2 @c This is part of the GNU Emacs Lisp Reference Manual.
3 @c Copyright (C) 1990-1995, 1998-1999, 2001-2012
4 @c Free Software Foundation, Inc.
5 @c See the file elisp.texi for copying conditions.
6 @setfilename ../../info/processes
7 @node Processes, Display, Abbrevs, Top
10 @cindex parent process
14 In the terminology of operating systems, a @dfn{process} is a space in
15 which a program can execute. Emacs runs in a process. Emacs Lisp
16 programs can invoke other programs in processes of their own. These are
17 called @dfn{subprocesses} or @dfn{child processes} of the Emacs process,
18 which is their @dfn{parent process}.
20 A subprocess of Emacs may be @dfn{synchronous} or @dfn{asynchronous},
21 depending on how it is created. When you create a synchronous
22 subprocess, the Lisp program waits for the subprocess to terminate
23 before continuing execution. When you create an asynchronous
24 subprocess, it can run in parallel with the Lisp program. This kind of
25 subprocess is represented within Emacs by a Lisp object which is also
26 called a ``process''. Lisp programs can use this object to communicate
27 with the subprocess or to control it. For example, you can send
28 signals, obtain status information, receive output from the process, or
31 @defun processp object
32 This function returns @code{t} if @var{object} represents an Emacs
33 subprocess, @code{nil} otherwise.
36 In addition to subprocesses of the current Emacs session, you can
37 also access other processes running on your machine. @xref{System
41 * Subprocess Creation:: Functions that start subprocesses.
42 * Shell Arguments:: Quoting an argument to pass it to a shell.
43 * Synchronous Processes:: Details of using synchronous subprocesses.
44 * Asynchronous Processes:: Starting up an asynchronous subprocess.
45 * Deleting Processes:: Eliminating an asynchronous subprocess.
46 * Process Information:: Accessing run-status and other attributes.
47 * Input to Processes:: Sending input to an asynchronous subprocess.
48 * Signals to Processes:: Stopping, continuing or interrupting
49 an asynchronous subprocess.
50 * Output from Processes:: Collecting output from an asynchronous subprocess.
51 * Sentinels:: Sentinels run when process run-status changes.
52 * Query Before Exit:: Whether to query if exiting will kill a process.
53 * System Processes:: Accessing other processes running on your system.
54 * Transaction Queues:: Transaction-based communication with subprocesses.
55 * Network:: Opening network connections.
56 * Network Servers:: Network servers let Emacs accept net connections.
57 * Datagrams:: UDP network connections.
58 * Low-Level Network:: Lower-level but more general function
59 to create connections and servers.
60 * Misc Network:: Additional relevant functions for net connections.
61 * Serial Ports:: Communicating with serial ports.
62 * Byte Packing:: Using bindat to pack and unpack binary data.
65 @node Subprocess Creation
66 @section Functions that Create Subprocesses
68 There are three primitives that create a new subprocess in which to run
69 a program. One of them, @code{start-process}, creates an asynchronous
70 process and returns a process object (@pxref{Asynchronous Processes}).
71 The other two, @code{call-process} and @code{call-process-region},
72 create a synchronous process and do not return a process object
73 (@pxref{Synchronous Processes}). There are various higher-level
74 functions that make use of these primitives to run particular types of
77 Synchronous and asynchronous processes are explained in the following
78 sections. Since the three functions are all called in a similar
79 fashion, their common arguments are described here.
81 @cindex execute program
82 @cindex @env{PATH} environment variable
83 @cindex @env{HOME} environment variable
84 In all cases, the function's @var{program} argument specifies the
85 program to be run. An error is signaled if the file is not found or
86 cannot be executed. If the file name is relative, the variable
87 @code{exec-path} contains a list of directories to search. Emacs
88 initializes @code{exec-path} when it starts up, based on the value of
89 the environment variable @env{PATH}. The standard file name
90 constructs, @samp{~}, @samp{.}, and @samp{..}, are interpreted as
91 usual in @code{exec-path}, but environment variable substitutions
92 (@samp{$HOME}, etc.) are not recognized; use
93 @code{substitute-in-file-name} to perform them (@pxref{File Name
94 Expansion}). @code{nil} in this list refers to
95 @code{default-directory}.
97 Executing a program can also try adding suffixes to the specified
100 @defvar exec-suffixes
101 This variable is a list of suffixes (strings) to try adding to the
102 specified program file name. The list should include @code{""} if you
103 want the name to be tried exactly as specified. The default value is
107 @strong{Please note:} The argument @var{program} contains only the
108 name of the program; it may not contain any command-line arguments. You
109 must use a separate argument, @var{args}, to provide those, as
112 Each of the subprocess-creating functions has a @var{buffer-or-name}
113 argument that specifies where the standard output from the program will
114 go. It should be a buffer or a buffer name; if it is a buffer name,
115 that will create the buffer if it does not already exist. It can also
116 be @code{nil}, which says to discard the output unless a filter function
117 handles it. (@xref{Filter Functions}, and @ref{Read and Print}.)
