Navigation
- index
- modules |
- next |
- previous |
- Python »
- 3.4.2 Documentation »
- The Python Standard Library »
- 21. Internet Protocols and Support »
Source code: Lib/ipaddress.py
ipaddress provides the capabilities to create, manipulate and operate on IPv4 and IPv6 addresses and networks.
The functions and classes in this module make it straightforward to handle various tasks related to IP addresses, including checking whether or not two hosts are on the same subnet, iterating over all hosts in a particular subnet, checking whether or not a string represents a valid IP address or network definition, and so on.
This is the full module API reference—for an overview and introduction, see An introduction to the ipaddress module.
New in version 3.3.
The ipaddress module provides factory functions to conveniently create IP addresses, networks and interfaces:
Return an IPv4Address or IPv6Address object depending on the IP address passed as argument. Either IPv4 or IPv6 addresses may be supplied; integers less than 2**32 will be considered to be IPv4 by default. A ValueError is raised if address does not represent a valid IPv4 or IPv6 address.
>>> ipaddress.ip_address('192.168.0.1')
IPv4Address('192.168.0.1')
>>> ipaddress.ip_address('2001:db8::')
IPv6Address('2001:db8::')
Return an IPv4Network or IPv6Network object depending on the IP address passed as argument. address is a string or integer representing the IP network. Either IPv4 or IPv6 networks may be supplied; integers less than 2**32 will be considered to be IPv4 by default. strict is passed to IPv4Network or IPv6Network constructor. A ValueError is raised if address does not represent a valid IPv4 or IPv6 address, or if the network has host bits set.
>>> ipaddress.ip_network('192.168.0.0/28')
IPv4Network('192.168.0.0/28')
Return an IPv4Interface or IPv6Interface object depending on the IP address passed as argument. address is a string or integer representing the IP address. Either IPv4 or IPv6 addresses may be supplied; integers less than 2**32 will be considered to be IPv4 by default. A ValueError is raised if address does not represent a valid IPv4 or IPv6 address.
One downside of these convenience functions is that the need to handle both IPv4 and IPv6 formats means that error messages provide minimal information on the precise error, as the functions don’t know whether the IPv4 or IPv6 format was intended. More detailed error reporting can be obtained by calling the appropriate version specific class constructors directly.
The IPv4Address and IPv6Address objects share a lot of common attributes. Some attributes that are only meaningful for IPv6 addresses are also implemented by IPv4Address objects, in order to make it easier to write code that handles both IP versions correctly.
Construct an IPv4 address. An AddressValueError is raised if address is not a valid IPv4 address.
The following constitutes a valid IPv4 address:
>>> ipaddress.IPv4Address('192.168.0.1')
IPv4Address('192.168.0.1')
>>> ipaddress.IPv4Address(3232235521)
IPv4Address('192.168.0.1')
>>> ipaddress.IPv4Address(b'\xC0\xA8\x00\x01')
IPv4Address('192.168.0.1')
The appropriate version number: 4 for IPv4, 6 for IPv6.
The total number of bits in the address representation for this version: 32 for IPv4, 128 for IPv6.
The prefix defines the number of leading bits in an address that are compared to determine whether or not an address is part of a network.
The string representation in dotted decimal notation. Leading zeroes are never included in the representation.
As IPv4 does not define a shorthand notation for addresses with octets set to zero, these two attributes are always the same as str(addr) for IPv4 addresses. Exposing these attributes makes it easier to write display code that can handle both IPv4 and IPv6 addresses.
The binary representation of this address - a bytes object of the appropriate length (most significant octet first). This is 4 bytes for IPv4 and 16 bytes for IPv6.
True if the address is reserved for multicast use. See RFC 3171 (for IPv4) or RFC 2373 (for IPv6).
True if the address is allocated for private networks. See iana-ipv4-special-registry (for IPv4) or iana-ipv6-special-registry (for IPv6).
True if the address is allocated for public networks. See iana-ipv4-special-registry (for IPv4) or iana-ipv6-special-registry (for IPv6).
New in version 3.4.
True if the address is otherwise IETF reserved.
Construct an IPv6 address. An AddressValueError is raised if address is not a valid IPv6 address.
The following constitutes a valid IPv6 address:
>>> ipaddress.IPv6Address('2001:db8::1000')
IPv6Address('2001:db8::1000')
The short form of the address representation, with leading zeroes in groups omitted and the longest sequence of groups consisting entirely of zeroes collapsed to a single empty group.
This is also the value returned by str(addr) for IPv6 addresses.
The long form of the address representation, with all leading zeroes and groups consisting entirely of zeroes included.
