pub struct Ipv6Addr { /* private fields */ }
Expand description
An IPv6 address.
IPv6 addresses are defined as 128-bit integers in IETF RFC 4291. They are usually represented as eight 16-bit segments.
§Embedding IPv4 Addresses
See IpAddr
for a type encompassing both IPv4 and IPv6 addresses.
To assist in the transition from IPv4 to IPv6 two types of IPv6 addresses that embed an IPv4 address were defined: IPv4-compatible and IPv4-mapped addresses. Of these IPv4-compatible addresses have been officially deprecated.
Both types of addresses are not assigned any special meaning by this implementation,
other than what the relevant standards prescribe. This means that an address like ::ffff:127.0.0.1
,
while representing an IPv4 loopback address, is not itself an IPv6 loopback address; only ::1
is.
To handle these so called “IPv4-in-IPv6” addresses, they have to first be converted to their canonical IPv4 address.
§IPv4-Compatible IPv6 Addresses
IPv4-compatible IPv6 addresses are defined in IETF RFC 4291 Section 2.5.5.1, and have been officially deprecated. The RFC describes the format of an “IPv4-Compatible IPv6 address” as follows:
| 80 bits | 16 | 32 bits |
+--------------------------------------+--------------------------+
|0000..............................0000|0000| IPv4 address |
+--------------------------------------+----+---------------------+
So ::a.b.c.d
would be an IPv4-compatible IPv6 address representing the IPv4 address a.b.c.d
.
To convert from an IPv4 address to an IPv4-compatible IPv6 address, use Ipv4Addr::to_ipv6_compatible
.
Use Ipv6Addr::to_ipv4
to convert an IPv4-compatible IPv6 address to the canonical IPv4 address.
§IPv4-Mapped IPv6 Addresses
IPv4-mapped IPv6 addresses are defined in IETF RFC 4291 Section 2.5.5.2. The RFC describes the format of an “IPv4-Mapped IPv6 address” as follows:
| 80 bits | 16 | 32 bits |
+--------------------------------------+--------------------------+
|0000..............................0000|FFFF| IPv4 address |
+--------------------------------------+----+---------------------+
So ::ffff:a.b.c.d
would be an IPv4-mapped IPv6 address representing the IPv4 address a.b.c.d
.
To convert from an IPv4 address to an IPv4-mapped IPv6 address, use Ipv4Addr::to_ipv6_mapped
.
Use Ipv6Addr::to_ipv4
to convert an IPv4-mapped IPv6 address to the canonical IPv4 address.
Note that this will also convert the IPv6 loopback address ::1
to 0.0.0.1
. Use
Ipv6Addr::to_ipv4_mapped
to avoid this.
§Textual representation
Ipv6Addr
provides a FromStr
implementation. There are many ways to represent
an IPv6 address in text, but in general, each segments is written in hexadecimal
notation, and segments are separated by :
. For more information, see
IETF RFC 5952.
§Examples
Implementations§
source§impl Ipv6Addr
impl Ipv6Addr
1.0.0 (const: 1.32.0) · sourcepub const fn new(
a: u16,
b: u16,
c: u16,
d: u16,
e: u16,
f: u16,
g: u16,
h: u16,
) -> Ipv6Addr
pub const fn new( a: u16, b: u16, c: u16, d: u16, e: u16, f: u16, g: u16, h: u16, ) -> Ipv6Addr
Creates a new IPv6 address from eight 16-bit segments.
The result will represent the IP address a:b:c:d:e:f:g:h
.
§Examples
1.80.0 (const: 1.80.0) · sourcepub const fn to_bits(self) -> u128
pub const fn to_bits(self) -> u128
Converts an IPv6 address into a u128
representation using native byte order.
Although IPv6 addresses are big-endian, the u128
value will use the target platform’s
native byte order. That is, the u128
value is an integer representation of the IPv6
address and not an integer interpretation of the IPv6 address’s big-endian bitstring. This
means that the u128
value masked with 0xffffffffffffffffffffffffffff0000_u128
will set
the last segment in the address to 0, regardless of the target platform’s endianness.