118 Normally, you should avoid having multiple processes send output to the
119 same buffer because their output would be intermixed randomly.
120 For synchronous processes, you can send the output to a file instead
123 @cindex program arguments
124 All three of the subprocess-creating functions have a @code{&rest}
125 argument, @var{args}. The @var{args} must all be strings, and they are
126 supplied to @var{program} as separate command line arguments. Wildcard
127 characters and other shell constructs have no special meanings in these
128 strings, since the strings are passed directly to the specified program.
130 @cindex environment variables, subprocesses
131 The subprocess inherits its environment from Emacs, but you can
132 specify overrides for it with @code{process-environment}. @xref{System
133 Environment}. The subprocess gets its current directory from the
134 value of @code{default-directory}.
136 @defvar exec-directory
138 The value of this variable is a string, the name of a directory that
139 contains programs that come with GNU Emacs and are intended for Emacs
140 to invoke. The program @code{movemail} is an example of such a program;
141 Rmail uses it to fetch new mail from an inbox.
145 The value of this variable is a list of directories to search for
146 programs to run in subprocesses. Each element is either the name of a
147 directory (i.e., a string), or @code{nil}, which stands for the default
148 directory (which is the value of @code{default-directory}).
149 @cindex program directories
151 The value of @code{exec-path} is used by @code{call-process} and
152 @code{start-process} when the @var{program} argument is not an absolute
155 Generally, you should not modify @code{exec-path} directly. Instead,
156 ensure that your @env{PATH} environment variable is set appropriately
157 before starting Emacs. Trying to modify @code{exec-path}
158 independently of @env{PATH} can lead to confusing results.
161 @node Shell Arguments
162 @section Shell Arguments
163 @cindex arguments for shell commands
164 @cindex shell command arguments
166 Lisp programs sometimes need to run a shell and give it a command
167 that contains file names that were specified by the user. These
168 programs ought to be able to support any valid file name. But the shell
169 gives special treatment to certain characters, and if these characters
170 occur in the file name, they will confuse the shell. To handle these
171 characters, use the function @code{shell-quote-argument}:
173 @defun shell-quote-argument argument
174 This function returns a string that represents, in shell syntax,
175 an argument whose actual contents are @var{argument}. It should
176 work reliably to concatenate the return value into a shell command
177 and then pass it to a shell for execution.
179 Precisely what this function does depends on your operating system. The
180 function is designed to work with the syntax of your system's standard
181 shell; if you use an unusual shell, you will need to redefine this
185 ;; @r{This example shows the behavior on GNU and Unix systems.}
186 (shell-quote-argument "foo > bar")
187 @result{} "foo\\ \\>\\ bar"
189 ;; @r{This example shows the behavior on MS-DOS and MS-Windows.}
190 (shell-quote-argument "foo > bar")
191 @result{} "\"foo > bar\""
194 Here's an example of using @code{shell-quote-argument} to construct
199 (shell-quote-argument oldfile)
201 (shell-quote-argument newfile))
205 @cindex quoting and unquoting command-line arguments
206 @cindex minibuffer input, and command-line arguments
207 @cindex @code{call-process}, command-line arguments from minibuffer
208 @cindex @code{start-process}, command-line arguments from minibuffer
209 The following two functions are useful for combining a list of
210 individual command-line argument strings into a single string, and
211 taking a string apart into a list of individual command-line
212 arguments. These functions are mainly intended for
213 converting user input in the minibuffer, a Lisp string, into a list of
214 string arguments to be passed to @code{call-process} or
215 @code{start-process}, or for converting such lists of arguments into
216 a single Lisp string to be presented in the minibuffer or echo area.
218 @defun split-string-and-unquote string &optional separators
219 This function splits @var{string} into substrings at matches for the
220 regular expression @var{separators}, like @code{split-string} does
221 (@pxref{Creating Strings}); in addition, it removes quoting from the
222 substrings. It then makes a list of the substrings and returns it.
224 If @var{separators} is omitted or @code{nil}, it defaults to
225 @code{"\\s-+"}, which is a regular expression that matches one or more
226 characters with whitespace syntax (@pxref{Syntax Class Table}).
228 This function supports two types of quoting: enclosing a whole string
229 in double quotes @code{"@dots{}"}, and quoting individual characters
230 with a backslash escape @samp{\}. The latter is also used in Lisp
231 strings, so this function can handle those as well.
234 @defun combine-and-quote-strings list-of-strings &optional separator
235 This function concatenates @var{list-of-strings} into a single string,
236 quoting each string as necessary. It also sticks the @var{separator}
237 string between each pair of strings; if @var{separator} is omitted or
238 @code{nil}, it defaults to @code{" "}. The return value is the
241 The strings in @var{list-of-strings} that need quoting are those that
242 include @var{separator} as their substring. Quoting a string encloses
243 it in double quotes @code{"@dots{}"}. In the simplest case, if you
244 are consing a command from the individual command-line arguments,
245 every argument that includes embedded blanks will be quoted.
