Removed orbz.gst-group.co.uk and added sbl.spamhaus.org

[squirrelmail.git] / functions / mime.php

<?php

/**
 * mime.php
 *
 * Copyright (c) 1999-2002 The SquirrelMail Project Team
 * Licensed under the GNU GPL. For full terms see the file COPYING.
 *
 * This contains the functions necessary to detect and decode MIME
 * messages.
 *
 * $Id$
 */

require_once('../functions/imap.php');
require_once('../functions/attachment_common.php');

/** Setting up the objects that have the structure for the message **/
class msg_header {
    /** msg_header contains generic variables for values that **/
    /** could be in a header.                                 **/

    var $type0 = '', $type1 = '', $boundary = '', $charset = '',
        $encoding = '', $size = 0, $to = array(), $from = '', $date = '',
        $cc = array(), $bcc = array(), $reply_to = '', $subject = '',
        $id = 0, $mailbox = '', $description = '', $filename = '',
        $entity_id = 0, $message_id = 0, $name = '', $priority = 3;
}

class message {
    /** message is the object that contains messages.  It is a recursive
      object in that through the $entities variable, it can contain
      more objects of type message.  See documentation in mime.txt for
      a better description of how this works.
    **/
    var $header = '';
    var $entities = array();
    
    function addEntity ($msg) {
        $this->entities[] = $msg;
    }
}

/* --------------------------------------------------------------------------------- */
/* MIME DECODING                                                                     */
/* --------------------------------------------------------------------------------- */

/* This function gets the structure of a message and stores it in the "message" class.
 * It will return this object for use with all relevant header information and
 * fully parsed into the standard "message" object format.
 */
function mime_structure ($imap_stream, $header) {

    sqimap_messages_flag ($imap_stream, $header->id, $header->id, 'Seen');
    $ssid = sqimap_session_id();
    $lsid = strlen( $ssid );
    $id = $header->id;
    fputs ($imap_stream, "$ssid FETCH $id BODYSTRUCTURE\r\n");
    //
    // This should use sqimap_read_data instead of reading it itself
    //
    $read = fgets ($imap_stream, 10000);
    $bodystructure = '';
    while ( substr($read, 0, $lsid) <> $ssid &&
         !feof( $imap_stream ) ) {
        $bodystructure .= $read;
        $read = fgets ($imap_stream, 10000);
    }
    $read = $bodystructure;
    
    // isolate the body structure and remove beginning and end parenthesis
    $read = trim(substr ($read, strpos(strtolower($read), 'bodystructure') + 13));
    $read = trim(substr ($read, 0, -1));
    $end = mime_match_parenthesis(0, $read);
    while ($end == strlen($read)-1) {
        $read = trim(substr ($read, 0, -1));
        $read = trim(substr ($read, 1));
        $end = mime_match_parenthesis(0, $read);
    }
    
    $msg = mime_parse_structure ($read, 0);
    $msg->header = $header;
    
    return( $msg );
}

/* this starts the parsing of a particular structure.  It is called recursively,
 * so it can be passed different structures.  It returns an object of type
 * $message.
 * First, it checks to see if it is a multipart message.  If it is, then it
 * handles that as it sees is necessary.  If it is just a regular entity,
 * then it parses it and adds the necessary header information (by calling out
 * to mime_get_elements()
 */
function mime_parse_structure ($structure, $ent_id) {

  $msg = new message();
  if ($structure{0} == '(') {
     $ent_id = mime_new_element_level($ent_id);
     $start = $end = -1;
     do {
        $start = $end+1;
        $end = mime_match_parenthesis ($start, $structure);

        $element = substr($structure, $start+1, ($end - $start)-1);
        $ent_id = mime_increment_id ($ent_id);
        $newmsg = mime_parse_structure ($element, $ent_id);
        $msg->addEntity ($newmsg);
     } while ($structure{$end+1} == '(');
  } else {
     // parse the elements
     $msg = mime_get_element ($structure, $msg, $ent_id);
  }
  return $msg;
}

