2 @c This is part of the GNU Emacs Lisp Reference Manual.
3 @c Copyright (C) 1990-1995, 1998-1999, 2001-2014 Free Software
5 @c See the file elisp.texi for copying conditions.
13 In the terminology of operating systems, a @dfn{process} is a space in
14 which a program can execute. Emacs runs in a process. Emacs Lisp
15 programs can invoke other programs in processes of their own. These are
16 called @dfn{subprocesses} or @dfn{child processes} of the Emacs process,
17 which is their @dfn{parent process}.
19 A subprocess of Emacs may be @dfn{synchronous} or @dfn{asynchronous},
20 depending on how it is created. When you create a synchronous
21 subprocess, the Lisp program waits for the subprocess to terminate
22 before continuing execution. When you create an asynchronous
23 subprocess, it can run in parallel with the Lisp program. This kind of
24 subprocess is represented within Emacs by a Lisp object which is also
25 called a ``process''. Lisp programs can use this object to communicate
26 with the subprocess or to control it. For example, you can send
27 signals, obtain status information, receive output from the process, or
30 @defun processp object
31 This function returns @code{t} if @var{object} represents an Emacs
32 subprocess, @code{nil} otherwise.
35 In addition to subprocesses of the current Emacs session, you can
36 also access other processes running on your machine. @xref{System
40 * Subprocess Creation:: Functions that start subprocesses.
41 * Shell Arguments:: Quoting an argument to pass it to a shell.
42 * Synchronous Processes:: Details of using synchronous subprocesses.
43 * Asynchronous Processes:: Starting up an asynchronous subprocess.
44 * Deleting Processes:: Eliminating an asynchronous subprocess.
45 * Process Information:: Accessing run-status and other attributes.
46 * Input to Processes:: Sending input to an asynchronous subprocess.
47 * Signals to Processes:: Stopping, continuing or interrupting
48 an asynchronous subprocess.
49 * Output from Processes:: Collecting output from an asynchronous subprocess.
50 * Sentinels:: Sentinels run when process run-status changes.
51 * Query Before Exit:: Whether to query if exiting will kill a process.
52 * System Processes:: Accessing other processes running on your system.
53 * Transaction Queues:: Transaction-based communication with subprocesses.
54 * Network:: Opening network connections.
55 * Network Servers:: Network servers let Emacs accept net connections.
56 * Datagrams:: UDP network connections.
57 * Low-Level Network:: Lower-level but more general function
58 to create connections and servers.
59 * Misc Network:: Additional relevant functions for net connections.
60 * Serial Ports:: Communicating with serial ports.
61 * Byte Packing:: Using bindat to pack and unpack binary data.
64 @node Subprocess Creation
65 @section Functions that Create Subprocesses
67 There are three primitives that create a new subprocess in which to run
68 a program. One of them, @code{start-process}, creates an asynchronous
69 process and returns a process object (@pxref{Asynchronous Processes}).
70 The other two, @code{call-process} and @code{call-process-region},
71 create a synchronous process and do not return a process object
72 (@pxref{Synchronous Processes}). There are various higher-level
73 functions that make use of these primitives to run particular types of
76 Synchronous and asynchronous processes are explained in the following
77 sections. Since the three functions are all called in a similar
78 fashion, their common arguments are described here.
80 @cindex execute program
81 @cindex @env{PATH} environment variable
82 @cindex @env{HOME} environment variable
83 In all cases, the function's @var{program} argument specifies the
84 program to be run. An error is signaled if the file is not found or
85 cannot be executed. If the file name is relative, the variable
86 @code{exec-path} contains a list of directories to search. Emacs
87 initializes @code{exec-path} when it starts up, based on the value of
88 the environment variable @env{PATH}. The standard file name
89 constructs, @samp{~}, @samp{.}, and @samp{..}, are interpreted as
90 usual in @code{exec-path}, but environment variable substitutions
91 (@samp{$HOME}, etc.)@: are not recognized; use
92 @code{substitute-in-file-name} to perform them (@pxref{File Name
93 Expansion}). @code{nil} in this list refers to
94 @code{default-directory}.
96 Executing a program can also try adding suffixes to the specified
100 This variable is a list of suffixes (strings) to try adding to the
101 specified program file name. The list should include @code{""} if you
102 want the name to be tried exactly as specified. The default value is
106 @strong{Please note:} The argument @var{program} contains only the
107 name of the program; it may not contain any command-line arguments. You
108 must use a separate argument, @var{args}, to provide those, as
111 Each of the subprocess-creating functions has a @var{buffer-or-name}
112 argument that specifies where the standard output from the program will
113 go. It should be a buffer or a buffer name; if it is a buffer name,
114 that will create the buffer if it does not already exist. It can also
115 be @code{nil}, which says to discard the output, unless a custom filter function
116 handles it. (@xref{Filter Functions}, and @ref{Read and Print}.)
117 Normally, you should avoid having multiple processes send output to the
118 same buffer because their output would be intermixed randomly.
119 For synchronous processes, you can send the output to a file instead
122 @cindex program arguments
123 All three of the subprocess-creating functions have a @code{&rest}
124 argument, @var{args}. The @var{args} must all be strings, and they are
125 supplied to @var{program} as separate command line arguments. Wildcard
126 characters and other shell constructs have no special meanings in these
127 strings, since the strings are passed directly to the specified program.
129 @cindex environment variables, subprocesses
130 The subprocess inherits its environment from Emacs, but you can
131 specify overrides for it with @code{process-environment}. @xref{System
132 Environment}. The subprocess gets its current directory from the
133 value of @code{default-directory}.
135 @defvar exec-directory
137 The value of this variable is a string, the name of a directory that
138 contains programs that come with GNU Emacs and are intended for Emacs
139 to invoke. The program @code{movemail} is an example of such a program;
140 Rmail uses it to fetch new mail from an inbox.
144 The value of this variable is a list of directories to search for
145 programs to run in subprocesses. Each element is either the name of a
146 directory (i.e., a string), or @code{nil}, which stands for the default
147 directory (which is the value of @code{default-directory}).
148 @cindex program directories
150 The value of @code{exec-path} is used by @code{call-process} and
151 @code{start-process} when the @var{program} argument is not an absolute
154 Generally, you should not modify @code{exec-path} directly. Instead,
155 ensure that your @env{PATH} environment variable is set appropriately
156 before starting Emacs. Trying to modify @code{exec-path}
157 independently of @env{PATH} can lead to confusing results.
160 @node Shell Arguments
161 @section Shell Arguments
162 @cindex arguments for shell commands
163 @cindex shell command arguments
165 Lisp programs sometimes need to run a shell and give it a command
166 that contains file names that were specified by the user. These
167 programs ought to be able to support any valid file name. But the shell
168 gives special treatment to certain characters, and if these characters
169 occur in the file name, they will confuse the shell. To handle these
170 characters, use the function @code{shell-quote-argument}:
172 @defun shell-quote-argument argument
173 This function returns a string that represents, in shell syntax,
174 an argument whose actual contents are @var{argument}. It should
175 work reliably to concatenate the return value into a shell command
176 and then pass it to a shell for execution.
178 Precisely what this function does depends on your operating system. The
179 function is designed to work with the syntax of your system's standard
180 shell; if you use an unusual shell, you will need to redefine this
184 ;; @r{This example shows the behavior on GNU and Unix systems.}
185 (shell-quote-argument "foo > bar")
186 @result{} "foo\\ \\>\\ bar"
188 ;; @r{This example shows the behavior on MS-DOS and MS-Windows.}
189 (shell-quote-argument "foo > bar")
190 @result{} "\"foo > bar\""
193 Here's an example of using @code{shell-quote-argument} to construct
198 (shell-quote-argument oldfile)
200 (shell-quote-argument newfile))
204 @cindex quoting and unquoting command-line arguments
205 @cindex minibuffer input, and command-line arguments
206 @cindex @code{call-process}, command-line arguments from minibuffer
207 @cindex @code{start-process}, command-line arguments from minibuffer
208 The following two functions are useful for combining a list of
209 individual command-line argument strings into a single string, and
210 taking a string apart into a list of individual command-line
211 arguments. These functions are mainly intended for
212 converting user input in the minibuffer, a Lisp string, into a list of
213 string arguments to be passed to @code{call-process} or
214 @code{start-process}, or for converting such lists of arguments into
215 a single Lisp string to be presented in the minibuffer or echo area.
217 @defun split-string-and-unquote string &optional separators
218 This function splits @var{string} into substrings at matches for the
219 regular expression @var{separators}, like @code{split-string} does
220 (@pxref{Creating Strings}); in addition, it removes quoting from the
221 substrings. It then makes a list of the substrings and returns it.
223 If @var{separators} is omitted or @code{nil}, it defaults to
224 @code{"\\s-+"}, which is a regular expression that matches one or more
225 characters with whitespace syntax (@pxref{Syntax Class Table}).
227 This function supports two types of quoting: enclosing a whole string
228 in double quotes @code{"@dots{}"}, and quoting individual characters
229 with a backslash escape @samp{\}. The latter is also used in Lisp
230 strings, so this function can handle those as well.
233 @defun combine-and-quote-strings list-of-strings &optional separator
234 This function concatenates @var{list-of-strings} into a single string,
235 quoting each string as necessary. It also sticks the @var{separator}
236 string between each pair of strings; if @var{separator} is omitted or
237 @code{nil}, it defaults to @code{" "}. The return value is the
240 The strings in @var{list-of-strings} that need quoting are those that
241 include @var{separator} as their substring. Quoting a string encloses
242 it in double quotes @code{"@dots{}"}. In the simplest case, if you
243 are consing a command from the individual command-line arguments,
244 every argument that includes embedded blanks will be quoted.
247 @node Synchronous Processes
248 @section Creating a Synchronous Process
249 @cindex synchronous subprocess
251 After a @dfn{synchronous process} is created, Emacs waits for the
252 process to terminate before continuing. Starting Dired on GNU or
253 Unix@footnote{On other systems, Emacs uses a Lisp emulation of
254 @code{ls}; see @ref{Contents of Directories}.} is an example of this: it
255 runs @code{ls} in a synchronous process, then modifies the output
256 slightly. Because the process is synchronous, the entire directory
257 listing arrives in the buffer before Emacs tries to do anything with it.
259 While Emacs waits for the synchronous subprocess to terminate, the
260 user can quit by typing @kbd{C-g}. The first @kbd{C-g} tries to kill
261 the subprocess with a @code{SIGINT} signal; but it waits until the
262 subprocess actually terminates before quitting. If during that time the
263 user types another @kbd{C-g}, that kills the subprocess instantly with
264 @code{SIGKILL} and quits immediately (except on MS-DOS, where killing
265 other processes doesn't work). @xref{Quitting}.
267 The synchronous subprocess functions return an indication of how the
270 The output from a synchronous subprocess is generally decoded using a
271 coding system, much like text read from a file. The input sent to a
272 subprocess by @code{call-process-region} is encoded using a coding
273 system, much like text written into a file. @xref{Coding Systems}.
275 @defun call-process program &optional infile destination display &rest args
276 This function calls @var{program} and waits for it to finish.
278 The current working directory of the subprocess is
279 @code{default-directory}.
281 The standard input for the new process comes from file @var{infile} if
282 @var{infile} is not @code{nil}, and from the null device otherwise.
283 The argument @var{destination} says where to put the process output.
284 Here are the possibilities:
288 Insert the output in that buffer, before point. This includes both the
289 standard output stream and the standard error stream of the process.
292 Insert the output in a buffer with that name, before point.
295 Insert the output in the current buffer, before point.
301 Discard the output, and return @code{nil} immediately without waiting
302 for the subprocess to finish.
304 In this case, the process is not truly synchronous, since it can run in
305 parallel with Emacs; but you can think of it as synchronous in that
306 Emacs is essentially finished with the subprocess as soon as this
309 MS-DOS doesn't support asynchronous subprocesses, so this option doesn't
312 @item @code{(:file @var{file-name})}
313 Send the output to the file name specified, overwriting it if it
316 @item @code{(@var{real-destination} @var{error-destination})}
317 Keep the standard output stream separate from the standard error stream;
318 deal with the ordinary output as specified by @var{real-destination},
319 and dispose of the error output according to @var{error-destination}.
320 If @var{error-destination} is @code{nil}, that means to discard the
321 error output, @code{t} means mix it with the ordinary output, and a
322 string specifies a file name to redirect error output into.
324 You can't directly specify a buffer to put the error output in; that is
325 too difficult to implement. But you can achieve this result by sending
326 the error output to a temporary file and then inserting the file into a
330 If @var{display} is non-@code{nil}, then @code{call-process} redisplays
331 the buffer as output is inserted. (However, if the coding system chosen
332 for decoding output is @code{undecided}, meaning deduce the encoding
333 from the actual data, then redisplay sometimes cannot continue once
334 non-@acronym{ASCII} characters are encountered. There are fundamental
335 reasons why it is hard to fix this; see @ref{Output from Processes}.)
337 Otherwise the function @code{call-process} does no redisplay, and the
338 results become visible on the screen only when Emacs redisplays that
339 buffer in the normal course of events.
341 The remaining arguments, @var{args}, are strings that specify command
342 line arguments for the program.
344 The value returned by @code{call-process} (unless you told it not to
345 wait) indicates the reason for process termination. A number gives the
346 exit status of the subprocess; 0 means success, and any other value
347 means failure. If the process terminated with a signal,
348 @code{call-process} returns a string describing the signal.
