- Reference manual
- SWI-Prolog binding to libarchive
- SWI-Prolog BerkeleyDB interface
- SWI-Prolog C-library
- library(process): Create processes and redirect I/O
- library(filesex): Extended operations on files
- library(uid): User and group management on Unix systems
- library(syslog): Unix syslog interface
- library(socket): Network socket (TCP and UDP) library
- The stream_pool library
- library(uri): Process URIs
- CGI Support library
- Password encryption library
- library(uuid): Universally Unique Identifier (UUID) Library
- SHA* Secure Hash Algorithms
- library(md5): MD5 hashes
- library(hash_stream): Maintain a hash on a stream
- Memory files
- Time and alarm library
- library(unix): Unix specific operations
- Limiting process resources
- library(udp_broadcast): A UDP Broadcast Bridge
- library(prolog_stream): A stream with Prolog callbacks
- Constraint Query Language A high level interface to SQL databases
- SWI-Prolog HTTP support
- JPL: A bidirectional Prolog/Java interface
- SWI-Prolog binding to BSD libedit
- SWI-Prolog Natural Language Processing Primitives
- SWI-Prolog ODBC Interface
- SWI-Prolog Regular Expression library
- SWI-Prolog support for PDT
- Pengines: Web Logic Programming Made Easy
- A C++ interface to SWI-Prolog
- SWI-Prolog Source Documentation Version 2
- Prolog Unit Tests
- Google's Protocol Buffers Library
- SWI-Prolog RDF parser
- SWI-Prolog binding to GNU readline
- SWI-Prolog Semantic Web Library 3.0
- SWI-Prolog SGML/XML parser
- SWI-Prolog SSL Interface
- Managing external tables for SWI-Prolog
- Transparent Inter-Process Communications (TIPC) libraries
- SWI-Prolog Unicode library
- SWI-Prolog binding to zlib
VU University of Amsterdam
On Windows systems, the
library(unix) library can only
be used if the whole SWI-Prolog suite is compiled using
Cygwin. The other
libraries have been ported to native Windows.
Many useful facilities offered by one or more of the operating systems supported by SWI-Prolog are not supported by the SWI-Prolog kernel distribution. Including these would enlarge the footprint and complicate portability matters while supporting only a limited part of the user-community.
This document describes
library(unix) to deal with the
Unix process API,
library(socket) to deal with inet-domain TCP and UDP
library(cgi) to deal with getting CGI form-data if
SWI-Prolog is used as a CGI scripting language,
to provide password encryption and verification,
providing cryptographic hash functions and
library(memfile) providing in-memorty pseudo files.
- SICStus 4
- To be done
- Implement detached option in process_create/3
library(process) implements interaction with
child processes and unifies older interfaces such as shell/[1,2],
etc. This library is modelled after SICStus 4.
In addition to the predicates, this module defines a file search path
(see user:file_search_path/2 and absolute_file_name/3)
path that locates files on the system's search path
for executables. E.g. the following finds the executable for
?- absolute_file_name(path(ls), Path, [access(execute)]).
Incompatibilities and current limitations
- Where SICStus distinguishes between an internal process id and the OS process id, this implementation does not make this distinction. This implies that is_process/1 is incomplete and unreliable.
- SICStus only supports ISO 8859-1 (latin-1). This implementation supports arbitrary OS multibyte interaction using the default locale.
- It is unclear what the
detached(true)option is supposed to do. Disable signals in the child? Use
setsid()to detach from the session? The current implementation uses
setsid()on Unix systems.
- An extra option
env([Name=Value, ...])is added to process_create/3.
- [det]process_create(+Exe, +Args:list, +Options)
- Create a new process running the file Exe and using arguments
from the given list. Exe is a file specification as handed to
absolute_file_name/3. Typically one use
pathfile alias to specify an executable file on the current PATH. Args is a list of arguments that are handed to the new process. On Unix systems, each element in the list becomes a seperate argument in the new process. In Windows, the arguments are simply concatenated to form the commandline. Each argument itself is either a primitive or a list of primitives. A primitive is either atomic or a term
file(Spec), the system inserts a filename using the OS filename conventions which is properly quoted if needed.
- Bind the standard streams of the new process. Spec is one of
the terms below. If
pipe(Pipe)is used, the Prolog stream is a stream in text-mode using the encoding of the default locale. The encoding can be changed using set_stream/2. The options
stderrmay use the same stream, in which case both output streams are connected to the same Prolog stream.
- Just share with the Prolog I/O streams
- Bind to a null stream. Reading from such a stream returns end-of-file, writing produces no output
- Attach input and/or output to a Prolog stream.
- Run the new process in Directory. Directory can be a compound specification, which is converted using absolute_file_name/3.
- Specify the environment for the new process. List is a list of Name=Value terms. Note that the current implementation does not pass any environment variables. If unspecified, the environment is inherited from the Prolog process.
- Unify PID with the process id of the created process.
- In Unix: If
true, detach the process from the terminal Currently mapped to
setsid(); Also creates a new process group for the child In Windows: If
true, detach the process from the current job via the CREATE_BREAKAWAY_FROM_JOB flag. In Vista and beyond, processes launched from the shell directly have the 'compatibility assistant' attached to them automatically unless they have a UAC manifest embedded in them. This means that you will get a permission denied error if you try and assign the newly-created PID to a job you create yourself.
true, create a window for the process (Windows only)
- In Unix: specifies the process priority for the newly created process. Priority must be an integer between -20 and 19. Positive values are nicer to others, and negative values are less so. The default is zero. Users are free to lower their own priority. Only the super-user may raise it to less-than zero.
If the user specifies the
process(-PID)option, he must call process_wait/2 to reclaim the process. Without this option, the system will wait for completion of the process after the last pipe stream is closed.
If the process is not waited for, it must succeed with status 0. If not, an process_error is raised.
On Windows this call is an interface to the CreateProcess() API. The commandline consists of the basename of Exe and the arguments formed from Args. Arguments are separated by a single space. If all characters satisfy
iswalnum()it is unquoted. If the argument contains a double-quote it is quoted using single quotes. If both single and double quotes appear a domain_error is raised, otherwise double-quote are used.
The CreateProcess() API has many options. Currently only the
CREATE_NO_WINDOWoptions is supported through the
window(+Bool)option. If omitted, the default is to use this option if the application has no console. Future versions are likely to support more window specific options and replace win_exec/2.
First, a very simple example that behaves the same as
shell('ls -l'), except for error handling:
?- process_create(path(ls), ['-l'], ).
The following example uses grep to find all matching lines in a file.
grep(File, Pattern, Lines) :- setup_call_cleanup( process_create(path(grep), [ Pattern, file(File) ], [ stdout(pipe(Out)) ]), read_lines(Out, Lines), close(Out)). read_lines(Out, Lines) :- read_line_to_codes(Out, Line1), read_lines(Line1, Out, Lines). read_lines(end_of_file, _, ) :- !. read_lines(Codes, Out, [Line|Lines]) :- atom_codes(Line, Codes), read_line_to_codes(Out, Line2), read_lines(Line2, Out, Lines).
process_error(Exe, Status)where Status is one of
killed(Signal). Raised if the process is waited for (i.e., Options does not include
process(-PID)), and does not exit with status 0.
- True if PID is the process id of the running Prolog process.
- [det]process_id(+Process, -PID)
- PID is the process id of Process. Given that they are united in SWI-Prolog, this is a simple unify.
- True if PID might be a process. Succeeds for any positive integer.
- Release process handle. In this implementation this is the same as
- [det]process_wait(+PID, -Status)
- [det]process_wait(+PID, -Status, +Options)
- True if PID completed with Status. This call
normally blocks until the process is finished. Options:
infinite. If this option is a number, the waits for a maximum of Timeout seconds and unifies Status with
timeoutif the process does not terminate within Timeout. In this case PID is not invalidated. On Unix systems only timeout 0 and
infiniteare supported. A 0-value can be used to poll the status of the process.
- Do/do not release the process. We do not support this flag and a
domain_error is raised if
Status is one of
killed(Signal), where Code and Signal are integers.
- [det]process_kill(+PID, +Signal)
- Send signal to process PID. Default is
term. Signal is an integer, Unix signal name (e.g.
