% This LaTeX document was generated using the LaTeX backend of PlDoc,
% The SWI-Prolog documentation system



\section{library(socket): Network socket (TCP and UDP) library}

\label{sec:socket}

The \file{library(socket)} provides TCP and UDP inet-domain sockets from
SWI-Prolog, both client and server-side communication. The interface of
this library is very close to the Unix socket interface, also supported
by the MS-Windows \textit{winsock} API. SWI-Prolog applications that wish to
communicate with multiple sources have two options:

\begin{itemize}
    \item Use I/O multiplexing based on \predref{wait_for_input}{3}. On Windows
systems this can only be used for sockets, not for general
(device-) file handles.
    \item Use multiple threads, handling either a single blocking socket
or a pool using I/O multiplexing as above.
\end{itemize}

\subsection{Client applications}

\label{sec:socket-server}

Using this library to establish a TCP connection to a server is as
simple as opening a file. See also \predref{http_open}{3}.

\begin{code}
dump_swi_homepage :-
    setup_call_cleanup(
        tcp_connect('www.swi-prolog.org':http, Stream, []),
        ( format(Stream,
                 'GET / HTTP/1.1~n\c
                  Host: www.swi-prolog.org~n\c
                  Connection: close~n~n', []),
          flush_output(Stream),
          copy_stream_data(Stream, current_output)
        ),
        close(Stream)).
\end{code}

To deal with timeouts and multiple connections, threads,
\predref{wait_for_input}{3} and/or non-blocking streams (see \predref{tcp_fcntl}{3}) can be
used.

\subsection{Server applications}

\label{sec:socket-client}

The typical sequence for generating a server application is given below.
To close the server, use \predref{close}{1} on \arg{AcceptFd}.

\begin{code}
create_server(Port) :-
      tcp_socket(Socket),
      tcp_bind(Socket, Port),
      tcp_listen(Socket, 5),
      tcp_open_socket(Socket, AcceptFd, _),
      <dispatch>
\end{code}

There are various options for $<$dispatch$>$. The most commonly used option
is to start a Prolog thread to handle the connection. Alternatively,
input from multiple clients can be handled in a single thread by
listening to these clients using \predref{wait_for_input}{3}. Finally, on Unix
systems, we can use \predref{fork}{1} to handle the connection in a new process.
Note that \predref{fork}{1} and threads do not cooperate well. Combinations can be
realised but require good understanding of POSIX thread and
fork-semantics.

Below is the typical example using a thread. Note the use of
\predref{setup_call_cleanup}{3} to guarantee that all resources are reclaimed, also
in case of failure or exceptions.

\begin{code}
dispatch(AcceptFd) :-
        tcp_accept(AcceptFd, Socket, Peer),
        thread_create(process_client(Socket, Peer), _,
                      [ detached(true)
                      ]),
        dispatch(AcceptFd).

process_client(Socket, Peer) :-
        setup_call_cleanup(
            tcp_open_socket(Socket, StreamPair),
            handle_service(StreamPair),
            close(StreamPair)).

handle_service(StreamPair) :-
        ...
\end{code}

\subsection{Socket exceptions}

\label{sec:socket-exceptions}

Errors that are trapped by the low-level library are mapped to an
exception of the shape below. In this term, \arg{Code} is a lower case atom
that corresponds to the C macro name, e.g., \const{epipe} for a broken pipe.
\arg{Message} is the human readable string for the error code returned by
the OS or the same as \arg{Code} if the OS does not provide this
functionality. Note that \arg{Code} is derived from a static set of macros
that may or may not be defines for the target OS. If the macro name is
not known, \arg{Code} is \verb$ERROR_nnn$, where \textit{nnn} is an integer.

\begin{code}
error(socket_error(Code, Message), _)
\end{code}

Note that on Windows \arg{Code} is a \verb$wsa*$ code which makes it hard to
write portable code that handles specific socket errors. Even on POSIX
systems the exact set of errors produced by the network stack is not
defined.

\subsection{Socket addresses (families)}

\label{sec:socket-domains}

The library supports both IP4 and IP6 addresses. On Unix systems it also
supports \textit{Unix domain sockets} (\verb$AF_UNIX$). The address of a Unix
domain sockets is a file name. Unix domain sockets are created using
\predref{socket_create}{2} or \predref{unix_domain_socket}{1}.

