(PHP 5 >= 5.3.0)
quoted_printable_encode — Convert a 8 bit string to a quoted-printable string
$str
)Returns a quoted printable string created according to » RFC2045, section 6.7.
This function is similar to imap_8bit(), except this one does not require the IMAP module to work.
str
The input string.
Returns the encoded string.
ericth at NOSPAM dot pennyworth dot com (2011-10-07 13:46:14)
I have re-written the PHP 5.3.8 function for quoted_printable_encode into PHP for use with PHP < 5.3. Tested with PHP 5.2.11.
<?php
define('PHP_QPRINT_MAXL', 75);
function php_quot_print_encode($str)
{
$lp = 0;
$ret = '';
$hex = "0123456789ABCDEF";
$length = strlen($str);
$str_index = 0;
while ($length--) {
if ((($c = $str[$str_index++]) == "\015") && ($str[$str_index] == "\012") && $length > 0) {
$ret .= "\015";
$ret .= $str[$str_index++];
$length--;
$lp = 0;
} else {
if (ctype_cntrl($c)
|| (ord($c) == 0x7f)
|| (ord($c) & 0x80)
|| ($c == '=')
|| (($c == ' ') && ($str[$str_index] == "\015")))
{
if (($lp += 3) > PHP_QPRINT_MAXL)
{
$ret .= '=';
$ret .= "\015";
$ret .= "\012";
$lp = 3;
}
$ret .= '=';
$ret .= $hex[ord($c) >> 4];
$ret .= $hex[ord($c) & 0xf];
}
else
{
if ((++$lp) > PHP_QPRINT_MAXL)
{
$ret .= '=';
$ret .= "\015";
$ret .= "\012";
$lp = 1;
}
$ret .= $c;
}
}
}
return $ret;
}
?>
arnaudv6 (2011-08-15 08:56:21)
One will like to know and clearly read that RFC2045 specifies a line shall not exceed 75 characters.
Accordingly, quoted_printable_encode() splits line at this limit.
senacle (2011-01-28 06:24:02)
If you need to encode the subject of your mail message, use as follow :
<?php
quoted_printable_encode($subject, 58, "subject");
function quoted_printable_encode($input, $line_max = 75, $subject) {
/** The Quoted Printable encodes only with 75 characters per ligne.
* For encoding the subject of message, we must split the subject every 58 characters,
* because of adding =?iso-8859-1?Q? at the beginning and ?= at the end of each split.
* @param string $input string to encode
* @param int $line_max max number of char per ligne
* @param string $subject specify if we encode a subject or any other string
* @access public
*/
$hex = array('0','1','2','3','4','5','6','7',
'8','9','A','B','C','D','E','F');
$lines = preg_split("/(?:\r\n|\r|\n)/", $input);
$linebreak = "\r\n";
/* the linebreak also counts as characters in the mime_qp_long_line
* rule of spam-assassin */
$line_max = $line_max - strlen($linebreak);
$escape = "=";
$output = "";
$cur_conv_line = "";
$length = 0;
$whitespace_pos = 0;
$addtl_chars = 0;
// iterate lines
for ($j=0; $j<count($lines); $j++) {
$line = $lines[$j];
$linlen = strlen($line);
// iterate chars
for ($i = 0; $i < $linlen; $i++) {
$c = substr($line, $i, 1);
$dec = ord($c);
$length++;
if ($dec == 32) {
// space occurring at end of line, need to encode
if (($i == ($linlen - 1))) {
$c = "=20";
$length += 2;
}
$addtl_chars = 0;
$whitespace_pos = $i;
} else if ( ($dec == 61) || ($dec < 32 ) || ($dec > 126) ) {
$h2 = floor($dec/16); $h1 = floor($dec%16);
$c = $escape . $hex["$h2"] . $hex["$h1"];
$length += 2;
$addtl_chars += 2;
}
// length for wordwrap exceeded, get a newline into the text
if ($length >= $line_max) {
$cur_conv_line .= $c;
// read only up to the whitespace for the current line
$whitesp_diff = $i - $whitespace_pos + $addtl_chars;
/* the text after the whitespace will have to be read
* again ( + any additional characters that came into
* existence as a result of the encoding process after the whitespace)
*
* Also, do not start at 0, if there was *no* whitespace in
* the whole line */
if (($i + $addtl_chars) > $whitesp_diff) {
if ($subject == "subject") {
$output .= "=?ISO-8859-1?Q?".substr($cur_conv_line, 0,
(strlen($cur_conv_line) - $whitesp_diff))."?=";
} else {
$output .= substr($cur_conv_line, 0,