SIGSTOP) or the more Prolog friendly variation one gets after removing
SIGand downcase the result:
stop. On Windows systems, Signal is ignored and the process is terminated using the TerminateProcess() API. On Windows systems PID must be obtained from process_create/3, while any PID is allowed on Unix systems.
- SICStus does not accept the prolog friendly version. We choose to do so for compatibility with on_signal/3.
- [det]process_group_kill(+PID, +Signal)
- Send signal to the group containing process PID. Default is
term. See process_wait/1 for a description of signal handling. In Windows, the same restriction on PID applies: it must have been created from process_create/3, and the the group is terminated via the TerminateJobObject API.
This module provides additional operations on files. This covers both more obscure and possible non-portable low-level operations and high-level utilities.
Using these Prolog primitives is typically to be preferred over using operating system primitives through shell/1 or process_create/3 because (1) there are no potential file name quoting issues, (2) there is no dependency on operating system commands and (3) using the implementations from this library is usually faster.
- [det]set_time_file(+File, -OldTimes, +NewTimes)
- Query and set POSIX time attributes of a file. Both OldTimes
NewTimes are lists of option-terms. Times are represented in
SWI-Prolog's standard floating point numbers. New times may be specified
nowto indicate the current time. Defined options are:
- Describes the time of last access of the file. This value can be read and written.
- Describes the time the contents of the file was last modified. This value can be read and written.
- Describes the time the file-structure itself was changed by adding (
link()) or removing (
Below are some example queries. The first retrieves the access-time, while the second sets the last-modified time to the current time.
?- set_time_file(foo, [access(Access)], ). ?- set_time_file(foo, , [modified(now)]).
- [det]link_file(+OldPath, +NewPath, +Type)
- Create a link in the filesystem from NewPath to OldPath. Type
defines the type of link and is one of
With some limitations, these functions also work on Windows. First of all, the unerlying filesystem must support links. This requires NTFS. Second, symbolic links are only supported in Vista and later.
domain_error(link_type, Type)if the requested link-type is unknown or not supported on the target OS.
- [det]relative_file_name(+Path:atom, +RelTo:atom, -RelPath:atom)
- [det]relative_file_name(-Path:atom, +RelTo:atom, +RelPath:atom)
- True when RelPath is Path, relative to RelTo. Path
and RelTo are first handed to absolute_file_name/2,
which makes the absolute
and canonical. Below are two examples:
?- relative_file_name('/home/janw/nice', '/home/janw/deep/dir/file', Path). Path = '../../nice'. ?- relative_file_name(Path, '/home/janw/deep/dir/file', '../../nice'). Path = '/home/janw/nice'.
All paths must be in canonical POSIX notation, i.e., using / to separate segments in the path. See prolog_to_os_filename/2.
- This predicate is defined as a syntactical operation.
- [det]directory_file_path(+Directory, +File, -Path)
- [det]directory_file_path(?Directory, ?File, +Path)
- True when Path is the full path-name for File in
Dir. This is comparable to
atom_concat(Directory, File, Path), but it ensures there is exactly one / between the two parts. Notes:
- [det]copy_file(From, To)
- Copy a file into a new file or directory. The data is copied as binary data.
- Create Dir and all required components (like mkdir -p). Can raise various file-specific exceptions.
- [det]copy_directory(+From, +To)
- Copy the contents of the directory From to To (recursively). If To is the name of an existing directory, the contents of From are copied into To. I.e., no subdirectory using the basename of From is created.
- Recursively remove the directory Dir and its contents. If Dir is a symbolic link or symbolic links inside Dir are encountered, the links are removed rather than their content. Use with care!
- Remove all content from directory Dir, without removing Dir itself. Similar to delete_directory_and_contents/2, if symbolic links are encountered in Dir, the links are removed rather than their content.
- [det]chmod(+File, +Spec)
- Set the mode of the target file. Spec is one of
-Modeor a plain Mode, which adds new permissions, revokes permissions or sets the exact permissions. Mode itself is an integer, a POSIX mode name or a list of POSIX mode names. Defines names are
svtxand the all names defined by the regular expression
[ugo]*[rwx]*. Specifying none of "ugo" is the same as specifying all of them. For example, to make a file executable for the owner (user) and group, we can use:
?- chmod(myfile, +ugx).
- See also
- Please check the documentation of your OS for details on the semantics of this predicates.
This module provides and interface to user and group information on Posix systems. In addition, it allows for changing user and group ids. When changing user and group settings for the calling process, bear in mind that:
- Changing user and groups of the calling process requires permission.
- The functions
initgroups()are not part of the POSIX standard and therefore the derived predicates may not be present.
- UID is the real user ID of the calling process.
- GID is the real group ID of the calling process.
- UID is the effective user ID of the calling process.
- GID is the effective group ID of the calling process.
- GroupsIDs is the set of supplementary group IDs of the calling process. Note that these are numeric identifiers. Use group_info/2 to obtain details on the returned group identifiers.
- [det]user_info(+User, -UserData)
- UserData represent the passwd information for User. User is either a numeric UID or a user name. The predicate user_data/3 can be used to extract information from UserData.
- user_data(?Field, ?UserData, ?Value)
- Value is the value for Field in UserData.
Defined fields are:
- Name of the user
- Password hash of the user (or
xif this is not accessible)
- Numeric user id of the user
- Numeric primary group id of the user
- The gecos field
- Home directory of the user
- Default (login) shell of the user.
- [det]group_info(+Group, -GroupData)
- GroupData represent the group information for Group. Group is either a numeric GID or a group name. The predicate group_data/3 can be used to extract information from GroupData.
- group_data(?Field, ?GroupData, ?Value)
- Value is the value for Field GroupData.
Defined fields are:
- Name of the user
- Password hash of the user (or
xif this is not accessible)
- Numeric group id of the group
- List of user-names that are member of this group.
- Set the user id of the calling process.
- Set the effective user id of the calling process.
- Set the group id of the calling process.
- Set the effective group id of the calling process.
- [det]initgroups(+User, +Group)
- Initialise the group access list of the calling process to the registered groups for User and the group Group. This predicate is only available if the underlying OS provides it.
- Set the group access list of the caling process to the indicated groups. This predicate is only available if the underlying OS provides it.
- [det]set_user_and_group(+User, +Group)
- Set the UID and GID to the User. User is either a UID or a user name. If Group is not specified, the primary group of User is used. If initgroups/2 is available, the resulting group access list of the calling process consists of the registered groups for User and the specified Group.
- See also
- - detach_IO/1 to detach normal
I/O of the process and remove it from the process group.
- fork/1 to create a daemon process.
library(uid)to manage user identifiers (e.g., drop root privileges).
This library provides an interface to the Unix
facility. The interface is an almost direct translation of the POSIX
syslog API, with two additions:
- syslog/3 exploits format/3 to format syslog messages
- The library integrates into
library(debug)using prolog:debug_print_hook/3, where debug topics are mapped to syslog priorities and remaining debug topics are mapped to the syslog priority
Note that this interface makes no attempt to abstract over logging
facilities of operating systems. We expect that such abstractions will
be implemented at the Prolog level using multiple integrations into
- [det]openlog(+Ident:atom, +Options:list(atom), +Facility:atom)
- Open system log. This predicate provides a direct interface into the
openlog()library call. If the library call is successful, it runs
at_halt(closelog)to ensure closing the system log on clean exit.
Ident prepended to every message, and is typically set to the program name. Options is a list of options. Values are corresponding C options, after removing =LOG_= and translation to lower case:
Facility is one of
- [det]syslog(+Priority, +Message)
- Send a message to the system log. Note that syslog/2
implicitly opens a connection to the system log if such a connection has
not been opened explicitly using openlog/3.
Priority is one of
- [det]syslog(+Priority, +Format, +Args)
- Send a formatted message to the system log if system logging is opened using openlog/3. This predicate combined format/3 with syslog/2. If there is no open syslog connection, syslog/3 calls print_message/2.
- Close the system log.
- [semidet,multifile]prolog:debug_print_hook(+Topic, +Format, +Args)
- Integration of debug/3 with the syslog
facility. If syslog is enabled, debug/3 is
re-routed to use the syslog facilities. If the topic of the debug
message matches one of the sylog
priority values (see syslog/2),
the message is sent with the corresponding syslog priority. Otherwise it
it sent with the
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 winsock API. SWI-Prolog
applications that wish to communicate with multiple sources have three
- Use I/O multiplexing based on wait_for_input/3. On Windows systems this can only be used for sockets, not for general (device-) file handles.