248 @node Synchronous Processes
249 @section Creating a Synchronous Process
250 @cindex synchronous subprocess
252 After a @dfn{synchronous process} is created, Emacs waits for the
253 process to terminate before continuing. Starting Dired on GNU or
254 Unix@footnote{On other systems, Emacs uses a Lisp emulation of
255 @code{ls}; see @ref{Contents of Directories}.} is an example of this: it
256 runs @code{ls} in a synchronous process, then modifies the output
257 slightly. Because the process is synchronous, the entire directory
258 listing arrives in the buffer before Emacs tries to do anything with it.
260 While Emacs waits for the synchronous subprocess to terminate, the
261 user can quit by typing @kbd{C-g}. The first @kbd{C-g} tries to kill
262 the subprocess with a @code{SIGINT} signal; but it waits until the
263 subprocess actually terminates before quitting. If during that time the
264 user types another @kbd{C-g}, that kills the subprocess instantly with
265 @code{SIGKILL} and quits immediately (except on MS-DOS, where killing
266 other processes doesn't work). @xref{Quitting}.
268 The synchronous subprocess functions return an indication of how the
271 The output from a synchronous subprocess is generally decoded using a
272 coding system, much like text read from a file. The input sent to a
273 subprocess by @code{call-process-region} is encoded using a coding
274 system, much like text written into a file. @xref{Coding Systems}.
276 @defun call-process program &optional infile destination display &rest args
277 This function calls @var{program} and waits for it to finish.
279 The standard input for the new process comes from file @var{infile} if
280 @var{infile} is not @code{nil}, and from the null device otherwise.
281 The argument @var{destination} says where to put the process output.
282 Here are the possibilities:
286 Insert the output in that buffer, before point. This includes both the
287 standard output stream and the standard error stream of the process.
290 Insert the output in a buffer with that name, before point.
293 Insert the output in the current buffer, before point.
299 Discard the output, and return @code{nil} immediately without waiting
300 for the subprocess to finish.
302 In this case, the process is not truly synchronous, since it can run in
303 parallel with Emacs; but you can think of it as synchronous in that
304 Emacs is essentially finished with the subprocess as soon as this
307 MS-DOS doesn't support asynchronous subprocesses, so this option doesn't
310 @item @code{(:file @var{file-name})}
311 Send the output to the file name specified, overwriting it if it
314 @item @code{(@var{real-destination} @var{error-destination})}
315 Keep the standard output stream separate from the standard error stream;
316 deal with the ordinary output as specified by @var{real-destination},
317 and dispose of the error output according to @var{error-destination}.
318 If @var{error-destination} is @code{nil}, that means to discard the
319 error output, @code{t} means mix it with the ordinary output, and a
320 string specifies a file name to redirect error output into.
322 You can't directly specify a buffer to put the error output in; that is
323 too difficult to implement. But you can achieve this result by sending
324 the error output to a temporary file and then inserting the file into a
328 If @var{display} is non-@code{nil}, then @code{call-process} redisplays
329 the buffer as output is inserted. (However, if the coding system chosen
330 for decoding output is @code{undecided}, meaning deduce the encoding
331 from the actual data, then redisplay sometimes cannot continue once
332 non-@acronym{ASCII} characters are encountered. There are fundamental
333 reasons why it is hard to fix this; see @ref{Output from Processes}.)
335 Otherwise the function @code{call-process} does no redisplay, and the
336 results become visible on the screen only when Emacs redisplays that
337 buffer in the normal course of events.
339 The remaining arguments, @var{args}, are strings that specify command
340 line arguments for the program.
342 The value returned by @code{call-process} (unless you told it not to
343 wait) indicates the reason for process termination. A number gives the
344 exit status of the subprocess; 0 means success, and any other value
345 means failure. If the process terminated with a signal,
346 @code{call-process} returns a string describing the signal.
348 In the examples below, the buffer @samp{foo} is current.
352 (call-process "pwd" nil t)
355 ---------- Buffer: foo ----------
357 ---------- Buffer: foo ----------
361 (call-process "grep" nil "bar" nil "lewis" "/etc/passwd")
364 ---------- Buffer: bar ----------
365 lewis:x:1001:1001:Bil Lewis,,,,:/home/lewis:/bin/bash
367 ---------- Buffer: bar ----------
371 Here is an example of the use of @code{call-process}, as used to
372 be found in the definition of the @code{insert-directory} function:
376 (call-process insert-directory-program nil t nil switches
378 (concat (file-name-as-directory file) ".")
384 @defun process-file program &optional infile buffer display &rest args
385 This function processes files synchronously in a separate process. It
386 is similar to @code{call-process}, but may invoke a file handler based
387 on the value of the variable @code{default-directory}, which specifies
388 the current working directory of the subprocess.
