((preg_match('/^&#(\d+);/',$substring,$match) &&
$match[1]>
127) ||
(preg_match('/^&x([0-9a-f]+);/i',$substring,$match) &&
$match[1]>
"\x7f") ||
(preg_match('/^&([a-z]+);/i',$substring,$match) &&
Documentation generated on Sat, 07 Oct 2006 16:14:01 +0300 by phpDocumentor 1.3.0RC6