X-Git-Url: https://vcs.fsf.org/?a=blobdiff_plain;f=class%2Fmime%2FRfc822Header.class.php;h=8b77b8bb96d29818550e9d0455aa46d68d735beb;hb=c4faef335b2362c81b8ebf026d4066c12d70536c;hp=d4d36604c1d1e76795314fb38af50bd6d2dfc87e;hpb=0b4d4be725340956522c4a968a375ee7eaa820a7;p=squirrelmail.git diff --git a/class/mime/Rfc822Header.class.php b/class/mime/Rfc822Header.class.php index d4d36604..8b77b8bb 100644 --- a/class/mime/Rfc822Header.class.php +++ b/class/mime/Rfc822Header.class.php @@ -3,53 +3,175 @@ /** * Rfc822Header.class.php * - * Copyright (c) 2003 The SquirrelMail Project Team - * Licensed under the GNU GPL. For full terms see the file COPYING. + * This file contains functions needed to handle headers in mime messages. * - * This contains functions needed to handle mime messages. - * - * $Id$ + * @copyright 2003-2020 The SquirrelMail Project Team + * @license http://opensource.org/licenses/gpl-license.php GNU Public License + * @version $Id$ + * @package squirrelmail + * @subpackage mime + * @since 1.3.2 */ -/* - * rdc822_header class +/** + * MIME header class * input: header_string or array + * You must call parseHeader() function after creating object in order to fill object's + * parameters. + * @todo FIXME: there is no constructor function and class should ignore all input args. + * @package squirrelmail + * @subpackage mime + * @since 1.3.0 */ class Rfc822Header { - var $date = '', - $subject = '', - $from = array(), - $sender = '', - $reply_to = array(), - $to = array(), - $cc = array(), - $bcc = array(), - $in_reply_to = '', - $message_id = '', - $references = '', - $mime = false, - $content_type = '', - $disposition = '', - $xmailer = '', - $priority = 3, - $dnt = '', - $encoding = '', - $mlist = array(), - $more_headers = array(); /* only needed for constructing headers - in smtp.php */ + /** + * All headers, unparsed + * @var array + */ + var $raw_headers = array(); + /** + * Date header + * @var mixed + */ + var $date = -1; + /** + * Original date header as fallback for unparsable dates + * @var mixed + */ + var $date_unparsed = ''; + /** + * Subject header + * @var string + */ + var $subject = ''; + /** + * From header + * @var array + */ + var $from = array(); + /** + * @var mixed + */ + var $sender = ''; + /** + * Reply-To header + * @var array + */ + var $reply_to = array(); + /** + * Mail-Followup-To header + * @var array + */ + var $mail_followup_to = array(); + /** + * To header + * @var array + */ + var $to = array(); + /** + * Cc header + * @var array + */ + var $cc = array(); + /** + * Bcc header + * @var array + */ + var $bcc = array(); + /** + * In-reply-to header + * @var string + */ + var $in_reply_to = ''; + /** + * Message-ID header + * @var string + */ + var $message_id = ''; + /** + * References header + * @var string + */ + var $references = ''; + /** + * @var mixed + */ + var $mime = false; + /** + * Content Type object + * @var object + */ + var $content_type = ''; + /** + * @var mixed + */ + var $disposition = ''; + /** + * X-Mailer header + * @var string + */ + var $xmailer = ''; + /** + * Priority header + * @var integer + */ + var $priority = 3; + /** + * Disposition notification for requesting message delivery notification (MDN) + * @var mixed + */ + var $dnt = ''; + /** + * Delivery notification (DR) + * @var mixed + */ + var $drnt = ''; + /** + * @var mixed + */ + var $encoding = ''; + /** + * @var mixed + */ + var $content_id = ''; + /** + * @var mixed + */ + var $content_desc = ''; + /** + * @var mixed + */ + var $mlist = array(); + /** + * SpamAssassin 'x-spam-status' header + * @var mixed + */ + var $x_spam_status = array(); + /** + * Extra header + * only needed for constructing headers in delivery class + * @var array + */ + var $more_headers = array(); + + /** + * @param mixed $hdr string or array with message headers + */ function parseHeader($hdr) { if (is_array($hdr)) { $hdr = implode('', $hdr); } - /* First we unfold the header */ - $hdr = trim(str_replace(array("\r\n\t", "\r\n "),array('', ''), $hdr)); + /* First