For the following attributes, see the corresponding documention of the IPv4Address class:
New in version 3.4: is_global
True if the address is reserved for site-local usage. Note that the site-local address space has been deprecated by RFC 3879. Use is_private to test if this address is in the space of unique local addresses as defined by RFC 4193.
For addresses that appear to be IPv4 mapped addresses (starting with ::FFFF/96), this property will report the embedded IPv4 address. For any other address, this property will be None.
To interoperate with networking interfaces such as the socket module, addresses must be converted to strings or integers. This is handled using the str() and int() builtin functions:
>>> str(ipaddress.IPv4Address('192.168.0.1'))
'192.168.0.1'
>>> int(ipaddress.IPv4Address('192.168.0.1'))
3232235521
>>> str(ipaddress.IPv6Address('::1'))
'::1'
>>> int(ipaddress.IPv6Address('::1'))
1
Address objects support some operators. Unless stated otherwise, operators can only be applied between compatible objects (i.e. IPv4 with IPv4, IPv6 with IPv6).
Address objects can be compared with the usual set of comparison operators. Some examples:
>>> IPv4Address('127.0.0.2') > IPv4Address('127.0.0.1')
True
>>> IPv4Address('127.0.0.2') == IPv4Address('127.0.0.1')
False
>>> IPv4Address('127.0.0.2') != IPv4Address('127.0.0.1')
True
Integers can be added to or subtracted from address objects. Some examples:
>>> IPv4Address('127.0.0.2') + 3
IPv4Address('127.0.0.5')
>>> IPv4Address('127.0.0.2') - 3
IPv4Address('126.255.255.255')
>>> IPv4Address('255.255.255.255') + 1
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ipaddress.AddressValueError: 4294967296 (>= 2**32) is not permitted as an IPv4 address
The IPv4Network and IPv6Network objects provide a mechanism for defining and inspecting IP network definitions. A network definition consists of a mask and a network address, and as such defines a range of IP addresses that equal the network address when masked (binary AND) with the mask. For example, a network definition with the mask 255.255.255.0 and the network address 192.168.1.0 consists of IP addresses in the inclusive range 192.168.1.0 to 192.168.1.255.
There are several equivalent ways to specify IP network masks. A prefix /<nbits> is a notation that denotes how many high-order bits are set in the network mask. A net mask is an IP address with some number of high-order bits set. Thus the prefix /24 is equivalent to the net mask 255.255.255.0 in IPv4, or ffff:ff00:: in IPv6. In addition, a host mask is the logical inverse of a net mask, and is sometimes used (for example in Cisco access control lists) to denote a network mask. The host mask equivalent to /24 in IPv4 is 0.0.0.255.
All attributes implemented by address objects are implemented by network objects as well. In addition, network objects implement additional attributes. All of these are common between IPv4Network and IPv6Network, so to avoid duplication they are only documented for IPv4Network.
Construct an IPv4 network definition. address can be one of the following:
A string consisting of an IP address and an optional mask, separated by a slash (/). The IP address is the network address, and the mask can be either a single number, which means it’s a prefix, or a string representation of an IPv4 address. If it’s the latter, the mask is interpreted as a net mask if it starts with a non-zero field, or as a host mask if it starts with a zero field. If no mask is provided, it’s considered to be /32.
For example, the following address specifications are equivalent: 192.168.1.0/24, 192.168.1.0/255.255.255.0 and 192.168.1.0/0.0.0.255.
An integer that fits into 32 bits. This is equivalent to a single-address network, with the network address being address and the mask being /32.
An integer packed into a bytes object of length 4, big-endian. The interpretation is similar to an integer address.
An AddressValueError is raised if address is not a valid IPv4 address. A NetmaskValueError is raised if the mask is not valid for an IPv4 address.
If strict is True and host bits are set in the supplied address, then ValueError is raised. Otherwise, the host bits are masked out to determine the appropriate network address.
Unless stated otherwise, all network methods accepting other network/address objects will raise TypeError if the argument’s IP version is incompatible to self
Refer to the corresponding attribute documentation in IPv4Address
These attributes are true for the network as a whole if they are true for both the network address and the broadcast address
The network address for the network. The network address and the prefix length together uniquely define a network.
The broadcast address for the network. Packets sent to the broadcast address should be received by every host on the network.
The host mask, as a string.
A string representation of the network, with the mask in prefix notation.
with_prefixlen and compressed are always the same as str(network). exploded uses the exploded form the network address.
A string representation of the network, with the mask in net mask notation.
A string representation of the network, with the mask in host mask notation.
The total number of addresses in the network.
Length of the network prefix, in bits.
Returns an iterator over the usable hosts in the network. The usable hosts are all the IP addresses that belong to the network, except the network address itself and the network broadcast address.