§Examples
use std::net::Ipv6Addr;
let addr = Ipv6Addr::new(
0x1020, 0x3040, 0x5060, 0x7080,
0x90A0, 0xB0C0, 0xD0E0, 0xF00D,
);
assert_eq!(0x102030405060708090A0B0C0D0E0F00D_u128, u128::from(addr));
use std::net::Ipv6Addr;
let addr = Ipv6Addr::new(
0x1020, 0x3040, 0x5060, 0x7080,
0x90A0, 0xB0C0, 0xD0E0, 0xF00D,
);
let addr_bits = addr.to_bits() & 0xffffffffffffffffffffffffffff0000_u128;
assert_eq!(
Ipv6Addr::new(
0x1020, 0x3040, 0x5060, 0x7080,
0x90A0, 0xB0C0, 0xD0E0, 0x0000,
),
Ipv6Addr::from_bits(addr_bits));
1.80.0 (const: 1.80.0) · sourcepub const fn from_bits(bits: u128) -> Ipv6Addr
pub const fn from_bits(bits: u128) -> Ipv6Addr
Converts a native byte order u128
into an IPv6 address.
See Ipv6Addr::to_bits
for an explanation on endianness.
§Examples
1.30.0 · sourcepub const LOCALHOST: Self = _
pub const LOCALHOST: Self = _
An IPv6 address representing localhost: ::1
.
This corresponds to constant IN6ADDR_LOOPBACK_INIT
or in6addr_loopback
in other
languages.
§Examples
1.30.0 · sourcepub const UNSPECIFIED: Self = _
pub const UNSPECIFIED: Self = _
An IPv6 address representing the unspecified address: ::
.
This corresponds to constant IN6ADDR_ANY_INIT
or in6addr_any
in other languages.
§Examples
1.0.0 (const: 1.50.0) · sourcepub const fn segments(&self) -> [u16; 8]
pub const fn segments(&self) -> [u16; 8]
Returns the eight 16-bit segments that make up this address.
§Examples
1.7.0 (const: 1.50.0) · sourcepub const fn is_unspecified(&self) -> bool
pub const fn is_unspecified(&self) -> bool
Returns true
for the special ‘unspecified’ address (::
).
This property is defined in IETF RFC 4291.
§Examples
1.7.0 (const: 1.50.0) · sourcepub const fn is_loopback(&self) -> bool
pub const fn is_loopback(&self) -> bool
Returns true
if this is the loopback address (::1
),
as defined in IETF RFC 4291 section 2.5.3.
Contrary to IPv4, in IPv6 there is only one loopback address.
§Examples
sourcepub const fn is_global(&self) -> bool
🔬This is a nightly-only experimental API. (ip
#27709)
pub const fn is_global(&self) -> bool
ip
#27709)Returns true
if the address appears to be globally reachable
as specified by the IANA IPv6 Special-Purpose Address Registry.
Whether or not an address is practically reachable will depend on your network configuration. Most IPv6 addresses are globally reachable, unless they are specifically defined as not globally reachable.
Non-exhaustive list of notable addresses that are not globally reachable:
- The unspecified address (
is_unspecified
) - The loopback address (
is_loopback
) - IPv4-mapped addresses
- Addresses reserved for benchmarking (
is_benchmarking
) - Addresses reserved for documentation (
is_documentation
) - Unique local addresses (
is_unique_local
) - Unicast addresses with link-local scope (
is_unicast_link_local
)
For the complete overview of which addresses are globally reachable, see the table at the IANA IPv6 Special-Purpose Address Registry.
Note that an address having global scope is not the same as being globally reachable, and there is no direct relation between the two concepts: There exist addresses with global scope that are not globally reachable (for example unique local addresses), and addresses that are globally reachable without having global scope (multicast addresses with non-global scope).
§Examples
#![feature(ip)]
use std::net::Ipv6Addr;
// Most IPv6 addresses are globally reachable:
assert_eq!(Ipv6Addr::new(0x26, 0, 0x1c9, 0, 0, 0xafc8, 0x10, 0x1).is_global(), true);
// However some addresses have been assigned a special meaning
// that makes them not globally reachable. Some examples are:
// The unspecified address (`::`)
assert_eq!(Ipv6Addr::UNSPECIFIED.is_global(), false);
// The loopback address (`::1`)
assert_eq!(Ipv6Addr::LOCALHOST.is_global(), false);
// IPv4-mapped addresses (`::ffff:0:0/96`)
assert_eq!(Ipv6Addr::new(0, 0, 0, 0, 0, 0xffff, 0xc00a, 0x2ff).is_global(), false);
// Addresses reserved for benchmarking (`2001:2::/48`)
assert_eq!(Ipv6Addr::new(0x2001, 2, 0, 0, 0, 0, 0, 1,).is_global(), false);
// Addresses reserved for documentation (`2001:db8::/32`)
assert_eq!(Ipv6Addr::new(0x2001, 0xdb8, 0, 0, 0, 0, 0, 1).is_global(), false);
// Unique local addresses (`fc00::/7`)
assert_eq!(Ipv6Addr::new(0xfc02, 0, 0, 0, 0, 0, 0, 1).is_global(), false);
// Unicast addresses with link-local scope (`fe80::/10`)
assert_eq!(Ipv6Addr::new(0xfe81, 0, 0, 0, 0, 0, 0, 1).is_global(), false);
// For a complete overview see the IANA IPv6 Special-Purpose Address Registry.