/* Increments the element ID.  An element id can look like any of
 * the following:  1, 1.2, 4.3.2.4.1, etc.  This function increments
 * the last number of the element id, changing 1.2 to 1.3.
 */
function mime_increment_id ($id) {

    if (strpos($id, ".")) {
        $first = substr($id, 0, strrpos($id, "."));
        $last = substr($id, strrpos($id, ".")+1);
        $last++;
        $new = $first . "." .$last;
    } else {
        $new = $id + 1;
    }
    
    return $new;
}

/*
 * See comment for mime_increment_id().
 * This adds another level on to the entity_id changing 1.3 to 1.3.0
 * NOTE:  1.3.0 is not a valid element ID.  It MUST be incremented
 *        before it can be used.  I left it this way so as not to have
 *        to make a special case if it is the first entity_id.  It
 *        always increments it, and that works fine.
 */
function mime_new_element_level ($id) {

  if (!$id) {
      $id = 0;
  } else {
      $id = $id . '.0';
  }

  return( $id );
}

function mime_get_element (&$structure, $msg, $ent_id) {

  $elem_num = 1;
  $msg->header = new msg_header();
  $msg->header->entity_id = $ent_id;
  $properties = array();

  while (strlen($structure) > 0) {
     $structure = trim($structure);
     $char = $structure{0};

     if (strtolower(substr($structure, 0, 3)) == 'nil') {
        $text = '';
        $structure = substr($structure, 3);
     } else if ($char == '"') {
        // loop through until we find the matching quote, and return that as a string
        $pos = 1;
        $text = '';
        while ( ($char = $structure{$pos} ) <> '"' && $pos < strlen($structure)) {
           $text .= $char;
           $pos++;
        }
        $structure = substr($structure, strlen($text) + 2);
     } else if ($char == '(') {
        // comment me
        $end = mime_match_parenthesis (0, $structure);
        $sub = substr($structure, 1, $end-1);
        $properties = mime_get_props($properties, $sub);
        $structure = substr($structure, strlen($sub) + 2);
     } else {
        // loop through until we find a space or an end parenthesis
        $pos = 0;
        $char = $structure{$pos};
        $text = '';
        while ($char != ' ' && $char != ')' && $pos < strlen($structure)) {
           $text .= $char;
           $pos++;
           $char = $structure{$pos};
        }
        $structure = substr($structure, strlen($text));
     }

     // This is where all the text parts get put into the header
     switch ($elem_num) {
        case 1:
           $msg->header->type0 = strtolower($text);
           break;
        case 2:
           $msg->header->type1 = strtolower($text);
           break;
        case 4: // Id
           // Invisimail enclose images with <>
           $msg->header->id = str_replace( '<', '', str_replace( '>', '', $text ) );
           break;
        case 5:
           $msg->header->description = $text;
           break;
        case 6:
           $msg->header->encoding = strtolower($text);
           break;
        case 7:
           $msg->header->size = $text;
           break;
        default:
           if ($msg->header->type0 == 'text' && $elem_num == 8) {
              // This is a plain text message, so lets get the number of lines
              // that it contains.
              $msg->header->num_lines = $text;

           } else if ($msg->header->type0 == 'message' && $msg->header->type1 == 'rfc822' && $elem_num == 8) {
              // This is an encapsulated message, so lets start all over again and
              // parse this message adding it on to the existing one.
              $structure = trim($structure);
              if ( $structure{0} == '(' ) {
                 $e = mime_match_parenthesis (0, $structure);
                 $structure = substr($structure, 0, $e);
                 $structure = substr($structure, 1);
                 $m = mime_parse_structure($structure, $msg->header->entity_id);

                 // the following conditional is there to correct a bug that wasn't
                 // incrementing the entity IDs correctly because of the special case
                 // that message/rfc822 is.  This fixes it fine.
                 if (substr($structure, 1, 1) != '(')
                    $m->header->entity_id = mime_increment_id(mime_new_element_level($ent_id));