we replace \r\n by \n and unfold the header */ + /* FIXME: unfolding header with multiple spaces "\n( +)" */ + $hdr = trim(str_replace(array("\r\n", "\n\t", "\n "),array("\n", ' ', ' '), $hdr)); /* Now we can make a new header array with */ /* each element representing a headerline */ - $hdr = explode("\r\n" , $hdr); + $hdr = explode("\n" , $hdr); foreach ($hdr as $line) { $pos = strpos($line, ':'); if ($pos > 0) { + $this->raw_headers[] = $line; $field = substr($line, 0, $pos); if (!strstr($field,' ')) { /* valid field */ $value = trim(substr($line, $pos+1)); @@ -57,11 +179,15 @@ class Rfc822Header { } } } - if ($this->content_type == '') { + if (!is_object($this->content_type)) { $this->parseContentType('text/plain; charset=us-ascii'); } } + /** + * @param string $value + * @return string + */ function stripComments($value) { $result = ''; $cnt = strlen($value); @@ -76,7 +202,9 @@ class Rfc822Header { } $result .= $value{$i}; } - $result .= $value{$i}; + if($i < $cnt) { + $result .= $value{$i}; + } break; case '(': $depth = 1; @@ -104,6 +232,11 @@ class Rfc822Header { return $result; } + /** + * Parse header field according to field type + * @param string $field field name + * @param string $value field value + */ function parseField($field, $value) { $field = strtolower($field); switch($field) { @@ -112,6 +245,7 @@ class Rfc822Header { $d = strtr($value, array(' ' => ' ')); $d = explode(' ', $d); $this->date = getTimeStamp($d); + $this->date_unparsed = strtr($value,'<>',' '); break; case 'subject': $this->subject = $value; @@ -125,6 +259,9 @@ class Rfc822Header { case 'reply-to': $this->reply_to = $this->parseAddress($value, true); break; + case 'mail-followup-to': + $this->mail_followup_to = $this->parseAddress($value, true); + break; case 'to': $this->to = $this->parseAddress($value, true); break; @@ -146,11 +283,14 @@ class Rfc822Header { $this->references = $value; break; case 'x-confirm-reading-to': - case 'return-receipt-to': case 'disposition-notification-to': $value = $this->stripComments($value); $this->dnt = $this->parseAddress($value); break; + case 'return-receipt-to': + $value = $this->stripComments($value); + $this->drnt = $this->parseAddress($value); + break; case 'mime-version': $value = $this->stripComments($value); $value = str_replace(' ', '', $value); @@ -164,23 +304,35 @@ class Rfc822Header { $value = $this->stripComments($value); $this->parseDisposition($value); break; + case 'content-transfer-encoding': + $this->encoding = $value; + break; + case 'content-description': + $this->content_desc = $value; + break; + case 'content-id': + $value = $this->stripComments($value); + $this->content_id = $value; + break; case 'user-agent': case 'x-mailer': $this->xmailer = $value; break; case 'x-priority': - $this->priority = $value; + case 'importance': + case 'priority': + $this->priority = $this->parsePriority($value); break; case 'list-post': $value = $this->stripComments($value); $this->mlist('post', $value); break; case 'list-reply': - $value = $this->stripComments($value); + $value = $this->stripComments($value); $this->mlist('reply', $value); break; case 'list-subscribe': - $value = $this->stripComments($value); + $value = $this->stripComments($value); $this->mlist('subscribe', $value); break; case 'list-unsubscribe': @@ -203,14 +355,24 @@ class Rfc822Header { $value = $this->stripComments($value); $this->mlist('id', $value); break; + case 'x-spam-status': + case 'x-spam-score': + $this->x_spam_status = $this->parseSpamStatus($value); + break; + case 'x-sm-flag-reply': + $this->x_sm_flag_reply = $value; + break; default: break; } } + /** + * @param string $address + * @return array + */ function getAddressTokens($address) { $aTokens = array(); - $aAddress = array(); $aSpecials = array('(' ,'<' ,',' ,';' ,':'); $aReplace = array(' (',' <',' ,',' ;',' :'); $address = str_replace($aSpecials,$aReplace,$address); @@ -230,7 +392,7 @@ class Rfc822Header { $i = $iEnd; } $sToken = str_replace($aReplace, $aSpecials,$sToken); - $aTokens[] = $sToken; + if ($sToken) $aTokens[] = $sToken; break; case '"': $iEnd = strpos($address,$cChar,$i+1); @@ -293,24 +455,24 @@ class Rfc822Header { // check the next token in case comments appear in the middle of email addresses $prevToken = end($aTokens); if (!in_array($prevToken,$aSpecials,true)) { - if (isset($address{$i+1}) && !in_array($address{$i+1},$aSpecials,true)) { + if ($i+1personal = encodeHeader($sPersonal); } else { @@ -369,21 +539,8 @@ class Rfc822Header { return $oAddr; } - /* - * parseAddress: recursive function for parsing address strings and store - * them in an address stucture object. - * input: $address = string - * $ar = boolean (return array instead of only the - * first element) - * $addr_ar = array with parsed addresses // obsolete - * $group = string // obsolete - * $host = string (default domainname in case of - * addresses without a domainname) - * $lookup = callback function (for lookup address - * strings which are probably nicks - * (without @ ) ) - * output: array with addressstructure objects or only one - * address_structure object. + /** + * recursive function for parsing address strings and storing them in an address stucture object. * personal name: encoded: =?charset?Q|B?string?= * quoted: "string" * normal: string @@ -391,11 +548,17 @@ class Rfc822Header { * : mailbox@host * This function is also used for validating addresses returned from compose * That's also the reason that the function became a little bit huge + * @param string $address + * @param boolean $ar return array instead of only the first element + * @param array $addr_ar (obsolete) array with parsed addresses + * @param string $group (obsolete) + * @param string $host default domainname in case of addresses without a domainname + * @param string $lookup (since) callback function for lookup of address strings which are probably nicks (without @) + * @return mixed array with AddressStructure objects or only one address_structure object. */ - function parseAddress($address,$ar=false,$aAddress=array(),$sGroup='',$sHost='',$lookup=false) { $aTokens = $this->getAddressTokens($address); - $sPersonal = $sEmail = $sComment = $sGroup = ''; + $sPersonal = $sEmail = $sGroup = ''; $aStack = $aComment = array(); foreach ($aTokens as $sToken) { $cChar = $sToken{0}; @@ -404,17 +567,18 @@ class Rfc822Header { case '=': case '"': case ' ': - $aStack[] = $sToken; + $aStack[] = $sToken; break; case '(': $aComment[] = substr($sToken,1,-1); break; case ';': if ($sGroup) { + $aAddress[] = $this->createAddressObject($aStack,$aComment,$sEmail,$sGroup); $oAddr = end($aAddress); - if(!$oAddr || ((isset($oAddr)) && !$oAddr->mailbox && !$oAddr->personal)) { + if(!$oAddr || ((isset($oAddr)) && !strlen($oAddr->mailbox) && !$oAddr->personal)) { $sEmail = $sGroup . ':;'; - } + } $aAddress[] = $this->createAddressObject($aStack,$aComment,$sEmail,$sGroup); $sGroup = ''; $aStack = $aComment = array(); @@ -423,8 +587,8 @@ class Rfc822Header { case ',': $aAddress[] = $this->createAddressObject($aStack,$aComment,$sEmail,$sGroup); break; - case ':': - $sGroup = trim(implode(' ',$aStack)); break; + case ':': + $sGroup = trim(implode(' ',$aStack)); $sGroup = preg_replace('/\s+/',' ',$sGroup); $aStack = array(); break; @@ -433,7 +597,7 @@ class Rfc822Header { break; case '>': /* skip */ - break; + break; default: $aStack[] = $sToken; break; } } @@ -453,8 +617,13 @@ class Rfc822Header { $aAddrBookAddress = $this->parseAddress($aAddr['email'],true); } else { $iPosAt = strpos($aAddr['email'], '@'); - $oAddr->mailbox = substr($aAddr['email'], 0, $iPosAt); - $oAddr->host = substr($aAddr['email'], $iPosAt+1); + if ($iPosAt === FALSE) { + $oAddr->mailbox = $aAddr['email']; + $oAddr->host = FALSE; + } else { + $oAddr->mailbox = substr($aAddr['email'], 0, $iPosAt); + $oAddr->host = substr($aAddr['email'], $iPosAt+1); + } if (isset($aAddr['name'])) { $oAddr->personal = $aAddr['name']; } else { @@ -463,26 +632,67 @@ class Rfc822Header { } } } - if (!$grouplookup && !