IP4 or IP6 sockets can be created using \predref{socket_create}{2} or \predref{tcp_connect}{3}
with the \const{inet} (default, IP3) or \const{inet6} domain option. Some of the
predicates produce or consume IP addresses as a Prolog term. The format
of this term is one of:

\begin{description}
    \termitem{ip}{A, B, C, D}
Represents an IP4 address. Each field is an integer in the range
0..255 (8 bit).
    \termitem{ip}{A, B, C, D, E, F, G, H}
Represents an IP6 address. Each field is an integer in the range
0..65535 (16 bit).
\end{description}

The predicate \predref{ip_name}{2} translates between the canonical textual
representation and the above defined address terms.

\subsection{Socket predicate reference}

\label{sec:socket-predicates}\vspace{0.7cm}

\begin{description}
    \predicate[det]{socket_create}{2}{-SocketId, +Options}
Create a socket according to \arg{Options}. Supported \arg{Options} are:

\begin{description}
    \termitem{domain}{+Domain}
One of \const{inet} (default), \const{inet6}, \const{unix} or \const{local} (same
as \const{unix})
    \termitem{type}{+Type}
One of \const{stream} (default) to create a TCP connection or
\const{dgram} to create a UDP socket.
\end{description}

This predicate subsumes tcp_socket/1m, \predref{udp_socket}{1} and
\predref{unix_domain_socket}{1}.

    \predicate[det]{tcp_socket}{1}{-SocketId}
Equivalent to \verb$socket_create(SocketId, [])$ or, explicit,
\verb$socket_create(SocketId, [domain(inet), type(stream)])$.

    \predicate[det]{unix_domain_socket}{1}{-SocketId}
Equivalent to \verb$socket_create(SocketId, [domain(unix)])$ or,
explicit, \verb$socket_create(SocketId, [domain(unix), type(stream)])$

Unix domain socket affect \predref{tcp_connect}{2} (for clients) and
\predref{tcp_bind}{2} and \predref{tcp_accept}{3} (for servers). The address is an atom
or string that is handled as a file name. On most systems the
length of this file name is limited to 128 bytes (including null
terminator), but according to the Linux documentation (\verb$unix(7)$),
portable applications must keep the address below 92 bytes. Note
that these lengths are in bytes. Non-ascii characters may be
represented as multiple bytes. If the length limit is exceeded a
\verb$representation_error(af_unix_name)$ exception is raised.

    \predicate[det]{tcp_close_socket}{1}{+SocketId}
Closes the indicated socket, making \arg{SocketId} invalid. Normally,
sockets are closed by closing both stream handles returned by
\predref{open_socket}{3}. There are two cases where \predref{tcp_close_socket}{1} is
used because there are no stream-handles:

\begin{itemize}
    \item If, after \predref{tcp_accept}{3}, the server uses \predref{fork}{1} to handle the
client in a sub-process. In this case the accepted socket is
not longer needed from the main server and must be discarded
using \predref{tcp_close_socket}{1}.
    \item If, after discovering the connecting client with
\predref{tcp_accept}{3}, the server does not want to accept the
connection, it should discard the accepted socket
immediately using \predref{tcp_close_socket}{1}.
\end{itemize}

    \predicate[det]{tcp_open_socket}{2}{+SocketId, -StreamPair}
Create streams to communicate to \arg{SocketId}. If \arg{SocketId} is a
master socket (see \predref{tcp_bind}{2}), \arg{StreamPair} should be used for
\predref{tcp_accept}{3}. If \arg{SocketId} is a connected (see \predref{tcp_connect}{2}) or
accepted socket (see \predref{tcp_accept}{3}), \arg{StreamPair} is unified to a
stream pair (see \predref{stream_pair}{3}) that can be used for reading and
writing. The stream or pair must be closed with \predref{close}{1}, which
also closes \arg{SocketId}.

    \predicate[det]{tcp_open_socket}{3}{+SocketId, -InStream, -OutStream}
Similar to \predref{tcp_open_socket}{2}, but creates two separate sockets
where \predref{tcp_open_socket}{2} would have created a stream pair.

\begin{tags}
    \tag{deprecated}
New code should use \predref{tcp_open_socket}{2} because
closing a stream pair is much easier to perform safely.
\end{tags}

    \predicate[det]{tcp_bind}{2}{SocketId, ?Address}
Bind the socket to \arg{Address} on the current machine. This
operation, together with \predref{tcp_listen}{2} and \predref{tcp_accept}{3} implement
the \textit{server-side} of the socket interface. \arg{Address} is either an
plain \arg{Port} or a term HostPort. The first form binds the socket
to the given port on all interfaces, while the second only binds
to the matching interface. A typical example is below, causing
the socket to listen only on port 8080 on the local machine's
network.

\begin{code}
  tcp_bind(Socket, localhost:8080)
\end{code}

If \arg{Port} is unbound, the system picks an arbitrary free port
and unifies \arg{Port} with the selected port number. \arg{Port} is
either an integer or the name of a registered service. See also
\predref{tcp_connect}{4}.

