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  * @package squirrelmail
  14  */
  15
  16 /**
  17  * Undocumented - complain, then patch.
  18  */
  19 function replaceBlock (&$in, $replace, $start, $end) {
  20     $begin = substr($in,0,$start);
  21     $end   = substr($in,$end,strlen($in)-$end);
  22     $in    = $begin.$replace.$end;
  23 }
  24
  25 /* Having this defined in just one spot could help when changes need
  26  * to be made to the pattern
  27  * Make sure that the expression is evaluated case insensitively
  28  *
  29  * Here's pretty sophisticated IP matching:
  30  * $IPMatch = '(2[0-5][0-9]|1?[0-9]{1,2})';
  31  * $IPMatch = '\[?' . $IPMatch . '(\.' . $IPMatch . '){3}\]?';
  32  */
  33 /* Here's enough: */
  34 global $IP_RegExp_Match, $Host_RegExp_Match, $Email_RegExp_Match;
  35 $IP_RegExp_Match = '\\[?[0-9]{1,3}(\\.[0-9]{1,3}){3}\\]?';
  36 $Host_RegExp_Match = '(' . $IP_RegExp_Match .
  37     '|[0-9a-z]([-.]?[0-9a-z])*\\.[a-z][a-z]+)';
  38 $Email_RegExp_Match = '[0-9a-z]([-_.+]?[0-9a-z])*(%' . $Host_RegExp_Match .
  39     ')?@' . $Host_RegExp_Match;
  40
  41 function parseEmail (&$body) {
  42     global $color, $Email_RegExp_Match;
  43     $sbody     = $body;
  44     $addresses = array();
  45
  46     /* Find all the email addresses in the body */
  47     while(eregi($Email_RegExp_Match, $sbody, $regs)) {
  48         $addresses[$regs[0]] = $regs[0];
  49         $start = strpos($sbody, $regs[0]) + strlen($regs[0]);
  50         $sbody = substr($sbody, $start);
  51     }
  52     /* Replace each email address with a compose URL */
  53     foreach ($addresses as $email) {
  54         $comp_uri = makeComposeLink('src/compose.php?send_to='.urlencode($email), $email);
  55         $body = str_replace($email, $comp_uri, $body);
  56     }
  57     /* Return number of unique addresses found */
  58     return count($addresses);
  59 }
  60
  61
  62 /* We don't want to re-initialize this stuff for every line.  Save work
  63  * and just do it once here.
  64  */
  65 global $url_parser_url_tokens;
  66 $url_parser_url_tokens = array(
  67     'http://',
  68     'https://',
  69     'ftp://',
  70     'telnet:',  // Special case -- doesn't need the slashes
  71     'gopher://',
  72     'news://');
  73
  74 global $url_parser_poss_ends;
  75 $url_parser_poss_ends = array(' ', "\n", "\r", '<', '>', ".\r", ".\n",
  76     '.&nbsp;', '&nbsp;', ')', '(', '&quot;', '&lt;', '&gt;', '.<',
  77     ']', '[', '{', '}', "\240", ', ', '. ', ",\n", ",\r");
  78
  79
  80 function parseUrl (&$body) {
  81     global $url_parser_poss_ends, $url_parser_url_tokens;;
  82     $start      = 0;
  83     $blength    = strlen($body);
  84
  85     while ($start != $blength) {
  86         $target_token = '';
  87         $target_pos = $blength;
  88
  89         /* Find the first token to replace */
  90         foreach ($url_parser_url_tokens as $the_token) {
  91             $pos = strpos(strtolower($body), $the_token, $start);
  92             if (is_int($pos) && $pos < $blength) {
  93                 $target_pos   = $pos;
  94                 $target_token = $the_token;
  95             }
  96         }
  97
  98         /* Look for email addresses between $start and $target_pos */
  99         $check_str = substr($body, $start, $target_pos-$start);
 100
 101         if (parseEmail($check_str)) {
 102             replaceBlock($body, $check_str, $start, $target_pos);
 103             $blength    = strlen($body);
 104             $target_pos = strlen($check_str) + $start;
 105         }
 106
 107         /* If there was a token to replace, replace it */
 108         if ($target_token != '') {
 109             /* Find the end of the URL */
 110             $end = $blength;
 111             foreach ($url_parser_poss_ends as $val) {
 112                 $enda = strpos($body, $val, $target_pos);
 113                 if (is_int($enda) && $enda < $end) {
 114                     $end = $enda;
 115                 }
 116             }
 117
 118             /* Extract URL */
 119             $url = substr($body, $target_pos, $end-$target_pos);
 120
 121             /* Needed since lines are not passed with \n or \r */
 122             while ( ereg("[,\.]$", $url) ) {
 123                 $url = substr( $url, 0, -1 );
 124                 $end--;
 125             }
 126
 127             /* Replace URL with HyperLinked Url, requires 1 char in link */
 128             if ($url != '' && $url != $target_token) {
 129                 $url_str = "<a href=\"$url\" target=\"_blank\">$url</a>";
 130                 replaceBlock($body,$url_str,$target_pos,$end);
 131                 $target_pos += strlen($url_str);
 132             }
 133             else {
 134                 // Not quite a valid link, skip ahead to next chance
 135                 $target_pos += strlen($target_token);
 136             }
 137         }
 138
 139         /* Move forward */
 140         $start   = $target_pos;
 141         $blength = strlen($body);
 142     }
 143 }
 144 ?>