- Use multiple threads, handling either a single blocking socket or a pool using I/O multiplexing as above.
- Using XPCE's class
socketwhich synchronises socket events in the GUI event-loop.
Using this library to establish a TCP connection to a server is as simple as opening a file. See also http_open/3.
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(S)).
The typical sequence for generating a server application is given below. To close the server, use close/1 on AcceptFd.
create_server(Port) :- tcp_socket(Socket), tcp_bind(Socket, Port), tcp_listen(Socket, 5), tcp_open_socket(Socket, AcceptFd, _), <dispatch>
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 wait_for_input/3. Finally, on Unix systems, we can use fork/1 to handle the connection in a new process. Note that 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 setup_call_cleanup/3 to guarantee that all resources are reclaimed, also in case of failure or exceptions.
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(In, StreamPair), close(StreamPair)). handle_service(StreamPair) :- ...
- Creates an INET-domain stream-socket and unifies an identifier to it with SocketId. On MS-Windows, if the socket library is not yet initialised, this will also initialise the library.
- Closes the indicated socket, making SocketId invalid.
Normally, sockets are closed by closing both stream handles returned by
open_socket/3. There are two cases where tcp_close_socket/1
is used because there are no stream-handles:
- If, after tcp_accept/3, the server uses 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 tcp_close_socket/1.
- If, after discovering the connecting client with tcp_accept/3, the server does not want to accept the connection, it should discard the accepted socket immediately using tcp_close_socket/1.
- [det]tcp_open_socket(+SocketId, -StreamPair)
- Create streams to communicate to SocketId. If SocketId is a master socket (see tcp_bind/2), StreamPair should be used for tcp_accept/3. If SocketId is a connected (see tcp_connect/2) or accepted socket (see tcp_accept/3), StreamPair is unified to a stream pair (see stream_pair/3) that can be used for reading and writing. The stream or pair must be closed with close/1, which also closes SocketId.
- [det]tcp_open_socket(+SocketId, -InStream, -OutStream)
- Similar to tcp_open_socket/2,
but creates two separate sockets where tcp_open_socket/2
would have created a stream pair.
- New code should use tcp_open_socket/2 because closing a stream pair is much easier to perform safely.
- [det]tcp_bind(SocketId, ?Address)
- Bind the socket to Address on the current machine. This
operation, together with tcp_listen/2
and tcp_accept/3 implement the server-side
of the socket interface. Address is either an plain 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.
If Port is unbound, the system picks an arbitrary free port and unifies Port with the selected port number. Port is either an integer or the name of a registered service. See also tcp_connect/4.
- [det]tcp_listen(+SocketId, +BackLog)
- Tells, after 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 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.
- [det]tcp_accept(+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 Slave. Peer is bound to the IP-address of the client.
- [det]tcp_connect(+SocketId, +HostAndPort)
- Connect SocketId. After successful completion, 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:
tcp_socket(Socket), tcp_connect(Socket, Host:Port), tcp_open_socket(Socket, StreamPair)
Typical client applications should use the high level interface provided by tcp_connect/3 which avoids resource leaking if a step in the process fails, and can be hooked to support proxies. For example:
setup_call_cleanup( tcp_connect(Host:Port, StreamPair, ), talk(StreamPair), close(StreamPair))
- [det]tcp_connect(+Socket, +Address, -Read, -Write)
- Connect a (client) socket to Address and return a
bi-directional connection through the stream-handles Read and Write.
This predicate may be hooked by defining socket:tcp_connect_hook/4
with the same signature. Hooking can be used to deal with proxy
:- 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).
- New code should use tcp_connect/3
tcp_connect(+Address, -StreamPair, +Options).
- [det]tcp_connect(+Address, -StreamPair, +Options)
- [det]tcp_connect(+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 proxy_for_url/3
must be defined. Permitted options are:
- Defaults to
true, do not attempt to use any proxies to obtain the connection
- Defaults to
true, set nodelay on the resulting socket using
proxy_error(tried(ResultList))is raised by mode (+,-,+) if proxies are defines by proxy_for_url/3 but no proxy can establsh the connection. ResultList contains one or more terms of the form
false(Proxy)for a hook that simply failed or
error(Proxy, ErrorTerm)for a hook that raised an exception.
- See also
library(http/http_proxy)defines a hook that allows to connect through HTTP proxies that support the
- tcp_select(+ListOfStreams, -ReadyList, +TimeOut)
- Same as the built-in wait_for_input/3, but integrates better with event processing and the various options of sockets for Windows. On non-windows systems this simply calls wait_for_input/3.
- [semidet,multifile]try_proxy(+Proxy, +TargetAddress, -Socket, -StreamPair)
- Attempt a socket-level connection via the given proxy to
TargetAddress. The Proxy argument must match the
output argument of proxy_for_url/3.
The predicate tcp_connect/3
(and http_open/3 from the
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 Proxy described below. The
proxy(Host,Port)which allows for HTTP proxies using the
- Do not use any proxy
- socks(Host, Port)
- Use a SOCKS5 proxy
- [nondet,multifile]proxy_for_url(+URL, +Hostname, -Proxy)
- This hook can be implemented to return a proxy to try when connecting to URL.
Returned proxies are tried in the order in which they are returned by
the multifile hook try_proxy/4.
Pre-defined proxy methods are:
- connect directly to the resource
- proxy(Host, Port)
- Connect to the resource using an HTTP proxy. If the resource is not an HTTP URL, then try to connect using the CONNECT verb, otherwise, use the GET verb.
- socks(Host, Port)
- Connect to the resource via a SOCKS5 proxy
- [det]tcp_setopt(+SocketId, +Option)
- Set options on the socket. Defined options are:
- Allow servers to reuse a port without the system being completely sure the port is no longer in use.
- Bind the socket to Device (an atom). For example, the code
below binds the socket to the loopback device that is typically
used to realise the localhost. See the manual pages for
setsockopt()and the socket interface (e.g.,
socket(7)on Linux) for details.
tcp_socket(Socket), tcp_setopt(Socket, bindtodevice(lo))
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
setsockopt()is not available on all platforms and systems may require additional privileges to change this option. If the option is not supported, tcp_setopt/2 raises a domain_error exception. See Wikipedia for details.
- 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 udp_send/4.
- In GUI environments (using XPCE or the Windows
swipl-win.exeexecutable) this flags defines whether or not any events are dispatched on behalf of the user interface. Default is
true. Only very specific situations require setting this to
- [det]tcp_fcntl(+Stream, +Action, ?Argument)
- Interface to the
fcntl()call. Currently only suitable to deal switch stream to non-blocking mode using:
tcp_fcntl(Stream, setfl, nonblock),
An attempt to read from a non-blocking stream while there is no data available returns -1 (or
end_of_filefor read/1), but at_end_of_stream/1 fails. On actual end-of-input, at_end_of_stream/1 succeeds.
- [det]tcp_host_to_address(?HostName, ?Address)
- Translate between a machines host-name and it's (IP-)address. If
HostName is an atom, it is resolved using
getaddrinfo()and the IP-number is unified to Address using a term of the format
ip(Byte1,Byte2,Byte3,Byte4). Otherwise, if Address is bound to an
ip(Byte1,Byte2,Byte3,Byte4)term, it is resolved by
gethostbyaddr()and the canonical hostname is unified with HostName.
- To be done
- This function should support more functionality provided by gethostbyaddr, probably by adding an option-list.
- Return the canonical fully qualified name of this host. This is achieved
gethostname()and return the canonical name returned by
- [det]negotiate_socks_connection(+DesiredEndpoint, +StreamPair)
- Negotiate a connection to DesiredEndpoint over StreamPair.
DesiredEndpoint should be in the form of either:
- hostname : port
socks_error(Details)if the SOCKS negotiation failed.
The current library provides limited support for UDP packets. The UDP protocol is a connection-less and unreliable datagram based protocol. That means that messages sent may or may not arrive at the client side and may arrive in a different order as they are sent. UDP messages are often used for streaming media or for service discovery using the broadcasting mechanism.