sourcepub const fn is_unique_local(&self) -> bool
🔬This is a nightly-only experimental API. (ip
#27709)
pub const fn is_unique_local(&self) -> bool
ip
#27709)Returns true
if this is a unique local address (fc00::/7
).
This property is defined in IETF RFC 4193.
§Examples
sourcepub const fn is_unicast(&self) -> bool
🔬This is a nightly-only experimental API. (ip
#27709)
pub const fn is_unicast(&self) -> bool
ip
#27709)Returns true
if this is a unicast address, as defined by IETF RFC 4291.
Any address that is not a multicast address (ff00::/8
) is unicast.
§Examples
#![feature(ip)]
use std::net::Ipv6Addr;
// The unspecified and loopback addresses are unicast.
assert_eq!(Ipv6Addr::UNSPECIFIED.is_unicast(), true);
assert_eq!(Ipv6Addr::LOCALHOST.is_unicast(), true);
// Any address that is not a multicast address (`ff00::/8`) is unicast.
assert_eq!(Ipv6Addr::new(0x2001, 0xdb8, 0, 0, 0, 0, 0, 0).is_unicast(), true);
assert_eq!(Ipv6Addr::new(0xff00, 0, 0, 0, 0, 0, 0, 0).is_unicast(), false);
sourcepub const fn is_unicast_link_local(&self) -> bool
🔬This is a nightly-only experimental API. (ip
#27709)
pub const fn is_unicast_link_local(&self) -> bool
ip
#27709)Returns true
if the address is a unicast address with link-local scope,
as defined in RFC 4291.
A unicast address has link-local scope if it has the prefix fe80::/10
, as per RFC 4291 section 2.4.
Note that this encompasses more addresses than those defined in RFC 4291 section 2.5.6,
which describes “Link-Local IPv6 Unicast Addresses” as having the following stricter format:
| 10 bits | 54 bits | 64 bits |
+----------+-------------------------+----------------------------+
|1111111010| 0 | interface ID |
+----------+-------------------------+----------------------------+
So while currently the only addresses with link-local scope an application will encounter are all in fe80::/64
,
this might change in the future with the publication of new standards. More addresses in fe80::/10
could be allocated,
and those addresses will have link-local scope.
Also note that while RFC 4291 section 2.5.3 mentions about the loopback address (::1
) that “it is treated as having Link-Local scope”,
this does not mean that the loopback address actually has link-local scope and this method will return false
on it.
§Examples
#![feature(ip)]
use std::net::Ipv6Addr;
// The loopback address (`::1`) does not actually have link-local scope.
assert_eq!(Ipv6Addr::LOCALHOST.is_unicast_link_local(), false);
// Only addresses in `fe80::/10` have link-local scope.
assert_eq!(Ipv6Addr::new(0x2001, 0xdb8, 0, 0, 0, 0, 0, 0).is_unicast_link_local(), false);
assert_eq!(Ipv6Addr::new(0xfe80, 0, 0, 0, 0, 0, 0, 0).is_unicast_link_local(), true);
// Addresses outside the stricter `fe80::/64` also have link-local scope.
assert_eq!(Ipv6Addr::new(0xfe80, 0, 0, 1, 0, 0, 0, 0).is_unicast_link_local(), true);
assert_eq!(Ipv6Addr::new(0xfe81, 0, 0, 0, 0, 0, 0, 0).is_unicast_link_local(), true);
sourcepub const fn is_documentation(&self) -> bool
🔬This is a nightly-only experimental API. (ip
#27709)
pub const fn is_documentation(&self) -> bool
ip
#27709)Returns true
if this is an address reserved for documentation
(2001:db8::/32
).
This property is defined in IETF RFC 3849.
§Examples
sourcepub const fn is_benchmarking(&self) -> bool
🔬This is a nightly-only experimental API. (ip
#27709)
pub const fn is_benchmarking(&self) -> bool
ip
#27709)Returns true
if this is an address reserved for benchmarking (2001:2::/48
).
