parse_http_host.php

<?php

// Parse HTTP HOST header hostname and port according to http://tools.ietf.org/html/rfc3986.html

$h16 = '[[:xdigit:]]{1,4}';
$subDelims = "[!$&'()*+,;=]";
$pctEncoded = '%[[:xdigit]][[:xdigit]]';
$unreserved = '[-._~[:alpha:][:digit:]]';
$decOctet = <<<EOS
  (?:
      [0-9]               # 0-9
    | (?: [1-9][0-9] )    # 10-99
    | (?: 1 [0-9][0-9] )  # 100-199
    | (?: 2 [0-4][0-9] )  # 200-249
    | (?: 25 [0-5] )      # 250-255
  )
EOS;

$regName = <<<EOS
  (?:
      $unreserved
    | $pctEncoded
    | $subDelims
  )*
EOS;

$ipv4Address = <<<EOS
(?:
  $decOctet \. $decOctet \. $decOctet \. $decOctet
)
EOS;

$ls32 = <<<EOS
  (?:
    (?: $h16 : $h16) | $ipv4Address
  )
EOS;

$h16c = "(?: $h16 : )";
$ipv6Address = <<<EOS
  (?:
      (?:                           $h16c{6} $ls32 )
    | (?:                        :: $h16c{5} $ls32 )
    | (?: (?:            $h16 )? :: $h16c{4} $ls32 )
    | (?: (?: $h16c{0,1} $h16 )? :: $h16c{3} $ls32 )
    | (?: (?: $h16c{0,2} $h16 )? :: $h16c{2} $ls32 )
    | (?: (?: $h16c{0,3} $h16 )? :: $h16c    $ls32 )
    | (?: (?: $h16c{0,4} $h16 )? ::          $ls32 )
    | (?: (?: $h16c{0,5} $h16 )? ::          $h16 )
    | (?: (?: $h16c{0,6} $h16 )? :: )
  )
EOS;

$ipvFuture = <<<EOS
  (?:
    v [[:xdigit:]]+ \. (?: $unreserved | $subDelims | : )+
  )
EOS;

$ipLiteral = <<<EOS
  (?: \[ (?: $ipv6Address | $ipvFuture )  \] )
EOS;


$hostRule = <<<EOS
/
  \A
  (?P<host>
      $ipLiteral
    | $ipv4Address
    | $regName
  )
  (?:
    : (?P<port> [0-9]*)
  )?
  \Z
/x
EOS;

$testUrls = array(
  'hello.com',
  'hello.com:90',
  '127.0.0.1',
  '127.0.0.1:90',
  '[::1]:90',
  '[2607:f0d0:1002:0051:0000:0000:0000:0004]:90',
  '[::ffff:192.0.2.128]:90',
  '',
  ':90',
);

echo $host."\n\n";

foreach ($testUrls as $t) {
  if (preg_match($hostRule, $t, $match)) {
    $host = $match['host'];
    $port = isset($match['port']) ? $match['port'] : '-';
    echo $t . '  ==>  ' . $host . ' : ' . $port . "\n";
  } else {
    echo '!! ' . $t . "\n";
  }
}