- Similar to tcp_socket/1,
but create a socket using the
SOCK_DGRAMprotocol, ready for UDP connections.
- udp_receive(+Socket, -Data, -From, +Options)
- Wait for and return the next datagram. The data is returned as a Prolog
string object (see string_to_list/2). From
is a term of the format ip(A,B,C,D):Port
indicating the sender of the message. Socket can be waited
for using wait_for_input/3.
- Defines the returned term-type. Type is one of
- Specify the maximum number of bytes to read from a UDP datagram. Size must be within the range 0-65535. If unspecified, a maximum of 4096 bytes will be read.
The typical sequence to receive UDP data is:
receive(Port) :- udp_socket(S), tcp_bind(S, Port), repeat, udp_receive(Socket, Data, From, [as(atom)]), format('Got ~q from ~q~n', [Data, From]), fail.
- udp_send(+Socket, +Data, +To, +Options)
- Send a UDP message. Data is a string, atom or code-list providing the
data. To is an address of the form Host:Port
where Host is either the hostname or a term ip/4. Options
is currently unused.
A simple example to send UDP data is:
send(Host, Port, Message) :- udp_socket(S), udp_send(S, Message, Host:Port, ), tcp_close_socket(S).
A broadcast is achieved by using
tcp_setopt(Socket, broadcast)prior to sending the datagram and using the local network broadcast address as a ip/4 term.
The normal mechanism to discover a service on the local network is for the client to send a broadcast message to an agreed port. The server receives this message and replies to the client with a message indicating further details to establish the communication.
library(streampool) library dispatches input from
multiple streams based on wait_for_input/3.
It is part of the clib package as it is used most of the time together
library(socket) library. On non-Unix systems it
often can only be used with socket streams.
With SWI-Prolog 5.1.x, multi-threading often provides a good
alternative to using this library. In this schema one thread watches the
listening socket waiting for connections and either creates a thread per
connection or processes the accepted connections with a pool of
worker threads. The library
provides an example realising a mult-threaded HTTP server.
- add_stream_to_pool(+Stream, :Goal)
- Add Stream, which must be an input stream and ---on non-unix systems--- connected to a socket to the pool. If input is available on Stream, Goal is called.
- Delete the given stream from the pool. Succeeds, even if Stream is no member of the pool. If Stream is unbound the entire pool is emtied but unlike close_stream_pool/0 the streams are not closed.
- Empty the pool, closing all streams that are part of it.
- Wait for maximum of TimeOut for input on any of the streams
in the pool. If there is input, call the Goal associated with
If Goal fails or raises an exception a message is printed. TimeOut
is described with wait_for_input/3.
If Goal is called, there is some input on the associated stream. Goal must be careful not to block as this will block the entire pool.1This is hard to achieve at the moment as none of the Prolog read-commands provide for a timeout.
- Calls dispatch_stream_pool/1 in a loop until the pool is empty.
Below is a very simple example that reads the first line of input and echos it back.
:- use_module(library(streampool)). server(Port) :- tcp_socket(Socket), tcp_bind(Socket, Port), tcp_listen(Socket, 5), tcp_open_socket(Socket, In, _Out), add_stream_to_pool(In, accept(Socket)), stream_pool_main_loop. accept(Socket) :- tcp_accept(Socket, Slave, Peer), tcp_open_socket(Slave, In, Out), add_stream_to_pool(In, client(In, Out, Peer)). client(In, Out, _Peer) :- read_line_to_codes(In, Command), close(In), format(Out, 'Please to meet you: ~s~n', [Command]), close(Out), delete_stream_from_pool(In).
This library provides high-performance C-based primitives for manipulating URIs. We decided for a C-based implementation for the much better performance on raw character manipulation. Notably, URI handling primitives are used in time-critical parts of RDF processing. This implementation is based on RFC-3986:
The URI processing in this library is rather liberal. That is, we break URIs according to the rules, but we do not validate that the components are valid. Also, percent-decoding for IRIs is liberal. It first tries UTF-8; then ISO-Latin-1 and finally accepts %-characters verbatim.
Earlier experience has shown that strict enforcement of the URI syntax results in many errors that are accepted by many other web-document processing tools.
- [det]uri_components(+URI, -Components)
- [det]uri_components(-URI, +Components)
- Break a URI into its 5 basic components according to the
RFC-3986 regular expression:
^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))? 12 3 4 5 6 7 8 9
Components is a term
uri_components(Scheme, Authority, Path, Search, Fragment). If a URI is parsed, i.e., using mode (+,-), components that are not found are left uninstantiated (variable). See uri_data/3 for accessing this structure.
- [semidet]uri_data(?Field, +Components, ?Data)
- Provide access the uri_component structure. Defined field-names are:
- [semidet]uri_data(+Field, +Components, +Data, -NewComponents)
- NewComponents is the same as Components with Field set to Data.
- [det]uri_normalized(+URI, -NormalizedURI)
- NormalizedURI is the normalized form of URI.
Normalization is syntactic and involves the following steps:
- 126.96.36.199. Case Normalization
- 188.8.131.52. Percent-Encoding Normalization
- 184.108.40.206. Path Segment Normalization
- [det]iri_normalized(+IRI, -NormalizedIRI)
- NormalizedIRI is the normalized form of IRI.
Normalization is syntactic and involves the following steps:
- 220.127.116.11. Case Normalization
- 18.104.22.168. Path Segment Normalization
- See also
- This is similar to uri_normalized/2, but does not do normalization of %-escapes.
- [det]uri_normalized_iri(+URI, -NormalizedIRI)
- As uri_normalized/2, but
percent-encoding is translated into IRI Unicode characters. The
translation is liberal: valid UTF-8 sequences of %-encoded bytes are
mapped to the Unicode character. Other %XX-sequences are mapped to the
corresponding ISO-Latin-1 character and sole % characters are left
- See also
- True if URI has a scheme. The semantics is the same as the
code below, but the implementation is more efficient as it does not need
to parse the other components, nor needs to bind the scheme. The
condition to demand a scheme of more than one character as added to
avoid confusion with DOS path names.
uri_is_global(URI) :- uri_components(URI, Components), uri_data(scheme, Components, Scheme), nonvar(Scheme), atom_length(Scheme, Len), Len > 1.
- [det]uri_resolve(+URI, +Base, -GlobalURI)
- Resolve a possibly local URI relative to Base. This implements http://labs.apache.org/webarch/uri/rfc/rfc3986.html#relative-transform
- [det]uri_normalized(+URI, +Base, -NormalizedGlobalURI)
- NormalizedGlobalURI is the normalized global version of URI.
Behaves as if defined by:
uri_normalized(URI, Base, NormalizedGlobalURI) :- uri_resolve(URI, Base, GlobalURI), uri_normalized(GlobalURI, NormalizedGlobalURI).
- [det]iri_normalized(+IRI, +Base, -NormalizedGlobalIRI)
- NormalizedGlobalIRI is the normalized global version of IRI. This is similar to uri_normalized/3, but does not do %-escape normalization.
- [det]uri_normalized_iri(+URI, +Base, -NormalizedGlobalIRI)
- NormalizedGlobalIRI is the normalized global IRI of URI.
Behaves as if defined by:
uri_normalized(URI, Base, NormalizedGlobalIRI) :- uri_resolve(URI, Base, GlobalURI), uri_normalized_iri(GlobalURI, NormalizedGlobalIRI).
- [det]uri_query_components(+String, -Query)
- [det]uri_query_components(-String, +Query)
- Perform encoding and decoding of an URI query string. Query
is a list of fully decoded (Unicode) Name=Value pairs. In mode (-,+),
query elements of the forms Name(Value) and Name-Value are also accepted
to enhance interoperability with the option and pairs libraries. E.g.
?- uri_query_components(QS, [a=b, c('d+w'), n-'VU Amsterdam']). QS = 'a=b&c=d%2Bw&n=VU%20Amsterdam'. ?- uri_query_components('a=b&c=d%2Bw&n=VU%20Amsterdam', Q). Q = [a=b, c='d+w', n='VU Amsterdam'].
- [det]uri_authority_components(+Authority, -Components)
- [det]uri_authority_components(-Authority, +Components)
- Break-down the authority component of a URI. The fields of the structure Components can be accessed using uri_authority_data/3.