                 // Now we'll go through and reformat the results.
                 if ($m->entities) {
                    for ($i=0; $i < count($m->entities); $i++) {
                       $msg->addEntity($m->entities[$i]);
                    }
                 } else {
                    $msg->addEntity($m);
                 }
                 $structure = "";
              }
           }
           break;
     }
     $elem_num++;
     $text = "";
  }
  // loop through the additional properties and put those in the various headers
  if ($msg->header->type0 != 'message') {
     for ($i=0; $i < count($properties); $i++) {
        $msg->header->{$properties[$i]['name']} = $properties[$i]['value'];
     }
  }

  return $msg;
}

/*
 * I did most of the MIME stuff yesterday (June 20, 2000), but I couldn't
 * figure out how to do this part, so I decided to go to bed.  I woke up
 * in the morning and had a flash of insight.  I went to the white-board
 * and scribbled it out, then spent a bit programming it, and this is the
 * result.  Nothing complicated, but I think my brain was fried yesterday.
 * Funny how that happens some times.
 *
 * This gets properties in a nested parenthesisized list.  For example,
 * this would get passed something like:  ("attachment" ("filename" "luke.tar.gz"))
 * This returns an array called $props with all paired up properties.
 * It ignores the "attachment" for now, maybe that should change later
 * down the road.  In this case, what is returned is:
 *    $props[0]["name"] = "filename";
 *    $props[0]["value"] = "luke.tar.gz";
 */
function mime_get_props ($props, $structure) {

  while (strlen($structure) > 0) {
     $structure = trim($structure);
     $char = $structure{0};

     if ($char == '"') {
        $pos = 1;
        $tmp = '';
        while ( ( $char = $structure{$pos} ) != '"' &&
                $pos < strlen($structure)) {
           $tmp .= $char;
           $pos++;
        }
        $structure = trim(substr($structure, strlen($tmp) + 2));
        $char = $structure{0};

        if ($char == '"') {
           $pos = 1;
           $value = '';
           while ( ( $char = $structure{$pos} ) != '"' &&
                   $pos < strlen($structure) ) {
              $value .= $char;
              $pos++;
           }
           $structure = trim(substr($structure, strlen($tmp) + 2));

           $k = count($props);
           $props[$k]['name'] = strtolower($tmp);
           $props[$k]['value'] = $value;
        } else if ($char == '(') {
           $end = mime_match_parenthesis (0, $structure);
           $sub = substr($structure, 1, $end-1);
       if (! isset($props))
           $props = array();
           $props = mime_get_props($props, $sub);
           $structure = substr($structure, strlen($sub) + 2);
        }
        return $props;
     } else if ($char == '(') {
        $end = mime_match_parenthesis (0, $structure);
        $sub = substr($structure, 1, $end-1);
        $props = mime_get_props($props, $sub);
        $structure = substr($structure, strlen($sub) + 2);
        return $props;
     } else {
        return $props;
     }
  }
}

/*
 *  Matches parenthesis.  It will return the position of the matching
 *  parenthesis in $structure.  For instance, if $structure was:
 *     ("text" "plain" ("val1name", "1") nil ... )
 *     x                                         x
 *  then this would return 42 to match up those two.
 */
function mime_match_parenthesis ($pos, $structure) {

    $j = strlen( $structure );
    