350 In the examples below, the buffer @samp{foo} is current.
354 (call-process "pwd" nil t)
357 ---------- Buffer: foo ----------
359 ---------- Buffer: foo ----------
363 (call-process "grep" nil "bar" nil "lewis" "/etc/passwd")
366 ---------- Buffer: bar ----------
367 lewis:x:1001:1001:Bil Lewis,,,,:/home/lewis:/bin/bash
369 ---------- Buffer: bar ----------
373 Here is an example of the use of @code{call-process}, as used to
374 be found in the definition of the @code{insert-directory} function:
378 (call-process insert-directory-program nil t nil switches
380 (concat (file-name-as-directory file) ".")
386 @defun process-file program &optional infile buffer display &rest args
387 This function processes files synchronously in a separate process. It
388 is similar to @code{call-process}, but may invoke a file handler based
389 on the value of the variable @code{default-directory}, which specifies
390 the current working directory of the subprocess.
392 The arguments are handled in almost the same way as for
393 @code{call-process}, with the following differences:
395 Some file handlers may not support all combinations and forms of the
396 arguments @var{infile}, @var{buffer}, and @var{display}. For example,
397 some file handlers might behave as if @var{display} were @code{nil},
398 regardless of the value actually passed. As another example, some
399 file handlers might not support separating standard output and error
400 output by way of the @var{buffer} argument.
402 If a file handler is invoked, it determines the program to run based
403 on the first argument @var{program}. For instance, suppose that a
404 handler for remote files is invoked. Then the path that is used for
405 searching for the program might be different from @code{exec-path}.
407 The second argument @var{infile} may invoke a file handler. The file
408 handler could be different from the handler chosen for the
409 @code{process-file} function itself. (For example,
410 @code{default-directory} could be on one remote host, and
411 @var{infile} on a different remote host. Or @code{default-directory}
412 could be non-special, whereas @var{infile} is on a remote host.)
414 If @var{buffer} is a list of the form @code{(@var{real-destination}
415 @var{error-destination})}, and @var{error-destination} names a file,
416 then the same remarks as for @var{infile} apply.
418 The remaining arguments (@var{args}) will be passed to the process
419 verbatim. Emacs is not involved in processing file names that are
420 present in @var{args}. To avoid confusion, it may be best to avoid
421 absolute file names in @var{args}, but rather to specify all file
422 names as relative to @code{default-directory}. The function
423 @code{file-relative-name} is useful for constructing such relative
427 @defvar process-file-side-effects
428 This variable indicates whether a call of @code{process-file} changes
431 By default, this variable is always set to @code{t}, meaning that a
432 call of @code{process-file} could potentially change any file on a
433 remote host. When set to @code{nil}, a file handler could optimize
434 its behavior with respect to remote file attribute caching.
436 You should only ever change this variable with a let-binding; never
440 @defun call-process-region start end program &optional delete destination display &rest args
441 This function sends the text from @var{start} to @var{end} as
442 standard input to a process running @var{program}. It deletes the text
443 sent if @var{delete} is non-@code{nil}; this is useful when
444 @var{destination} is @code{t}, to insert the output in the current
445 buffer in place of the input.
447 The arguments @var{destination} and @var{display} control what to do
448 with the output from the subprocess, and whether to update the display
449 as it comes in. For details, see the description of
450 @code{call-process}, above. If @var{destination} is the integer 0,
451 @code{call-process-region} discards the output and returns @code{nil}
452 immediately, without waiting for the subprocess to finish (this only
453 works if asynchronous subprocesses are supported; i.e., not on MS-DOS).
455 The remaining arguments, @var{args}, are strings that specify command
456 line arguments for the program.
458 The return value of @code{call-process-region} is just like that of
459 @code{call-process}: @code{nil} if you told it to return without
460 waiting; otherwise, a number or string which indicates how the
461 subprocess terminated.
463 In the following example, we use @code{call-process-region} to run the
464 @code{cat} utility, with standard input being the first five characters
465 in buffer @samp{foo} (the word @samp{input}). @code{cat} copies its
466 standard input into its standard output. Since the argument
467 @var{destination} is @code{t}, this output is inserted in the current
472 ---------- Buffer: foo ----------
474 ---------- Buffer: foo ----------
478 (call-process-region 1 6 "cat" nil t)
481 ---------- Buffer: foo ----------
483 ---------- Buffer: foo ----------
487 For example, the @code{shell-command-on-region} command uses
488 @code{call-process-region} in a manner similar to this:
494 shell-file-name ; @r{name of program}
495 nil ; @r{do not delete region}
496 buffer ; @r{send output to @code{buffer}}
497 nil ; @r{no redisplay during output}
498 "-c" command) ; @r{arguments for the shell}
501 @c It actually uses shell-command-switch, but no need to mention that here.
504 @defun call-process-shell-command command &optional infile destination display &rest args
505 This function executes the shell command @var{command} synchronously.
506 The final arguments @var{args} are additional arguments to add at the
507 end of @var{command}. The other arguments are handled as in
511 @defun process-file-shell-command command &optional infile destination display &rest args
512 This function is like @code{call-process-shell-command}, but uses
513 @code{process-file} internally. Depending on @code{default-directory},
514 @var{command} can be executed also on remote hosts.
517 @defun shell-command-to-string command
518 This function executes @var{command} (a string) as a shell command,
519 then returns the command's output as a string.
522 @c There is also shell-command-on-region, but that is more of a user
523 @c command, not something to use in programs.
525 @defun process-lines program &rest args
526 This function runs @var{program}, waits for it to finish, and returns
527 its output as a list of strings. Each string in the list holds a
528 single line of text output by the program; the end-of-line characters
529 are stripped from each line. The arguments beyond @var{program},
530 @var{args}, are strings that specify command-line arguments with which
533 If @var{program} exits with a non-zero exit status, this function
536 This function works by calling @code{call-process}, so program output
537 is decoded in the same way as for @code{call-process}.
540 @node Asynchronous Processes
541 @section Creating an Asynchronous Process
542 @cindex asynchronous subprocess
544 In this section, we describe how to create an @dfn{asynchronous
545 process}. After an asynchronous process is created, it runs in
546 parallel with Emacs, and Emacs can communicate with it using the
547 functions described in the following sections (@pxref{Input to
548 Processes}, and @pxref{Output from Processes}). Note that process
549 communication is only partially asynchronous: Emacs sends data to the
550 process only when certain functions are called, and Emacs accepts data
551 from the process only while waiting for input or for a time delay.
555 An asynchronous process is controlled either via a @dfn{pty}
556 (pseudo-terminal) or a @dfn{pipe}. The choice of pty or pipe is made
557 when creating the process, based on the value of the variable
558 @code{process-connection-type} (see below). Ptys are usually
559 preferable for processes visible to the user, as in Shell mode,
560 because they allow for job control (@kbd{C-c}, @kbd{C-z}, etc.)@:
561 between the process and its children, whereas pipes do not. For
562 subprocesses used for internal purposes by programs, it is often
563 better to use a pipe, because they are more efficient, and because
564 they are immune to stray character injections that ptys introduce for
565 large (around 500 byte) messages. Also, the total number of ptys is
566 limited on many systems and it is good not to waste them.
568 @defun start-process name buffer-or-name program &rest args
569 This function creates a new asynchronous subprocess and starts the
570 program @var{program} running in it. It returns a process object that
571 stands for the new subprocess in Lisp. The argument @var{name}
572 specifies the name for the process object; if a process with this name
573 already exists, then @var{name} is modified (by appending @samp{<1>},
574 etc.)@: to be unique. The buffer @var{buffer-or-name} is the buffer to
575 associate with the process.
577 If @var{program} is @code{nil}, Emacs opens a new pseudoterminal (pty)
578 and associates its input and output with @var{buffer-or-name}, without
579 creating a subprocess. In that case, the remaining arguments
580 @var{args} are ignored.
582 The remaining arguments, @var{args}, are strings that specify command
583 line arguments for the subprocess.
585 In the example below, the first process is started and runs (rather,
586 sleeps) for 100 seconds (the output buffer @samp{foo} is created
587 immediately). Meanwhile, the second process is started, and
588 given the name @samp{my-process<1>} for the sake of uniqueness. It
589 inserts the directory listing at the end of the buffer @samp{foo},
590 before the first process finishes. Then it finishes, and a message to
591 that effect is inserted in the buffer. Much later, the first process
592 finishes, and another message is inserted in the buffer for it.
596 (start-process "my-process" "foo" "sleep" "100")
597 @result{} #<process my-process>
601 (start-process "my-process" "foo" "ls" "-l" "/bin")
602 @result{} #<process my-process<1>>
604 ---------- Buffer: foo ----------
606 -rwxr-xr-x 1 root root 971384 Mar 30 10:14 bash
607 -rwxr-xr-x 1 root root 146920 Jul 5 2011 bsd-csh
609 -rwxr-xr-x 1 root root 696880 Feb 28 15:55 zsh4
611 Process my-process<1> finished
613 Process my-process finished
614 ---------- Buffer: foo ----------
619 @defun start-file-process name buffer-or-name program &rest args
620 Like @code{start-process}, this function starts a new asynchronous
621 subprocess running @var{program} in it, and returns its process
624 The difference from @code{start-process} is that this function may
625 invoked a file handler based on the value of @code{default-directory}.
626 This handler ought to run @var{program}, perhaps on the local host,
627 perhaps on a remote host that corresponds to @code{default-directory}.
628 In the latter case, the local part of @code{default-directory} becomes
629 the working directory of the process.
631 This function does not try to invoke file name handlers for
632 @var{program} or for the @var{program-args}.
634 Depending on the implementation of the file handler, it might not be
635 possible to apply @code{process-filter} or @code{process-sentinel} to
636 the resulting process object. @xref{Filter Functions}, and @ref{Sentinels}.
638 @c FIXME Can we find a better example (i.e., a more modern function
639 @c that is actually documented).
640 Some file handlers may not support @code{start-file-process} (for
641 example the function @code{ange-ftp-hook-function}). In such cases,
642 this function does nothing and returns @code{nil}.
645 @defun start-process-shell-command name buffer-or-name command
646 This function is like @code{start-process}, except that it uses a shell
647 to execute the specified command. The argument @var{command} is a shell
648 command name. The variable @code{shell-file-name} specifies which shell to
651 The point of running a program through the shell, rather than directly
652 with @code{start-process}, is so that you can employ shell features such
653 as wildcards in the arguments. It follows that if you include any
654 arbitrary user-specified arguments in the command, you should quote them
655 with @code{shell-quote-argument} first, so that any special shell
656 characters do @emph{not} have their special shell meanings. @xref{Shell
657 Arguments}. Of course, when executing commands based on user input
658 you should also consider the security implications.
661 @defun start-file-process-shell-command name buffer-or-name command
662 This function is like @code{start-process-shell-command}, but uses
663 @code{start-file-process} internally. Because of this, @var{command}
664 can also be executed on remote hosts, depending on @code{default-directory}.
667 @defvar process-connection-type
668 This variable controls the type of device used to communicate with
669 asynchronous subprocesses. If it is non-@code{nil}, then ptys are
670 used, when available. Otherwise, pipes are used.
672 The value of @code{process-connection-type} takes effect when
673 @code{start-process} is called. So you can specify how to communicate
674 with one subprocess by binding the variable around the call to
675 @code{start-process}.
679 (let ((process-connection-type nil)) ; @r{use a pipe}
680 (start-process @dots{}))
684 To determine whether a given subprocess actually got a pipe or a pty,
685 use the function @code{process-tty-name} (@pxref{Process
689 @node Deleting Processes
690 @section Deleting Processes
691 @cindex deleting processes
693 @dfn{Deleting a process} disconnects Emacs immediately from the
694 subprocess. Processes are deleted automatically after they terminate,
695 but not necessarily right away. You can delete a process explicitly
696 at any time. If you explicitly delete a terminated process before it
697 is deleted automatically, no harm results. Deleting a running
698 process sends a signal to terminate it (and its child processes, if
699 any), and calls the process sentinel. @xref{Sentinels}.
701 When a process is deleted, the process object itself continues to
702 exist as long as other Lisp objects point to it. All the Lisp
703 primitives that work on process objects accept deleted processes, but
704 those that do I/O or send signals will report an error. The process
705 mark continues to point to the same place as before, usually into a
706 buffer where output from the process was being inserted.
708 @defopt delete-exited-processes
709 This variable controls automatic deletion of processes that have
710 terminated (due to calling @code{exit} or to a signal). If it is
711 @code{nil}, then they continue to exist until the user runs
712 @code{list-processes}. Otherwise, they are deleted immediately after
716 @defun delete-process process
717 This function deletes a process, killing it with a @code{SIGKILL}
718 signal. The argument may be a process, the name of a process, a
719 buffer, or the name of a buffer. (A buffer or buffer-name stands for
720 the process that @code{get-buffer-process} returns.) Calling
721 @code{delete-process} on a running process terminates it, updates the
722 process status, and runs the sentinel immediately. If the
723 process has already terminated, calling @code{delete-process} has no
724 effect on its status, or on the running of its sentinel (which will
725 happen sooner or later).
729 (delete-process "*shell*")
735 @node Process Information
736 @section Process Information
738 Several functions return information about processes.
740 @deffn Command list-processes &optional query-only buffer
741 This command displays a listing of all living processes. In addition,
742 it finally deletes any process whose status was @samp{Exited} or
743 @samp{Signaled}. It returns @code{nil}.
745 The processes are shown in a buffer named @file{*Process List*}
746 (unless you specify otherwise using the optional argument @var{buffer}),
747 whose major mode is Process Menu mode.