- [semidet]uri_authority_data(+Field, ?Components, ?Data)
- Provide access the uri_authority structure. Defined field-names are:
- [det]uri_encoded(+Component, +Value, -Encoded)
- [det]uri_encoded(+Component, -Value, +Encoded)
- Encoded is the URI encoding for Value. When
->Encoded), Component specifies the URI component where the value is used. It is one of
path. Besides alphanumerical characters, the following characters are passed verbatim (the set is split in logical groups according to RFC3986).
- query_value, fragment
- [det]uri_iri(+URI, -IRI)
- [det]uri_iri(-URI, +IRI)
- Convert between a URI, encoded in US-ASCII and an IRI.
An IRI is a fully expanded Unicode string. Unicode strings
are first encoded into UTF-8, after which %-encoding takes place.
syntax_error(Culprit)in mode (+,-) if URI is not a legally percent-encoded UTF-8 string.
- [semidet]uri_file_name(+URI, -FileName)
- [det]uri_file_name(-URI, +FileName)
- Convert between a URI and a local file_name. This protocol is covered by RFC 1738. Please note that file-URIs use absolute paths. The mode (-, +) translates a possible relative path into an absolute one.
This is currently a very simple library, providing support for obtaining the form-data for a CGI script:
- Decodes standard input and the environment variables to obtain a list of
arguments passed to the CGI script. This predicate both deals with the
CGI GET method as well as the POST method. If the data
cannot be obtained, an
existence_errorexception is raised.
Below is a very simple CGI script that prints the passed parameters.
To test it, compile this program using the command below, copy it to
your cgi-bin directory (or make it otherwise known as a CGI-script) and
make the query
% pl -o cgidemo --goal=main --toplevel=halt -c cgidemo.pl
:- use_module(library(cgi)). main :- set_stream(current_output, encoding(utf8)), cgi_get_form(Arguments), format('Content-type: text/html; charset=UTF-8~n~n', ), format('<html>~n', ), format('<head>~n', ), format('<title>Simple SWI-Prolog CGI script</title>~n', ), format('</head>~n~n', ), format('<body>~n', ), format('<p>', ), print_args(Arguments), format('</body>~n</html>~n', ). print_args(). print_args([A0|T]) :- A0 =.. [Name, Value], format('<b>~w</b>=<em>~w</em><br>~n', [Name, Value]), print_args(T).
Printing an HTML document using format/2
is not a neat way of producing HTML because it is vulnerable to required
escape sequences. A high-level alternative is provided by
from the HTTP library.
The startup-time of Prolog is relatively long, in particular if the program is large. In many cases it is much better to use the SWI-Prolog HTTP server library and make the main web-server relay requests to the SWI-Prolog webserver. See the SWI-Prolog HTTP package for details.
The CGI standard is unclear about handling Unicode data. The above two declarations ensure the CGI script will send all data in UTF-8 and thus provide full support of Unicode. It is assumed that browsers generally send form-data using the same encoding as the page in which the form appears, UTF-8 or ISO Latin-1. The current version of cgi_get_form/1 assumes the CGI data is in UTF-8.
- crypt(+Plain, ?Encrypted)
- This predicate can be used in three modes. To test whether a password
matches an encrypted version thereof, simply run with both arguments
fully instantiated. To generate a default encrypted version of
Plain, run with unbound Encrypted and this
argument is unified to a list of character codes holding an encrypted
The library supports two encryption formats: traditional Unix DES-hashes2On non-Unix systems, crypt() is provided by the NetBSD library. The license header is added at the end of this document. and FreeBSD compatible MD5 hashes (all platforms). MD5 hashes start with the magic sequence
$1$, followed by an up to 8 character salt. DES hashes start with a 2 character salt. Note that a DES hash considers only the first 8 characters. The MD5 considers the whole string.
Salt and algorithm can be forced by instantiating the start of Encrypted with it. This is typically used to force MD5 hashes:
?- phrase("$1$", E, _), crypt("My password", E), format('~s~n', [E]). $1$qdaDeDZn$ZUxSQEESEHIDCHPNc3fxZ1
Encrypted is always a list of ASCII character codes. Plain only supports ISO-Latin-1 passwords in the current implementation.
Plain is either an atom, SWI-Prolog string, list of characters or list of character-codes. It is not advised to use atoms, as this implies the password will be available from the Prolog heap as a defined atom.
NOTE: crypt/2 provides an interface to the Unix password hashing API. Above we already introduced support for classical DES and MD5 hashes, both hashes that are considered insecure by today's standards.3Insecure means that the password can realistically be derived from the password hash using a brute-force attack. This implies that leaking the password database is an immediate security risk. The crypt() API of modern Unix systems typically support more secure hashes. Using crypt/2 is suitable if compatibility with OS passwords is required. If strong hashes and platform independence are important to you, use crypto_password_hash/2 provided by library
library(crypto)from the ssl package.
- See also
- To be done
- Compare UUIDs, extract time and version from UUIDs
The library provides operations on UUIDs. Please consult other sources for understanding UUIDs and the implications of the different UUID versions. Some typical calls are given below:
?- uuid(X). X = 'ea6589fa-19dd-11e2-8a49-001d92e1879d'. ?- uuid(X, [url('http://www.swi-prolog.org')]). X = '73a07870-6a90-3f2e-ae2b-ffa538dc7c2c'.
- UUID is an atom representing a new UUID. This is
the same as calling
uuid(UUID, ). See uuid/2 for options.
- [det]uuid(-UUID, +Options)
- Create a new UUID according to Options. The
following options are defined:
- Integer in the range 1..5, which specifies the UUID version that is created. Default is 1.
- Provide additional context information for UUIDs using version 3 or 5. If there is no explicit version option, UUID version 3 is used.
- Representation of the UUID. Default is
atom, yielding atoms such as
8304efdd-bd6e-5b7c-a27f-83f3f05c64e0. The alternative is
integer, returning a large integer that represents the 128 bits of the UUID.
library(sha) provides Secure Hash
Algorihms approved by FIPS (Federal Information Processing
SHA (Secure Hash Algorithm) hash functions refer to five FIPS-approved
algorithms for computing a condensed digital representation (known as a
message digest) that is, to a high degree of probability, unique for a
given input data sequence (the message). These algorithms are called
`secure' because (in the words of the standard), ``for a given
algorithm, it is computationally infeasible 1) to find a message that
corresponds to a given message digest, or 2) to find two different
messages that produce the same message digest. Any change to a message
will, with a very high probability, result in a different message
The current library supports all 5 approved algorithms, both computing the hash-key from data and the hash Message Authentication Code (HMAC).
A general secure hash interface is provided by
part of the ssl package.
Input is text, represented as an atom, packed string object or code-list. Note that these functions operate on byte-sequences and therefore are not meaningful on Unicode text. The result is returned as a list of byte-values. This is the most general format that is comfortable supported by standard Prolog and can easily be transformed in other formats. Commonly used text formats are ASCII created by encoding each byte as two hexadecimal digits and ASCII created using base64 encoding. Representation as a large integer can be desirable for computational processing.
- sha_hash(+Data, -Hash, +Options)
- Hash is the SHA hash of Data. Data is either an atom, packed
string or list of character codes. Hash is unified with a
list of bytes (integers in the range 0..255) representing the hash. See
to convert this into the more commonly seen hexadecimal representation.
The conversion is controlled by Options:
- One of
- This option defines the mapping from Prolog (Unicode) text to bytes on
which the SHA algorithm is performed. It has two values. The defualt is
utf8, which implies that Unicode text is encoded as UTF-8 bytes. This option can deal with any atom. The alternative is
octet, which implies that the text is considered as a sequence of bytes. This is suitable for e.g., atoms that represent binary data. An error is raised if the text contains code-points outside the range 0..255.
- hmac_sha(+Key, +Data, -HMAC, +Options)
- Quoting Wikipedia:
``A keyed-hash message authentication code, or HMAC, is a type of
message authentication code (MAC) calculated using a cryptographic hash
function in combination with a secret key. As with any MAC, it may be
used to simultaneously verify both the data integrity and the
authenticity of a message. Any iterative cryptographic hash function,
such as MD5 or SHA-1, may be used in the calculation of an HMAC; the
resulting MAC algorithm is termed HMAC-MD5 or HMAC-SHA-1 accordingly.