This property is defined in IETF RFC 5180, where it is mistakenly specified as covering the range 2001:0200::/48
.
This is corrected in IETF RFC Errata 1752 to 2001:0002::/48
.
sourcepub const fn is_unicast_global(&self) -> bool
🔬This is a nightly-only experimental API. (ip
#27709)
pub const fn is_unicast_global(&self) -> bool
ip
#27709)Returns true
if the address is a globally routable unicast address.
The following return false:
- the loopback address
- the link-local addresses
- unique local addresses
- the unspecified address
- the address range reserved for documentation
This method returns true
for site-local addresses as per RFC 4291 section 2.5.7
The special behavior of [the site-local unicast] prefix defined in [RFC3513] must no longer
be supported in new implementations (i.e., new implementations must treat this prefix as
Global Unicast).
§Examples
sourcepub const fn multicast_scope(&self) -> Option<Ipv6MulticastScope>
🔬This is a nightly-only experimental API. (ip
#27709)
pub const fn multicast_scope(&self) -> Option<Ipv6MulticastScope>
ip
#27709)Returns the address’s multicast scope if the address is multicast.
§Examples
1.7.0 (const: 1.50.0) · sourcepub const fn is_multicast(&self) -> bool
pub const fn is_multicast(&self) -> bool
Returns true
if this is a multicast address (ff00::/8
).
This property is defined by IETF RFC 4291.
§Examples
sourcepub const fn is_ipv4_mapped(&self) -> bool
🔬This is a nightly-only experimental API. (ip
#27709)
pub const fn is_ipv4_mapped(&self) -> bool
ip
#27709)Returns true
if the address is an IPv4-mapped address (::ffff:0:0/96
).
IPv4-mapped addresses can be converted to their canonical IPv4 address with
to_ipv4_mapped
.
§Examples
#![feature(ip)]
use std::net::{Ipv4Addr, Ipv6Addr};
let ipv4_mapped = Ipv4Addr::new(192, 0, 2, 255).to_ipv6_mapped();
assert_eq!(ipv4_mapped.is_ipv4_mapped(), true);
assert_eq!(Ipv6Addr::new(0, 0, 0, 0, 0, 0xffff, 0xc000, 0x2ff).is_ipv4_mapped(), true);
assert_eq!(Ipv6Addr::new(0x2001, 0xdb8, 0, 0, 0, 0, 0, 0).is_ipv4_mapped(), false);
1.63.0 (const: 1.75.0) · sourcepub const fn to_ipv4_mapped(&self) -> Option<Ipv4Addr>
pub const fn to_ipv4_mapped(&self) -> Option<Ipv4Addr>
Converts this address to an IPv4
address if it’s an IPv4-mapped address,
as defined in IETF RFC 4291 section 2.5.5.2, otherwise returns None
.
::ffff:a.b.c.d
becomes a.b.c.d
.
All addresses not starting with ::ffff
will return None
.
§Examples
use std::net::{Ipv4Addr, Ipv6Addr};
assert_eq!(Ipv6Addr::new(0xff00, 0, 0, 0, 0, 0, 0, 0).to_ipv4_mapped(), None);
assert_eq!(Ipv6Addr::new(0, 0, 0, 0, 0, 0xffff, 0xc00a, 0x2ff).to_ipv4_mapped(),
Some(Ipv4Addr::new(192, 10, 2, 255)));
assert_eq!(Ipv6Addr::new(0, 0, 0, 0, 0, 0, 0, 1).to_ipv4_mapped(), None);
1.0.0 (const: 1.50.0) · sourcepub const fn to_ipv4(&self) -> Option<Ipv4Addr>
pub const fn to_ipv4(&self) -> Option<Ipv4Addr>
Converts this address to an IPv4
address if it is either
an IPv4-compatible address as defined in IETF RFC 4291 section 2.5.5.1,
or an IPv4-mapped address as defined in IETF RFC 4291 section 2.5.5.2,
otherwise returns None
.
Note that this will return an IPv4
address for the IPv6 loopback address ::1
. Use
Ipv6Addr::to_ipv4_mapped
to avoid this.
::a.b.c.d
and ::ffff:a.b.c.d
become a.b.c.d
. ::1
becomes 0.0.0.1
.
All addresses not starting with either all zeroes or ::ffff
will return None
.