749 If @var{query-only} is non-@code{nil}, it only lists processes
750 whose query flag is non-@code{nil}. @xref{Query Before Exit}.
754 This function returns a list of all processes that have not been deleted.
759 @result{} (#<process display-time> #<process shell>)
764 @defun get-process name
765 This function returns the process named @var{name} (a string), or
766 @code{nil} if there is none.
770 (get-process "shell")
771 @result{} #<process shell>
776 @defun process-command process
777 This function returns the command that was executed to start
778 @var{process}. This is a list of strings, the first string being the
779 program executed and the rest of the strings being the arguments that
780 were given to the program.
784 (process-command (get-process "shell"))
785 @result{} ("bash" "-i")
790 @defun process-contact process &optional key
792 This function returns information about how a network or serial
793 process was set up. When @var{key} is @code{nil}, it returns
794 @code{(@var{hostname} @var{service})} for a network process, and
795 @code{(@var{port} @var{speed})} for a serial process.
796 For an ordinary child process, this function always returns @code{t}.
798 If @var{key} is @code{t}, the value is the complete status information
799 for the connection, server, or serial port; that is, the list of
800 keywords and values specified in @code{make-network-process} or
801 @code{make-serial-process}, except that some of the values represent
802 the current status instead of what you specified.
804 For a network process, the values include (see
805 @code{make-network-process} for a complete list):
809 The associated value is the process buffer.
811 The associated value is the process filter function.
813 The associated value is the process sentinel function.
815 In a connection, the address in internal format of the remote peer.
817 The local address, in internal format.
819 In a server, if you specified @code{t} for @var{service},
820 this value is the actual port number.
823 @code{:local} and @code{:remote} are included even if they were not
824 specified explicitly in @code{make-network-process}.
826 For a serial process, see @code{make-serial-process} and
827 @code{serial-process-configure} for a list of keys.
829 If @var{key} is a keyword, the function returns the value corresponding
833 @defun process-id process
834 This function returns the @acronym{PID} of @var{process}. This is an
835 integer that distinguishes the process @var{process} from all other
836 processes running on the same computer at the current time. The
837 @acronym{PID} of a process is chosen by the operating system kernel when the
838 process is started and remains constant as long as the process exists.
841 @defun process-name process
842 This function returns the name of @var{process}, as a string.
845 @defun process-status process-name
846 This function returns the status of @var{process-name} as a symbol.
847 The argument @var{process-name} must be a process, a buffer, or a
848 process name (a string).
850 The possible values for an actual subprocess are:
854 for a process that is running.
856 for a process that is stopped but continuable.
858 for a process that has exited.
860 for a process that has received a fatal signal.
862 for a network connection that is open.
864 for a network connection that is closed. Once a connection
865 is closed, you cannot reopen it, though you might be able to open
866 a new connection to the same place.
868 for a non-blocking connection that is waiting to complete.
870 for a non-blocking connection that has failed to complete.
872 for a network server that is listening.
874 if @var{process-name} is not the name of an existing process.
879 (process-status (get-buffer "*shell*"))
884 For a network connection, @code{process-status} returns one of the symbols
885 @code{open} or @code{closed}. The latter means that the other side
886 closed the connection, or Emacs did @code{delete-process}.
889 @defun process-live-p process
890 This function returns non-@code{nil} if @var{process} is alive. A
891 process is considered alive if its status is @code{run}, @code{open},
892 @code{listen}, @code{connect} or @code{stop}.
895 @defun process-type process
896 This function returns the symbol @code{network} for a network
897 connection or server, @code{serial} for a serial port connection, or
898 @code{real} for a real subprocess.
901 @defun process-exit-status process
902 This function returns the exit status of @var{process} or the signal
903 number that killed it. (Use the result of @code{process-status} to
904 determine which of those it is.) If @var{process} has not yet
905 terminated, the value is 0.
908 @defun process-tty-name process
909 This function returns the terminal name that @var{process} is using for
910 its communication with Emacs---or @code{nil} if it is using pipes
911 instead of a terminal (see @code{process-connection-type} in
912 @ref{Asynchronous Processes}). If @var{process} represents a program
913 running on a remote host, the terminal name used by that program on
914 the remote host is provided as process property @code{remote-tty}.
917 @defun process-coding-system process
918 @anchor{Coding systems for a subprocess}
919 This function returns a cons cell @code{(@var{decode} . @var{encode})},
920 describing the coding systems in use for decoding output from, and
921 encoding input to, @var{process} (@pxref{Coding Systems}).
924 @defun set-process-coding-system process &optional decoding-system encoding-system
925 This function specifies the coding systems to use for subsequent output
926 from and input to @var{process}. It will use @var{decoding-system} to
927 decode subprocess output, and @var{encoding-system} to encode subprocess
931 Every process also has a property list that you can use to store
932 miscellaneous values associated with the process.
934 @defun process-get process propname
935 This function returns the value of the @var{propname} property
939 @defun process-put process propname value
940 This function sets the value of the @var{propname} property
941 of @var{process} to @var{value}.
944 @defun process-plist process
945 This function returns the process plist of @var{process}.
948 @defun set-process-plist process plist
949 This function sets the process plist of @var{process} to @var{plist}.
952 @node Input to Processes
953 @section Sending Input to Processes
954 @cindex process input
956 Asynchronous subprocesses receive input when it is sent to them by
957 Emacs, which is done with the functions in this section. You must
958 specify the process to send input to, and the input data to send. The
959 data appears on the ``standard input'' of the subprocess.
962 Some operating systems have limited space for buffered input in a
963 pty. On these systems, Emacs sends an @acronym{EOF} periodically
964 amidst the other characters, to force them through. For most
965 programs, these @acronym{EOF}s do no harm.
967 Subprocess input is normally encoded using a coding system before the
968 subprocess receives it, much like text written into a file. You can use
969 @code{set-process-coding-system} to specify which coding system to use
970 (@pxref{Process Information}). Otherwise, the coding system comes from
971 @code{coding-system-for-write}, if that is non-@code{nil}; or else from
972 the defaulting mechanism (@pxref{Default Coding Systems}).
974 Sometimes the system is unable to accept input for that process,
975 because the input buffer is full. When this happens, the send functions
976 wait a short while, accepting output from subprocesses, and then try
977 again. This gives the subprocess a chance to read more of its pending
978 input and make space in the buffer. It also allows filters, sentinels
979 and timers to run---so take account of that in writing your code.
981 In these functions, the @var{process} argument can be a process or
982 the name of a process, or a buffer or buffer name (which stands
983 for a process via @code{get-buffer-process}). @code{nil} means
984 the current buffer's process.
986 @defun process-send-string process string
987 This function sends @var{process} the contents of @var{string} as
988 standard input. It returns @code{nil}. For example, to make a
989 Shell buffer list files:
993 (process-send-string "shell<1>" "ls\n")
999 @defun process-send-region process start end
1000 This function sends the text in the region defined by @var{start} and
1001 @var{end} as standard input to @var{process}.
1003 An error is signaled unless both @var{start} and @var{end} are
1004 integers or markers that indicate positions in the current buffer. (It
1005 is unimportant which number is larger.)
1008 @defun process-send-eof &optional process
1009 This function makes @var{process} see an end-of-file in its
1010 input. The @acronym{EOF} comes after any text already sent to it.
1011 The function returns @var{process}.
1015 (process-send-eof "shell")
1021 @defun process-running-child-p &optional process
1022 This function will tell you whether a @var{process} has given control of
1023 its terminal to its own child process. The value is @code{t} if this is
1024 true, or if Emacs cannot tell; it is @code{nil} if Emacs can be certain
1025 that this is not so.
1028 @node Signals to Processes
1029 @section Sending Signals to Processes
1030 @cindex process signals
1031 @cindex sending signals
1034 @dfn{Sending a signal} to a subprocess is a way of interrupting its
1035 activities. There are several different signals, each with its own
1036 meaning. The set of signals and their names is defined by the operating
1037 system. For example, the signal @code{SIGINT} means that the user has
1038 typed @kbd{C-c}, or that some analogous thing has happened.
1040 Each signal has a standard effect on the subprocess. Most signals
1041 kill the subprocess, but some stop (or resume) execution instead. Most
1042 signals can optionally be handled by programs; if the program handles
1043 the signal, then we can say nothing in general about its effects.
1045 You can send signals explicitly by calling the functions in this
1046 section. Emacs also sends signals automatically at certain times:
1047 killing a buffer sends a @code{SIGHUP} signal to all its associated
1048 processes; killing Emacs sends a @code{SIGHUP} signal to all remaining
1049 processes. (@code{SIGHUP} is a signal that usually indicates that the
1050 user ``hung up the phone'', i.e., disconnected.)
1052 Each of the signal-sending functions takes two optional arguments:
1053 @var{process} and @var{current-group}.
1055 The argument @var{process} must be either a process, a process
1056 name, a buffer, a buffer name, or @code{nil}. A buffer or buffer name
1057 stands for a process through @code{get-buffer-process}. @code{nil}
1058 stands for the process associated with the current buffer. An error
1059 is signaled if @var{process} does not identify a process.
1061 The argument @var{current-group} is a flag that makes a difference
1062 when you are running a job-control shell as an Emacs subprocess. If it
1063 is non-@code{nil}, then the signal is sent to the current process-group
1064 of the terminal that Emacs uses to communicate with the subprocess. If
1065 the process is a job-control shell, this means the shell's current
1066 subjob. If it is @code{nil}, the signal is sent to the process group of
1067 the immediate subprocess of Emacs. If the subprocess is a job-control
1068 shell, this is the shell itself.
1070 The flag @var{current-group} has no effect when a pipe is used to
1071 communicate with the subprocess, because the operating system does not
1072 support the distinction in the case of pipes. For the same reason,
1073 job-control shells won't work when a pipe is used. See
1074 @code{process-connection-type} in @ref{Asynchronous Processes}.
1076 @defun interrupt-process &optional process current-group
1077 This function interrupts the process @var{process} by sending the
1078 signal @code{SIGINT}. Outside of Emacs, typing the ``interrupt
1079 character'' (normally @kbd{C-c} on some systems, and @key{DEL} on
1080 others) sends this signal. When the argument @var{current-group} is
1081 non-@code{nil}, you can think of this function as ``typing @kbd{C-c}''
1082 on the terminal by which Emacs talks to the subprocess.
1085 @defun kill-process &optional process current-group
1086 This function kills the process @var{process} by sending the
1087 signal @code{SIGKILL}. This signal kills the subprocess immediately,
1088 and cannot be handled by the subprocess.
1091 @defun quit-process &optional process current-group
1092 This function sends the signal @code{SIGQUIT} to the process
1093 @var{process}. This signal is the one sent by the ``quit
1094 @c FIXME? Never heard of C-b being used for this. In readline, e.g.,
1095 @c bash, that is backward-word.
1096 character'' (usually @kbd{C-b} or @kbd{C-\}) when you are not inside
1100 @defun stop-process &optional process current-group
1101 This function stops the process @var{process} by sending the
1102 signal @code{SIGTSTP}. Use @code{continue-process} to resume its
1105 Outside of Emacs, on systems with job control, the ``stop character''
1106 (usually @kbd{C-z}) normally sends this signal. When
1107 @var{current-group} is non-@code{nil}, you can think of this function as
1108 ``typing @kbd{C-z}'' on the terminal Emacs uses to communicate with the
1112 @defun continue-process &optional process current-group
1113 This function resumes execution of the process @var{process} by sending
1114 it the signal @code{SIGCONT}. This presumes that @var{process} was
1118 @deffn Command signal-process process signal
1119 This function sends a signal to process @var{process}. The argument
1120 @var{signal} specifies which signal to send; it should be an integer,
1121 or a symbol whose name is a signal.
1123 The @var{process} argument can be a system process @acronym{ID} (an
1124 integer); that allows you to send signals to processes that are not
1125 children of Emacs. @xref{System Processes}.
1128 @node Output from Processes
1129 @section Receiving Output from Processes
1130 @cindex process output
1131 @cindex output from processes
1133 The output that a subprocess writes to its standard output stream
1134 is passed to a function called the @dfn{filter function}. The default
1135 filter function simply inserts the output into a buffer, which is
1136 called the associated buffer of the process (@pxref{Process
1137 Buffers}). If the process has no buffer then the default filter
1138 discards the output.
1140 When a subprocess terminates, Emacs reads any pending output,
1141 then stops reading output from that subprocess. Therefore, if the
1142 subprocess has children that are still live and still producing
1143 output, Emacs won't receive that output.
1145 Output from a subprocess can arrive only while Emacs is waiting: when
1146 reading terminal input (see the function @code{waiting-for-user-input-p}),
1147 in @code{sit-for} and @code{sleep-for} (@pxref{Waiting}), and in
1148 @code{accept-process-output} (@pxref{Accepting Output}). This
1149 minimizes the problem of timing errors that usually plague parallel
1150 programming. For example, you can safely create a process and only
1151 then specify its buffer or filter function; no output can arrive
1152 before you finish, if the code in between does not call any primitive
1155 @defvar process-adaptive-read-buffering
1156 On some systems, when Emacs reads the output from a subprocess, the
1157 output data is read in very small blocks, potentially resulting in
1158 very poor performance. This behavior can be remedied to some extent
1159 by setting the variable @code{process-adaptive-read-buffering} to a
1160 non-@code{nil} value (the default), as it will automatically delay reading
1161 from such processes, thus allowing them to produce more output before
1162 Emacs tries to read it.
1165 It is impossible to separate the standard output and standard error
1166 streams of the subprocess, because Emacs normally spawns the subprocess
1167 inside a pseudo-TTY, and a pseudo-TTY has only one output channel. If
1168 you want to keep the output to those streams separate, you should
1169 redirect one of them to a file---for example, by using an appropriate
1173 * Process Buffers:: By default, output is put in a buffer.