The cryptographic strength of the HMAC depends upon the cryptographic
strength of the underlying hash function, on the size and quality of the
key and the size of the hash output length in bits.''
Key and Data are either an atom, packed string or list of character codes. HMAC is unified with a list of integers representing the authentication code. Options is the same as for sha_hash/3, but currently only
- hash_atom(+Hash, -HexAtom)
- True when HexAtom is the commonly used hexadecimal encoding
of the hash code. E.g.,
?- sha_hash('SWI-Prolog', Hash, ), hash_atom(Hash, Hex). Hash = [61, 128, 252, 38, 121, 69, 229, 85, 199|...], Hex = '3d80fc267945e555c730403bd0ab0716e2a68c68'.
The underlying SHA-2 library is an unmodified copy created by Dr Brian Gladman, Worcester, UK. It is distributed under the license conditions below.
The free distribution and use of this software in both source and binary form is allowed (with or without changes) provided that:
- distributions of this source code include the above copyright
notice, this list of conditions and the following disclaimer;
- distributions in binary form include the above copyright notice,
this list of conditions and the following disclaimer in the
documentation and/or other associated materials;
- the copyright holder's name is not used to endorse products built using this software without specific written permission.
ALTERNATIVELY, provided that this notice is retained in full, this product may be distributed under the terms of the GNU General Public License (GPL), in which case the provisions of the GPL apply INSTEAD OF those given above.
- See also
Compute MD5 hashes from a Prolog string. This library provides a
lightweight alternative to the general secure hash interface provided by
library(crypto) from the
- [det]md5_hash(+Data, -Hash, +Options)
- Hash is the MD5 hash of Data, The conversion is
- If Data is a sequence of character codes, this must be
translated into a sequence of bytes, because that is what the
hashing requires. The default encoding is
utf8. The other meaningful value is
octet, claiming that Data contains raw bytes.
Data is either an atom, string, code-list or char-list. Hash is an atom holding 32 characters, representing the hash in hexadecimal notation
- See also
- In addition to this hash library, SWI-Prolog provides
library(sha)and hash functions through
library(crypto), part of the
This library defines a filter stream that maintains a hash of the data that passes through the stream. It can be used to compute the hash of input data while it is being processed. This is notably interesting if data is processed from a socket as it avoids the need for collecting the data first in a temporary file.
A typical processing sequence is illustrated below, where process/2 somehow processed the data and save_result/3 records the result as obtained from URL with content digest SHA256 its Result.
..., http_open(URL, In0, ), open_hash_stream(In0, In, [algorithm(sha256)]), process(In, Result), stream_hash(In, SHA256), close(In), save_result(URL, SHA256, Result)
This library can also be used to compute the hash for the content of a file. The advantage is that this code doesn't rely on external tools. It is considerably faster for short files, but considerably slower on large files because Prolog I/O is based on character streams rather than blocks.
file_hash(Algorithm, File, Hash) :- setup_call_cleanup( open(File, read, In0, [type(binary)]), setup_call_cleanup( open_hash_stream(In0, In, [ algorithm(Algorithm), close_parent(false) ]), ( setup_call_cleanup( open_null_stream(Null), copy_stream_data(In, Null), close(Null)), stream_hash(In, Hash) ), close(In)), close(In0)).
- [det]open_hash_stream(+OrgStream, -HashStream, +Options)
- Open a filter stream on OrgStream that maintains a hash. The
hash can be retrieved at any time using stream_hash/2.
- One of
sha512. Default is
true(default), closing the filter stream also closes the original (parent) stream.
- [det]stream_hash(+HashStream, -Digest:atom)
- Unify Digest with a hash for the bytes send to or read from HashStream. Note that the hash is computed on the stream buffers. If the stream is an output stream, it is first flushed and the Digest represents the hash at the current location. If the stream is an input stream the Digest represents the hash of the processed input including the already buffered data.
library(memfile) provides an alternative to
temporary files, intended for temporary buffering of data. Memory files
in general are faster than temporary files and do not suffer from
security risks or naming conflicts associated with temporary-file
There is no limit to the number of memory streams, nor the size of them. However, a single memory file cannot have multiple streams at the same time, i.e., a memory file cannot be opened multiple times, not even for reading. Memory files are thread-safe and subject to (atom) garbage collection.
- Create a new memory file and return a unique opaque handle to it.
- Discard the memory file and its contents. If the file is open it is first closed.
- open_memory_file(+Handle, +Mode, -Stream)
- Open the memory-file. Mode is one of
insert. The resulting Stream must be closed using close/1. When opened for
insert, the current location is initialized at the start of the data and can be modified using seek/2 or set_stream_position/2. In
updatemode, existing content is replaced, while the size is enlarged after hitting the end of the data. In
insertmode, the new data is inserted at the current point.
- open_memory_file(+Handle, +Mode, -Stream, +Options)
- Open a memory-file as open_memory_file/3.
- Set the encoding for a memory file and the created stream. Encoding
names are the same as used with open/4.
By default, memoryfiles represent UTF-8 streams, making them capable of
storing arbitrary Unicode text. In practice the only alternative is
octet, turning the memoryfile into binary mode. Please study SWI-Prolog Unicode and encoding issues before using this option.
falseand the memory file is opened for reading, discard the file (see free_memory_file/1) if the input is closed. This is used to realise open_chars_stream/2 in library(charsio).
- size_memory_file(+Handle, -Size)
- Return the content-length of the memory-file in characters in the current encoding of the memory file. The file should be closed and contain data.
- size_memory_file(+Handle, -Size, +Encoding)
- Return the content-length of the memory-file in characters in the given Encoding. The file should be closed and contain data.
- atom_to_memory_file(+Atom, -Handle)
- Turn an atom into a read-only memory-file containing the (shared)
characters of the atom. Opening this memory-file in mode
writeyields a permission error.
- insert_memory_file(+Handle, +Offset, +Data)
- Insert Data into the memory file at location Offset. The offset is specified in characters. Data can be an atom, string, code or character list. Other terms are first serialized using writeq/1. This predicate raises a domain_error exception if Offset is out of range and a permission_error if the memory file is read-only or opened.
- delete_memory_file(+Handle, +Offset, +Length)
- Delete a Length characters from the memory file, starting at Offset. This predicate raises a domain_error exception if Offset or Offset+Length is out of range and a permission_error if the memory file is read-only or opened.
- memory_file_to_atom(+Handle, -Atom)
- Return the content of the memory-file in Atom.
- memory_file_to_atom(+Handle, -Atom, +Encoding)
- Return the content of the memory-file in Atom, pretending the
data is in the given Encoding. This can be used to convert
from one encoding into another, typically from/to bytes. For example, if
we must convert a set of bytes that contain text in UTF-8, open the
memory file as octet stream, fill it, and get the result using Encoding
- memory_file_to_codes(+Handle, -Codes)
- Return the content of the memory-file as a list of character-codes in Codes.
- memory_file_to_codes(+Handle, -Codes, +Encoding)
- Return the content of the memory-file as a list of character-codes in Codes, pretending the data is in the given Encoding.
- memory_file_to_string(+Handle, -String)
- Return the content of the memory-file as a string in -String.
- memory_file_to_string(+Handle, -String, +Encoding)
- Return the content of the memory-file as a string in String, pretending the data is in the given Encoding.
- memory_file_substring(+Handle, ?Before, ?Length, ?After, -SubString)
- SubString is a substring of the memory file. There are Before characters in the memory file before SubString, SubString contains Length character and is followed by After characters in the memory file. The signature is the same as sub_string/5 and sub_atom/5, but currently at least two of the 3 position arguments must be specified. Future versions might implement the full functionality of sub_string/5.
- memory_file_line_position(+MF, ?Line, ?LinePos, ?Offset)
- True if the character offset Offset corresponds with the LinePos character on line Line. Lines are counted from one (1). Note that LinePos is not the column as each character counts for one, including backspace and tab.
library(time) provides timing and alarm functions.
Alarms are thread-specific, i.e., creating an alarm causes the alarm
goal to be called in the thread that created it. The predicate current_alarm/4
only reports alarms that are related to the calling thread. If a thread
terminates, all remaining alarms are silently removed. Most applications
- alarm(+Time, :Callable, -Id, +Options)
- Schedule Callable to be called Time seconds from
Time is a number (integer or float). Callable is
called on the next pass through a call- or redo-port of the Prolog
engine, or a call to the PL_handle_signals() routine from
SWI-Prolog. Id is unified with a reference to the timer.
