.TH IPSEC_TTOADDR 3 "28 Sept 2001"
.\" RCSID $Id$
.SH NAME
ipsec ttoaddr, tnatoaddr, addrtot \- convert Internet addresses to and from text
.br
ipsec ttosubnet, subnettot \- convert subnet/mask text form to and from addresses
.SH SYNOPSIS
.B "#include <freeswan.h>
.sp
.B "const char *ttoaddr(const char *src, size_t srclen,"
.ti +1c
.B "int af, ip_address *addr);"
.br
.B "const char *tnatoaddr(const char *src, size_t srclen,"
.ti +1c
.B "int af, ip_address *addr);"
.br
.B "size_t addrtot(const ip_address *addr, int format,"
.ti +1c
.B "char *dst, size_t dstlen);"
.sp
.B "const char *ttosubnet(const char *src, size_t srclen,"
.ti +1c
.B "int af, ip_subnet *dst);"
.br
.B "size_t subnettot(const ip_subnet *sub, int format,"
.ti +1c
.B "char *dst, size_t dstlen);"
.SH DESCRIPTION
.I Ttoaddr
converts a text-string name or numeric address into a binary address
(in network byte order).
.I Tnatoaddr
does the same conversion,
but the only text forms it accepts are
the ``official'' forms of
numeric address (dotted-decimal for IPv4, colon-hex for IPv6).
.I Addrtot
does the reverse conversion, from binary address back to a text form.
.I Ttosubnet
and
.I subnettot
do likewise for the ``address/mask'' form used to write a
specification of a subnet.
.PP
An IPv4 address is specified in text as a
dotted-decimal address (e.g.
.BR 1.2.3.4 ),
an eight-digit network-order hexadecimal number with the usual C prefix (e.g.
.BR 0x01020304 ,
which is synonymous with
.BR 1.2.3.4 ),
an eight-digit host-order hexadecimal number with a
.B 0h
prefix (e.g.
.BR 0h01020304 ,
which is synonymous with
.B 1.2.3.4
on a big-endian host and
.B 4.3.2.1
on a little-endian host),
a DNS name to be looked up via
.IR gethostbyname (3),
or an old-style network name to be looked up via
.IR getnetbyname (3).
.PP
A dotted-decimal address may be incomplete, in which case
text-to-binary conversion implicitly appends
as many instances of
.B .0
as necessary to bring it up to four components.
The components of a dotted-decimal address are always taken as
decimal, and leading zeros are ignored.
For example,
.B 10
is synonymous with
.BR 10.0.0.0 ,
and
.B 128.009.000.032
is synonymous with
.BR 128.9.0.32
(the latter example is verbatim from RFC 1166).
The result of applying
.I addrtot
to an IPv4 address is always complete and does not contain leading zeros.
.PP
Use of hexadecimal addresses is
.B strongly
.BR discouraged ;
they are included only to save hassles when dealing with
the handful of perverted programs which already print 
network addresses in hexadecimal.
.PP
An IPv6 address is specified in text with
colon-hex notation (e.g.
.BR 0:56:78ab:22:33:44:55:66 ),
colon-hex with
.B ::
abbreviating at most one subsequence of multiple zeros (e.g.
.BR 99:ab::54:068 ,
which is synonymous with
.BR 99:ab:0:0:0:0:54:68 ),
or a DNS name to be looked up via
.IR gethostbyname (3).
The result of applying
.I addrtot
to an IPv6 address will use
.B ::
abbreviation if possible,
and will not contain leading zeros.
.PP
The letters in hexadecimal
may be uppercase or lowercase or any mixture thereof.
.PP
DNS names may be complete (optionally terminated with a ``.'')
or incomplete, and are looked up as specified by local system configuration
(see
.IR resolver (5)).
The
.I h_addr
value returned by
.IR gethostbyname2 (3)
is used,
so with current DNS implementations,
the result when the name corresponds to more than one address is
difficult to predict.
IPv4 name lookup resorts to
.IR getnetbyname (3)
only if
.IR gethostbyname2 (3)
fails.
.PP
A subnet specification is of the form \fInetwork\fB/\fImask\fR.
The
.I network
and
.I mask
can be any form acceptable to
.IR ttoaddr .
In addition, and preferably, the
.I mask
can be a decimal integer (leading zeros ignored) giving a bit count,
in which case
it stands for a mask with that number of high bits on and all others off
(e.g.,
.B 24
in IPv4 means
.BR 255.255.255.0 ).
In any case, the mask must be contiguous
(a sequence of high bits on and all remaining low bits off).
As a special case, the subnet specification
.B %default
is a synonym for
.B 0.0.0.0/0
or
.B ::/0
in IPv4 or IPv6 respectively.
.PP
.I Ttosubnet
ANDs the mask with the address before returning,
so that any non-network bits in the address are turned off
(e.g.,
.B 10.1.2.3/24
is synonymous with
.BR 10.1.2.0/24 ).
.I Subnettot
always generates the decimal-integer-bit-count
form of the mask,
with no leading zeros.