    // ignore all extra characters
    // If inside of a string, skip string -- Boundary IDs and other
    // things can have ) in them.
    if ( $structure{$pos} != '(' ) {
        return( $j );
    }
    
    while ( $pos < $j ) {
        $pos++;
        if ($structure{$pos} == ')') {
            return $pos;
        } elseif ($structure{$pos} == '"') {
            $pos++;
            while ( $structure{$pos} != '"' &&
                    $pos < $j ) {
               if (substr($structure, $pos, 2) == '\\"') {
                  $pos++;
               } elseif (substr($structure, $pos, 2) == '\\\\') {
                  $pos++;
               }
               $pos++;
            }
        } elseif ( $structure{$pos} == '(' ) {
            $pos = mime_match_parenthesis ($pos, $structure);
        }
    }
    echo _("Error decoding mime structure.  Report this as a bug!") . '<br>';
    return( $pos );
}

function mime_fetch_body ($imap_stream, $id, $ent_id ) {
    // do a bit of error correction.  If we couldn't find the entity id, just guess
    // that it is the first one.  That is usually the case anyway.
    if (!$ent_id)
        $ent_id = 1;
    $data = sqimap_run_command ($imap_stream, "FETCH $id BODY[$ent_id]", true, $response, $message);
    $topline = array_shift($data);
    while (! ereg('\\* [0-9]+ FETCH ', $topline) && $data)
        $topline = array_shift($data);
    $wholemessage = implode('', $data);
    if (ereg('\\{([^\\}]*)\\}', $topline, $regs)) {
        $ret = substr( $wholemessage, 0, $regs[1] );
        /*
            There is some information in the content info header that could be important
            in order to parse html messages. Let's get them here.
        */
        if ( $ret{0} == '<' ) {
            $data = sqimap_run_command ($imap_stream, "FETCH $id BODY[$ent_id.MIME]", true, $response, $message);
            $base = '';
            $k = 10;
            foreach( $data as $d ) {
                if ( substr( $d, 0, 13 ) == 'Content-Base:' ) {
                    $j = strlen( $d );
                    $i = 13;
                    $base = '';
                    while ( $i < $j &&
                           ( !isNoSep( $d{$i} ) || $d{$i} == '"' )  )
                        $i++;
                    while ( $i < $j ) {
                        if ( isNoSep( $d{$i} ) )
                            $base .= $d{$i};
                        $i++;
                    }
                    $k = 0;
                } elseif ( $k == 1 && !isnosep( $d{0} ) ) {
                    $base .= substr( $d, 1 );
                }
                $k++;
            }
            if ( $base <> '' )
                
                $ret = "<base href=\"$base\">" . $ret;
        }
    } else if (ereg('"([^"]*)"', $topline, $regs)) {
        $ret = $regs[1];
    } else {
        global $where, $what, $mailbox, $passed_id, $startMessage;
        $par = "mailbox=".urlencode($mailbox)."&passed_id=$passed_id";
        if (isset($where) && isset($what)) {
            $par .= "&where=".urlencode($where)."&what=".urlencode($what);
        } else {
            $par .= "&startMessage=$startMessage&show_more=0";
        }
        $par .= '&response='.urlencode($response).'&message='.urlencode($message).
                '&topline='.urlencode($topline);

        echo   '<b><font color=$color[2]>' .
               _("Body retrieval error. The reason for this is most probably that the message is malformed. Please help us making future versions better by submitting this message to the developers knowledgebase!") .
               "<A HREF=\"../src/retrievalerror.php?$par\">Submit message</A><BR>" .
               '<tt>' . _("Response:") . "$response<BR>" .
               _("Message:") . " $message<BR>" .
               _("FETCH line:") . " $topline<BR></tt></font></b>";

        $data = sqimap_run_command ($imap_stream, "FETCH $passed_id BODY[]", true, $response, $message);
        array_shift($data);
        $wholemessage = implode('', $data);

        $ret = "---------------\n$wholemessage";

    }
    return( $ret );
}

function mime_print_body_lines ($imap_stream, $id, $ent_id, $encoding) {
    // do a bit of error correction.  If we couldn't find the entity id, just guess
    // that it is the first one.  That is usually the case anyway.
    if (!$ent_id) {
        $ent_id = 1;
    }
    $sid = sqimap_session_id();
    // Don't kill the connection if the browser is over a dialup
    // and it would take over 30 seconds to download it.