The resolution of the alarm depends on the underlying implementation, which is based on pthread_cond_timedwait() (on Windows on the pthread emulation thereof). Long-running foreign predicates that do not call PL_handle_signals() may further delay the alarm. The relation to blocking system calls (sleep, reading from slow devices, etc.) is undefined and varies between implementations.
Options is a list of
Name(Value)terms. Defined options are:
- alarm(+Time, :Callable, -Id)
- Same as
alarm(Time, Callable, Id,).
- alarm_at(+Time, :Callable, -Id, +Options)
- as alarm/3, but Time is the specification of an absolute point in time. Absolute times are specified in seconds after the Jan 1, 1970 epoch. See also date_time_stamp/2.
- Activate an alarm allocated using alarm/4
with the option
install(false)or stopped using uninstall_alarm/1.
- install_alarm(+Id, +Time)
- As install_alarm/1, but specifies a new (relative) timeout value.
- Deactivate a running alarm, but do not invalidate the alarm identifier. Later, the alarm can be reactivated using either install_alarm/1 or install_alarm/2. Reinstalled using install_alarm/1, it will fire at the originally scheduled time. Reinstalled using install_alarm/2 causes the alarm to fire at the specified time from now.
- Remove an alarm. If it is not yet fired, it will not be fired any more.
- current_alarm(?At, ?:Callable, ?Id, ?Status)
- Enumerate the not-yet-removed alarms. Status is one of
doneif the alarm has been called,
nextif it is the next to be fired and scheduled otherwise.
- call_with_time_limit(+Time, :Goal)
- True if Goal completes within Time seconds. Goal
is executed as in once/1.
If Goal doesn't complete within Time seconds (wall
time), exit using the exception
time_limit_exceeded. See catch/3.
- See also
library(process)provides a portable high level interface to create and manage processes.
library(unix) library provides the commonly used
Unix primitives to deal with process management. These primitives are
useful for many tasks, including server management, parallel
computation, exploiting and controlling other processes, etc.
The predicates in this library are modelled closely after their native Unix counterparts.
- Clone the current process into two branches. In the child, Pid
is unified to child. In the original process, Pid is unified
to the process identifier of the created child. Both parent and child
are fully functional Prolog processes running the same program. The
processes share open I/O streams that refer to Unix native streams, such
as files, sockets and pipes. Data is not shared, though on most Unix
systems data is initially shared and duplicated only if one of the
programs attempts to modify the data.
fork()is the only way to create new processes and fork/1 is a simple direct interface to it.
permission_error(fork, process, main)is raised if the calling thread is not the only thread in the process. Forking a Prolog process with threads will typically deadlock because only the calling thread is cloned in the fork, while all thread synchronization are cloned.
- Fork (as fork/1) and exec (using exec/1)
the child immediately. This behaves as the code below, but bypasses the
check for the existence of other threads because this is a safe
fork_exec(Command) :- ( fork(child) -> exec(Command) ; true ).
- Replace the running program by starting Command. Command
is a callable term. The functor is the command and the arguments provide
the command-line arguments for the command. Each command-line argument
must be atomic and is converted to a string before passed to the Unix
execvp(). Here are some examples:
exec()is the only way to start an executable file executing. It is commonly used together with fork/1. For example to start netscape on an URL in the background, do:
run_netscape(URL) :- ( fork(child), exec(netscape(URL)) ; true ).
Using this code, netscape remains part of the process-group of the invoking Prolog process and Prolog does not wait for netscape to terminate. The predicate wait/2 allows waiting for a child, while detach_IO/0 disconnects the child as a deamon process.
- [det]wait(?Pid, -Status)
- Wait for a child to change status. Then report the child that changed
status as well as the reason. If Pid is bound on entry then
the status of the specified child is reported. If not, then the status
of any child is reported. Status is unified with
exited(ExitCode)if the child with pid Pid was terminated by calling
exit()(Prolog halt/1). ExitCode is the return status. Status is unified with
signaled(Signal)if the child died due to a software interrupt (see kill/2). Signal contains the signal number. Finally, if the process suspended execution due to a signal, Status is unified with
- [det]kill(+Pid, +Signal)
- Deliver a software interrupt to the process with identifier Pid
using software-interrupt number Signal. See also on_signal/2.
Signals can be specified as an integer or signal name, where signal
names are derived from the C constant by dropping the
SIGprefix and mapping to lowercase. E.g.
intis the same as
SIGINTin C. The meaning of the signal numbers can be found in the Unix manual.
- [det]pipe(-InSream, -OutStream)
- Create a communication-pipe. This is normally used to make a child
communicate to its parent. After pipe/2,
the process is cloned and, depending on the desired direction, both
processes close the end of the pipe they do not use. Then they use the
remaining stream to communicate. Here is a simple example:
:- use_module(library(unix)). fork_demo(Result) :- pipe(Read, Write), fork(Pid), ( Pid == child -> close(Read), format(Write, '~q.~n', [hello(world)]), flush_output(Write), halt ; close(Write), read(Read, Result), close(Read) ).
- [det]dup(+FromStream, +ToStream)
- Interface to Unix
dup2(), copying the underlying filedescriptor and thus making both streams point to the same underlying object. This is normally used together with fork/1 and pipe/2 to talk to an external program that is designed to communicate using standard I/O.
Both FromStream and ToStream either refer to a Prolog stream or an integer descriptor number to refer directly to OS descriptors. See also
demo/pipe.plin the source-distribution of this package.
- This predicate is intended to create Unix deamon processes. It
performs two actions.
- The I/O streams
user_errorare closed if they are connected to a terminal (see
ttyproperty in stream_property/2). Input streams are rebound to a dummy stream that returns EOF. Output streams are reboud to forward their output to Stream.
- The process is detached from the current process-group and its
controlling terminal. This is achieved using
setsid()if provided or using
To ignore all output, it may be rebound to a null stream. For example:
..., open_null_stream(Out), detach_IO(Out).
The detach_IO/1 should be called only once per process. Subsequent calls silently succeed without any side effects.
- See also
- detach_IO/0 and
- The I/O streams
- Detach I/O similar to detach_IO/1.
The output streams are bound to a file
/tmp/pl-out.<pid>. Output is line buffered (see set_stream/2).
- See also
library(syslog)allows for sending output to the Unix logging service.
- Older versions of this predicate only created this file if there was output.
- Access to Linux process control operations. Defines values for
- Control whether the process is allowed to dump core. This right is dropped under several uid and gid conditions.
- Get the value of the dumpable flag.
- Access system configuration. See
sysconf(1)for details. Conf is a term Config(Value), where Value is always an integer. Config is the
sysconf()name after removing =_SC_= and conversion to lowercase. Currently support the following configuration info:
nprocessors_onln. Note that not all values may be supported on all operating systems.
library(rlimit) library provides an interface to the
POSIX getrlimit()/setrlimit() API that control the maximum
resource-usage of a process or group of processes. This call is
especially useful for servers such as CGI scripts and inetd-controlled
servers to avoid an uncontrolled script claiming too much resources.
- rlimit(+Resource, -Old, +New)
- Query and/or set the limit for Resource. Time-values are in
seconds and size-values are counted in bytes. The following values are
supported by this library. Please note that not all resources may be
available and accessible on all platforms. This predicate can throw a
variety of exceptions. In portable code this should be guarded with catch/3.
The defined resources are:
Max address space
CPU time in seconds
max data size
max stack size
max core file size
max resident set size
max number of processes
max number of open files
max locked-in-memory address
When the process hits a limit POSIX systems normally send the process a signal that terminates it. These signals may be caught using SWI-Prolog's on_signal/3 primitive. The code below illustrates this behaviour. Please note that asynchronous signal handling is dangerous, especially when using threads. 100% fail-safe operation cannot be guaranteed, but this procedure will inform the user properly `most of the time'.
rlimit_demo :- rlimit(cpu, _, 2), on_signal(xcpu, _, cpu_exceeded), ( repeat, fail ). cpu_exceeded(_Sig) :- format(user_error, 'CPU time exceeded~n', ), halt(1).
