Net::IPv6Addr - Check and manipulate IPv6 addresses
This documents version 1.02 of Net::IPv6Addr corresponding to git commit f9065cb7b044da442df16443d65593a5a3fc6baa released on Wed Mar 31 11:11:47 2021 +0900.
use Net::IPv6Addr; my $addr = "dead:beef:cafe:babe::f0ad"; Net::IPv6Addr::ipv6_parse($addr); my $x = Net::IPv6Addr->new($addr); print $x->to_string_preferred(), "\n";
produces output
dead:beef:cafe:babe:0:0:0:f0ad
(This example is included as synopsis.pl in the distribution.)
Net::IPv6Addr checks whether strings contain valid IPv6 addresses, and converts IPv6 addresses into various formats.
Net::IPv6Addr
All of "new", "is_ipv6", and "ipv6_parse" can process the following formats:
2001:db8:0:0:0:ff00:42:8329
This is the form described as the "preferred form" in section 2.2 of "RFC1884" et al. Output with "to_string_preferred".
2001:db8::ff00:42:8329
This is the "canonical text representation format" of "RFC5952". Output with "to_string_compressed".
2001:db8:0:0:0:ff00:0.66.131.41
Output with "to_string_ipv4".
2001:db8::ff00:0.66.131.41
Output with "to_string_ipv4_compressed".
An IPv6 can be changed to a Math::BigInt object or a digit string using "to_bigint". Big integers can also be input with "from_bigint".
9R}vSQ9RqiCvG6zn?Zyh
This encoding was given in "RFC1924" as an April Fool's joke. Output with "to_string_base85".
In addition, the following formats can be output:
An IPv6 can be processed into its component pieces with "to_array" or "to_intarray".
An IPv6 can be processed into its reverse-address pointer, as defined by "RFC1886", using "to_string_ip6_int".
The methods and functions are listed in alphabetical order. All except "new" serve as both object methods and standalone functions.
use Net::IPv6Addr 'from_bigint'; print from_bigint ('12345678901234567890')->to_string_compressed ();
::ab54:a98c:eb1f:ad2
(This example is included as from-bigint.pl in the distribution.)
Given a string or a Math::BigInt object containing a number, this converts it into a Net::IPv6Addr object.
A string or a Math::BigInt object. If the input is a scalar, it's converted into a Math::BigInt object.
A Net::IPv6Addr object
Invalid input will generate an exception.
This function was added in "0.95".
use Net::IPv6Addr; my $obj = Net::IPv6Addr->new ('dead:beef:cafe:babe:dead:beef:cafe:babe'); if ($obj->in_network ('dead:beef:ca0::/21')) { print $obj->to_string_compressed, " is in network.\n"; }
dead:beef:cafe:babe:dead:beef:cafe:babe is in network.
(This example is included as inet.pl in the distribution.)
If used as an object method, a network and its size in bits
my $ok = $x->in_network ("aa:bb:cc:dd::", 64);
If used as a subroutine, an IPv6 address string in any format, followed by a network address string and its size in bits.
my $addr = 'fd00::54:20c:29fe:fe14:ab4b'; my $ok = Net::IPv6Addr::in_network ($addr, "aa:bb:cc:dd::", 64);
The network size may also be given with the / notation after the network address string:
my $ok = $x->in_network("aa:bb:cc:dd::/64");
A true value if the address $x is a member of the network given as the argument, or false otherwise.
$x
Prior to version "0.9", this did not work correctly unless the net size was a multiple of sixteen.
use Net::IPv6Addr 'in_network_of_size'; my $obj = in_network_of_size ('dead:beef:cafe:babe:dead:beef:cafe:babe', 42); print $obj->to_string_compressed ();
dead:beef:cac0::
(This example is included as inos.pl in the distribution.)
Given an input IPv6 address $x, this returns the $n most-significant bits of $x as a new Net::IPv6Addr object.
$n
If used as an object method, network size in bits:
my $obj = $x->in_network_of_size (64);
If used as a subroutine, an IPv6 address string in any format and a network size in bits:
my $obj = in_network_of_size ($addr, 64);
Network size may also be given with / notation:
/
my $obj = in_network_of_size ("$addr/64");
The $n most-significant bits of $x as a new Net::IPv6Addr object.
my $niok = ipv6_chkip ('dead:beef:cafe:babe::f0ad');
An IPv6 address string, without a prefix.
A true value (a code reference for the parser for this IP) if it's a valid address; a false value (undef) if not.
undef
my ($ni, $pl) = ipv6_parse ('dead:beef:cafe:babe::f0ad');
Either a string containing an IPv6 address string, which may also include a / character and a numeric prefix length,
my ($x, $y) = ipv6_parse ("a::/24");
or an IPv6 address string, with an optional second argument consisting of a numeric prefix length:
my ($x, $y) = ipv6_parse('a::', '24');
Called in array context, the return value is a list consisting of the address string and the prefix, if it parses correctly. Called in scalar context, the address and prefix are concatenated with "/".
