(PHP 5 >= 5.3.0, PHP 7, PECL intl >= 1.0.2, PHP 7, PECL idn >= 0.1)
idn_to_utf8 — Convert domain name from IDNA ASCII to Unicode.
过程化风格
$domain
[, int $options = 0
[, int $variant = INTL_IDNA_VARIANT_2003
[, array &$idna_info
]]] )This function converts a Unicode domain name from an IDNA ASCII-compatible format to plain Unicode, encoded in UTF-8.
domainDomain to convert in an IDNA ASCII-compatible format.
optionsConversion options - combination of IDNA_* constants (except IDNA_ERROR_* constants).
variant
Either INTL_IDNA_VARIANT_2003 for IDNA 2003 or
INTL_IDNA_VARIANT_UTS46 for UTS #46.
idna_info
This parameter can be used only if
INTL_IDNA_VARIANT_UTS46 was used for
variant. In that case, it will be filled with an
array with the keys 'result', the possibly illegal
result of the transformation,
'isTransitionalDifferent', a boolean indicating
whether the usage of the transitional mechanisms of UTS #46 either has
or would have changed the result and 'errors',
which is an int representing a bitset of the error
constants IDNA_ERROR_*.
The domain name in Unicode, encoded in UTF-8, 或者在失败时返回 FALSE
| 版本 | 说明 |
|---|---|
| 7.2.0 |
INTL_IDNA_VARIANT_2003 has been deprecated; use
INTL_IDNA_VARIANT_UTS46 instead.
|
| 5.4.0/PECL 2.0.0b1 |
Added the parameters variant and
idna_info; UTS #46 support.
|
Example #1 idn_to_utf8() example
<?php
echo idn_to_utf8('xn--tst-qla.de');
?>
以上例程会输出：
t?st.de
@kushik.com:
punyAdapt should be patched:
<?php
private static function punyAdapt($delta, $numpoints, $firsttime)
{
$delta = (int)($firsttime ? $delta / 700 : $delta / 2);
$delta += (int)($delta / $numpoints);
for($k = 0; $delta > 455; $k+=36)
$delta = (int)($delta / 35);
}
return $k + (int)((36) * $delta / ($delta + 38));
}
?>
<?php
// for those who has PHP older than version 5.3
class IDN {
// adapt bias for punycode algorithm
private static function punyAdapt(
$delta,
$numpoints,
$firsttime
) {
$delta = $firsttime ? $delta / 700 : $delta / 2;
$delta += $delta / $numpoints;
for ($k = 0; $delta > 455; $k += 36)
$delta = intval($delta / 35);
return $k + (36 * $delta) / ($delta + 38);
}
// translate character to punycode number
private static function decodeDigit($cp) {
$cp = strtolower($cp);
if ($cp >= 'a' && $cp <= 'z')
return ord($cp) - ord('a');
elseif ($cp >= '0' && $cp <= '9')
return ord($cp) - ord('0')+26;
}
// make utf8 string from unicode codepoint number
private static function utf8($cp) {
if ($cp < 128) return chr($cp);
if ($cp < 2048)
return chr(192+($cp >> 6)).chr(128+($cp & 63));
if ($cp < 65536) return
chr(224+($cp >> 12)).
chr(128+(($cp >> 6) & 63)).
chr(128+($cp & 63));
if ($cp < 2097152) return
chr(240+($cp >> 18)).
chr(128+(($cp >> 12) & 63)).
chr(128+(($cp >> 6) & 63)).
chr(128+($cp & 63));
// it should never get here
}
// main decoding function
private static function decodePart($input) {
if (substr($input,0,4) != "xn--") // prefix check...
return $input;
$input = substr($input,4); // discard prefix
$a = explode("-",$input);
if (count($a) > 1) {
$input = str_split(array_pop($a));
$output = str_split(implode("-",$a));
} else {
$output = array();
$input = str_split($input);
}
$n = 128; $i = 0; $bias = 72; // init punycode vars
while (!empty($input)) {
$oldi = $i;
$w = 1;
for ($k = 36;;$k += 36) {
$digit = IDN::decodeDigit(array_shift($input));
$i += $digit * $w;
if ($k <= $bias) $t = 1;
elseif ($k >= $bias + 26) $t = 26;
else $t = $k - $bias;
if ($digit < $t) break;
$w *= intval(36 - $t);
}
$bias = IDN::punyAdapt(
$i-$oldi,
count($output)+1,
$oldi == 0
);
$n += intval($i / (count($output) + 1));
$i %= count($output) + 1;
array_splice($output,$i,0,array(IDN::utf8($n)));
$i++;
}
return implode("",$output);
}
public static function decodeIDN($name) {
// split it, parse it and put it back together
return
implode(
".",
array_map("IDN::decodePart",explode(".",$name))
);
}
}
echo IDN::decodeIDN($_SERVER['HTTP_HOST']);
?>