390 The arguments are handled in almost the same way as for
391 @code{call-process}, with the following differences:
393 Some file handlers may not support all combinations and forms of the
394 arguments @var{infile}, @var{buffer}, and @var{display}. For example,
395 some file handlers might behave as if @var{display} were @code{nil},
396 regardless of the value actually passed. As another example, some
397 file handlers might not support separating standard output and error
398 output by way of the @var{buffer} argument.
400 If a file handler is invoked, it determines the program to run based
401 on the first argument @var{program}. For instance, suppose that a
402 handler for remote files is invoked. Then the path that is used for
403 searching for the program might be different from @code{exec-path}.
405 The second argument @var{infile} may invoke a file handler. The file
406 handler could be different from the handler chosen for the
407 @code{process-file} function itself. (For example,
408 @code{default-directory} could be on one remote host, and
409 @var{infile} on a different remote host. Or @code{default-directory}
410 could be non-special, whereas @var{infile} is on a remote host.)
412 If @var{buffer} is a list of the form @code{(@var{real-destination}
413 @var{error-destination})}, and @var{error-destination} names a file,
414 then the same remarks as for @var{infile} apply.
416 The remaining arguments (@var{args}) will be passed to the process
417 verbatim. Emacs is not involved in processing file names that are
418 present in @var{args}. To avoid confusion, it may be best to avoid
419 absolute file names in @var{args}, but rather to specify all file
420 names as relative to @code{default-directory}. The function
421 @code{file-relative-name} is useful for constructing such relative
425 @defvar process-file-side-effects
426 This variable indicates whether a call of @code{process-file} changes
429 By default, this variable is always set to @code{t}, meaning that a
430 call of @code{process-file} could potentially change any file on a
431 remote host. When set to @code{nil}, a file handler could optimize
432 its behavior with respect to remote file attribute caching.
434 You should only ever change this variable with a let-binding; never
438 @defun call-process-region start end program &optional delete destination display &rest args
439 This function sends the text from @var{start} to @var{end} as
440 standard input to a process running @var{program}. It deletes the text
441 sent if @var{delete} is non-@code{nil}; this is useful when
442 @var{destination} is @code{t}, to insert the output in the current
443 buffer in place of the input.
445 The arguments @var{destination} and @var{display} control what to do
446 with the output from the subprocess, and whether to update the display
447 as it comes in. For details, see the description of
448 @code{call-process}, above. If @var{destination} is the integer 0,
449 @code{call-process-region} discards the output and returns @code{nil}
450 immediately, without waiting for the subprocess to finish (this only
451 works if asynchronous subprocesses are supported; i.e. not on MS-DOS).
453 The remaining arguments, @var{args}, are strings that specify command
454 line arguments for the program.
456 The return value of @code{call-process-region} is just like that of
457 @code{call-process}: @code{nil} if you told it to return without
458 waiting; otherwise, a number or string which indicates how the
459 subprocess terminated.
461 In the following example, we use @code{call-process-region} to run the
462 @code{cat} utility, with standard input being the first five characters
463 in buffer @samp{foo} (the word @samp{input}). @code{cat} copies its
464 standard input into its standard output. Since the argument
465 @var{destination} is @code{t}, this output is inserted in the current
470 ---------- Buffer: foo ----------
472 ---------- Buffer: foo ----------
476 (call-process-region 1 6 "cat" nil t)
479 ---------- Buffer: foo ----------
481 ---------- Buffer: foo ----------
485 For example, the @code{shell-command-on-region} command uses
486 @code{call-process-region} in a manner similar to this:
492 shell-file-name ; @r{name of program}
493 nil ; @r{do not delete region}
494 buffer ; @r{send output to @code{buffer}}
495 nil ; @r{no redisplay during output}
496 "-c" command) ; @r{arguments for the shell}
499 @c It actually uses shell-command-switch, but no need to mention that here.
502 @defun call-process-shell-command command &optional infile destination display &rest args
503 This function executes the shell command @var{command} synchronously.
504 The final arguments @var{args} are additional arguments to add at the
505 end of @var{command}. The other arguments are handled as in
509 @defun process-file-shell-command command &optional infile destination display &rest args
510 This function is like @code{call-process-shell-command}, but uses
511 @code{process-file} internally. Depending on @code{default-directory},
512 @var{command} can be executed also on remote hosts.
515 @defun shell-command-to-string command
516 This function executes @var{command} (a string) as a shell command,
517 then returns the command's output as a string.
520 @c There is also shell-command-on-region, but that is more of a user
521 @c command, not something to use in programs.
523 @defun process-lines program &rest args
524 This function runs @var{program}, waits for it to finish, and returns
525 its output as a list of strings. Each string in the list holds a
526 single line of text output by the program; the end-of-line characters
527 are stripped from each line. The arguments beyond @var{program},
528 @var{args}, are strings that specify command-line arguments with which
531 If @var{program} exits with a non-zero exit status, this function
534 This function works by calling @code{call-process}, so program output
535 is decoded in the same way as for @code{call-process}.