    \predicate[det]{tcp_listen}{2}{+SocketId, +BackLog}
Tells, after \predref{tcp_bind}{2}, the socket to listen for incoming
requests for connections. Backlog indicates how many pending
connection requests are allowed. Pending requests are requests
that are not yet acknowledged using \predref{tcp_accept}{3}. If the
indicated number is exceeded, the requesting client will be
signalled that the service is currently not available. A
commonly used default value for Backlog is 5.

    \predicate[det]{tcp_accept}{3}{+Socket, -Slave, -Peer}
This predicate waits on a server socket for a connection request by
a client. On success, it creates a new socket for the client and
binds the identifier to \arg{Slave}. \arg{Peer} is bound to the IP-address of
the client or the atom \verb$af_unix$ if \arg{Socket} is an AF_UNIX socket (see
\predref{unix_domain_socket}{1}).

    \predicate[det]{tcp_connect}{2}{+SocketId, +Address}
Connect \arg{SocketId}. After successful completion, \predref{tcp_open_socket}{3}
can be used to create I/O-Streams to the remote socket. This
predicate is part of the low level client API. A connection to a
particular host and port is realised using these steps:

\begin{code}
    tcp_socket(Socket),
    tcp_connect(Socket, Host:Port),
    tcp_open_socket(Socket, StreamPair)
\end{code}

Typical client applications should use the high level interface
provided by \predref{tcp_connect}{3} which avoids resource leaking if a
step in the process fails, and can be hooked to support proxies.
For example:

\begin{code}
    setup_call_cleanup(
        tcp_connect(Host:Port, StreamPair, []),
        talk(StreamPair),
        close(StreamPair))
\end{code}

If \arg{SocketId} is an AF_UNIX socket (see \predref{unix_domain_socket}{1}), \arg{Address}
is an atom or string denoting a file name.

    \predicate[det]{tcp_connect}{4}{+Socket, +Address, -Read, -Write}
Connect a (client) socket to \arg{Address} and return a bi-directional
connection through the stream-handles \arg{Read} and \arg{Write}. This
predicate may be hooked by defining \qpredref{socket}{tcp_connect_hook}{4}
with the same signature. Hooking can be used to deal with proxy
connections. E.g.,

\begin{code}
:- multifile socket:tcp_connect_hook/4.

socket:tcp_connect_hook(Socket, Address, Read, Write) :-
    proxy(ProxyAdress),
    tcp_connect(Socket, ProxyAdress),
    tcp_open_socket(Socket, Read, Write),
    proxy_connect(Address, Read, Write).
\end{code}

\begin{tags}
    \tag{deprecated}
New code should use \predref{tcp_connect}{3} called as
\verb$tcp_connect(+Address, -StreamPair, +Options)$.
\end{tags}

    \predicate[det]{tcp_connect}{3}{+Address, -StreamPair, +Options}
\nodescription
    \predicate[det]{tcp_connect}{3}{+Socket, +Address, -StreamPair}
Establish a TCP communication as a client. The +,-,+ mode is the
preferred way for a client to establish a connection. This predicate
can be hooked to support network proxies. To use a proxy, the hook
\predref{proxy_for_url}{3} must be defined. Permitted options are:

\begin{description}
    \termitem{bypass_proxy}{+Boolean}
Defaults to \const{false}. If \const{true}, do not attempt to use any
proxies to obtain the connection
    \termitem{nodelay}{+Boolean}
Defaults to \const{false}. If \const{true}, set nodelay on the
resulting socket using \verb$tcp_setopt(Socket, nodelay)$
    \termitem{domain}{+Domain}
One of `inet' or \const{inet6}. When omitted we use \predref{host_address}{2}
with \verb$type(stream)$ and try the returned addresses in order.
\end{description}

The +,+,- mode is deprecated and does not support proxies. It
behaves like \predref{tcp_connect}{4}, but creates a stream pair (see
\predref{stream_pair}{3}).

