Net::IP - Perl extension for manipulating IPv4/IPv6 addresses

  use Net::IP;

  my $ip = new Net::IP ('193.0.1/24') or die (Net::IP::Error());
  print ("IP  : ".$ip->ip()."\n");
  print ("Sho : ".$ip->short()."\n");
  print ("Bin : ".$ip->binip()."\n");
  print ("Int : ".$ip->intip()."\n");
  print ("Mask: ".$ip->mask()."\n");
  print ("Last: ".$ip->last_ip()."\n");
  print ("Len : ".$ip->prefixlen()."\n");
  print ("Size: ".$ip->size()."\n");
  print ("Type: ".$ip->iptype()."\n");
  print ("Rev:  ".$ip->reverse_ip()."\n");

This module provides functions to deal with IPv4/IPv6 addresses. The module can be used as a class, allowing the user to instantiate IP objects, which can be single IP addresses, prefixes, or ranges of addresses. There is also a procedural way of accessing most of the functions. Most subroutines can take either IPv4 or IPv6 addresses transparently.

A Net::IP object can be created from a single IP address:

  $ip = new Net::IP ('193.0.1.46') || die ...

Or from a Classless Prefix (a /24 prefix is equivalent to a C class):

  $ip = new Net::IP ('195.114.80/24') || die ...

Or from a range of addresses:

  $ip = new Net::IP ('20.34.101.207 - 201.3.9.99') || die ...

Or from a address plus a number:

  $ip = new Net::IP ('20.34.10.0 + 255') || die ...

The new() function accepts IPv4 and IPv6 addresses:

  $ip = new Net::IP ('dead:beef::/32') || die ...

Optionnaly, the function can be passed the version of the IP. Otherwise, it tries to guess what the version is (see BI_is_ipv4() and BI_is_ipv6()).

  $ip = new Net::IP ('195/8',4); # Class A

Most of these methods are front-ends for the real functions, which use a procedural interface. Most functions return undef on failure, and a true value on success. A detailed description of the procedural interface is provided below.

Set an IP address in an existing IP object. This method has the same functionality as the new() method, except that it reuses an existing object to store the new IP.

CW$ip->set('130.23.1/24',4);

Like new(), set() takes two arguments - a string used to build an IP address, prefix, or range, and optionally, the IP version of the considered address.

It returns an IP object on success, and undef on failure.

Return the current object error string. The error string is set whenever one of the methods produces an error. Also, a global, class-wide BIError() function is avaliable.

CWwarn ($ip->error());

Return the current object error number. The error number is set whenever one of the methods produces an error. Also, a global $ERRNO variable is set when an error is produced.

CWwarn ($ip->errno());

Return the IP address (or first IP of the prefix or range) in quad format, as a string.

CWprint ($ip->ip());

Return the IP address as a binary string of 0s and 1s.

CWprint ($ip->binip());

Return the length in bits of the current prefix.

CWprint ($ip->prefixlen());

Return the version of the current IP object (4 or 6).

CWprint ($ip->version());

Return the number of IP addresses in the current prefix or range. Use of this function requires Math::BigInt.

CWprint ($ip->size());

Return the binary mask of the current prefix, if applicable.

CWprint ($ip->binmask());

Return the mask in quad format of the current prefix.

CWprint ($ip->mask());

Return the full prefix (ip+prefix length) in quad (standard) format.

CWprint ($ip->prefix());

Print the IP object (IP/Prefix or First - Last)

CWprint ($ip->print());

Convert the IP in integer format and return it as a Math::BigInt object.

CWprint ($ip->intip());

Return the IP in short format: IPv4 addresses: 194.5/16 IPv6 addresses: ab32:f000::

CWprint ($ip->short());

Return the IP Type - this describes the type of an IP (Public, Private, Reserved, etc.)

CWprint ($ip->iptype());

Return the reverse IP for a given IP address (in.addr. format).

CWprint ($ip->reserve_ip());

Return the last IP of a prefix/range in quad format.

CWprint ($ip->last_ip());

Return the last IP of a prefix/range in binary format.

CWprint ($ip->last_bin());

Return the last IP of a prefix/range in integer format.

CWprint ($ip->last_int());

This function finds all the prefixes that can be found between the two addresses of a range. The function returns a list of prefixes.