>>> list(ip_network('192.0.2.0/29').hosts())
[IPv4Address('192.0.2.1'), IPv4Address('192.0.2.2'),
IPv4Address('192.0.2.3'), IPv4Address('192.0.2.4'),
IPv4Address('192.0.2.5'), IPv4Address('192.0.2.6')]
True if this network is partly or wholly contained in other or other is wholly contained in this network.
Computes the network definitions resulting from removing the given network from this one. Returns an iterator of network objects. Raises ValueError if network is not completely contained in this network.
>>> n1 = ip_network('192.0.2.0/28')
>>> n2 = ip_network('192.0.2.1/32')
>>> list(n1.address_exclude(n2))
[IPv4Network('192.0.2.8/29'), IPv4Network('192.0.2.4/30'),
IPv4Network('192.0.2.2/31'), IPv4Network('192.0.2.0/32')]
The subnets that join to make the current network definition, depending on the argument values. prefixlen_diff is the amount our prefix length should be increased by. new_prefix is the desired new prefix of the subnets; it must be larger than our prefix. One and only one of prefixlen_diff and new_prefix must be set. Returns an iterator of network objects.
>>> list(ip_network('192.0.2.0/24').subnets())
[IPv4Network('192.0.2.0/25'), IPv4Network('192.0.2.128/25')]
>>> list(ip_network('192.0.2.0/24').subnets(prefixlen_diff=2))
[IPv4Network('192.0.2.0/26'), IPv4Network('192.0.2.64/26'),
IPv4Network('192.0.2.128/26'), IPv4Network('192.0.2.192/26')]
>>> list(ip_network('192.0.2.0/24').subnets(new_prefix=26))
[IPv4Network('192.0.2.0/26'), IPv4Network('192.0.2.64/26'),
IPv4Network('192.0.2.128/26'), IPv4Network('192.0.2.192/26')]
>>> list(ip_network('192.0.2.0/24').subnets(new_prefix=23))
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
raise ValueError('new prefix must be longer')
ValueError: new prefix must be longer
>>> list(ip_network('192.0.2.0/24').subnets(new_prefix=25))
[IPv4Network('192.0.2.0/25'), IPv4Network('192.0.2.128/25')]
The supernet containing this network definition, depending on the argument values. prefixlen_diff is the amount our prefix length should be decreased by. new_prefix is the desired new prefix of the supernet; it must be smaller than our prefix. One and only one of prefixlen_diff and new_prefix must be set. Returns a single network object.
>>> ip_network('192.0.2.0/24').supernet()
IPv4Network('192.0.2.0/23')
>>> ip_network('192.0.2.0/24').supernet(prefixlen_diff=2)
IPv4Network('192.0.0.0/22')
>>> ip_network('192.0.2.0/24').supernet(new_prefix=20)
IPv4Network('192.0.0.0/20')
Compare this network to other. In this comparison only the network addresses are considered; host bits aren’t. Returns either -1, 0 or 1.
>>> ip_network('192.0.2.1/32').compare_networks(ip_network('192.0.2.2/32'))
-1
>>> ip_network('192.0.2.1/32').compare_networks(ip_network('192.0.2.0/32'))
1
>>> ip_network('192.0.2.1/32').compare_networks(ip_network('192.0.2.1/32'))
0
Construct an IPv6 network definition. address can be one of the following:
A string consisting of an IP address and an optional mask, separated by a slash (/). The IP address is the network address, and the mask can be either a single number, which means it’s a prefix, or a string representation of an IPv6 address. If it’s the latter, the mask is interpreted as a net mask. If no mask is provided, it’s considered to be /128.
For example, the following address specifications are equivalent: 2001:db00::0/24 and 2001:db00::0/ffff:ff00::.
An integer that fits into 128 bits. This is equivalent to a single-address network, with the network address being address and the mask being /128.
An integer packed into a bytes object of length 16, bit-endian. The interpretation is similar to an integer address.
An AddressValueError is raised if address is not a valid IPv6 address. A NetmaskValueError is raised if the mask is not valid for an IPv6 address.
If strict is True and host bits are set in the supplied address, then ValueError is raised. Otherwise, the host bits are masked out to determine the appropriate network address.
Refer to the corresponding attribute documentation in IPv4Network
These attribute is true for the network as a whole if it is true for both the network address and the broadcast address
Network objects support some operators. Unless stated otherwise, operators can only be applied between compatible objects (i.e. IPv4 with IPv4, IPv6 with IPv6).
Network objects can be compared with the usual set of logical operators, similarly to address objects.
Network objects can be iterated to list all the addresses belonging to the network. For iteration, all hosts are returned, including unusable hosts (for usable hosts, use the hosts() method). An example:
>>> for addr in IPv4Network('192.0.2.0/28'):
... addr
...