1174 * Filter Functions:: Filter functions accept output from the process.
1175 * Decoding Output:: Filters can get unibyte or multibyte strings.
1176 * Accepting Output:: How to wait until process output arrives.
1179 @node Process Buffers
1180 @subsection Process Buffers
1182 A process can (and usually does) have an @dfn{associated buffer},
1183 which is an ordinary Emacs buffer that is used for two purposes: storing
1184 the output from the process, and deciding when to kill the process. You
1185 can also use the buffer to identify a process to operate on, since in
1186 normal practice only one process is associated with any given buffer.
1187 Many applications of processes also use the buffer for editing input to
1188 be sent to the process, but this is not built into Emacs Lisp.
1190 By default, process output is inserted in the associated buffer.
1191 (You can change this by defining a custom filter function,
1192 @pxref{Filter Functions}.) The position to insert the output is
1193 determined by the @code{process-mark}, which is then updated to point
1194 to the end of the text just inserted. Usually, but not always, the
1195 @code{process-mark} is at the end of the buffer.
1197 @findex process-kill-buffer-query-function
1198 Killing the associated buffer of a process also kills the process.
1199 Emacs asks for confirmation first, if the process's
1200 @code{process-query-on-exit-flag} is non-@code{nil} (@pxref{Query
1201 Before Exit}). This confirmation is done by the function
1202 @code{process-kill-buffer-query-function}, which is run from
1203 @code{kill-buffer-query-functions} (@pxref{Killing Buffers}).
1205 @defun process-buffer process
1206 This function returns the associated buffer of the process
1211 (process-buffer (get-process "shell"))
1212 @result{} #<buffer *shell*>
1217 @defun process-mark process
1218 This function returns the process marker for @var{process}, which is the
1219 marker that says where to insert output from the process.
1221 If @var{process} does not have a buffer, @code{process-mark} returns a
1222 marker that points nowhere.
1224 The default filter function uses this marker to decide where to
1225 insert process output, and updates it to point after the inserted text.
1226 That is why successive batches of output are inserted consecutively.
1228 Custom filter functions normally should use this marker in the same fashion.
1229 For an example of a filter function that uses @code{process-mark},
1230 @pxref{Process Filter Example}.
1232 When the user is expected to enter input in the process buffer for
1233 transmission to the process, the process marker separates the new input
1234 from previous output.
1237 @defun set-process-buffer process buffer
1238 This function sets the buffer associated with @var{process} to
1239 @var{buffer}. If @var{buffer} is @code{nil}, the process becomes
1240 associated with no buffer.
1243 @defun get-buffer-process buffer-or-name
1244 This function returns a nondeleted process associated with the buffer
1245 specified by @var{buffer-or-name}. If there are several processes
1246 associated with it, this function chooses one (currently, the one most
1247 recently created, but don't count on that). Deletion of a process
1248 (see @code{delete-process}) makes it ineligible for this function to
1251 It is usually a bad idea to have more than one process associated with
1256 (get-buffer-process "*shell*")
1257 @result{} #<process shell>
1261 Killing the process's buffer deletes the process, which kills the
1262 subprocess with a @code{SIGHUP} signal (@pxref{Signals to Processes}).
1265 @node Filter Functions
1266 @subsection Process Filter Functions
1267 @cindex filter function
1268 @cindex process filter
1270 A process @dfn{filter function} is a function that receives the
1271 standard output from the associated process. @emph{All} output from
1272 that process is passed to the filter. The default filter simply
1273 outputs directly to the process buffer.
1275 The filter function can only be called when Emacs is waiting for
1276 something, because process output arrives only at such times. Emacs
1277 waits when reading terminal input (see the function
1278 @code{waiting-for-user-input-p}), in @code{sit-for} and
1279 @code{sleep-for} (@pxref{Waiting}), and in
1280 @code{accept-process-output} (@pxref{Accepting Output}).
1282 A filter function must accept two arguments: the associated process
1283 and a string, which is output just received from it. The function is
1284 then free to do whatever it chooses with the output.
1286 @c Note this text is duplicated in the sentinels section.
1287 Quitting is normally inhibited within a filter function---otherwise,
1288 the effect of typing @kbd{C-g} at command level or to quit a user
1289 command would be unpredictable. If you want to permit quitting inside
1290 a filter function, bind @code{inhibit-quit} to @code{nil}. In most
1291 cases, the right way to do this is with the macro
1292 @code{with-local-quit}. @xref{Quitting}.
1294 If an error happens during execution of a filter function, it is
1295 caught automatically, so that it doesn't stop the execution of whatever
1296 program was running when the filter function was started. However, if
1297 @code{debug-on-error} is non-@code{nil}, errors are not caught.
1298 This makes it possible to use the Lisp debugger to debug the
1299 filter function. @xref{Debugger}.
1301 Many filter functions sometimes (or always) insert the output in the
1302 process's buffer, mimicking the actions of the default filter.
1303 Such filter functions need to make sure that they save the
1304 current buffer, select the correct buffer (if different) before
1305 inserting output, and then restore the original buffer.
1306 They should also check whether the buffer is still alive, update the
1307 process marker, and in some cases update the value of point. Here is
1308 how to do these things:
1310 @anchor{Process Filter Example}
1313 (defun ordinary-insertion-filter (proc string)
1314 (when (buffer-live-p (process-buffer proc))
1315 (with-current-buffer (process-buffer proc)
1316 (let ((moving (= (point) (process-mark proc))))
1320 ;; @r{Insert the text, advancing the process marker.}
1321 (goto-char (process-mark proc))
1323 (set-marker (process-mark proc) (point)))
1324 (if moving (goto-char (process-mark proc)))))))
1328 To make the filter force the process buffer to be visible whenever new
1329 text arrives, you could insert a line like the following just before the
1330 @code{with-current-buffer} construct:
1333 (display-buffer (process-buffer proc))
1336 To force point to the end of the new output, no matter where it was
1337 previously, eliminate the variable @code{moving} and call
1338 @code{goto-char} unconditionally.
1341 In earlier Emacs versions, every filter function that did regular
1342 expression searching or matching had to explicitly save and restore the
1343 match data. Now Emacs does this automatically for filter functions;
1344 they never need to do it explicitly.
1346 Note that Emacs automatically saves and restores the match data
1347 while executing filter functions. @xref{Match Data}.
1349 The output to the filter may come in chunks of any size. A program
1350 that produces the same output twice in a row may send it as one batch of
1351 200 characters one time, and five batches of 40 characters the next. If
1352 the filter looks for certain text strings in the subprocess output, make
1353 sure to handle the case where one of these strings is split across two
1354 or more batches of output; one way to do this is to insert the
1355 received text into a temporary buffer, which can then be searched.
1357 @defun set-process-filter process filter
1358 This function gives @var{process} the filter function @var{filter}. If
1359 @var{filter} is @code{nil}, it gives the process the default filter,
1360 which inserts the process output into the process buffer.
1363 @defun process-filter process
1364 This function returns the filter function of @var{process}.
1367 Here is an example of the use of a filter function:
1371 (defun keep-output (process output)
1372 (setq kept (cons output kept)))
1373 @result{} keep-output
1380 (set-process-filter (get-process "shell") 'keep-output)
1381 @result{} keep-output
1384 (process-send-string "shell" "ls ~/other\n")
1387 @result{} ("lewis@@slug:$ "
1390 "FINAL-W87-SHORT.MSS backup.otl kolstad.mss~
1391 address.txt backup.psf kolstad.psf
1392 backup.bib~ david.mss resume-Dec-86.mss~
1393 backup.err david.psf resume-Dec.psf
1394 backup.mss dland syllabus.mss
1396 "#backups.mss# backup.mss~ kolstad.mss
1401 @ignore @c The code in this example doesn't show the right way to do things.
1402 Here is another, more realistic example, which demonstrates how to use
1403 the process mark to do insertion in the same fashion as the default filter:
1407 ;; @r{Insert input in the buffer specified by @code{my-shell-buffer}}
1408 ;; @r{and make sure that buffer is shown in some window.}
1409 (defun my-process-filter (proc str)
1410 (let ((cur (selected-window))
1412 (pop-to-buffer my-shell-buffer)
1415 (goto-char (point-max))
1417 (set-marker (process-mark proc) (point-max))
1418 (select-window cur)))
1423 @node Decoding Output
1424 @subsection Decoding Process Output
1425 @cindex decode process output
1427 When Emacs writes process output directly into a multibyte buffer,
1428 it decodes the output according to the process output coding system.
1429 If the coding system is @code{raw-text} or @code{no-conversion}, Emacs
1430 converts the unibyte output to multibyte using
1431 @code{string-to-multibyte}, and inserts the resulting multibyte text.
1433 You can use @code{set-process-coding-system} to specify which coding
1434 system to use (@pxref{Process Information}). Otherwise, the coding
1435 system comes from @code{coding-system-for-read}, if that is
1436 non-@code{nil}; or else from the defaulting mechanism (@pxref{Default
1437 Coding Systems}). If the text output by a process contains null
1438 bytes, Emacs by default uses @code{no-conversion} for it; see
1439 @ref{Lisp and Coding Systems, inhibit-null-byte-detection}, for how to
1440 control this behavior.
1442 @strong{Warning:} Coding systems such as @code{undecided}, which
1443 determine the coding system from the data, do not work entirely
1444 reliably with asynchronous subprocess output. This is because Emacs
1445 has to process asynchronous subprocess output in batches, as it
1446 arrives. Emacs must try to detect the proper coding system from one
1447 batch at a time, and this does not always work. Therefore, if at all
1448 possible, specify a coding system that determines both the character
1449 code conversion and the end of line conversion---that is, one like
1450 @code{latin-1-unix}, rather than @code{undecided} or @code{latin-1}.
1452 @c Let's keep the index entries that were there for
1453 @c set-process-filter-multibyte and process-filter-multibyte-p,
1454 @cindex filter multibyte flag, of process
1455 @cindex process filter multibyte flag
1456 When Emacs calls a process filter function, it provides the process
1457 output as a multibyte string or as a unibyte string according to the
1458 process's filter coding system. Emacs
1459 decodes the output according to the process output coding system,
1460 which usually produces a multibyte string, except for coding systems
1461 such as @code{binary} and @code{raw-text}.
1463 @node Accepting Output
1464 @subsection Accepting Output from Processes
1465 @cindex accept input from processes
1467 Output from asynchronous subprocesses normally arrives only while
1468 Emacs is waiting for some sort of external event, such as elapsed time
1469 or terminal input. Occasionally it is useful in a Lisp program to
1470 explicitly permit output to arrive at a specific point, or even to wait
1471 until output arrives from a process.
1473 @defun accept-process-output &optional process seconds millisec just-this-one
1474 This function allows Emacs to read pending output from processes. The
1475 output is given to their filter functions. If @var{process} is
1476 non-@code{nil} then this function does not return until some output
1477 has been received from @var{process}.
1479 The arguments @var{seconds} and @var{millisec} let you specify timeout
1480 periods. The former specifies a period measured in seconds and the
1481 latter specifies one measured in milliseconds. The two time periods
1482 thus specified are added together, and @code{accept-process-output}
1483 returns after that much time, whether or not there has been any
1486 The argument @var{millisec} is obsolete (and should not be used),
1487 because @var{seconds} can be floating point to specify
1488 waiting a fractional number of seconds. If @var{seconds} is 0, the
1489 function accepts whatever output is pending but does not wait.
1491 @c Emacs 22.1 feature
1492 If @var{process} is a process, and the argument @var{just-this-one} is
1493 non-@code{nil}, only output from that process is handled, suspending output
1494 from other processes until some output has been received from that
1495 process or the timeout expires. If @var{just-this-one} is an integer,
1496 also inhibit running timers. This feature is generally not
1497 recommended, but may be necessary for specific applications, such as
1500 The function @code{accept-process-output} returns non-@code{nil} if it
1501 did get some output, or @code{nil} if the timeout expired before output
1506 @section Sentinels: Detecting Process Status Changes
1507 @cindex process sentinel
1508 @cindex sentinel (of process)
1510 A @dfn{process sentinel} is a function that is called whenever the
1511 associated process changes status for any reason, including signals
1512 (whether sent by Emacs or caused by the process's own actions) that
1513 terminate, stop, or continue the process. The process sentinel is
1514 also called if the process exits. The sentinel receives two
1515 arguments: the process for which the event occurred, and a string
1516 describing the type of event.
1518 The string describing the event looks like one of the following:
1520 @c FIXME? Also "killed\n" - see example below?
1523 @code{"finished\n"}.
1526 @code{"exited abnormally with code @var{exitcode}\n"}.
1529 @code{"@var{name-of-signal}\n"}.
1532 @code{"@var{name-of-signal} (core dumped)\n"}.
1535 A sentinel runs only while Emacs is waiting (e.g., for terminal
1536 input, or for time to elapse, or for process output). This avoids the
1537 timing errors that could result from running sentinels at random places in
1538 the middle of other Lisp programs. A program can wait, so that
1539 sentinels will run, by calling @code{sit-for} or @code{sleep-for}
1540 (@pxref{Waiting}), or @code{accept-process-output} (@pxref{Accepting
1541 Output}). Emacs also allows sentinels to run when the command loop is
1542 reading input. @code{delete-process} calls the sentinel when it
1543 terminates a running process.