Throws an exception on malformed input.
my $niok = is_ipv6 ('dead:beef:cafe:babe::f0ad');
Identical to "ipv6_parse".
This returns the return value of "ipv6_parse", called in scalar context, if it does parse out correctly, otherwise it returns undef. Unlike "ipv6_parse", is_ipv6 does not throw exceptions.
is_ipv6
my $ni = Net::IPv6Addr->new ('dead:beef:cafe:babe::f0ad');
Create a new Net::IPv6Addr object from a string. Internally, the object is a blessed array reference containing the eight parts of the address as integers.
A string to be interpreted as an IPv6 address.
A Net::IPv6Addr object if successful.
Throws an exception if the string isn't a valid address.
use Net::IPv6Addr 'to_array'; my @int = to_array ('dead::beef'); my $ipobj = Net::IPv6Addr->new ('dead::beef'); my @int2 = $ipobj->to_array (); print "@int\n@int2\n";
dead 0000 0000 0000 0000 0000 0000 beef dead 0000 0000 0000 0000 0000 0000 beef
(This example is included as array.pl in the distribution.)
Convert an IPv6 address into an array of eight hexadecimal numbers.
If used as an object method, none; if used as a subroutine, an IPv6 address string in any format.
An array [0..7] of 16-bit hexadecimal numbers (strings).
See also "to_intarray" and "to_bigint".
use Net::IPv6Addr 'to_bigint'; my $int = to_bigint ('dead::beef'); my $ipobj = Net::IPv6Addr->new ('dead::beef'); my $int2 = $ipobj->to_bigint (); print "$int\n$int2\n";
295986882420777848964380943247191621359 295986882420777848964380943247191621359
(This example is included as bigint.pl in the distribution.)
Convert an IPv6 address into a Math::BigInt object containing the IP address as a single number.
The BigInt representation of the given IPv6 address.
See also "from_bigint", "to_intarray" and "to_array".
Convert an IPv6 address into an array of eight integer numbers.
An array [0..7] of numbers.
See also "to_array" and "to_bigint".
The IPv6 address in the style detailed by "RFC1924".
The base 85 encoding described in "RFC1924" was an April Fool's joke.
use Net::IPv6Addr 'to_string_compressed'; print to_string_compressed ('dead:beef:0000:0000:0000:0000:cafe:babe');
dead:beef::cafe:babe
(This example is included as compressed.pl in the distribution.)
This provides the "canonical text representation format" of "RFC5952".
The IPv6 address in the "compressed" ("RFC1884" et al.) or "canonical" ("RFC5952") format. Hexadecimal numbers are reduced to lower case, consecutive zero elements are reduced to double colons, and leading zeros are removed from strings of hexadecimal digits. All treatment of ambiguities is as per RFC5952. (See t/rfc5952.t for tests.)
use Net::IPv6Addr 'to_string_ip6_int'; my $s = to_string_ip6_int ('dead::beef'); my $ipobj = Net::IPv6Addr->new ('dead::beef'); my $s2 = $ipobj->to_string_ip6_int (); print "$s\n$s2\n";
f.e.e.b.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.d.a.e.d.IP6.INT. f.e.e.b.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.d.a.e.d.IP6.INT.
(This example is included as string-ip6-int.pl in the distribution.)
The reverse-address pointer as defined by "RFC1886".
The reverse process of converting these into Net::IPv6Addr objects is not supported.
use Net::IPv6Addr ':all'; print to_string_ipv4_compressed ('dead:beef:0:3:2:1:cafe:babe');
dead:beef::3:2:1:202.254.186.190
(This example is included as to-string-ipv4.pl in the distribution.)
The IPv6 address in the IPv4 format detailed by "RFC1884" et al.
When used as a subroutine, invalid input will generate an exception.
From version "0.95", this allows any IPv6 address to be produced, not just the restricted forms allowed previously.
(This example is included as to-string-ipv4-comp.pl in the distribution.)
The IPv6 address in the compressed IPv4 format detailed by "RFC1884" et al.
use Net::IPv6Addr 'to_string_preferred'; print to_string_preferred ('dead:beef:cafe:babe::f0ad');
(This example is included as preferred.pl in the distribution.)
The IPv6 address, formatted in the "preferred" way (as detailed by "RFC1884" et al).