\begin{arguments}
\arg{Address} & is either a Host:Port term or a file name (atom or
string). The latter connects to an AF_UNIX socket and requires
\predref{unix_domain_socket}{1}. \\
\end{arguments}

\begin{tags}
    \tag{Errors}
\verb$proxy_error(tried(ResultList))$ is raised by mode (+,-,+) if
proxies are defines by \predref{proxy_for_url}{3} but no proxy can establsh the
connection. \arg{ResultList} contains one or more terms of the form
\verb$false(Proxy)$ for a hook that simply failed or \verb$error(Proxy, ErrorTerm)$ for a hook that raised an exception.
    \tag{See also}
\file{library(http/http_proxy)} defines a hook that allows to connect
through HTTP proxies that support the \verb$CONNECT$ method.
\end{tags}

    \predicate{tcp_select}{3}{+ListOfStreams, -ReadyList, +TimeOut}
Same as the built-in \predref{wait_for_input}{3}. Used to allow for interrupts
and timeouts on Windows. A redesign of the Windows socket interface
makes it impossible to do better than Windows \verb$select()$ call
underlying \predref{wait_for_input}{3}. As input multiplexing typically happens
in a background thread anyway we accept the loss of timeouts and
interrupts.

\begin{tags}
    \tag{deprecated}
Use \predref{wait_for_input}{3}
\end{tags}

    \predicate[semidet,multifile]{try_proxy}{4}{+Proxy, +TargetAddress, -Socket, -StreamPair}
Attempt a socket-level connection via the given proxy to
\arg{TargetAddress}. The \arg{Proxy} argument must match the output argument
of \predref{proxy_for_url}{3}. The predicate \predref{tcp_connect}{3} (and \predref{http_open}{3}
from the \file{library(http/http_open)}) collect the results of failed
proxies and raise an exception no proxy is capable of realizing
the connection.

The default implementation recognises the values for \arg{Proxy}
described below. The \file{library(http/http_proxy)} adds
\verb$proxy(Host,Port)$ which allows for HTTP proxies using the
\verb$CONNECT$ method.

\begin{description}
    \termitem{direct}{}
Do not use any proxy
    \termitem{socks}{Host, Port}
Use a SOCKS5 proxy
\end{description}

    \predicate[nondet,multifile]{proxy_for_url}{3}{+URL, +Hostname, -Proxy}
This hook can be implemented to return a proxy to try when
connecting to \arg{URL}. Returned proxies are tried in the order in
which they are returned by the multifile hook \predref{try_proxy}{4}.
Pre-defined proxy methods are:

\begin{description}
    \termitem{direct}{}
connect directly to the resource
    \termitem{proxy}{Host, Port}
Connect to the resource using an HTTP proxy. If the
resource is not an HTTP \arg{URL}, then try to connect using the
CONNECT verb, otherwise, use the GET verb.
    \termitem{socks}{Host, Port}
Connect to the resource via a SOCKS5 proxy
\end{description}

These correspond to the proxy methods defined by PAC \href{http://en.wikipedia.org/wiki/Proxy_auto-config}{\arg{Proxy}
auto-config}.
Additional methods can be returned if suitable clauses for
\qpredref{http}{http_connection_over_proxy}{6} or \predref{try_proxy}{4} are defined.

    \predicate[det]{udp_socket}{1}{-SocketId}
Equivalent to \verb$socket_create(SocketId, [type(dgram)])$ or, explicit,
\verb$socket_create(SocketId, [domain(inet), type(dgram)])$.

    \predicate[det]{udp_receive}{4}{+Socket, -Data, -From, +Options}
Wait for and return the next datagram. The \arg{Data} is returned as a
Prolog term depending on \arg{Options}. \arg{From} is a term of the format
Ip:Port indicating the sender of the message. Here, \arg{Ip} is either
an ip4 or ip6 structure. \arg{Socket} can be waited for using
\predref{wait_for_input}{3}. Defined \arg{Options}:

\begin{description}
    \termitem{as}{+Type}
Defines the type for \arg{Data}. Possible values are \const{atom}, \const{codes},
\const{string} (default) or \const{term} (parse as Prolog term).
    \termitem{encoding}{+Encoding}
Specify the encoding used to interpret the message. It is one of
\const{octet}. \verb$iso_latin_1$, \const{text} or \const{utf8}.
    \termitem{max_message_size}{+Size}
Specify the maximum number of bytes to read from a UDP
datagram. \arg{Size} must be within the range 0-65535. If unspecified,
a maximum of 4096 bytes will be read.
\end{description}