1545 Emacs does not keep a queue of multiple reasons to call the sentinel
1546 of one process; it records just the current status and the fact that
1547 there has been a change. Therefore two changes in status, coming in
1548 quick succession, can call the sentinel just once. However, process
1549 termination will always run the sentinel exactly once. This is
1550 because the process status can't change again after termination.
1552 Emacs explicitly checks for output from the process before running
1553 the process sentinel. Once the sentinel runs due to process
1554 termination, no further output can arrive from the process.
1556 A sentinel that writes the output into the buffer of the process
1557 should check whether the buffer is still alive. If it tries to insert
1558 into a dead buffer, it will get an error. If the buffer is dead,
1559 @code{(buffer-name (process-buffer @var{process}))} returns @code{nil}.
1561 @c Note this text is duplicated in the filter functions section.
1562 Quitting is normally inhibited within a sentinel---otherwise, the
1563 effect of typing @kbd{C-g} at command level or to quit a user command
1564 would be unpredictable. If you want to permit quitting inside a
1565 sentinel, bind @code{inhibit-quit} to @code{nil}. In most cases, the
1566 right way to do this is with the macro @code{with-local-quit}.
1569 If an error happens during execution of a sentinel, it is caught
1570 automatically, so that it doesn't stop the execution of whatever
1571 programs was running when the sentinel was started. However, if
1572 @code{debug-on-error} is non-@code{nil}, errors are not caught.
1573 This makes it possible to use the Lisp debugger to debug the
1574 sentinel. @xref{Debugger}.
1576 While a sentinel is running, the process sentinel is temporarily
1577 set to @code{nil} so that the sentinel won't run recursively.
1578 For this reason it is not possible for a sentinel to specify
1582 In earlier Emacs versions, every sentinel that did regular expression
1583 searching or matching had to explicitly save and restore the match data.
1584 Now Emacs does this automatically for sentinels; they never need to do
1587 Note that Emacs automatically saves and restores the match data
1588 while executing sentinels. @xref{Match Data}.
1590 @defun set-process-sentinel process sentinel
1591 This function associates @var{sentinel} with @var{process}. If
1592 @var{sentinel} is @code{nil}, then the process will have the default
1593 sentinel, which inserts a message in the process's buffer when the
1594 process status changes.
1596 Changes in process sentinels take effect immediately---if the sentinel
1597 is slated to be run but has not been called yet, and you specify a new
1598 sentinel, the eventual call to the sentinel will use the new one.
1602 (defun msg-me (process event)
1604 (format "Process: %s had the event `%s'" process event)))
1605 (set-process-sentinel (get-process "shell") 'msg-me)
1609 (kill-process (get-process "shell"))
1610 @print{} Process: #<process shell> had the event `killed'
1611 @result{} #<process shell>
1616 @defun process-sentinel process
1617 This function returns the sentinel of @var{process}.
1620 @defun waiting-for-user-input-p
1621 While a sentinel or filter function is running, this function returns
1622 non-@code{nil} if Emacs was waiting for keyboard input from the user at
1623 the time the sentinel or filter function was called, or @code{nil} if it
1627 @node Query Before Exit
1628 @section Querying Before Exit
1630 When Emacs exits, it terminates all its subprocesses by sending them
1631 the @code{SIGHUP} signal. Because subprocesses may be doing
1632 valuable work, Emacs normally asks the user to confirm that it is ok
1633 to terminate them. Each process has a query flag, which, if
1634 non-@code{nil}, says that Emacs should ask for confirmation before
1635 exiting and thus killing that process. The default for the query flag
1636 is @code{t}, meaning @emph{do} query.
1638 @defun process-query-on-exit-flag process
1639 This returns the query flag of @var{process}.
1642 @defun set-process-query-on-exit-flag process flag
1643 This function sets the query flag of @var{process} to @var{flag}. It
1646 Here is an example of using @code{set-process-query-on-exit-flag} on a
1647 shell process to avoid querying:
1651 (set-process-query-on-exit-flag (get-process "shell") nil)
1657 @node System Processes
1658 @section Accessing Other Processes
1659 @cindex system processes
1661 In addition to accessing and manipulating processes that are
1662 subprocesses of the current Emacs session, Emacs Lisp programs can
1663 also access other processes running on the same machine. We call
1664 these @dfn{system processes}, to distinguish them from Emacs
1667 Emacs provides several primitives for accessing system processes.
1668 Not all platforms support these primitives; on those which don't,
1669 these primitives return @code{nil}.
1671 @defun list-system-processes
1672 This function returns a list of all the processes running on the
1673 system. Each process is identified by its @acronym{PID}, a numerical
1674 process ID that is assigned by the OS and distinguishes the process
1675 from all the other processes running on the same machine at the same
1679 @defun process-attributes pid
1680 This function returns an alist of attributes for the process specified
1681 by its process ID @var{pid}. Each association in the alist is of the
1682 form @code{(@var{key} . @var{value})}, where @var{key} designates the
1683 attribute and @var{value} is the value of that attribute. The various
1684 attribute @var{key}s that this function can return are listed below.
1685 Not all platforms support all of these attributes; if an attribute is
1686 not supported, its association will not appear in the returned alist.
1687 Values that are numbers can be either integer or floating point,
1688 depending on the magnitude of the value.
1692 The effective user ID of the user who invoked the process. The
1693 corresponding @var{value} is a number. If the process was invoked by
1694 the same user who runs the current Emacs session, the value is
1695 identical to what @code{user-uid} returns (@pxref{User
1699 User name corresponding to the process's effective user ID, a string.
1702 The group ID of the effective user ID, a number.
1705 Group name corresponding to the effective user's group ID, a string.
1708 The name of the command that runs in the process. This is a string
1709 that usually specifies the name of the executable file of the process,
1710 without the leading directories. However, some special system
1711 processes can report strings that do not correspond to an executable
1715 The state code of the process. This is a short string that encodes
1716 the scheduling state of the process. Here's a list of the most
1717 frequently seen codes:
1721 uninterruptible sleep (usually I/O)
1725 interruptible sleep (waiting for some event)
1727 stopped, e.g., by a job control signal
1729 ``zombie'': a process that terminated, but was not reaped by its parent
1733 For the full list of the possible states, see the manual page of the
1734 @command{ps} command.
1737 The process ID of the parent process, a number.
1740 The process group ID of the process, a number.
1743 The session ID of the process. This is a number that is the process
1744 ID of the process's @dfn{session leader}.
1747 A string that is the name of the process's controlling terminal. On
1748 Unix and GNU systems, this is normally the file name of the
1749 corresponding terminal device, such as @file{/dev/pts65}.
1752 The numerical process group ID of the foreground process group that
1753 uses the process's terminal.
1756 The number of minor page faults caused by the process since its
1757 beginning. (Minor page faults are those that don't involve reading
1761 The number of major page faults caused by the process since its
1762 beginning. (Major page faults require a disk to be read, and are thus
1763 more expensive than minor page faults.)
1767 Like @code{minflt} and @code{majflt}, but include the number of page
1768 faults for all the child processes of the given process.
1771 Time spent by the process in the user context, for running the
1772 application's code. The corresponding @var{value} is in the
1773 @w{@code{(@var{high} @var{low} @var{microsec} @var{picosec})}} format, the same
1774 format used by functions @code{current-time} (@pxref{Time of Day,
1775 current-time}) and @code{file-attributes} (@pxref{File Attributes}).
1778 Time spent by the process in the system (kernel) context, for
1779 processing system calls. The corresponding @var{value} is in the same
1780 format as for @code{utime}.
1783 The sum of @code{utime} and @code{stime}. The corresponding
1784 @var{value} is in the same format as for @code{utime}.
1789 Like @code{utime}, @code{stime}, and @code{time}, but include the
1790 times of all the child processes of the given process.
1793 The numerical priority of the process.
1796 The @dfn{nice value} of the process, a number. (Processes with smaller
1797 nice values get scheduled more favorably.)
1800 The number of threads in the process.
1803 The time when the process was started, in the same
1804 @code{(@var{high} @var{low} @var{microsec} @var{picosec})} format used by
1805 @code{file-attributes} and @code{current-time}.
1808 The time elapsed since the process started, in the format @code{(@var{high}
1809 @var{low} @var{microsec} @var{picosec})}.
1812 The virtual memory size of the process, measured in kilobytes.
1815 The size of the process's @dfn{resident set}, the number of kilobytes
1816 occupied by the process in the machine's physical memory.
1819 The percentage of the CPU time used by the process since it started.
1820 The corresponding @var{value} is a floating-point number between 0 and
1824 The percentage of the total physical memory installed on the machine
1825 used by the process's resident set. The value is a floating-point
1826 number between 0 and 100.
1829 The command-line with which the process was invoked. This is a string
1830 in which individual command-line arguments are separated by blanks;
1831 whitespace characters that are embedded in the arguments are quoted as
1832 appropriate for the system's shell: escaped by backslash characters on
1833 GNU and Unix, and enclosed in double quote characters on Windows.
1834 Thus, this command-line string can be directly used in primitives such
1835 as @code{shell-command}.
1841 @node Transaction Queues
1842 @section Transaction Queues
1843 @cindex transaction queue
1845 @c That's not very informative. What is a transaction, and when might
1846 @c I want to use one?
1847 You can use a @dfn{transaction queue} to communicate with a subprocess
1848 using transactions. First use @code{tq-create} to create a transaction
1849 queue communicating with a specified process. Then you can call
1850 @code{tq-enqueue} to send a transaction.
1852 @defun tq-create process
1853 This function creates and returns a transaction queue communicating with
1854 @var{process}. The argument @var{process} should be a subprocess
1855 capable of sending and receiving streams of bytes. It may be a child
1856 process, or it may be a TCP connection to a server, possibly on another
1860 @defun tq-enqueue queue question regexp closure fn &optional delay-question
1861 This function sends a transaction to queue @var{queue}. Specifying the
1862 queue has the effect of specifying the subprocess to talk to.
1864 The argument @var{question} is the outgoing message that starts the
1865 transaction. The argument @var{fn} is the function to call when the
1866 corresponding answer comes back; it is called with two arguments:
1867 @var{closure}, and the answer received.
1869 The argument @var{regexp} is a regular expression that should match
1870 text at the end of the entire answer, but nothing before; that's how
1871 @code{tq-enqueue} determines where the answer ends.
1873 If the argument @var{delay-question} is non-@code{nil}, delay sending
1874 this question until the process has finished replying to any previous
1875 questions. This produces more reliable results with some processes.
1878 @c Let's not mention it then.
1879 The return value of @code{tq-enqueue} itself is not meaningful.
1883 @defun tq-close queue
1884 Shut down transaction queue @var{queue}, waiting for all pending transactions
1885 to complete, and then terminate the connection or child process.
1888 Transaction queues are implemented by means of a filter function.
1889 @xref{Filter Functions}.
1892 @section Network Connections
1893 @cindex network connection
1897 Emacs Lisp programs can open stream (TCP) and datagram (UDP) network
1898 connections (@pxref{Datagrams}) to other processes on the same machine
1900 A network connection is handled by Lisp much like a subprocess, and is
1901 represented by a process object. However, the process you are
1902 communicating with is not a child of the Emacs process, has no
1903 process @acronym{ID}, and you can't kill it or send it signals. All you
1904 can do is send and receive data. @code{delete-process} closes the
1905 connection, but does not kill the program at the other end; that
1906 program must decide what to do about closure of the connection.
1908 Lisp programs can listen for connections by creating network
1909 servers. A network server is also represented by a kind of process
1910 object, but unlike a network connection, the network server never
1911 transfers data itself. When it receives a connection request, it
1912 creates a new network connection to represent the connection just
1913 made. (The network connection inherits certain information, including
1914 the process plist, from the server.) The network server then goes
1915 back to listening for more connection requests.
1917 Network connections and servers are created by calling
1918 @code{make-network-process} with an argument list consisting of
1919 keyword/argument pairs, for example @code{:server t} to create a
1920 server process, or @code{:type 'datagram} to create a datagram
1921 connection. @xref{Low-Level Network}, for details. You can also use
1922 the @code{open-network-stream} function described below.
1924 To distinguish the different types of processes, the
1925 @code{process-type} function returns the symbol @code{network} for a
1926 network connection or server, @code{serial} for a serial port
1927 connection, or @code{real} for a real subprocess.
1929 The @code{process-status} function returns @code{open},
1930 @code{closed}, @code{connect}, or @code{failed} for network
1931 connections. For a network server, the status is always
1932 @code{listen}. None of those values is possible for a real
1933 subprocess. @xref{Process Information}.
1935 You can stop and resume operation of a network process by calling
1936 @code{stop-process} and @code{continue-process}. For a server
1937 process, being stopped means not accepting new connections. (Up to 5
1938 connection requests will be queued for when you resume the server; you
1939 can increase this limit, unless it is imposed by the operating
1940 system---see the @code{:server} keyword of @code{make-network-process},
1941 @ref{Network Processes}.) For a network stream connection, being
1942 stopped means not processing input (any arriving input waits until you
1943 resume the connection). For a datagram connection, some number of
1944 packets may be queued but input may be lost. You can use the function
1945 @code{process-command} to determine whether a network connection or
1946 server is stopped; a non-@code{nil} value means yes.