CW@list = $ip->find_prefixes($other_ip));

Binary comparaison of two IP objects. The function takes an operation and an IP object as arguments. It returns a boolean value.

The operation can be one of: lt: less than (smaller than) le: smaller or equal to gt: greater than ge: greater or equal to

CWif ($ip->bincomp('lt',$ip2) {...}

Binary addition of two IP objects. The value returned is an IP object.

CWmy $sum = $ip->binadd($ip2);

Aggregate 2 IPs - Append one range/prefix of IPs to another. The last address of the first range must be the one immediately preceding the first address of the second range. A new IP object is returned.

CWmy $total = $ip->aggregate($ip2);

Check if two IP ranges/prefixes overlap each other. The value returned by the function should be one of: CW$IP_PARTIAL_OVERLAP (ranges overlap) CW$IP_NO_OVERLAP (no overlap) CW$IP_A_IN_B_OVERLAP (range2 contains range1) CW$IP_B_IN_A_OVERLAP (range1 contains range2) CW$IP_IDENTICAL (ranges are identical) undef (problem)

CWif ($ip->overlaps($ip2)==$IP_A_IN_B_OVERLAP) {...};

The CW+ operator is overloaded in order to allow looping though a whole range of IP addresses:

  my $ip = new Net::IP ('195.45.6.7 - 195.45.6.19') || die;
  # Loop
  do {
      print $ip->ip(), "\n";
  } while (++$ip);

The ++ operator returns undef when the last address of the range is reached.

Return IP authority information from the IP::Authority module

CW$auth = ip-auth ();>

Note: IPv4 only

These functions do the real work in the module. Like the OO methods, most of these return undef on failure. In order to access error codes and strings, instead of using CW$ip->error() and CW$ip->errno(), use the global functions CWError() and CWErrno().

The functions of the procedural interface are not exported by default. In order to import these functions, you need to modify the use statement for the module:

CWuse Net::IP qw(:PROC);

Returns the error string corresponding to the last error generated in the module. This is also useful for the OO interface, as if the new() function fails, we cannot call CW$ip->error() and so we have to use Error().

warn Error();

Returns a numeric error code corresponding to the error string returned by Error.

Transform an IP address into a bit string.

    Params  : IP address, IP version
    Returns : binary IP string on success, undef otherwise

CW$binip = ip_iptobin ($ip,6);

Transform a bit string into an IP address

    Params  : binary IP, IP version
    Returns : IP address on success, undef otherwise

CW$ip = ip_bintoip ($binip,6);

Transform a bit string into a BigInt.

    Params  : binary IP
    Returns : BigInt

CW$bigint = new Math::BigInt (ip_bintoint($binip));

Transform a BigInt into a bit string. Warning: sets warnings (CW-w) off. This is necessary because Math::BigInt is not compliant.

    Params  : BigInt, IP version
    Returns : binary IP

CW$binip = ip_inttobin ($bigint);

Try to guess the IP version of an IP address.

    Params  : IP address
    Returns : 4, 6, undef(unable to determine)

CW$version = ip_get_version ($ip)

Check if an IP address is of type 4.

    Params  : IP address
    Returns : 1 (yes) or 0 (no)

CWip_is_ipv4($ip) and print "$ip is IPv4";

Check if an IP address is of type 6.

    Params            : IP address
    Returns           : 1 (yes) or 0 (no)

CWip_is_ipv6($ip) and print "$ip is IPv6";

Expand an IP address from compact notation.

    Params  : IP address, IP version
    Returns : expanded IP address or undef on failure

CW$ip = ip_expand_address ($ip,4);

Get IP mask from prefix length.

    Params  : Prefix length, IP version
    Returns : Binary Mask

CW$mask = ip_get_mask ($len,6);

Return the last binary address of a prefix.

    Params  : First binary IP, prefix length, IP version
    Returns : Binary IP

CW$lastbin = ip_last_address_bin ($ip,$len,6);

Split a prefix into IP and prefix length. If it was passed a simple IP, it just returns it.

    Params  : Prefix
    Returns : IP, optionnaly length of prefix

CW($ip,$len) = ip_splitprefix ($prefix)

Get a range of IPs from a prefix.