538 @node Asynchronous Processes
539 @section Creating an Asynchronous Process
540 @cindex asynchronous subprocess
542 After an @dfn{asynchronous process} is created, Emacs and the subprocess
543 both continue running immediately. The process thereafter runs
544 in parallel with Emacs, and the two can communicate with each other
545 using the functions described in the following sections. However,
546 communication is only partially asynchronous: Emacs sends data to the
547 process only when certain functions are called, and Emacs accepts data
548 from the process only when Emacs is waiting for input or for a time
551 Here we describe how to create an asynchronous process.
553 @defun start-process name buffer-or-name program &rest args
554 This function creates a new asynchronous subprocess and starts the
555 program @var{program} running in it. It returns a process object that
556 stands for the new subprocess in Lisp. The argument @var{name}
557 specifies the name for the process object; if a process with this name
558 already exists, then @var{name} is modified (by appending @samp{<1>},
559 etc.) to be unique. The buffer @var{buffer-or-name} is the buffer to
560 associate with the process.
562 The remaining arguments, @var{args}, are strings that specify command
563 line arguments for the program.
565 In the example below, the first process is started and runs (rather,
566 sleeps) for 100 seconds (the output buffer @samp{foo} is created
567 immediately). Meanwhile, the second process is started, and
568 given the name @samp{my-process<1>} for the sake of uniqueness. It
569 inserts the directory listing at the end of the buffer @samp{foo},
570 before the first process finishes. Then it finishes, and a message to
571 that effect is inserted in the buffer. Much later, the first process
572 finishes, and another message is inserted in the buffer for it.
576 (start-process "my-process" "foo" "sleep" "100")
577 @result{} #<process my-process>
581 (start-process "my-process" "foo" "ls" "-l" "/bin")
582 @result{} #<process my-process<1>>
584 ---------- Buffer: foo ----------
586 -rwxr-xr-x 1 root root 971384 Mar 30 10:14 bash
587 -rwxr-xr-x 1 root root 146920 Jul 5 2011 bsd-csh
589 -rwxr-xr-x 1 root root 696880 Feb 28 15:55 zsh4
591 Process my-process<1> finished
593 Process my-process finished
594 ---------- Buffer: foo ----------
599 @defun start-file-process name buffer-or-name program &rest args
600 Like @code{start-process}, this function starts a new asynchronous
601 subprocess running @var{program} in it, and returns its process
604 The difference from @code{start-process} is that this function may
605 invoked a file handler based on the value of @code{default-directory}.
606 This handler ought to run @var{program}, perhaps on the local host,
607 perhaps on a remote host that corresponds to @code{default-directory}.
608 In the latter case, the local part of @code{default-directory} becomes
609 the working directory of the process.
611 This function does not try to invoke file name handlers for
612 @var{program} or for the @var{program-args}.
614 Depending on the implementation of the file handler, it might not be
615 possible to apply @code{process-filter} or @code{process-sentinel} to
616 the resulting process object. @xref{Filter Functions}, and @ref{Sentinels}.
618 @c FIXME Can we find a better example (i.e. a more modern function
619 @c that is actually documented).
620 Some file handlers may not support @code{start-file-process} (for
621 example the function @code{ange-ftp-hook-function}). In such cases,
622 this function does nothing and returns @code{nil}.
625 @defun start-process-shell-command name buffer-or-name command
626 This function is like @code{start-process}, except that it uses a shell
627 to execute the specified command. The argument @var{command} is a shell
628 command name. The variable @code{shell-file-name} specifies which shell to
631 The point of running a program through the shell, rather than directly
632 with @code{start-process}, is so that you can employ shell features such
633 as wildcards in the arguments. It follows that if you include any
634 arbitrary user-specified arguments in the command, you should quote them
635 with @code{shell-quote-argument} first, so that any special shell
636 characters do @emph{not} have their special shell meanings. @xref{Shell
637 Arguments}. Of course, when executing commands based on user input
638 you should also consider the security implications.
641 @defun start-file-process-shell-command name buffer-or-name command
642 This function is like @code{start-process-shell-command}, but uses
643 @code{start-file-process} internally. Because of this, @var{command}
644 can also be executed on remote hosts, depending on @code{default-directory}.
647 @defvar process-connection-type
649 @cindex @acronym{PTY}s
650 This variable controls the type of device used to communicate with
651 asynchronous subprocesses. If it is non-@code{nil}, then @acronym{PTY}s are
652 used, when available. Otherwise, pipes are used.
654 @acronym{PTY}s are usually preferable for processes visible to the user, as
655 in Shell mode, because they allow job control (@kbd{C-c}, @kbd{C-z},
656 etc.) to work between the process and its children, whereas pipes do
657 not. For subprocesses used for internal purposes by programs, it is
658 often better to use a pipe, because they are more efficient. In
659 addition, the total number of @acronym{PTY}s is limited on many systems and
660 it is good not to waste them.
662 The value of @code{process-connection-type} takes effect when
663 @code{start-process} is called. So you can specify how to communicate
664 with one subprocess by binding the variable around the call to
665 @code{start-process}.