.PP
The
.I srclen
parameter of
.I ttoaddr
and
.I ttosubnet
specifies the length of the text string pointed to by
.IR src ;
it is an error for there to be anything else
(e.g., a terminating NUL) within that length.
As a convenience for cases where an entire NUL-terminated string is
to be converted,
a
.I srclen
value of
.B 0
is taken to mean
.BR strlen(src) .
.PP
The
.I af
parameter of
.I ttoaddr
and
.I ttosubnet
specifies the address family of interest.
It should be either
.B AF_INET
or
.BR AF_INET6 .
.PP
The
.I dstlen
parameter of
.I addrtot
and
.I subnettot
specifies the size of the
.I dst
parameter;
under no circumstances are more than
.I dstlen
bytes written to
.IR dst .
A result which will not fit is truncated.
.I Dstlen
can be zero, in which case
.I dst
need not be valid and no result is written,
but the return value is unaffected;
in all other cases, the (possibly truncated) result is NUL-terminated.
The
.I freeswan.h
header file defines constants,
.B ADDRTOT_BUF
and
.BR SUBNETTOT_BUF ,
which are the sizes of buffers just large enough for worst-case results.
.PP
The
.I format
parameter of
.I addrtot
and
.I subnettot
specifies what format is to be used for the conversion.
The value
.B 0
(not the character
.BR '0' ,
but a zero value)
specifies a reasonable default,
and is in fact the only format currently available in
.IR subnettot .
.I Addrtot
also accepts format values
.B 'r'
(signifying a text form suitable for DNS reverse lookups,
e.g.
.B 4.3.2.1.IN-ADDR.ARPA.
for IPv4 and
RFC 2874 format for IPv6),
and
.B 'R'
(signifying an alternate reverse-lookup form,
an error for IPv4 and RFC 1886 format for IPv6).
Reverse-lookup names always end with a ``.''.
.PP
The text-to-binary functions return NULL for success and
a pointer to a string-literal error message for failure;
see DIAGNOSTICS.
The binary-to-text functions return
.B 0
for a failure, and otherwise
always return the size of buffer which would 
be needed to
accommodate the full conversion result, including terminating NUL;
it is the caller's responsibility to check this against the size of
the provided buffer to determine whether truncation has occurred.
.SH SEE ALSO
inet(3)
.SH DIAGNOSTICS
Fatal errors in
.I ttoaddr
are:
empty input;
unknown address family;
attempt to allocate temporary storage for a very long name failed;
name lookup failed;
syntax error in dotted-decimal or colon-hex form;
dotted-decimal or colon-hex component too large.
.PP
Fatal errors in
.I ttosubnet
are:
no
.B /
in
.IR src ;
.I ttoaddr
error in conversion of
.I network
or
.IR mask ;
bit-count mask too big;
mask non-contiguous.
.PP
Fatal errors in
.I addrtot
and
.I subnettot
are:
unknown format.
.SH HISTORY
Written for the FreeS/WAN project by Henry Spencer.
.SH BUGS
The interpretation of incomplete dotted-decimal addresses
(e.g.
.B 10/24
means
.BR 10.0.0.0/24 )
differs from that of some older conversion
functions, e.g. those of
.IR inet (3).
The behavior of the older functions has never been
particularly consistent or particularly useful.
.PP
Ignoring leading zeros in dotted-decimal components and bit counts
is arguably the most useful behavior in this application,
but it might occasionally cause confusion with the historical use of leading 
zeros to denote octal numbers.
.PP
.I Ttoaddr
does not support the mixed colon-hex-dotted-decimal
convention used to embed an IPv4 address in an IPv6 address.
.PP
.I Addrtot
always uses the
.B ::
abbreviation (which can appear only once in an address) for the
.I first
sequence of multiple zeros in an IPv6 address.
One can construct addresses (unlikely ones) in which this is suboptimal.
.PP
.I Addrtot
.B 'r'
conversion of an IPv6 address uses lowercase hexadecimal,
not the uppercase used in RFC 2874's examples.
It takes careful reading of RFCs 2874, 2673, and 2234 to realize
that lowercase is technically legitimate here,
and there may be software which botches this
and hence would have trouble with lowercase hex.
.PP
Possibly
.I subnettot
ought to recognize the
.B %default
case and generate that string as its output.
Currently it doesn't.
.PP
It is barely possible that somebody, somewhere,
might have a legitimate use for non-contiguous subnet masks.
.PP
.IR Getnetbyname (3)
is a historical dreg.
.PP
.I Tnatoaddr
probably should enforce completeness of dotted-decimal addresses.
.PP
The restriction of text-to-binary error reports to literal strings
(so that callers don't need to worry about freeing them or copying them)
does limit the precision of error reporting.
.PP
The text-to-binary error-reporting convention lends itself
to slightly obscure code,
because many readers will not think of NULL as signifying success.
A good way to make it clearer is to write something like:
.PP
.RS
.nf
.B "const char *error;"
.sp
.B "error = ttoaddr( /* ... */ );"
.B "if (error != NULL) {"
.B "        /* something went wrong */"
.fi
.RE