1948 @cindex network connection, encrypted
1949 @cindex encrypted network connections
1950 @cindex @acronym{TLS} network connections
1951 @cindex @acronym{STARTTLS} network connections
1952 Emacs can create encrypted network connections, using either built-in
1953 or external support. The built-in support uses the GnuTLS
1954 (``Transport Layer Security'') library; see
1955 @uref{http://www.gnu.org/software/gnutls/, the GnuTLS project page}.
1956 If your Emacs was compiled with GnuTLS support, the function
1957 @code{gnutls-available-p} is defined and returns non-@code{nil}. For
1958 more details, @pxref{Top,, Overview, emacs-gnutls, The Emacs-GnuTLS manual}.
1959 The external support uses the @file{starttls.el} library, which
1960 requires a helper utility such as @command{gnutls-cli} to be installed
1961 on the system. The @code{open-network-stream} function can
1962 transparently handle the details of creating encrypted connections for
1963 you, using whatever support is available.
1965 @defun open-network-stream name buffer host service &rest parameters
1966 This function opens a TCP connection, with optional encryption, and
1967 returns a process object that represents the connection.
1969 The @var{name} argument specifies the name for the process object. It
1970 is modified as necessary to make it unique.
1972 The @var{buffer} argument is the buffer to associate with the
1973 connection. Output from the connection is inserted in the buffer,
1974 unless you specify your own filter function to handle the output. If
1975 @var{buffer} is @code{nil}, it means that the connection is not
1976 associated with any buffer.
1978 The arguments @var{host} and @var{service} specify where to connect to;
1979 @var{host} is the host name (a string), and @var{service} is the name of
1980 a defined network service (a string) or a port number (an integer).
1982 The remaining arguments @var{parameters} are keyword/argument pairs
1983 that are mainly relevant to encrypted connections:
1987 @item :nowait @var{boolean}
1988 If non-@code{nil}, try to make an asynchronous connection.
1990 @item :type @var{type}
1991 The type of connection. Options are:
1995 An ordinary, unencrypted connection.
1998 A @acronym{TLS} (``Transport Layer Security'') connection.
2001 Start with a plain connection, and if parameters @samp{:success}
2002 and @samp{:capability-command} are supplied, try to upgrade to an encrypted
2003 connection via @acronym{STARTTLS}. If that fails, retain the
2004 unencrypted connection.
2006 As for @code{nil}, but if @acronym{STARTTLS} fails drop the connection.
2011 @item :always-query-capabilities @var{boolean}
2012 If non-@code{nil}, always ask for the server's capabilities, even when
2013 doing a @samp{plain} connection.
2015 @item :capability-command @var{capability-command}
2016 Command string to query the host capabilities.
2018 @item :end-of-command @var{regexp}
2019 @itemx :end-of-capability @var{regexp}
2020 Regular expression matching the end of a command, or the end of the
2021 command @var{capability-command}. The latter defaults to the former.
2023 @item :starttls-function @var{function}
2024 Function of one argument (the response to @var{capability-command}),
2025 which returns either @code{nil}, or the command to activate @acronym{STARTTLS}
2028 @item :success @var{regexp}
2029 Regular expression matching a successful @acronym{STARTTLS} negotiation.
2031 @item :use-starttls-if-possible @var{boolean}
2032 If non-@code{nil}, do opportunistic @acronym{STARTTLS} upgrades even if Emacs
2033 doesn't have built-in @acronym{TLS} support.
2035 @item :client-certificate @var{list-or-t}
2036 Either a list of the form @code{(@var{key-file} @var{cert-file})},
2037 naming the certificate key file and certificate file itself, or
2038 @code{t}, meaning to query @code{auth-source} for this information
2039 (@pxref{Top,,Overview, auth, The Auth-Source Manual}).
2040 Only used for @acronym{TLS} or @acronym{STARTTLS}.
2042 @item :return-list @var{cons-or-nil}
2043 The return value of this function. If omitted or @code{nil}, return a
2044 process object. Otherwise, a cons of the form @code{(@var{process-object}
2045 . @var{plist})}, where @var{plist} has keywords:
2048 @item :greeting @var{string-or-nil}
2049 If non-@code{nil}, the greeting string returned by the host.
2050 @item :capabilities @var{string-or-nil}
2051 If non-@code{nil}, the host's capability string.
2052 @item :type @var{symbol}
2053 The connection type: @samp{plain} or @samp{tls}.
2060 @node Network Servers
2061 @section Network Servers
2062 @cindex network servers
2064 You create a server by calling @code{make-network-process}
2065 (@pxref{Network Processes}) with @code{:server t}. The server will
2066 listen for connection requests from clients. When it accepts a client
2067 connection request, that creates a new network connection, itself a
2068 process object, with the following parameters:
2072 The connection's process name is constructed by concatenating the
2073 server process's @var{name} with a client identification string. The
2074 @c FIXME? What about IPv6? Say briefly what the difference is?
2075 client identification string for an IPv4 connection looks like
2076 @samp{<@var{a}.@var{b}.@var{c}.@var{d}:@var{p}>}, which represents an
2077 address and port number. Otherwise, it is a
2078 unique number in brackets, as in @samp{<@var{nnn}>}. The number
2079 is unique for each connection in the Emacs session.
2082 If the server has a non-default filter, the connection process does
2083 not get a separate process buffer; otherwise, Emacs creates a new
2084 buffer for the purpose. The buffer name is the server's buffer name
2085 or process name, concatenated with the client identification string.
2087 The server's process buffer value is never used directly, but the log
2088 function can retrieve it and use it to log connections by inserting
2092 The communication type and the process filter and sentinel are
2093 inherited from those of the server. The server never directly
2094 uses its filter and sentinel; their sole purpose is to initialize
2095 connections made to the server.
2098 The connection's process contact information is set according to the client's
2099 addressing information (typically an IP address and a port number).
2100 This information is associated with the @code{process-contact}
2101 keywords @code{:host}, @code{:service}, @code{:remote}.
2104 The connection's local address is set up according to the port
2105 number used for the connection.
2108 The client process's plist is initialized from the server's plist.
2115 A @dfn{datagram} connection communicates with individual packets rather
2116 than streams of data. Each call to @code{process-send} sends one
2117 datagram packet (@pxref{Input to Processes}), and each datagram
2118 received results in one call to the filter function.
2120 The datagram connection doesn't have to talk with the same remote
2121 peer all the time. It has a @dfn{remote peer address} which specifies
2122 where to send datagrams to. Each time an incoming datagram is passed
2123 to the filter function, the peer address is set to the address that
2124 datagram came from; that way, if the filter function sends a datagram,
2125 it will go back to that place. You can specify the remote peer
2126 address when you create the datagram connection using the
2127 @code{:remote} keyword. You can change it later on by calling
2128 @code{set-process-datagram-address}.
2130 @defun process-datagram-address process
2131 If @var{process} is a datagram connection or server, this function
2132 returns its remote peer address.
2135 @defun set-process-datagram-address process address
2136 If @var{process} is a datagram connection or server, this function
2137 sets its remote peer address to @var{address}.
2140 @node Low-Level Network
2141 @section Low-Level Network Access
2143 You can also create network connections by operating at a lower
2144 level than that of @code{open-network-stream}, using
2145 @code{make-network-process}.
2148 * Proc: Network Processes. Using @code{make-network-process}.
2149 * Options: Network Options. Further control over network connections.
2150 * Features: Network Feature Testing.
2151 Determining which network features work on
2152 the machine you are using.
2155 @node Network Processes
2156 @subsection @code{make-network-process}
2158 The basic function for creating network connections and network
2159 servers is @code{make-network-process}. It can do either of those
2160 jobs, depending on the arguments you give it.
2162 @defun make-network-process &rest args
2163 This function creates a network connection or server and returns the
2164 process object that represents it. The arguments @var{args} are a
2165 list of keyword/argument pairs. Omitting a keyword is always
2166 equivalent to specifying it with value @code{nil}, except for
2167 @code{:coding}, @code{:filter-multibyte}, and @code{:reuseaddr}. Here
2168 are the meaningful keywords (those corresponding to network options
2169 are listed in the following section):
2172 @item :name @var{name}
2173 Use the string @var{name} as the process name. It is modified if
2174 necessary to make it unique.
2176 @item :type @var{type}
2177 Specify the communication type. A value of @code{nil} specifies a
2178 stream connection (the default); @code{datagram} specifies a datagram
2179 connection; @code{seqpacket} specifies a ``sequenced packet stream''
2180 connection. Both connections and servers can be of these types.
2182 @item :server @var{server-flag}
2183 If @var{server-flag} is non-@code{nil}, create a server. Otherwise,
2184 create a connection. For a stream type server, @var{server-flag} may
2185 be an integer, which then specifies the length of the queue of pending
2186 connections to the server. The default queue length is 5.
2188 @item :host @var{host}
2189 Specify the host to connect to. @var{host} should be a host name or
2190 Internet address, as a string, or the symbol @code{local} to specify
2191 the local host. If you specify @var{host} for a server, it must
2192 specify a valid address for the local host, and only clients
2193 connecting to that address will be accepted.
2195 @item :service @var{service}
2196 @var{service} specifies a port number to connect to; or, for a server,
2197 the port number to listen on. It should be a service name that
2198 translates to a port number, or an integer specifying the port number
2199 directly. For a server, it can also be @code{t}, which means to let
2200 the system select an unused port number.
2202 @item :family @var{family}
2203 @var{family} specifies the address (and protocol) family for
2204 communication. @code{nil} means determine the proper address family
2205 automatically for the given @var{host} and @var{service}.
2206 @code{local} specifies a Unix socket, in which case @var{host} is
2207 ignored. @code{ipv4} and @code{ipv6} specify to use IPv4 and IPv6,
2210 @item :local @var{local-address}
2211 For a server process, @var{local-address} is the address to listen on.
2212 It overrides @var{family}, @var{host} and @var{service}, so you
2213 might as well not specify them.
2215 @item :remote @var{remote-address}
2216 For a connection, @var{remote-address} is the address to connect to.
2217 It overrides @var{family}, @var{host} and @var{service}, so you
2218 might as well not specify them.
2220 For a datagram server, @var{remote-address} specifies the initial
2221 setting of the remote datagram address.
2223 The format of @var{local-address} or @var{remote-address} depends on
2228 An IPv4 address is represented as a five-element vector of four 8-bit
2229 integers and one 16-bit integer
2230 @code{[@var{a} @var{b} @var{c} @var{d} @var{p}]} corresponding to
2231 numeric IPv4 address @var{a}.@var{b}.@var{c}.@var{d} and port number
2235 An IPv6 address is represented as a nine-element vector of 16-bit
2236 integers @code{[@var{a} @var{b} @var{c} @var{d} @var{e} @var{f}
2237 @var{g} @var{h} @var{p}]} corresponding to numeric IPv6 address
2238 @var{a}:@var{b}:@var{c}:@var{d}:@var{e}:@var{f}:@var{g}:@var{h} and
2239 port number @var{p}.
2242 A local address is represented as a string, which specifies the address
2243 in the local address space.
2246 An ``unsupported family'' address is represented by a cons
2247 @code{(@var{f} . @var{av})}, where @var{f} is the family number and
2248 @var{av} is a vector specifying the socket address using one element
2249 per address data byte. Do not rely on this format in portable code,
2250 as it may depend on implementation defined constants, data sizes, and
2251 data structure alignment.
2254 @item :nowait @var{bool}
2255 If @var{bool} is non-@code{nil} for a stream connection, return
2256 without waiting for the connection to complete. When the connection
2257 succeeds or fails, Emacs will call the sentinel function, with a
2258 second argument matching @code{"open"} (if successful) or
2259 @code{"failed"}. The default is to block, so that
2260 @code{make-network-process} does not return until the connection
2261 has succeeded or failed.
2263 @item :stop @var{stopped}
2264 If @var{stopped} is non-@code{nil}, start the network connection or
2265 server in the ``stopped'' state.
2267 @item :buffer @var{buffer}
2268 Use @var{buffer} as the process buffer.
2270 @item :coding @var{coding}
2271 Use @var{coding} as the coding system for this process. To specify
2272 different coding systems for decoding data from the connection and for
2273 encoding data sent to it, specify @code{(@var{decoding} .
2274 @var{encoding})} for @var{coding}.
2276 If you don't specify this keyword at all, the default
2277 is to determine the coding systems from the data.
2279 @item :noquery @var{query-flag}
2280 Initialize the process query flag to @var{query-flag}.
2281 @xref{Query Before Exit}.
2283 @item :filter @var{filter}
2284 Initialize the process filter to @var{filter}.
2286 @item :filter-multibyte @var{multibyte}
2287 If @var{multibyte} is non-@code{nil}, strings given to the process
2288 filter are multibyte, otherwise they are unibyte. The default is the
2289 default value of @code{enable-multibyte-characters}.
2291 @item :sentinel @var{sentinel}
2292 Initialize the process sentinel to @var{sentinel}.
2294 @item :log @var{log}
2295 Initialize the log function of a server process to @var{log}. The log
2296 function is called each time the server accepts a network connection
2297 from a client. The arguments passed to the log function are
2298 @var{server}, @var{connection}, and @var{message}; where @var{server}
2299 is the server process, @var{connection} is the new process for the
2300 connection, and @var{message} is a string describing what has
2303 @item :plist @var{plist}
2304 Initialize the process plist to @var{plist}.
2307 The original argument list, modified with the actual connection
2308 information, is available via the @code{process-contact} function.
2311 @node Network Options
2312 @subsection Network Options
2314 The following network options can be specified when you create a
2315 network process. Except for @code{:reuseaddr}, you can also set or
2316 modify these options later, using @code{set-network-process-option}.