669 (let ((process-connection-type nil)) ; @r{use a pipe}
670 (start-process @dots{}))
674 To determine whether a given subprocess actually got a pipe or a
675 @acronym{PTY}, use the function @code{process-tty-name} (@pxref{Process
679 @node Deleting Processes
680 @section Deleting Processes
681 @cindex deleting processes
683 @dfn{Deleting a process} disconnects Emacs immediately from the
684 subprocess. Processes are deleted automatically after they terminate,
685 but not necessarily right away. You can delete a process explicitly
686 at any time. If you explicitly delete a terminated process before it
687 is deleted automatically, no harm results. Deleting a running
688 process sends a signal to terminate it (and its child processes, if
689 any), and calls the process sentinel if it has one. @xref{Sentinels}.
691 When a process is deleted, the process object itself continues to
692 exist as long as other Lisp objects point to it. All the Lisp
693 primitives that work on process objects accept deleted processes, but
694 those that do I/O or send signals will report an error. The process
695 mark continues to point to the same place as before, usually into a
696 buffer where output from the process was being inserted.
698 @defopt delete-exited-processes
699 This variable controls automatic deletion of processes that have
700 terminated (due to calling @code{exit} or to a signal). If it is
701 @code{nil}, then they continue to exist until the user runs
702 @code{list-processes}. Otherwise, they are deleted immediately after
706 @defun delete-process process
707 This function deletes a process, killing it with a @code{SIGKILL}
708 signal. The argument may be a process, the name of a process, a
709 buffer, or the name of a buffer. (A buffer or buffer-name stands for
710 the process that @code{get-buffer-process} returns.) Calling
711 @code{delete-process} on a running process terminates it, updates the
712 process status, and runs the sentinel (if any) immediately. If the
713 process has already terminated, calling @code{delete-process} has no
714 effect on its status, or on the running of its sentinel (which will
715 happen sooner or later).
719 (delete-process "*shell*")
725 @node Process Information
726 @section Process Information
728 Several functions return information about processes.
729 @code{list-processes} is provided for interactive use.
731 @deffn Command list-processes &optional query-only
732 This command displays a listing of all living processes. In addition,
733 it finally deletes any process whose status was @samp{Exited} or
734 @samp{Signaled}. It returns @code{nil}.
736 The processes are shown in a buffer named @file{*Process List*}, whose
737 major mode is named Process Menu mode.
739 If @var{query-only} is non-@code{nil} then it lists only processes
740 whose query flag is non-@code{nil}. @xref{Query Before Exit}.
744 This function returns a list of all processes that have not been deleted.
749 @result{} (#<process display-time> #<process shell>)
754 @defun get-process name
755 This function returns the process named @var{name}, or @code{nil} if
756 there is none. An error is signaled if @var{name} is not a string.
760 (get-process "shell")
761 @result{} #<process shell>
766 @defun process-command process
767 This function returns the command that was executed to start
768 @var{process}. This is a list of strings, the first string being the
769 program executed and the rest of the strings being the arguments that
770 were given to the program.
774 (process-command (get-process "shell"))
775 @result{} ("/bin/csh" "-i")
780 @defun process-contact process &optional key
782 This function returns information about how a network or serial
783 process was set up. For a network process, when @var{key} is
784 @code{nil}, it returns @code{(@var{hostname} @var{service})} which
785 specifies what you connected to. For a serial process, when @var{key}
786 is @code{nil}, it returns @code{(@var{port} @var{speed})}. For an
787 ordinary child process, this function always returns @code{t}.
789 If @var{key} is @code{t}, the value is the complete status information
790 for the connection, server, or serial port; that is, the list of
791 keywords and values specified in @code{make-network-process} or
792 @code{make-serial-process}, except that some of the values represent
793 the current status instead of what you specified.
795 For a network process:
799 The associated value is the process buffer.
801 The associated value is the process filter function.
803 The associated value is the process sentinel function.
805 In a connection, the address in internal format of the remote peer.
807 The local address, in internal format.
809 In a server, if you specified @code{t} for @var{service},
810 this value is the actual port number.
813 @code{:local} and @code{:remote} are included even if they were not
814 specified explicitly in @code{make-network-process}.
816 For a serial process, see @code{make-serial-process} and
817 @code{serial-process-configure} for a list of keys.
819 If @var{key} is a keyword, the function returns the value corresponding
823 @defun process-id process
824 This function returns the @acronym{PID} of @var{process}. This is an
825 integer that distinguishes the process @var{process} from all other
826 processes running on the same computer at the current time. The
827 @acronym{PID} of a process is chosen by the operating system kernel when the
828 process is started and remains constant as long as the process exists.
831 @defun process-name process
832 This function returns the name of @var{process}.
835 @defun process-status process-name
836 This function returns the status of @var{process-name} as a symbol.
837 The argument @var{process-name} must be a process, a buffer, or a
838 process name (a string).
840 The possible values for an actual subprocess are:
844 for a process that is running.