$oAddr->mailbox) { + if (!$grouplookup && !strlen($oAddr->mailbox)) { $oAddr->mailbox = trim($sEmail); - if ($sHost && $oAddr->mailbox) { + if ($sHost && strlen($oAddr->mailbox)) { + $oAddr->host = $sHost; + } + } else if (!$grouplookup && !$oAddr->host) { + if ($sHost && strlen($oAddr->mailbox)) { $oAddr->host = $sHost; } } } - if (!$aAddrBookAddress && $oAddr->mailbox) { + if (!$aAddrBookAddress && strlen($oAddr->mailbox)) { $aProcessedAddress[] = $oAddr; } else { - $aProcessedAddress = array_merge($aProcessedAddress,$aAddrBookAddress); + $aProcessedAddress = array_merge($aProcessedAddress,$aAddrBookAddress); } } - if ($ar) { + if ($ar) { return $aProcessedAddress; } else { - return $aProcessedAddress[0]; + if (isset($aProcessedAddress[0])) + return $aProcessedAddress[0]; + else + return ''; } - } + } + /** + * Normalise the different Priority headers into a uniform value, + * namely that of the X-Priority header (1, 3, 5). Supports: + * Priority, X-Priority, Importance. + * X-MS-Mail-Priority is not parsed because it always coincides + * with one of the other headers. + * + * NOTE: this is actually a duplicate from the code in + * functions/imap_messages:parseFetch(). + * I'm not sure if it's ok here to call + * that function? + * @param string $sValue literal priority name + * @return integer + */ + function parsePriority($sValue) { + // don't use function call inside array_shift. + $aValue = preg_split('/\s/',trim($sValue)); + $value = strtolower(array_shift($aValue)); + + if ( is_numeric($value) ) { + return $value; + } + if ( $value == 'urgent' || $value == 'high' ) { + return 1; + } elseif ( $value == 'non-urgent' || $value == 'low' ) { + return 5; + } + // default is normal priority + return 3; + } + + /** + * @param string $value content type header + */ function parseContentType($value) { $pos = strpos($value, ';'); $props = ''; @@ -503,6 +713,52 @@ class Rfc822Header { $this->content_type = $content_type; } + /** + * RFC2184 + * @param array $aParameters + * @return array + */ + function processParameters($aParameters) { + $aResults = array(); + $aCharset = array(); + // handle multiline parameters + foreach($aParameters as $key => $value) { + if ($iPos = strpos($key,'*')) { + $sKey = substr($key,0,$iPos); + if (!isset($aResults[$sKey])) { + $aResults[$sKey] = $value; + if (substr($key,-1) == '*') { // parameter contains language/charset info + $aCharset[] = $sKey; + } + } else { + $aResults[$sKey] .= $value; + } + } else { + $aResults[$key] = $value; + } + } + foreach ($aCharset as $key) { + $value = $aResults[$key]; + // extract the charset & language + $charset = substr($value,0,strpos($value,"'")); + $value = substr($value,strlen($charset)+1); + $language = substr($value,0,strpos($value,"'")); + $value = substr($value,strlen($charset)+1); + /* FIXME: What's the status of charset decode with language information ???? + * Maybe language information contains only ascii text and charset_decode() + * only runs sm_encode_html_special_chars() on it. If it contains 8bit information, you + * get html encoded text in charset used by selected translation. + */ + $value = charset_decode($charset,$value); + $aResults[$key] = $value; + } + return $aResults; + } + + /** + * @param string $value + * @return array + */ function parseProperties($value) { $propArray = explode(';', $value); $propResultArray = array(); @@ -512,15 +768,19 @@ class Rfc822Header { if ($pos > 0) { $key = trim(substr($prop, 0, $pos)); $val = trim(substr($prop, $pos+1)); - if ($val{0} == '"') { + if (strlen($val) > 0 && $val{0} == '"') { $val = substr($val, 1, -1); } $propResultArray[$key] = $val; } } - return $propResultArray; + return $this->processParameters($propResultArray); } + /** + * Fills disposition object in rfc822Header object + * @param string $value + */ function parseDisposition($value) { $pos = strpos($value, ';'); $props = ''; @@ -536,6 +796,11 @@ class Rfc822Header { $this->disposition = $disp; } + /** + * Fills mlist array keys in rfc822Header object + * @param string $field + * @param string $value + */ function mlist($field, $value) { $res_a = array(); $value_a = explode(',', $value); @@ -553,19 +818,90 @@ class Rfc822Header { $this->mlist[$field] = $res_a; } - /* - * function to get the addres strings out of the header. - * Arguments: string or array of strings ! + /** + * Parses the X-Spam-Status or X-Spam-Score header + * @param string $value + */ + function parseSpamStatus($value) { + // Header value looks like this: + // No, score=1.5 required=5.0 tests=MSGID_FROM_MTA_ID,NO_REAL_NAME,UPPERCASE_25_50 autolearn=disabled version=3.1.0-gr0 + // Update circa 2018, this header can also be simply: + // No, score=1.5 + // So we make the rest of the line optional (there are likely other permutations, so + // each element is made optional except the first two... maybe even that's not flexible enough) + // + // Also now allow parsing of X-Spam-Score header, whose value is just a float + + $spam_status = array(); + + if (preg_match ('/^(?:(No|Yes),\s+score=)?(-?\d+\.\d+)(?:\s+required=(-?\d+\.\d+))?(?:\s+tests=(.*?))?(?:\s+autolearn=(.*?))?(?:\s+version=(.+?))?$/i', $value, $matches)) { + + // full header + $spam_status['bad_format'] = 0; + $spam_status['value'] = $matches[0]; + + // is_spam + if (!empty($matches[1])) { + if (strtolower($matches[1]) == 'yes') + $spam_status['is_spam'] = true; + else + $spam_status['is_spam'] = false; + } + + // score + if (!empty($matches[2])) + $spam_status['score'] = $matches[2]; + + // required + if (!empty($matches[3])) + $spam_status['required'] = $matches[3]; + + // tests + if (isset($matches[4])) { + $tests = array(); + $tests = explode(',', $matches[4]); + foreach ($tests as $test) { + $spam_status['tests'][] = trim($test); + } + } + + // autolearn + if (isset($matches[5])) + $spam_status['autolearn'] = $matches[5]; + + // version + if (isset($matches[6])) + $spam_status['version'] = $matches[6]; + + } else { + $spam_status['bad_format'] = 1; + $spam_status['value'] = $value; + } + return $spam_status; + } + + /** + * function to get the address strings out of the header. * example1: header->getAddr_s('to'). * example2: header->getAddr_s(array('to', 'cc', 'bcc')) + * @param mixed $arr string or array of strings + * @param string $separator + * @param boolean $encoded (since 1.4.0) return encoded or plain text addresses + * @param boolean $unconditionally_quote (since 1.4.21/1.5.2) When TRUE, always + * quote the personal part, + * whether or not it is + * encoded, otherwise quoting + * is only added if the + * personal part is not encoded + * @return string */ - function getAddr_s($arr, $separator = ',',$encoded=false) { + function getAddr_s($arr, $separator = ', ', $encoded=false, $unconditionally_quote=FALSE) { $s = ''; if (is_array($arr)) { foreach($arr as $arg) { - if ($this->getAddr_s($arg, $separator, $encoded)) { - $s .= $separator . $result; + if ($this->getAddr_s($arg, $separator, $encoded, $unconditionally_quote)) { + $s .= $separator; } } $s = ($s ? substr($s, 2) : $s); @@ -575,9 +911,9 @@ class Rfc822Header { foreach ($addr as $addr_o) { if (is_object($addr_o)) { if ($encoded) { - $s .= $addr_o->getEncodedAddress() . $separator; + $s .= $addr_o->getEncodedAddress($unconditionally_quote) . $separator; } else { - $s .= $addr_o->getAddress() . $separator; + $s .= $addr_o->getAddress(TRUE, FALSE, $unconditionally_quote) . $separator; } } } @@ -585,9 +921,9 @@ class Rfc822Header { } else { if (is_object($addr)) { if ($encoded) { - $s .= $addr->getEncodedAddress(); + $s .= $addr->getEncodedAddress($unconditionally_quote); } else { - $s .= $addr->getAddress(); + $s .