As of version 1.02, "from_bigint", "in_network", "in_network_of_size", "ipv6_chkip", "ipv6_parse", "is_ipv6", "to_array", "to_bigint", "to_intarray", "to_string_base85", "to_string_compressed", "to_string_ip6_int", "to_string_ipv4", "to_string_ipv4_compressed", "to_string_preferred" may be exported on demand. All the exported functions may be exported using
use Net::IPv6Addr ':all';
This is used by "to_bigint" and "from_bigint".
Search grep.cpan.me for uses of this module
The following RFCs (requests for comment, internet standards documentation) contain information on IPv6.
These are all the same standard, with updates. The most recent one is the active one.
IPv6 Addressing Architecture - December 1995
IP Version 6 Addressing Architecture - July 1998
Internet Protocol Version 6 (IPv6) Addressing Architecture - April 2003
IP Version 6 Addressing Architecture - February 2006
DNS Extensions to support IP version 6 - December 1995
A Compact Representation of IPv6 Addresses - 1 April 1996
This was an April Fool's joke.
A Recommendation for IPv6 Address Text Representation - August 2010
This contains a "recommendation for a canonical text representation format of IPv6 addresses" which corresponds to the output of "to_string_compressed" in this module.
The links go to the plain text online versions of the RFCs.
There are a very large number of CPAN modules which deal with IPv6 addresses. The following list gives all the ones I know about which overlap with this module, in alphabetical order.
This module uses Socket to validate IP addresses. It offers a number of facilities for special-purpose sub networks, like is_discard_ipv6, which are not offered in Net::IPv6Addr.
is_discard_ipv6
Its description says "A pure Perl IPv6 address manipulation library. Emphasis on manipulation of prefixes and addresses."
It insists on having a prefix with the IP address, so
my $ipv6 = IPv6::Address->new ('2001:0:0:1:0:0:0:1');
actually fails, you have to use
my $ipv6 = IPv6::Address->new ('2001:0:0:1:0:0:0:1/64');
Features binary IPs (strings like '101001'), etc.
It's a simplified version of "Net::IP".
It's a "Version-agnostic representation of an IP address". I have not tried this module.
This module is broken and strongly not recommended.
These are two things in the same distribution. The documentation is quite offputting, but there are a lot of users of the module and stars on metacpan.
This module consists of a regex for validating IPv6s. Because this module had a lot more and better tests than Net::IPv6Addr, I included the tests and one regex from Regexp::IPv6 in this module. (See t/Regexp-IPv6.t) Unlike Net::IPv6Addr, Regexp::IPv6 disallows ::, "the unspecified addresses". See the module's documentation for details.
Regexp::IPv6
::
https://www.helpsystems.com/intermapper/ipv6-test-address-validation
This module was originally written by Tony Monroe in 2001 to simplify the task of maintaining DNS records after he set himself up with Freenet6.
In 2017 the module was adopted by Ben Bullock with the help of Neil Bowers as part of "CPAN day". Significant changes to the module since then include the following:
Net::IPv4Addr dependence removed. This module suffered from the "octal bug". This module had only been used to validate ipv4 addresses and was easy to remove.
Math::Base85 dependence removed. This module is only needed to support the April Fool's joke method of IPv6 addresses, so if the user doesn't already have Math::Base85, the module now disables support for the April Fool addresses.
Checking of base 85 addresses and prefixes was made stricter in response to user complaints.
The "from_bigint" method was added and the documentation updated to reflect the current internet standards.
The restriction on mixed address inputs removed in "0.92" was also removed in the output routines "to_string_ipv4" and "to_string_ipv4_compressed".
The valid format consisting of a compressed-but-non-zero six-element IPv6 followed by an IPv4, such as fe80::204:61ff:254.157.241.86, is accepted by the module.
fe80::204:61ff:254.157.241.86
"in_network" and "in_network_of_size" were fixed to allow more kinds of previxes.
Exporting of some functions was added. Prior to this, everything had to be done fully-qualified, as in Net::IPv6Addr::to_string_compressed.
Net::IPv6Addr::to_string_compressed
Tony Monroe(*)
The module's interface resembles Net::IPv4Addr by Francis J. Lacoste <francis dot lacoste at iNsu dot COM>.
Some fixes and subroutines from Jyrki Soini <jyrki dot soini at sonera dot com>.
(*) The current module maintainer (BKB) does not have any contact information for Tony Monroe. Those wishing to contact him can do so via Neil Bowers (see his CPAN user page for contact details).
This distribution is copyright (c) 2001-2002 Tony Monroe. All rights reserved. This software is distributed under the same license terms as Perl itself. This software comes with NO WARRANTIES WHATSOEVER, express, implied, or otherwise.
To install Net::IPv6Addr, copy and paste the appropriate command in to your terminal.
cpanm
cpanm Net::IPv6Addr
CPAN shell
perl -MCPAN -e shell install Net::IPv6Addr
For more information on module installation, please visit the detailed CPAN module installation guide.