846 for a process that is stopped but continuable.
848 for a process that has exited.
850 for a process that has received a fatal signal.
852 for a network connection that is open.
854 for a network connection that is closed. Once a connection
855 is closed, you cannot reopen it, though you might be able to open
856 a new connection to the same place.
858 for a non-blocking connection that is waiting to complete.
860 for a non-blocking connection that has failed to complete.
862 for a network server that is listening.
864 if @var{process-name} is not the name of an existing process.
869 (process-status (get-buffer "*shell*"))
874 @result{} #<process xx<1>>
880 For a network connection, @code{process-status} returns one of the symbols
881 @code{open} or @code{closed}. The latter means that the other side
882 closed the connection, or Emacs did @code{delete-process}.
885 @defun process-live-p process
886 This function returns non-@code{nil} if @var{process} is alive. A
887 process is considered alive if its status is @code{run}, @code{open},
888 @code{listen}, @code{connect} or @code{stop}.
891 @defun process-type process
892 This function returns the symbol @code{network} for a network
893 connection or server, @code{serial} for a serial port connection, or
894 @code{real} for a real subprocess.
897 @defun process-exit-status process
898 This function returns the exit status of @var{process} or the signal
899 number that killed it. (Use the result of @code{process-status} to
900 determine which of those it is.) If @var{process} has not yet
901 terminated, the value is 0.
904 @defun process-tty-name process
905 This function returns the terminal name that @var{process} is using for
906 its communication with Emacs---or @code{nil} if it is using pipes
907 instead of a terminal (see @code{process-connection-type} in
908 @ref{Asynchronous Processes}). If @var{process} represents a program
909 running on a remote host, the terminal name used by that program on
910 the remote host is provided as process property @code{remote-tty}.
913 @defun process-coding-system process
914 @anchor{Coding systems for a subprocess}
915 This function returns a cons cell describing the coding systems in use
916 for decoding output from @var{process} and for encoding input to
917 @var{process} (@pxref{Coding Systems}). The value has this form:
920 (@var{coding-system-for-decoding} . @var{coding-system-for-encoding})
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.
961 Some operating systems have limited space for buffered input in a
962 @acronym{PTY}. On these systems, Emacs sends an @acronym{EOF}
963 periodically amidst the other characters, to force them through. For
964 most programs, these @acronym{EOF}s do no harm.
966 Subprocess input is normally encoded using a coding system before the
967 subprocess receives it, much like text written into a file. You can use
968 @code{set-process-coding-system} to specify which coding system to use
969 (@pxref{Process Information}). Otherwise, the coding system comes from
970 @code{coding-system-for-write}, if that is non-@code{nil}; or else from
971 the defaulting mechanism (@pxref{Default Coding Systems}).
973 Sometimes the system is unable to accept input for that process,
974 because the input buffer is full. When this happens, the send functions
975 wait a short while, accepting output from subprocesses, and then try
976 again. This gives the subprocess a chance to read more of its pending
977 input and make space in the buffer. It also allows filters, sentinels
978 and timers to run---so take account of that in writing your code.
980 In these functions, the @var{process} argument can be a process or
981 the name of a process, or a buffer or buffer name (which stands
982 for a process via @code{get-buffer-process}). @code{nil} means
983 the current buffer's process.
985 @defun process-send-string process string
986 This function sends @var{process} the contents of @var{string} as
987 standard input. If it is @code{nil}, the current buffer's process is used.
989 The function returns @code{nil}.
993 (process-send-string "shell<1>" "ls\n")
999 ---------- Buffer: *shell* ----------
1001 introduction.texi syntax-tables.texi~
1002 introduction.texi~ text.texi
1003 introduction.txt text.texi~
1005 ---------- Buffer: *shell* ----------
1010 @defun process-send-region process start end
1011 This function sends the text in the region defined by @var{start} and
1012 @var{end} as standard input to @var{process}.
1014 An error is signaled unless both @var{start} and @var{end} are
1015 integers or markers that indicate positions in the current buffer. (It
1016 is unimportant which number is larger.)
1019 @defun process-send-eof &optional process
1020 This function makes @var{process} see an end-of-file in its
1021 input. The @acronym{EOF} comes after any text already sent to it.
1023 The function returns @var{process}.
1027 (process-send-eof "shell")
1033 @defun process-running-child-p &optional process
1034 This function will tell you whether a @var{process} has given control of
1035 its terminal to its own child process. The value is @code{t} if this is
1036 true, or if Emacs cannot tell; it is @code{nil} if Emacs can be certain
1037 that this is not so.
1040 @node Signals to Processes
1041 @section Sending Signals to Processes
1042 @cindex process signals
1043 @cindex sending signals
1046 @dfn{Sending a signal} to a subprocess is a way of interrupting its
1047 activities. There are several different signals, each with its own
1048 meaning. The set of signals and their names is defined by the operating
1049 system. For example, the signal @code{SIGINT} means that the user has
1050 typed @kbd{C-c}, or that some analogous thing has happened.