For example:

\begin{code}
receive(Port) :-
    udp_socket(Socket),
    tcp_bind(Socket, Port),
    repeat,
        udp_receive(Socket, Data, From, [as(atom)]),
        format('Got ~q from ~q~n', [Data, From]),
        fail.
\end{code}

    \predicate[det]{udp_send}{4}{+Socket, +Data, +To, +Options}
Send a UDP message. \arg{Data} is a string, atom or code-list providing
the data. \arg{To} is an address of the form Host:Port where Host is
either the hostname or an IP address. Defined \arg{Options} are:

\begin{description}
    \termitem{encoding}{+Encoding}
Specifies the encoding to use for the string. See
\predref{udp_receive}{4} for details
    \termitem{as}{+Type}
This uses the same values for \arg{Type} as the \verb$as(Type)$ option of
\predref{udp_receive}{4}. The are interpreted differently though. No \arg{Type}
corresponds to CVT_ALL of PL_get_chars(). Using atom
corresponds to CVT_ATOM and any of string or codes is mapped
to CVT_STRING\Sbar{}CVT_LIST, allowing for a SWI-Prolog string
object, list of character codes or list of characters.
Finally, \const{term} maps to CVT_WRITE_CANONICAL. This implies that
arbitrary Prolog terms can be sent reliably using the option
list `[\verb$as(term)$,\verb$encoding(utf8)$])`, using the same option list
for \predref{udp_receive}{4}.
\end{description}

For example

\begin{code}
send(Host, Port, Message) :-
    udp_socket(S),
    udp_send(S, Message, Host:Port, []),
    tcp_close_socket(S).
\end{code}

A broadcast is achieved by using \verb$tcp_setopt(Socket, broadcast)$
prior to sending the datagram and using the local network
broadcast address as a \predref{ip}{4} term.

    \predicate[det]{tcp_setopt}{2}{+SocketId, +Option}
Set options on the socket. Defined options are:

\begin{description}
    \termitem{reuseaddr}{}
Allow servers to reuse a port without the system being
completely sure the port is no longer in use.
    \termitem{bindtodevice}{+Device}
Bind the socket to \arg{Device} (an atom). For example, the code
below binds the socket to the \textit{loopback} device that is
typically used to realise the \textit{localhost}. See the manual
pages for \verb$setsockopt()$ and the socket interface (e.g.,
\verb$socket(7)$ on Linux) for details.

\begin{code}
tcp_socket(Socket),
tcp_setopt(Socket, bindtodevice(lo))
\end{code}

    \termitem{nodelay}{}
    \termitem{nodelay}{true}
If \const{true}, disable the Nagle optimization on this socket,
which is enabled by default on almost all modern TCP/IP
stacks. The Nagle optimization joins small packages, which is
generally desirable, but sometimes not. Please note that the
underlying TCP_NODELAY setting to \verb$setsockopt()$ is not
available on all platforms and systems may require additional
privileges to change this option. If the option is not
supported, \predref{tcp_setopt}{2} raises a domain_error exception. See
\href{http://en.wikipedia.org/wiki/Nagle's_algorithm}{Wikipedia}
for details.
    \termitem{broadcast}{}
UDP sockets only: broadcast the package to all addresses
matching the address. The address is normally the address of
the local subnet (i.e. 192.168.1.255). See \predref{udp_send}{4}.
    \termitem{ip_add_membership}{+MultiCastGroup}
    \termitem{ip_add_membership}{+MultiCastGroup, +LocalInterface}
    \termitem{ip_add_membership}{+MultiCastGroup, +LocalInterface, +InterfaceIndex}
    \termitem{ip_drop_membership}{+MultiCastGroup}
    \termitem{ip_drop_membership}{+MultiCastGroup, +LocalInterface}
    \termitem{ip_drop_membership}{+MultiCastGroup, +LocalInterface, +InterfaceIndex}
Join/leave a multicast group. Calls \verb$setsockopt()$ with the
corresponding arguments.
    \termitem{dispatch}{+Boolean}
In GUI environments (using XPCE or the Windows \verb$swipl-win.exe$
executable) this flags defines whether or not any events are
dispatched on behalf of the user interface. Default is
\const{true}. Only very specific situations require setting
this to \const{false}.
    \termitem{sndbuf}{+Integer}
Sets the send buffer size to \arg{Integer} (bytes). On Windows this defaults
(now) to 64kb. Higher latency links may benefit from increasing this
further since the maximum theoretical throughput on a link is given by
buffer-size / latency.
See \url{https://support.microsoft.com/en-gb/help/823764/slow-performance-occurs-when-you-copy-data-to-a-tcp-server-by-using-a}
for Microsoft's discussion
\end{description}

