Move Encoding class away from the Language namespace

Language\Encoding::__mbMimeEncode -> Convert\MimeEncode::__mbMimeEncode
Langauge\Encoding::checkConvertEncoding -> Check\Encoding::checkConvertEncoding
Langauge\Encoding::setErrorChar -> Check\Encoding::setErrorChar
Langauge\Encoding::getErrorChar -> Encoding::getErrorChar
Langauge\Encoding::convertEncoding -> Convert\Encoding::convertEncoding

Also fixed encoding check that not only a code point but a string can
also be used as a parameter.

Update phpunit tests and split them out for each class

Normal test page is still combined for all classes but updated to
correctly use each class
This commit is contained in:
Clemens Schwaighofer
2022-04-13 09:25:42 +09:00
parent a3c49e408a
commit 6f4c5e36e6
12 changed files with 618 additions and 401 deletions

View File

@@ -0,0 +1,60 @@
<?php
/*
* check if string is valid in target encoding
*/
declare(strict_types=1);
namespace CoreLibs\Convert;
class Encoding
{
/**
* detects the source encoding of the string and if doesn't match
* to the given target encoding it convert is
* if source encoding is set and auto check is true (default) a second
* check is done so that the source string encoding actually matches
* will be skipped if source encoding detection is ascii
*
* @param string $string string to convert
* @param string $to_encoding target encoding
* @param string $source_encoding optional source encoding, will try to auto detect
* @param bool $auto_check default true, if source encoding is set
* check that the source is actually matching
* to what we sav the source is
* @return string encoding converted string
*/
public static function convertEncoding(
string $string,
string $to_encoding,
string $source_encoding = '',
bool $auto_check = true
): string {
// set if not given
if (!$source_encoding) {
$source_encoding = mb_detect_encoding($string);
} else {
$_source_encoding = mb_detect_encoding($string);
}
if (
$auto_check === true &&
isset($_source_encoding) &&
$_source_encoding == $source_encoding
) {
// trigger check if we have override source encoding.
// if different (_source is all but not ascii) then trigger
// skip if matching
}
if ($source_encoding != $to_encoding) {
if ($source_encoding) {
$string = mb_convert_encoding($string, $to_encoding, $source_encoding);
} else {
$string = mb_convert_encoding($string, $to_encoding);
}
}
return $string;
}
}
// __END__

View File

@@ -0,0 +1,67 @@
<?php
/*
* alternate for header mime encode to void problems with long strings and
* spaces/strange encoding problems.
* Orignal issues during PHP5/7
*/
declare(strict_types=1);
namespace CoreLibs\Convert;
class MimeEncode
{
/**
* wrapper function for mb mime convert
* for correct conversion with long strings
*
* @param string $string string to encode
* @param string $encoding target encoding
* @param string $line_break default line break is \r\n
* @return string encoded string
*/
public static function __mbMimeEncode(
string $string,
string $encoding,
string $line_break = "\r\n"
): string {
// set internal encoding, so the mimeheader encode works correctly
mb_internal_encoding($encoding);
// if a subject, make a work around for the broken mb_mimencode
$pos = 0;
// after 36 single bytes characters,
// if then comes MB, it is broken
// has to 2 x 36 < 74 so the mb_encode_mimeheader
// 74 hardcoded split does not get triggered
$split = 36;
$_string = '';
while ($pos < mb_strlen($string, $encoding)) {
$output = mb_strimwidth($string, $pos, $split, "", $encoding);
$pos += mb_strlen($output, $encoding);
// if the strinlen is 0 here, get out of the loop
if (!mb_strlen($output, $encoding)) {
$pos += mb_strlen($string, $encoding);
}
$_string_encoded = mb_encode_mimeheader($output, $encoding);
// only make linebreaks if we have mime encoded code inside
// the space only belongs in the second line
if ($_string && preg_match("/^=\?/", $_string_encoded)) {
$_string .= $line_break . " ";
} elseif (
// hack for plain text with space at the end
mb_strlen($output, $encoding) == $split &&
mb_substr($output, -1, 1, $encoding) == " "
) {
// if output ends with space, add one more
$_string_encoded .= " ";
}
$_string .= $_string_encoded;
}
// strip out any spaces BEFORE a line break
$string = str_replace(" " . $line_break, $line_break, $_string);
return $string;
}
}
// __END__