    Params  : Prefix, IP version
    Returns : First IP, last IP

CW($ip1,$ip2) = ip_prefix_to_range ($prefix,6);

Compare binary Ips with <, >, <=, >=. Operators are lt(<), le(<=), gt(>), and ge(>=)

    Params  : First binary IP, operator, Last binary IP
    Returns : 1 (yes), 0 (no), or undef (problem)

CWip_bincomp ($ip1,'lt',$ip2) == 1 or do {}

Add two binary IPs.

    Params  : First binary IP, Last binary IP
    Returns : Binary sum or undef (problem)

CW$binip = ip_binadd ($bin1,$bin2);

Get the prefix length for a given range of 2 IPs.

    Params  : First binary IP, Last binary IP
    Returns : Length of prefix or undef (problem)

CW$len = ip_get_prefix_length ($ip1,$ip2);

Return all prefixes between two IPs.

    Params  : First IP, Last IP, IP version
    Returns : List of Prefixes or undef (problem)

The prefixes returned have the form q.q.q.q/nn.

CW@prefix = ip_range_to_prefix ($ip1,$ip2,6);

Compress an IPv4 Prefix.

    Params  : IP, Prefix length
    Returns : Compressed Prefix

CW$ip = ip_compress_v4_prefix ($ip, $len);

Compress an IPv6 address. Just returns the IP if it is an IPv4.

    Params  : IP, IP version
    Returns : Compressed IP or undef (problem)

CW$ip = ip_compress_adress ($ip, $version);

Check if two ranges of IPs overlap.

    Params  : Four binary IPs (begin of range 1,end1,begin2,end2), IP version
        $IP_PARTIAL_OVERLAP (ranges overlap) 
        $IP_NO_OVERLAP      (no overlap)
        $IP_A_IN_B_OVERLAP  (range2 contains range1)
        $IP_B_IN_A_OVERLAP  (range1 contains range2)
        $IP_IDENTICAL       (ranges are identical)
        undef               (problem)

CW(ip_is_overlap($rb1,$re1,$rb2,$re2,4) eq $IP_A_IN_B_OVERLAP) and do {};

Get an IPv4 embedded in an IPv6 address

    Params  : IPv6
    Returns : IPv4 string or undef (not found)

CW$ip4 = ip_get_embedded($ip6);

Check the validity of a binary IP mask

    Params  : Mask
    Returns : 1 or undef (invalid)

CWip_check_mask($binmask) or do {};

Checks if mask has only 1s followed by 0s.

Aggregate 2 ranges of binary IPs

    Params  : 1st range (1st IP, Last IP), last range (1st IP, last IP), IP version
    Returns : prefix or undef (invalid)

CW$prefix = ip_aggregate ($bip1,$eip1,$bip2,$eip2) || die ...

Return the type of an IP (Public, Private, Reserved)

    Params  : IP to test, IP version
    Returns : type or undef (invalid)

CW$type = ip_iptype ($ip);

Check the validity of a prefix

    Params  : binary IP, length of prefix, IP version
    Returns : 1 or undef (invalid)

Checks if the variant part of a prefix only has 0s, and the length is correct.

CWip_check_prefix ($ip,$len,$ipv) or do {};

Get a reverse name from a prefix

    Params  : IP, length of prefix, IP version
    Returns : Reverse name or undef (error)

CW$reverse = ip_reverse ($ip);

Normalize data to a range/prefix of IP addresses

    Params  : Data String (Single IP, Range, Prefix)
    Returns : ip1, ip2 (if range/prefix) or undef (error)

CW($ip1,$ip2) = ip_normalize ($data);

Return IP authority information from the IP::Authority module

    Params  : IP, version
    Returns : Auth info (RI for RIPE, AR for ARIN, etc)

CW$auth = ip_auth ($ip,4);

Note: IPv4 only

The Math::BigInt library is needed for functions that use integers. These are ip_inttobin, ip_bintoint, and the size method. In a next version, Math::BigInt will become optionnal.

Manuel Valente <manuel.valente@gmail.com>.

Original IPv4 code by Monica Cortes Sack <mcortes@ripe.net>.

Original IPv6 code by Lee Wilmot <lee@ripe.net>.

ipv4pack.pm, iplib.pm, iplibncc.pm.

perl(1), IP::Authority