    \predicate[det]{tcp_fcntl}{3}{+Stream, +Action, ?Argument}
Interface to the \verb$fcntl()$ call. Currently only suitable to deal
switch stream to non-blocking mode using:

\begin{code}
  tcp_fcntl(Stream, setfl, nonblock),
\end{code}

An attempt to read from a non-blocking stream while there is no
data available returns -1 (or \verb$end_of_file$ for \predref{read}{1}), but
\predref{at_end_of_stream}{1} fails. On actual end-of-input,
\predref{at_end_of_stream}{1} succeeds.

    \predicate[semidet]{tcp_getopt}{2}{+Socket, ?Option}
Get information about \arg{Socket}. Defined properties are below.
Requesting an unknown option results in a \verb$domain_error$ exception.

\begin{description}
    \termitem{file_no}{-File}
Get the OS file handle as an integer. This may be used for
debugging and integration.
\end{description}

    \predicate[nondet]{host_address}{3}{+HostName, -Address, +Options}
\nodescription
    \predicate[det]{host_address}{3}{-HostName, +Address, +Options}
Translate between a machines host-name and it's (IP-)address.
Supported options:

\begin{description}
    \termitem{domain}{+Domain}
One of \const{inet} or \const{inet6} to limit the results to the given
family.
    \termitem{type}{+Type}
One of \const{stream} or \const{dgram}.
    \termitem{canonname}{+Boolean}
If \const{true} (default \const{false}), return the canonical host name
in the frist answer
\end{description}

In mode (+,-,+) \arg{Address} is unified to a dict with the following keys:

\begin{description}
    \termitem{address}{}
A Prolog terms describing the ip address.
    \termitem{domain}{}
One of \const{inet} or \const{inet6}. The underlying \verb$getaddrinfo()$ calls
this \const{family}. We use \const{domain} for consistency with
\predref{socket_create}{2}.
    \termitem{type}{}
Currently one of \const{stream} or \const{dgram}.
    \termitem{host}{}
Available if \verb$canonname(true)$ is specified on the first
returned address. Holds the official canonical host name.
\end{description}

    \predicate[det]{tcp_host_to_address}{2}{?HostName, ?Address}
Translate between a machines host-name and it's (IP-)address. If
\arg{HostName} is an atom, it is resolved using \verb$getaddrinfo()$ and the
IP-number is unified to \arg{Address} using a term of the format
\verb$ip(Byte1,Byte2,Byte3,Byte4)$. Otherwise, if \arg{Address} is bound to an
\verb$ip(Byte1,Byte2,Byte3,Byte4)$ term, it is resolved by \verb$gethostbyaddr()$
and the canonical hostname is unified with \arg{HostName}.

\begin{tags}
    \tag{deprecated}
New code should use \predref{host_address}{3}. This version is
bootstrapped from \predref{host_address}{3} and only searches for IP4 addresses
that support TCP connections.
\end{tags}

    \predicate[det]{gethostname}{1}{-Hostname}
Return the canonical fully qualified name of this host. This is
achieved by calling \verb$gethostname()$ and return the canonical name
returned by \verb$getaddrinfo()$.

    \predicate[det]{ip_name}{2}{?IP, ?Name}
Translate between the textual representation of an \arg{IP} address and
the Prolog data structure. Prolog represents ip4 addresses as
\verb$ip(A,B,C,D)$ and ip6 addresses as \verb$ip(A,B,C,D,E,F,H)$. For example:

\begin{code}
?- ip_name(ip(1,2,3,4), Name)
Name = '1.2.3.4'.
?- ip_name(IP, '::').
IP = ip(0,0,0,0,0,0,0,0).
?- ip_name(IP, '1:2::3').
IP = ip(1,2,0,0,0,0,0,3).
\end{code}

    \predicate[det]{negotiate_socks_connection}{2}{+DesiredEndpoint, +StreamPair}
Negotiate a connection to \arg{DesiredEndpoint} over \arg{StreamPair}.
\arg{DesiredEndpoint} should be in the form of either:

\begin{shortlist}
    \item hostname : port
    \item \verb$ip(A,B,C,D)$ : port
\end{shortlist}

\begin{tags}
    \tag{Errors}
\verb$socks_error(Details)$ if the SOCKS negotiation failed.
\end{tags}
\end{description}