(strlen($cur_conv_line) - $whitesp_diff)).$linebreak;
}
$i = $i - $whitesp_diff + $addtl_chars;
} else {
/* emit continuation --mirabilos */
if ($subject == "subject") {
$output .= "=?ISO-8859-1?Q?".$cur_conv_line."?=";
} else {
$output .= $cur_conv_line. '=' . $linebreak;
}
}
$cur_conv_line = "";
$length = 0;
$whitespace_pos = 0;
} else {
// length for wordwrap not reached, continue reading
$cur_conv_line .= $c;
}
} // end of for
$length = 0;
$whitespace_pos = 0;
if ($subject == "subject") {
$output .= "=?ISO-8859-1?Q?".$cur_conv_line."?=";
} else {
$output .= $cur_conv_line;
if ($j<=count($lines)-1) {
$output .= $linebreak;
}
}
$cur_conv_line = "";
} // end for
return trim($output);
} // end quoted_printable_encode
?>
Thorsten Glaser (2010-09-24 06:05:29)
Two bugs:
1) your linebreak is wrong
$linebreak = "\r\n";
2) continuation of lines with no whitespace is broken
/*
* the text after the whitespace will have to
* be read again ( + any additional characters
* that came into existence as a result of the
* encoding process after the whitespace)
*
* Also, do not start at 0, if there was *no*
* whitespace in the whole line
*/
if (($i + $addtl_chars) > $whitesp_diff) {
$output .= substr($cur_conv_line, 0,
(strlen($cur_conv_line) - $whitesp_diff)) .
$linebreak;
$i = $i - $whitesp_diff + $addtl_chars;
} else {
/* emit continuation --mirabilos */
$output .= $cur_conv_line .
'=' . $linebreak;
}
tzangerl [dot] pdc {dot} kth dot se (2010-04-09 02:05:20)
A function that QP-encodes an input string (written for PHP < 5.3) and
wordwraps it at the same time, in order to avoid classification according to the MIME QP LONG LINE rule of SpamAssassin. Thanks for Matt Jeffers to point out errors in the below quoted_printable script!
<?php
function quoted_printable_encode($input, $line_max = 75) {
$hex = array('0','1','2','3','4','5','6','7',
'8','9','A','B','C','D','E','F');
$lines = preg_split("/(?:\r\n|\r|\n)/", $input);
$linebreak = "=0D=0A=\r\n";
/* the linebreak also counts as characters in the mime_qp_long_line
* rule of spam-assassin */
$line_max = $line_max - strlen($linebreak);
$escape = "=";
$output = "";
$cur_conv_line = "";
$length = 0;
$whitespace_pos = 0;
$addtl_chars = 0;
// iterate lines
for ($j=0; $j<count($lines); $j++) {
$line = $lines[$j];
$linlen = strlen($line);
// iterate chars
for ($i = 0; $i < $linlen; $i++) {
$c = substr($line, $i, 1);
$dec = ord($c);
$length++;
if ($dec == 32) {
// space occurring at end of line, need to encode
if (($i == ($linlen - 1))) {
$c = "=20";
$length += 2;
}
$addtl_chars = 0;
$whitespace_pos = $i;
} elseif ( ($dec == 61) || ($dec < 32 ) || ($dec > 126) ) {
$h2 = floor($dec/16); $h1 = floor($dec%16);
$c = $escape . $hex["$h2"] . $hex["$h1"];
$length += 2;
$addtl_chars += 2;
}
// length for wordwrap exceeded, get a newline into the text
if ($length >= $line_max) {
$cur_conv_line .= $c;
// read only up to the whitespace for the current line
$whitesp_diff = $i - $whitespace_pos + $addtl_chars;
/* the text after the whitespace will have to be read
* again ( + any additional characters that came into
* existence as a result of the encoding process after the whitespace)
*
* Also, do not start at 0, if there was *no* whitespace in
* the whole line */
if (($i + $addtl_chars) > $whitesp_diff) {
$output .= substr($cur_conv_line, 0, (strlen($cur_conv_line) -
$whitesp_diff)) . $linebreak;
$i = $i - $whitesp_diff + $addtl_chars;
} else {
$output .= $cur_conv_line . $linebreak;
}
$cur_conv_line = "";
$length = 0;
$whitespace_pos = 0;
} else {
// length for wordwrap not reached, continue reading
$cur_conv_line .= $c;
}
} // end of for
$length = 0;
$whitespace_pos = 0;
$output .= $cur_conv_line;
$cur_conv_line = "";
if ($j<=count($lines)-1) {
$output .= $linebreak;
}
} // end for
return trim($output);
} // end quoted_printable_encode
?>