IPv4Address('192.0.2.0')
IPv4Address('192.0.2.1')
IPv4Address('192.0.2.2')
IPv4Address('192.0.2.3')
IPv4Address('192.0.2.4')
IPv4Address('192.0.2.5')
IPv4Address('192.0.2.6')
IPv4Address('192.0.2.7')
IPv4Address('192.0.2.8')
IPv4Address('192.0.2.9')
IPv4Address('192.0.2.10')
IPv4Address('192.0.2.11')
IPv4Address('192.0.2.12')
IPv4Address('192.0.2.13')
IPv4Address('192.0.2.14')
IPv4Address('192.0.2.15')
Network objects can act as containers of addresses. Some examples:
>>> IPv4Network('192.0.2.0/28')[0]
IPv4Address('192.0.2.0')
>>> IPv4Network('192.0.2.0/28')[15]
IPv4Address('192.0.2.15')
>>> IPv4Address('192.0.2.6') in IPv4Network('192.0.2.0/28')
True
>>> IPv4Address('192.0.3.6') in IPv4Network('192.0.2.0/28')
False
Construct an IPv4 interface. The meaning of address is as in the constructor of IPv4Network, except that arbitrary host addresses are always accepted.
IPv4Interface is a subclass of IPv4Address, so it inherits all the attributes from that class. In addition, the following attributes are available:
The address (IPv4Address) without network information.
>>> interface = IPv4Interface('192.0.2.5/24')
>>> interface.ip
IPv4Address('192.0.2.5')
The network (IPv4Network) this interface belongs to.
>>> interface = IPv4Interface('192.0.2.5/24')
>>> interface.network
IPv4Network('192.0.2.0/24')
A string representation of the interface with the mask in prefix notation.
>>> interface = IPv4Interface('192.0.2.5/24')
>>> interface.with_prefixlen
'192.0.2.5/24'
A string representation of the interface with the network as a net mask.
>>> interface = IPv4Interface('192.0.2.5/24')
>>> interface.with_netmask
'192.0.2.5/255.255.255.0'
A string representation of the interface with the network as a host mask.
>>> interface = IPv4Interface('192.0.2.5/24')
>>> interface.with_hostmask
'192.0.2.5/0.0.0.255'
Construct an IPv6 interface. The meaning of address is as in the constructor of IPv6Network, except that arbitrary host addresses are always accepted.
IPv6Interface is a subclass of IPv6Address, so it inherits all the attributes from that class. In addition, the following attributes are available:
Refer to the corresponding attribute documentation in IPv4Interface.
The module also provides the following module level functions:
Represent an address as 4 packed bytes in network (big-endian) order. address is an integer representation of an IPv4 IP address. A ValueError is raised if the integer is negative or too large to be an IPv4 IP address.
>>> ipaddress.ip_address(3221225985)
IPv4Address('192.0.2.1')
>>> ipaddress.v4_int_to_packed(3221225985)
b'\xc0\x00\x02\x01'
Represent an address as 16 packed bytes in network (big-endian) order. address is an integer representation of an IPv6 IP address. A ValueError is raised if the integer is negative or too large to be an IPv6 IP address.
Return an iterator of the summarized network range given the first and last IP addresses. first is the first IPv4Address or IPv6Address in the range and last is the last IPv4Address or IPv6Address in the range. A TypeError is raised if first or last are not IP addresses or are not of the same version. A ValueError is raised if last is not greater than first or if first address version is not 4 or 6.
>>> [ipaddr for ipaddr in ipaddress.summarize_address_range(
... ipaddress.IPv4Address('192.0.2.0'),
... ipaddress.IPv4Address('192.0.2.130'))]
[IPv4Network('192.0.2.0/25'), IPv4Network('192.0.2.128/31'), IPv4Network('192.0.2.130/32')]
Return an iterator of the collapsed IPv4Network or IPv6Network objects. addresses is an iterator of IPv4Network or IPv6Network objects. A TypeError is raised if addresses contains mixed version objects.
>>> [ipaddr for ipaddr in
... ipaddress.collapse_addresses([ipaddress.IPv4Network('192.0.2.0/25'),
... ipaddress.IPv4Network('192.0.2.128/25')])]
[IPv4Network('192.0.2.0/24')]
Return a key suitable for sorting between networks and addresses. Address and Network objects are not sortable by default; they’re fundamentally different, so the expression:
IPv4Address('192.0.2.0') <= IPv4Network('192.0.2.0/24')
doesn’t make sense. There are some times however, where you may wish to have ipaddress sort these anyway. If you need to do this, you can use this function as the key argument to sorted().
obj is either a network or address object.
To support more specific error reporting from class constructors, the module defines the following exceptions:
Any value error related to the address.
Any value error related to the netmask.