2318 For a server process, the options specified with
2319 @code{make-network-process} are not inherited by the client
2320 connections, so you will need to set the necessary options for each
2321 child connection as it is created.
2324 @item :bindtodevice @var{device-name}
2325 If @var{device-name} is a non-empty string identifying a network
2326 interface name (see @code{network-interface-list}), only handle
2327 packets received on that interface. If @var{device-name} is @code{nil}
2328 (the default), handle packets received on any interface.
2330 Using this option may require special privileges on some systems.
2332 @item :broadcast @var{broadcast-flag}
2333 If @var{broadcast-flag} is non-@code{nil} for a datagram process, the
2334 process will receive datagram packet sent to a broadcast address, and
2335 be able to send packets to a broadcast address. This is ignored for a stream
2338 @item :dontroute @var{dontroute-flag}
2339 If @var{dontroute-flag} is non-@code{nil}, the process can only send
2340 to hosts on the same network as the local host.
2342 @item :keepalive @var{keepalive-flag}
2343 If @var{keepalive-flag} is non-@code{nil} for a stream connection,
2344 enable exchange of low-level keep-alive messages.
2346 @item :linger @var{linger-arg}
2347 If @var{linger-arg} is non-@code{nil}, wait for successful
2348 transmission of all queued packets on the connection before it is
2349 deleted (see @code{delete-process}). If @var{linger-arg} is an
2350 integer, it specifies the maximum time in seconds to wait for queued
2351 packets to be sent before closing the connection. The default is
2352 @code{nil}, which means to discard unsent queued packets when the
2355 @c FIXME Where out-of-band data is ...?
2356 @item :oobinline @var{oobinline-flag}
2357 If @var{oobinline-flag} is non-@code{nil} for a stream connection,
2358 receive out-of-band data in the normal data stream. Otherwise, ignore
2361 @item :priority @var{priority}
2362 Set the priority for packets sent on this connection to the integer
2363 @var{priority}. The interpretation of this number is protocol
2364 specific; such as setting the TOS (type of service) field on IP
2365 packets sent on this connection. It may also have system dependent
2366 effects, such as selecting a specific output queue on the network
2369 @item :reuseaddr @var{reuseaddr-flag}
2370 If @var{reuseaddr-flag} is non-@code{nil} (the default) for a stream
2371 server process, allow this server to reuse a specific port number (see
2372 @code{:service}), unless another process on this host is already
2373 listening on that port. If @var{reuseaddr-flag} is @code{nil}, there
2374 may be a period of time after the last use of that port (by any
2375 process on the host) where it is not possible to make a new server on
2379 @defun set-network-process-option process option value &optional no-error
2380 This function sets or modifies a network option for network process
2381 @var{process}. The accepted options and values are as for
2382 @code{make-network-process}. If @var{no-error} is non-@code{nil},
2383 this function returns @code{nil} instead of signaling an error if
2384 @var{option} is not a supported option. If the function successfully
2385 completes, it returns @code{t}.
2387 The current setting of an option is available via the
2388 @code{process-contact} function.
2391 @node Network Feature Testing
2392 @subsection Testing Availability of Network Features
2394 To test for the availability of a given network feature, use
2395 @code{featurep} like this:
2398 (featurep 'make-network-process '(@var{keyword} @var{value}))
2402 The result of this form is @code{t} if it works to specify
2403 @var{keyword} with value @var{value} in @code{make-network-process}.
2404 Here are some of the @var{keyword}---@var{value} pairs you can test in
2409 Non-@code{nil} if non-blocking connect is supported.
2410 @item (:type datagram)
2411 Non-@code{nil} if datagrams are supported.
2412 @item (:family local)
2413 Non-@code{nil} if local (a.k.a.@: ``UNIX domain'') sockets are supported.
2414 @item (:family ipv6)
2415 Non-@code{nil} if IPv6 is supported.
2417 Non-@code{nil} if the system can select the port for a server.
2420 To test for the availability of a given network option, use
2421 @code{featurep} like this:
2424 (featurep 'make-network-process '@var{keyword})
2428 The accepted @var{keyword} values are @code{:bindtodevice}, etc.
2429 For the complete list, @pxref{Network Options}. This form returns
2430 non-@code{nil} if that particular network option is supported by
2431 @code{make-network-process} (or @code{set-network-process-option}).
2434 @section Misc Network Facilities
2436 These additional functions are useful for creating and operating
2437 on network connections. Note that they are supported only on some
2440 @defun network-interface-list
2441 This function returns a list describing the network interfaces
2442 of the machine you are using. The value is an alist whose
2443 elements have the form @code{(@var{name} . @var{address})}.
2444 @var{address} has the same form as the @var{local-address}
2445 and @var{remote-address} arguments to @code{make-network-process}.
2448 @defun network-interface-info ifname
2449 This function returns information about the network interface named
2450 @var{ifname}. The value is a list of the form
2451 @code{(@var{addr} @var{bcast} @var{netmask} @var{hwaddr} @var{flags})}.
2455 The Internet protocol address.
2457 The broadcast address.
2461 The layer 2 address (Ethernet MAC address, for instance).
2463 The current flags of the interface.
2467 @defun format-network-address address &optional omit-port
2468 This function converts the Lisp representation of a network address to
2471 A five-element vector @code{[@var{a} @var{b} @var{c} @var{d} @var{p}]}
2472 represents an IPv4 address @var{a}.@var{b}.@var{c}.@var{d} and port
2473 number @var{p}. @code{format-network-address} converts that to the
2474 string @code{"@var{a}.@var{b}.@var{c}.@var{d}:@var{p}"}.
2476 A nine-element vector @code{[@var{a} @var{b} @var{c} @var{d} @var{e}
2477 @var{f} @var{g} @var{h} @var{p}]} represents an IPv6 address along
2478 with a port number. @code{format-network-address} converts that to
2480 @code{"[@var{a}:@var{b}:@var{c}:@var{d}:@var{e}:@var{f}:@var{g}:@var{h}]:@var{p}"}.
2482 If the vector does not include the port number, @var{p}, or if
2483 @var{omit-port} is non-@code{nil}, the result does not include the
2484 @code{:@var{p}} suffix.
2488 @section Communicating with Serial Ports
2489 @cindex @file{/dev/tty}
2491 @cindex serial connections
2493 Emacs can communicate with serial ports. For interactive use,
2494 @kbd{M-x serial-term} opens a terminal window. In a Lisp program,
2495 @code{make-serial-process} creates a process object.
2497 The serial port can be configured at run-time, without having to
2498 close and re-open it. The function @code{serial-process-configure}
2499 lets you change the speed, bytesize, and other parameters. In a
2500 terminal window created by @code{serial-term}, you can click on the
2501 mode line for configuration.
2503 A serial connection is represented by a process object, which can be
2504 used in a similar way to a subprocess or network process. You can send and
2505 receive data, and configure the serial port. A serial process object
2506 has no process ID, however, and you can't send signals to it, and the
2507 status codes are different from other types of processes.
2508 @code{delete-process} on the process object or @code{kill-buffer} on
2509 the process buffer close the connection, but this does not affect the
2510 device connected to the serial port.
2512 The function @code{process-type} returns the symbol @code{serial}
2513 for a process object representing a serial port connection.
2515 Serial ports are available on GNU/Linux, Unix, and MS Windows systems.
2517 @deffn Command serial-term port speed
2518 Start a terminal-emulator for a serial port in a new buffer.
2519 @var{port} is the name of the serial port to connect to. For
2520 example, this could be @file{/dev/ttyS0} on Unix. On MS Windows, this
2521 could be @file{COM1}, or @file{\\.\COM10} (double the backslashes in
2524 @c FIXME is 9600 still the most common value, or is it 115200 now?
2525 @c (Same value, 9600, appears below as well.)
2526 @var{speed} is the speed of the serial port in bits per second. 9600
2527 is a common value. The buffer is in Term mode; see @ref{Term Mode,,,
2528 emacs, The GNU Emacs Manual}, for the commands to use in that buffer.
2529 You can change the speed and the configuration in the mode line menu.
2532 @defun make-serial-process &rest args
2533 This function creates a process and a buffer. Arguments are specified
2534 as keyword/argument pairs. Here's the list of the meaningful
2535 keywords, with the first two (@var{port} and @var{speed}) being mandatory:
2538 @item :port @var{port}
2539 This is the name of the serial port. On Unix and GNU systems, this is
2540 a file name such as @file{/dev/ttyS0}. On Windows, this could be
2541 @file{COM1}, or @file{\\.\COM10} for ports higher than @file{COM9}
2542 (double the backslashes in Lisp strings).
2544 @item :speed @var{speed}
2545 The speed of the serial port in bits per second. This function calls
2546 @code{serial-process-configure} to handle the speed; see the
2547 following documentation of that function for more details.
2549 @item :name @var{name}
2550 The name of the process. If @var{name} is not given, @var{port} will
2551 serve as the process name as well.
2553 @item :buffer @var{buffer}
2554 The buffer to associate with the process. The value can be either a
2555 buffer or a string that names a buffer. Process output goes at the
2556 end of that buffer, unless you specify an output stream or filter
2557 function to handle the output. If @var{buffer} is not given, the
2558 process buffer's name is taken from the value of the @code{:name}
2561 @item :coding @var{coding}
2562 If @var{coding} is a symbol, it specifies the coding system used for
2563 both reading and writing for this process. If @var{coding} is a cons
2564 @code{(@var{decoding} . @var{encoding})}, @var{decoding} is used for
2565 reading, and @var{encoding} is used for writing. If not specified,
2566 the default is to determine the coding systems from the data itself.
2568 @item :noquery @var{query-flag}
2569 Initialize the process query flag to @var{query-flag}. @xref{Query
2570 Before Exit}. The flags defaults to @code{nil} if unspecified.
2572 @item :stop @var{bool}
2573 Start process in the ``stopped'' state if @var{bool} is
2574 non-@code{nil}. In the stopped state, a serial process does not
2575 accept incoming data, but you can send outgoing data. The stopped
2576 state is cleared by @code{continue-process} and set by
2577 @code{stop-process}.
2579 @item :filter @var{filter}
2580 Install @var{filter} as the process filter.
2582 @item :sentinel @var{sentinel}
2583 Install @var{sentinel} as the process sentinel.
2585 @item :plist @var{plist}
2586 Install @var{plist} as the initial plist of the process.
2592 These are handled by @code{serial-process-configure}, which is called
2593 by @code{make-serial-process}.
2596 The original argument list, possibly modified by later configuration,
2597 is available via the function @code{process-contact}.
2602 (make-serial-process :port "/dev/ttyS0" :speed 9600)
2606 @defun serial-process-configure &rest args
2607 @cindex baud, in serial connections
2608 @cindex bytesize, in serial connections
2609 @cindex parity, in serial connections
2610 @cindex stopbits, in serial connections
2611 @cindex flowcontrol, in serial connections
2613 This function configures a serial port connection. Arguments are
2614 specified as keyword/argument pairs. Attributes that are not given
2615 are re-initialized from the process's current configuration (available
2616 via the function @code{process-contact}), or set to reasonable default
2617 values. The following arguments are defined:
2620 @item :process @var{process}
2621 @itemx :name @var{name}
2622 @itemx :buffer @var{buffer}
2623 @itemx :port @var{port}
2624 Any of these arguments can be given to identify the process that is to
2625 be configured. If none of these arguments is given, the current
2626 buffer's process is used.
2628 @item :speed @var{speed}
2629 The speed of the serial port in bits per second, a.k.a.@: @dfn{baud
2630 rate}. The value can be any number, but most serial ports work only
2631 at a few defined values between 1200 and 115200, with 9600 being the
2632 most common value. If @var{speed} is @code{nil}, the function ignores
2633 all other arguments and does not configure the port. This may be
2634 useful for special serial ports such as Bluetooth-to-serial converters,
2635 which can only be configured through @samp{AT} commands sent through the
2636 connection. The value of @code{nil} for @var{speed} is valid only for
2637 connections that were already opened by a previous call to
2638 @code{make-serial-process} or @code{serial-term}.
2640 @item :bytesize @var{bytesize}
2641 The number of bits per byte, which can be 7 or 8. If @var{bytesize}
2642 is not given or @code{nil}, it defaults to 8.
2644 @item :parity @var{parity}
2645 The value can be @code{nil} (don't use parity), the symbol
2646 @code{odd} (use odd parity), or the symbol @code{even} (use even
2647 parity). If @var{parity} is not given, it defaults to no parity.
2649 @item :stopbits @var{stopbits}
2650 The number of stopbits used to terminate a transmission
2651 of each byte. @var{stopbits} can be 1 or 2. If @var{stopbits} is not
2652 given or @code{nil}, it defaults to 1.
2654 @item :flowcontrol @var{flowcontrol}
2655 The type of flow control to use for this connection, which is either
2656 @code{nil} (don't use flow control), the symbol @code{hw} (use RTS/CTS
2657 hardware flow control), or the symbol @code{sw} (use XON/XOFF software
2658 flow control). If @var{flowcontrol} is not given, it defaults to no
2662 Internally, @code{make-serial-process} calls
2663 @code{serial-process-configure} for the initial configuration of the
2668 @section Packing and Unpacking Byte Arrays
2669 @cindex byte packing and unpacking
2671 This section describes how to pack and unpack arrays of bytes,
2672 usually for binary network protocols. These functions convert byte arrays
2673 to alists, and vice versa. The byte array can be represented as a
2674 @c FIXME? No multibyte?
2675 unibyte string or as a vector of integers, while the alist associates
2676 symbols either with fixed-size objects or with recursive sub-alists.
