functions/url_parser.php

   1 <?php
   2
   3 /**
   4  * url_parser.php
   5  *
   6  * Copyright (c) 1999-2003 The SquirrelMail Project Team
   7  * Licensed under the GNU GPL. For full terms see the file COPYING.
   8  *
   9  * This code provides various string manipulation functions that are
  10  * used by the rest of the Squirrelmail code.
  11  *
  12  * $Id$
  13  */
  14
  15 function replaceBlock (&$in, $replace, $start, $end) {
  16     $begin = substr($in,0,$start);
  17     $end   = substr($in,$end,strlen($in)-$end);
  18     $in    = $begin.$replace.$end;
  19 }
  20
  21 /* Having this defined in just one spot could help when changes need
  22  * to be made to the pattern
  23  * Make sure that the expression is evaluated case insensitively
  24  *
  25  * Here's pretty sophisticated IP matching:
  26  * $IPMatch = '(2[0-5][0-9]|1?[0-9]{1,2})';
  27  * $IPMatch = '\[?' . $IPMatch . '(\.' . $IPMatch . '){3}\]?';
  28  */
  29 /* Here's enough: */
  30 global $IP_RegExp_Match, $Host_RegExp_Match, $Email_RegExp_Match;
  31 $IP_RegExp_Match = '\\[?[0-9]{1,3}(\\.[0-9]{1,3}){3}\\]?';
  32 $Host_RegExp_Match = '(' . $IP_RegExp_Match .
  33     '|[0-9a-z]([-.]?[0-9a-z])*\\.[a-z][a-z]+)';
  34 $Email_RegExp_Match = '[0-9a-z]([-_.+]?[0-9a-z])*(%' . $Host_RegExp_Match .
  35     ')?@' . $Host_RegExp_Match;
  36
  37 function parseEmail (&$body) {
  38     global $color, $Email_RegExp_Match;
  39     $sbody     = $body;
  40     $addresses = array();
  41
  42     /* Find all the email addresses in the body */
  43     while(eregi($Email_RegExp_Match, $sbody, $regs)) {
  44         $addresses[$regs[0]] = $regs[0];
  45         $start = strpos($sbody, $regs[0]) + strlen($regs[0]);
  46         $sbody = substr($sbody, $start);
  47     }
  48     /* Replace each email address with a compose URL */
  49     foreach ($addresses as $email) {
  50         $comp_uri = makeComposeLink('src/compose.php?send_to='.urlencode($email), $email);
  51         $body = str_replace($email, $comp_uri, $body);
  52     }
  53     /* Return number of unique addresses found */
  54     return count($addresses);
  55 }
  56
  57
  58 /* We don't want to re-initialize this stuff for every line.  Save work
  59  * and just do it once here.
  60  */
  61 global $url_parser_url_tokens;
  62 $url_parser_url_tokens = array(
  63     'http://',
  64     'https://',
  65     'ftp://',
  66     'telnet:',  // Special case -- doesn't need the slashes
  67     'gopher://',
  68     'news://');
  69
  70 global $url_parser_poss_ends;
  71 $url_parser_poss_ends = array(' ', "\n", "\r", '<', '>', ".\r", ".\n",
  72     '.&nbsp;', '&nbsp;', ')', '(', '&quot;', '&lt;', '&gt;', '.<',
  73     ']', '[', '{', '}', "\240", ', ', '. ', ",\n", ",\r");
  74
  75
  76 function parseUrl (&$body) {
  77     global $url_parser_poss_ends, $url_parser_url_tokens;;
  78     $start      = 0;
  79     $blength    = strlen($body);
  80
  81     while ($start != $blength) {
  82         $target_token = '';
  83         $target_pos = $blength;
  84
  85         /* Find the first token to replace */
  86         foreach ($url_parser_url_tokens as $the_token) {
  87             $pos = strpos(strtolower($body), $the_token, $start);
  88             if (is_int($pos) && $pos < $blength) {
  89                 $target_pos   = $pos;
  90                 $target_token = $the_token;
  91             }
  92         }
  93
  94         /* Look for email addresses between $start and $target_pos */
  95         $check_str = substr($body, $start, $target_pos-$start);
  96
  97         if (parseEmail($check_str)) {
  98             replaceBlock($body, $check_str, $start, $target_pos);
  99             $blength    = strlen($body);
 100             $target_pos = strlen($check_str) + $start;
 101         }
 102
 103         /* If there was a token to replace, replace it */
 104         if ($target_token != '') {
 105             /* Find the end of the URL */
 106             $end = $blength;
 107             foreach ($url_parser_poss_ends as $val) {
 108                 $enda = strpos($body, $val, $target_pos);
 109                 if (is_int($enda) && $enda < $end) {
 110                     $end = $enda;
 111                 }
 112             }
 113
 114             /* Extract URL */
 115             $url = substr($body, $target_pos, $end-$target_pos);
 116
 117             /* Needed since lines are not passed with \n or \r */
 118             while ( ereg("[,\.]$", $url) ) {
 119                 $url = substr( $url, 0, -1 );
 120                 $end--;
 121             }
 122
 123             /* Replace URL with HyperLinked Url, requires 1 char in link */
 124             if ($url != '' && $url != $target_token) {
 125                 $url_str = "<a href=\"$url\" target=\"_blank\">$url</a>";
 126                 replaceBlock($body,$url_str,$target_pos,$end);
 127                 $target_pos += strlen($url_str);
 128             }
 129             else {
 130                 // Not quite a valid link, skip ahead to next chance
 131                 $target_pos += strlen($target_token);
 132             }
 133         }
 134
 135         /* Move forward */
 136         $start   = $target_pos;
 137         $blength = strlen($body);
 138     }
 139 }
 140 ?>