§Examples
use std::net::{Ipv4Addr, Ipv6Addr};
assert_eq!(Ipv6Addr::new(0xff00, 0, 0, 0, 0, 0, 0, 0).to_ipv4(), None);
assert_eq!(Ipv6Addr::new(0, 0, 0, 0, 0, 0xffff, 0xc00a, 0x2ff).to_ipv4(),
Some(Ipv4Addr::new(192, 10, 2, 255)));
assert_eq!(Ipv6Addr::new(0, 0, 0, 0, 0, 0, 0, 1).to_ipv4(),
Some(Ipv4Addr::new(0, 0, 0, 1)));
1.75.0 (const: 1.75.0) · sourcepub const fn to_canonical(&self) -> IpAddr
pub const fn to_canonical(&self) -> IpAddr
Converts this address to an IpAddr::V4
if it is an IPv4-mapped address,
otherwise returns self wrapped in an IpAddr::V6
.
§Examples
source§impl Ipv6Addr
impl Ipv6Addr
sourcepub fn parse_ascii(b: &[u8]) -> Result<Self, AddrParseError>
🔬This is a nightly-only experimental API. (addr_parse_ascii
#101035)
pub fn parse_ascii(b: &[u8]) -> Result<Self, AddrParseError>
addr_parse_ascii
#101035)Trait Implementations§
1.75.0 · source§impl BitAndAssign<&Ipv6Addr> for Ipv6Addr
impl BitAndAssign<&Ipv6Addr> for Ipv6Addr
source§fn bitand_assign(&mut self, rhs: &Ipv6Addr)
fn bitand_assign(&mut self, rhs: &Ipv6Addr)
&=
operation. Read more1.75.0 · source§impl BitAndAssign for Ipv6Addr
impl BitAndAssign for Ipv6Addr
source§fn bitand_assign(&mut self, rhs: Ipv6Addr)
fn bitand_assign(&mut self, rhs: Ipv6Addr)
&=
operation. Read more1.75.0 · source§impl BitOrAssign<&Ipv6Addr> for Ipv6Addr
impl BitOrAssign<&Ipv6Addr> for Ipv6Addr
source§fn bitor_assign(&mut self, rhs: &Ipv6Addr)
fn bitor_assign(&mut self, rhs: &Ipv6Addr)
|=
operation. Read more1.75.0 · source§impl BitOrAssign for Ipv6Addr
impl BitOrAssign for Ipv6Addr
source§fn bitor_assign(&mut self, rhs: Ipv6Addr)
fn bitor_assign(&mut self, rhs: Ipv6Addr)
|=
operation. Read more1.0.0 · source§impl Display for Ipv6Addr
impl Display for Ipv6Addr
Writes an Ipv6Addr, conforming to the canonical style described by RFC 5952.
1.16.0 · source§impl From<[u16; 8]> for Ipv6Addr
impl From<[u16; 8]> for Ipv6Addr
1.9.0 · source§impl From<[u8; 16]> for Ipv6Addr
impl From<[u8; 16]> for Ipv6Addr
1.0.0 · source§impl Ord for Ipv6Addr
impl Ord for Ipv6Addr
1.16.0 · source§impl PartialOrd<IpAddr> for Ipv6Addr
impl PartialOrd<IpAddr> for Ipv6Addr
1.16.0 · source§impl PartialOrd<Ipv6Addr> for IpAddr
impl PartialOrd<Ipv6Addr> for IpAddr
1.0.0 · source§impl PartialOrd for Ipv6Addr
impl PartialOrd for Ipv6Addr
source§impl Step for Ipv6Addr
impl Step for Ipv6Addr
source§fn steps_between(start: &Ipv6Addr, end: &Ipv6Addr) -> Option<usize>
fn steps_between(start: &Ipv6Addr, end: &Ipv6Addr) -> Option<usize>
step_trait
#42168)source§fn forward_checked(start: Ipv6Addr, count: usize) -> Option<Ipv6Addr>
fn forward_checked(start: Ipv6Addr, count: usize) -> Option<Ipv6Addr>
step_trait
#42168)source§fn backward_checked(start: Ipv6Addr, count: usize) -> Option<Ipv6Addr>
fn backward_checked(start: Ipv6Addr, count: usize) -> Option<Ipv6Addr>
step_trait
#42168)source§unsafe fn forward_unchecked(start: Ipv6Addr, count: usize) -> Ipv6Addr
unsafe fn forward_unchecked(start: Ipv6Addr, count: usize) -> Ipv6Addr
step_trait
#42168)source§unsafe fn backward_unchecked(start: Ipv6Addr, count: usize) -> Ipv6Addr
unsafe fn backward_unchecked(start: Ipv6Addr, count: usize) -> Ipv6Addr
step_trait
#42168)