2677 To use the functions referred to in this section, load the
2678 @code{bindat} library.
2679 @c It doesn't have any autoloads.
2682 @cindex deserializing
2685 Conversion from byte arrays to nested alists is also known as
2686 @dfn{deserializing} or @dfn{unpacking}, while going in the opposite
2687 direction is also known as @dfn{serializing} or @dfn{packing}.
2690 * Bindat Spec:: Describing data layout.
2691 * Bindat Functions:: Doing the unpacking and packing.
2692 * Bindat Examples:: Samples of what bindat.el can do for you!
2696 @subsection Describing Data Layout
2698 To control unpacking and packing, you write a @dfn{data layout
2699 specification}, a special nested list describing named and typed
2700 @dfn{fields}. This specification controls the length of each field to be
2701 processed, and how to pack or unpack it. We normally keep bindat specs
2702 in variables whose names end in @samp{-bindat-spec}; that kind of name
2703 is automatically recognized as ``risky''.
2707 @cindex little endian
2708 @cindex network byte ordering
2709 A field's @dfn{type} describes the size (in bytes) of the object
2710 that the field represents and, in the case of multibyte fields, how
2711 the bytes are ordered within the field. The two possible orderings
2712 are ``big endian'' (also known as ``network byte ordering'') and
2713 ``little endian''. For instance, the number @code{#x23cd} (decimal
2714 9165) in big endian would be the two bytes @code{#x23} @code{#xcd};
2715 and in little endian, @code{#xcd} @code{#x23}. Here are the possible
2721 Unsigned byte, with length 1.
2726 Unsigned integer in network byte order, with length 2.
2729 Unsigned integer in network byte order, with length 3.
2734 Unsigned integer in network byte order, with length 4.
2735 Note: These values may be limited by Emacs's integer implementation limits.
2740 Unsigned integer in little endian order, with length 2, 3 and 4, respectively.
2743 String of length @var{len}.
2745 @item strz @var{len}
2746 Zero-terminated string, in a fixed-size field with length @var{len}.
2748 @item vec @var{len} [@var{type}]
2749 Vector of @var{len} elements of type @var{type}, defaulting to bytes.
2750 The @var{type} is any of the simple types above, or another vector
2751 specified as a list of the form @code{(vec @var{len} [@var{type}])}.
2755 Four-byte vector representing an Internet address. For example:
2756 @code{[127 0 0 1]} for localhost.
2758 @item bits @var{len}
2759 List of set bits in @var{len} bytes. The bytes are taken in big
2760 endian order and the bits are numbered starting with @code{8 *
2761 @var{len} @minus{} 1} and ending with zero. For example: @code{bits
2762 2} unpacks @code{#x28} @code{#x1c} to @code{(2 3 4 11 13)} and
2763 @code{#x1c} @code{#x28} to @code{(3 5 10 11 12)}.
2765 @item (eval @var{form})
2766 @var{form} is a Lisp expression evaluated at the moment the field is
2767 unpacked or packed. The result of the evaluation should be one of the
2768 above-listed type specifications.
2771 For a fixed-size field, the length @var{len} is given as an integer
2772 specifying the number of bytes in the field.
2774 When the length of a field is not fixed, it typically depends on the
2775 value of a preceding field. In this case, the length @var{len} can be
2776 given either as a list @code{(@var{name} ...)} identifying a
2777 @dfn{field name} in the format specified for @code{bindat-get-field}
2778 below, or by an expression @code{(eval @var{form})} where @var{form}
2779 should evaluate to an integer, specifying the field length.
2781 A field specification generally has the form @code{([@var{name}]
2782 @var{handler})}, where @var{name} is optional. Don't use names that
2783 are symbols meaningful as type specifications (above) or handler
2784 specifications (below), since that would be ambiguous. @var{name} can
2785 be a symbol or an expression @code{(eval @var{form})}, in which case
2786 @var{form} should evaluate to a symbol.
2788 @var{handler} describes how to unpack or pack the field and can be one
2793 Unpack/pack this field according to the type specification @var{type}.
2795 @item eval @var{form}
2796 Evaluate @var{form}, a Lisp expression, for side-effect only. If the
2797 field name is specified, the value is bound to that field name.
2799 @item fill @var{len}
2800 Skip @var{len} bytes. In packing, this leaves them unchanged,
2801 which normally means they remain zero. In unpacking, this means
2804 @item align @var{len}
2805 Skip to the next multiple of @var{len} bytes.
2807 @item struct @var{spec-name}
2808 Process @var{spec-name} as a sub-specification. This describes a
2809 structure nested within another structure.
2811 @item union @var{form} (@var{tag} @var{spec})@dots{}
2812 @c ??? I don't see how one would actually use this.
2813 @c ??? what kind of expression would be useful for @var{form}?
2814 Evaluate @var{form}, a Lisp expression, find the first @var{tag}
2815 that matches it, and process its associated data layout specification
2816 @var{spec}. Matching can occur in one of three ways:
2820 If a @var{tag} has the form @code{(eval @var{expr})}, evaluate
2821 @var{expr} with the variable @code{tag} dynamically bound to the value
2822 of @var{form}. A non-@code{nil} result indicates a match.
2825 @var{tag} matches if it is @code{equal} to the value of @var{form}.
2828 @var{tag} matches unconditionally if it is @code{t}.
2831 @item repeat @var{count} @var{field-specs}@dots{}
2832 Process the @var{field-specs} recursively, in order, then repeat
2833 starting from the first one, processing all the specifications @var{count}
2834 times overall. The @var{count} is given using the same formats as a
2835 field length---if an @code{eval} form is used, it is evaluated just once.
2836 For correct operation, each specification in @var{field-specs} must
2840 For the @code{(eval @var{form})} forms used in a bindat specification,
2841 the @var{form} can access and update these dynamically bound variables
2846 Value of the last field processed.
2849 The data as a byte array.
2852 Current index (within @code{bindat-raw}) for unpacking or packing.
2855 The alist containing the structured data that have been unpacked so
2856 far, or the entire structure being packed. You can use
2857 @code{bindat-get-field} to access specific fields of this structure.
2861 Inside a @code{repeat} block, these contain the maximum number of
2862 repetitions (as specified by the @var{count} parameter), and the
2863 current repetition number (counting from 0). Setting @code{count} to
2864 zero will terminate the inner-most repeat block after the current
2865 repetition has completed.
2868 @node Bindat Functions
2869 @subsection Functions to Unpack and Pack Bytes
2871 In the following documentation, @var{spec} refers to a data layout
2872 specification, @code{bindat-raw} to a byte array, and @var{struct} to an
2873 alist representing unpacked field data.
2875 @defun bindat-unpack spec bindat-raw &optional bindat-idx
2876 @c FIXME? Again, no multibyte?
2877 This function unpacks data from the unibyte string or byte
2878 array @code{bindat-raw}
2879 according to @var{spec}. Normally, this starts unpacking at the
2880 beginning of the byte array, but if @var{bindat-idx} is non-@code{nil}, it
2881 specifies a zero-based starting position to use instead.
2883 The value is an alist or nested alist in which each element describes
2887 @defun bindat-get-field struct &rest name
2888 This function selects a field's data from the nested alist
2889 @var{struct}. Usually @var{struct} was returned by
2890 @code{bindat-unpack}. If @var{name} corresponds to just one argument,
2891 that means to extract a top-level field value. Multiple @var{name}
2892 arguments specify repeated lookup of sub-structures. An integer name
2893 acts as an array index.
2895 For example, if @var{name} is @code{(a b 2 c)}, that means to find
2896 field @code{c} in the third element of subfield @code{b} of field
2897 @code{a}. (This corresponds to @code{struct.a.b[2].c} in C.)
2900 Although packing and unpacking operations change the organization of
2901 data (in memory), they preserve the data's @dfn{total length}, which is
2902 the sum of all the fields' lengths, in bytes. This value is not
2903 generally inherent in either the specification or alist alone; instead,
2904 both pieces of information contribute to its calculation. Likewise, the
2905 length of a string or array being unpacked may be longer than the data's
2906 total length as described by the specification.
2908 @defun bindat-length spec struct
2909 This function returns the total length of the data in @var{struct},
2910 according to @var{spec}.
2913 @defun bindat-pack spec struct &optional bindat-raw bindat-idx
2914 This function returns a byte array packed according to @var{spec} from
2915 the data in the alist @var{struct}. It normally creates and fills a
2916 new byte array starting at the beginning. However, if @var{bindat-raw}
2917 is non-@code{nil}, it specifies a pre-allocated unibyte string or vector to
2918 pack into. If @var{bindat-idx} is non-@code{nil}, it specifies the starting
2919 offset for packing into @code{bindat-raw}.
2921 When pre-allocating, you should make sure @code{(length @var{bindat-raw})}
2922 meets or exceeds the total length to avoid an out-of-range error.
2925 @defun bindat-ip-to-string ip
2926 Convert the Internet address vector @var{ip} to a string in the usual
2928 @c FIXME? Does it do IPv6?
2931 (bindat-ip-to-string [127 0 0 1])
2932 @result{} "127.0.0.1"
2936 @node Bindat Examples
2937 @subsection Examples of Byte Unpacking and Packing
2938 @c FIXME? This seems a very long example for something that is not used
2939 @c very often. As of 24.1, gdb-mi.el is the only user of bindat.el in Emacs.
2940 @c Maybe one or both of these examples should just be moved to the
2941 @c commentary of bindat.el.
2943 Here is a complete example of byte unpacking and packing:
2948 (defvar fcookie-index-spec
2956 (:offset repeat (:count) (:foo u32)))
2957 "Description of a fortune cookie index file's contents.")
2959 (defun fcookie (cookies &optional index)
2960 "Display a random fortune cookie from file COOKIES.
2961 Optional second arg INDEX specifies the associated index
2962 filename, by default \"COOKIES.dat\". Display cookie text
2963 in buffer \"*Fortune Cookie: BASENAME*\", where BASENAME
2964 is COOKIES without the directory part."
2965 (interactive "fCookies file: ")
2966 (let* ((info (with-temp-buffer
2967 (insert-file-contents-literally
2968 (or index (concat cookies ".dat")))
2969 (bindat-unpack fcookie-index-spec
2971 (sel (random (bindat-get-field info :count)))
2972 (beg (cdar (bindat-get-field info :offset sel)))
2973 (end (or (cdar (bindat-get-field info
2975 (nth 7 (file-attributes cookies)))))
2978 (format "*Fortune Cookie: %s*"
2979 (file-name-nondirectory cookies))))
2981 (insert-file-contents-literally
2982 cookies nil beg (- end 3))))
2984 (defun fcookie-create-index (cookies &optional index delim)
2985 "Scan file COOKIES, and write out its index file.
2986 Optional arg INDEX specifies the index filename, which by
2987 default is \"COOKIES.dat\". Optional arg DELIM specifies the
2988 unibyte character that, when found on a line of its own in
2989 COOKIES, indicates the border between entries."
2990 (interactive "fCookies file: ")
2991 (setq delim (or delim ?%))
2992 (let ((delim-line (format "\n%c\n" delim))
2995 min p q len offsets)
2996 (unless (= 3 (string-bytes delim-line))
2997 (error "Delimiter cannot be represented in one byte"))
2999 (insert-file-contents-literally cookies)
3000 (while (and (setq p (point))
3001 (search-forward delim-line (point-max) t)
3002 (setq len (- (point) 3 p)))
3003 (setq count (1+ count)
3005 min (min (or min max) len)
3006 offsets (cons (1- p) offsets))))
3008 (set-buffer-multibyte nil)
3018 (:offset . ,(mapcar (lambda (o)
3019 (list (cons :foo o)))
3020 (nreverse offsets))))))
3021 (let ((coding-system-for-write 'raw-text-unix))
3022 (write-file (or index (concat cookies ".dat")))))))
3025 The following is an example of defining and unpacking a complex
3026 structure. Consider the following C structures:
3030 unsigned long dest_ip;
3031 unsigned long src_ip;
3032 unsigned short dest_port;
3033 unsigned short src_port;
3038 unsigned char opcode;
3039 unsigned short length; /* in network byte order */
3040 unsigned char id[8]; /* null-terminated string */
3041 unsigned char data[/* (length + 3) & ~3 */];
3045 struct header header;
3046 unsigned long counters[2]; /* in little endian order */
3047 unsigned char items;
3048 unsigned char filler[3];
3049 struct data item[/* items */];
3054 The corresponding data layout specification is:
3066 (length u16) ; network byte order
3072 '((header struct header-spec)
3073 (counters vec 2 u32r) ; little endian order
3076 (item repeat (items)
3077 (struct data-spec))))
3080 A binary data representation is:
3084 [ 192 168 1 100 192 168 1 101 01 28 21 32
3085 160 134 1 0 5 1 0 0 2 0 0 0
3086 2 3 0 5 ?A ?B ?C ?D ?E ?F 0 0 1 2 3 4 5 0 0 0
3087 1 4 0 7 ?B ?C ?D ?E ?F ?G 0 0 6 7 8 9 10 11 12 0 ])
3090 The corresponding decoded structure is:
3093 (setq decoded (bindat-unpack packet-spec binary-data))
3096 (dest-ip . [192 168 1 100])
3097 (src-ip . [192 168 1 101])
3100 (counters . [100000 261])
3102 (item ((data . [1 2 3 4 5])
3107 ((data . [6 7 8 9 10 11 12])
3114 An example of fetching data from this structure:
3117 (bindat-get-field decoded 'item 1 'id)