= $addr->getAddress(TRUE, FALSE, $unconditionally_quote); } } } @@ -595,6 +931,13 @@ class Rfc822Header { return $s; } + /** + * function to get the array of addresses out of the header. + * @param mixed $arg string or array of strings + * @param array $excl_arr array of excluded email addresses + * @param array $arr array of added email addresses + * @return array + */ function getAddr_a($arg, $excl_arr = array(), $arr = array()) { if (is_array($arg)) { foreach($arg as $argument) { @@ -629,32 +972,80 @@ class Rfc822Header { } return $arr; } - + + /** +//FIXME: This needs some documentation (inside the function too)! Don't code w/out comments! + * Looking at the code years after it was written, + * this is my (Paul) best guess as to what this + * function does (note that docs previously claimed + * that this function returns boolean or an array, + * but it no longer appears to return an array - an + * integer instead): + * + * Inspects the TO and CC headers of the message + * represented by this object, looking for the + * address(es) given by $address + * + * If $address is a string: + * Serves as a test (returns boolean) as to + * whether or not the given address is found + * anywhere in the TO or CC headers + * + * If $address is an array: + * Looks through this list of addresses and + * returns the array index (an integer even + * if the array is given with keys of a + * different type) of the first matching + * $address found in this message's + * TO or CC headers, unless there is an exact + * match (meaning that the "personal + * information" in addition to the email + * address also matches), in which case that + * index (the first one found) is returned + * + * @param mixed $address Address(es) to search for in this + * message's TO and CC headers - please + * see above how the format of this + * argument affects the return value + * of this function + * @param boolean $recurs FOR INTERNAL USE ONLY + * + * @return mixed Boolean when $address is a scalar, + * indicating whether or not the address + * was found in the TO or CC headers. + * An integer when $address is an array, + * containing the index of the value in + * that array that was found in the TO + * or CC headers, or boolean FALSE if + * there were no matches at all + * + * @since 1.3.2 + */ function findAddress($address, $recurs = false) { $result = false; if (is_array($address)) { $i=0; foreach($address as $argument) { $match = $this->findAddress($argument, true); - $last = end($match); - if ($match[1]) { + if ($match[1]) { // this indicates when the personal information matched return $i; } else { - if (count($match[0]) && !$result) { + if (count($match[0]) && $result === FALSE) { $result = $i; } } - ++$i; + ++$i; } } else { if (!is_array($this->cc)) $this->cc = array(); + if (!is_array($this->to)) $this->to = array(); $srch_addr = $this->parseAddress($address); $results = array(); foreach ($this->to as $to) { - if ($to->host == $srch_addr->host) { - if ($to->mailbox == $srch_addr->mailbox) { + if (strtolower($to->host) == strtolower($srch_addr->host)) { + if (strtolower($to->mailbox) == strtolower($srch_addr->mailbox)) { $results[] = $srch_addr; - if ($to->personal == $srch_addr->personal) { + if (strtolower($to->personal) == strtolower($srch_addr->personal)) { if ($recurs) { return array($results, true); } else { @@ -664,11 +1055,11 @@ class Rfc822Header { } } } - foreach ($this->cc as $cc) { - if ($cc->host == $srch_addr->host) { - if ($cc->mailbox == $srch_addr->mailbox) { + foreach ($this->cc as $cc) { + if (strtolower($cc->host) == strtolower($srch_addr->host)) { + if (strtolower($cc->mailbox) == strtolower($srch_addr->mailbox)) { $results[] = $srch_addr; - if ($cc->personal == $srch_addr->personal) { + if (strtolower($cc->personal) == strtolower($srch_addr->personal)) { if ($recurs) { return array($results, true); } else { @@ -680,21 +1071,25 @@ class Rfc822Header { } if ($recurs) { return array($results, false); - } elseif (count($result)) { + } elseif (count($results)) { return true; } else { return false; - } + } } //exit; return $result; } + /** + * @param string $type0 media type + * @param string $type1 media subtype + * @return array media properties + * @todo check use of media type arguments + */ function getContentType($type0, $type1) { $type0 = $this->content_type->type0; $type1 = $this->content_type->type1; return $this->content_type->properties; } } - -?>