- Jeffrey Rosenwald (JeffRose@acm.org)
- See also
SWI-Prolog's broadcast library provides a means that may be used to facilitate publish and subscribe communication regimes between anonymous members of a community of interest. The members of the community are however, necessarily limited to a single instance of Prolog. The UDP broadcast library removes that restriction. With this library loaded, any member on your local IP subnetwork that also has this library loaded may hear and respond to your broadcasts.
This module has only two public predicates. When the module is initialized, it starts a two listener threads that listen for broadcasts from others, received as UDP datagrams.
Unlike TIPC broadcast, UDP broadcast has only one scope,
broadcast/1 or broadcast_request/1
that is not directed to the listener above, behaves as usual and is
confined to the instance of Prolog that originated it. But when so
directed, the broadcast will be sent to all participating systems,
including itself, by way of UDP's multicast addressing facility. A UDP
broadcast or broadcast request takes the typical form:
To prevent the potential for feedback loops, the scope qualifier is
stripped from the message before transmission. The timeout is optional.
It specifies the amount to time to wait for replies to arrive in
response to a broadcast_request. The default period is 0.250 seconds.
The timeout is ignored for broadcasts.
An example of three separate processes cooperating on the same Node:
Process A: ?- listen(number(X), between(1, 5, X)). true. ?- Process B: ?- listen(number(X), between(7, 9, X)). true. ?- Process C: ?- findall(X, broadcast_request(udp_subnet(number(X))), Xs). Xs = [1, 2, 3, 4, 5, 7, 8, 9]. ?-
It is also possible to carry on a private dialog with a single responder. To do this, you supply a compound of the form, Term:PortId, to a UDP scoped broadcast/1 or broadcast_request/1, where PortId is the ip-address and port-id of the intended listener. If you supply an unbound variable, PortId, to broadcast_request, it will be unified with the address of the listener that responds to Term. You may send a directed broadcast to a specific member by simply providing this address in a similarly structured compound to a UDP scoped broadcast/1. The message is sent via unicast to that member only by way of the member's broadcast listener. It is received by the listener just as any other broadcast would be. The listener does not know the difference.
For example, in order to discover who responded with a particular value:
Host B Process 1: ?- listen(number(X), between(1, 5, X)). true. ?- Host A Process 1: ?- listen(number(X), between(7, 9, X)). true. ?- Host A Process 2: ?- listen(number(X), between(1, 5, X)). true. ?- bagof(X, broadcast_request(udp_subnet(number(X):From,1)), Xs). From = ip(192, 168, 1, 103):34855, Xs = [7, 8, 9] ; From = ip(192, 168, 1, 103):56331, Xs = [1, 2, 3, 4, 5] ; From = ip(192, 168, 1, 104):3217, Xs = [1, 2, 3, 4, 5].
While the implementation is mostly transparent, there are some important and subtle differences that must be taken into consideration:
- UDP broadcast requires an initialization step in order to launch the broadcast listener daemon. See udp_broadcast_initialize/2.
- Prolog's broadcast_request/1 is nondet. It sends the request, then evaluates the replies synchronously, backtracking as needed until a satisfactory reply is received. The remaining potential replies are not evaluated. This is not so when UDP is involved.
- A UDP broadcast/1 is completely asynchronous.
- A UDP broadcast_request/1 is partially synchronous. A broadcast_request/1 is sent, then the sender balks for a period of time (default: 250 ms) while the replies are collected. Any reply that is received after this period is silently discarded. A optional second argument is provided so that a sender may specify more (or less) time for replies.
- Replies are presented to the user as a choice point on arrival, until the broadcast request timer finally expires. This allows traffic to propagate through the system faster and provides the requestor with the opportunity to terminate a broadcast request early if desired, by simply cutting choice points.
- Please beware that broadcast request transactions remain active and resources consumed until broadcast_request finally fails on backtracking, an uncaught exception occurs, or until choice points are cut. Failure to properly manage this will likely result in chronic exhaustion of UDP sockets.
- If a listener is connected to a generator that always succeeds (e.g. a random number generator), then the broadcast request will never terminate and trouble is bound to ensue.
- broadcast_request/1 with
udp_subnetscope is not reentrant. If a listener performs a broadcast_request/1 with UDP scope recursively, then disaster looms certain. This caveat does not apply to a UDP scoped broadcast/1, which can safely be performed from a listener context.
- UDP broadcast's capacity is not infinite. While it can tolerate substantial bursts of activity, it is designed for short bursts of small messages. Unlike TIPC, UDP is unreliable and has no QOS protections. Congestion is likely to cause trouble in the form of non-Byzantine failure. That is, late, lost (e.g. infinitely late), or duplicate datagrams. Caveat emptor.
- A UDP broadcast_request/1 term that is grounded is considered to be a broadcast only. No replies are collected unless the there is at least one unbound variable to unify.
- A UDP broadcast/1 always succeeds, even if there are no listeners.
- A UDP broadcast_request/1 that receives no replies will fail.
- Replies may be coming from many different places in the network (or none at all). No ordering of replies is implied.
- Prolog terms are sent to others after first converting them to atoms using term_to_atom/2. Passing real numbers this way may result in a substantial truncation of precision.
- The broadcast model is based on anonymity and a presumption of trust--a perfect recipe for compromise. UDP is an Internet protocol. A UDP broadcast listener exposes a public port (20005), which is static and shared by all listeners, and a private port, which is semi-static and unique to the listener instance. Both can be seen from off-cluster nodes and networks. Usage of this module exposes the node and consequently, the cluster to significant security risks. So have a care when designing your application. You must talk only to those who share and contribute to your concerns using a carefully prescribed protocol.
- UDP broadcast categorically and silently ignores all message traffic originating from or terminating on nodes that are not members of the local subnet. This security measure only keeps honest people honest!
- [nondet]udp_broadcast_service(?Domain, ?Address)
- provides the UDP broadcast address for a given Domain. At
present, only one domain is supported,
- [nondet]udp_host_to_address(?Service, ?Address)
- locates a UDP service by name. Service is an atom or grounded
term representing the common name of the service. Address is
a UDP address structure. A server may advertise its services by name by
including the fact, udp:
host_to_address(+Service, +Address), somewhere in its source. This predicate can also be used to perform reverse searches. That is it will also resolve an Address to a Service name.
- [semidet]udp_broadcast_initialize(+IPAddress, +SubnetMask)
- causes any required runtime initialization to occur. At present, proper operation of UDP broadcast depends on local information that is not easily obtained mechanically. In order to determine the appropriate UDP broadcast address, you must supply the IPAddress and SubnetMask for the node that is running this module. These data are supplied in the form of ip/4 terms. This is now required to be included in an applications intialization directive.
This library defines a Prolog stream that realises its low-level I/O with callbacks to Prolog. The library was developed to bind normal Prolog I/O to Pengines I/O. This type of I/O redirection is probably the primary use case.
- open_prolog_stream(+Module, +Mode, -Stream, +Options)
- Create a new stream that implements its I/O by calling predicates in Module.
The called predicates are:
- Called for a
Mode = writestream if data is available. String contains the (textual) data that is written to Stream. The callback is called if the buffer of Stream overflows, the user calls
flush_output(Stream)or Stream is closed and there is buffered data.
- Called for a
Mode == readstream to get new data. On success the stream extracts text from the provided Term. Term is typically a string, atom, code or character list. If term is not one of the above, it is handed to writeq/1. To signal end-of-file, unify stream with an empty text, e.g.,
- Called when the stream is closed. This predicate must succeed. The callback can be used to cleanup associated resources.
The current implementation only deals with text streams. The stream uses the
wchar_tencoding. The buffer size must be a multiple of
wchar_t, i.e., a multiple of four for portability. The newline mode of the stream is
posixon all platforms, disabling the translation
"\n" --> "\r\n".
Options is currently ignored.
- Futher versions might require additional callbacks. As we demand all callbacks to be defined, existing code needs to implement the new callbacks.
NetBSD Crypt license
* Copyright (c) 1989, 1993 * The Regents of the University of California. All rights reserved. * * This code is derived from software contributed to Berkeley by * Tom Truscott. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE.
- 16 16
- 16 18
- 10 10 10
- 16 16 16 16
- 16 16
- 6.4 6.4
- 15 15
- 6.4 7 7