1052 Each signal has a standard effect on the subprocess. Most signals
1053 kill the subprocess, but some stop or resume execution instead. Most
1054 signals can optionally be handled by programs; if the program handles
1055 the signal, then we can say nothing in general about its effects.
1057 You can send signals explicitly by calling the functions in this
1058 section. Emacs also sends signals automatically at certain times:
1059 killing a buffer sends a @code{SIGHUP} signal to all its associated
1060 processes; killing Emacs sends a @code{SIGHUP} signal to all remaining
1061 processes. (@code{SIGHUP} is a signal that usually indicates that the
1062 user hung up the phone.)
1064 Each of the signal-sending functions takes two optional arguments:
1065 @var{process} and @var{current-group}.
1067 The argument @var{process} must be either a process, a process
1068 name, a buffer, a buffer name, or @code{nil}. A buffer or buffer name
1069 stands for a process through @code{get-buffer-process}. @code{nil}
1070 stands for the process associated with the current buffer. An error
1071 is signaled if @var{process} does not identify a process.
1073 The argument @var{current-group} is a flag that makes a difference
1074 when you are running a job-control shell as an Emacs subprocess. If it
1075 is non-@code{nil}, then the signal is sent to the current process-group
1076 of the terminal that Emacs uses to communicate with the subprocess. If
1077 the process is a job-control shell, this means the shell's current
1078 subjob. If it is @code{nil}, the signal is sent to the process group of
1079 the immediate subprocess of Emacs. If the subprocess is a job-control
1080 shell, this is the shell itself.
1082 The flag @var{current-group} has no effect when a pipe is used to
1083 communicate with the subprocess, because the operating system does not
1084 support the distinction in the case of pipes. For the same reason,
1085 job-control shells won't work when a pipe is used. See
1086 @code{process-connection-type} in @ref{Asynchronous Processes}.
1088 @defun interrupt-process &optional process current-group
1089 This function interrupts the process @var{process} by sending the
1090 signal @code{SIGINT}. Outside of Emacs, typing the ``interrupt
1091 character'' (normally @kbd{C-c} on some systems, and @code{DEL} on
1092 others) sends this signal. When the argument @var{current-group} is
1093 non-@code{nil}, you can think of this function as ``typing @kbd{C-c}''
1094 on the terminal by which Emacs talks to the subprocess.
1097 @defun kill-process &optional process current-group
1098 This function kills the process @var{process} by sending the
1099 signal @code{SIGKILL}. This signal kills the subprocess immediately,
1100 and cannot be handled by the subprocess.
1103 @defun quit-process &optional process current-group
1104 This function sends the signal @code{SIGQUIT} to the process
1105 @var{process}. This signal is the one sent by the ``quit
1106 character'' (usually @kbd{C-b} or @kbd{C-\}) when you are not inside
1110 @defun stop-process &optional process current-group
1111 This function stops the process @var{process} by sending the
1112 signal @code{SIGTSTP}. Use @code{continue-process} to resume its
1115 Outside of Emacs, on systems with job control, the ``stop character''
1116 (usually @kbd{C-z}) normally sends this signal. When
1117 @var{current-group} is non-@code{nil}, you can think of this function as
1118 ``typing @kbd{C-z}'' on the terminal Emacs uses to communicate with the
1122 @defun continue-process &optional process current-group
1123 This function resumes execution of the process @var{process} by sending
1124 it the signal @code{SIGCONT}. This presumes that @var{process} was
1128 @defun signal-process process signal
1129 This function sends a signal to process @var{process}. The argument
1130 @var{signal} specifies which signal to send; it should be an integer.
1132 The @var{process} argument can be a system process @acronym{ID}; that
1133 allows you to send signals to processes that are not children of
1134 Emacs. @xref{System Processes}.
1137 @node Output from Processes
1138 @section Receiving Output from Processes
1139 @cindex process output
1140 @cindex output from processes
1142 There are two ways to receive the output that a subprocess writes to
1143 its standard output stream. The output can be inserted in a buffer,
1144 which is called the associated buffer of the process, or a function
1145 called the @dfn{filter function} can be called to act on the output. If
1146 the process has no buffer and no filter function, its output is
1149 When a subprocess terminates, Emacs reads any pending output,
1150 then stops reading output from that subprocess. Therefore, if the
1151 subprocess has children that are still live and still producing
1152 output, Emacs won't receive that output.
1154 Output from a subprocess can arrive only while Emacs is waiting: when
1155 reading terminal input, in @code{sit-for} and @code{sleep-for}
1156 (@pxref{Waiting}), and in @code{accept-process-output} (@pxref{Accepting
1157 Output}). This minimizes the problem of timing errors that usually
1158 plague parallel programming. For example, you can safely create a
1159 process and only then specify its buffer or filter function; no output
1160 can arrive before you finish, if the code in between does not call any
1161 primitive that waits.
1163 @defvar process-adaptive-read-buffering
1164 On some systems, when Emacs reads the output from a subprocess, the
1165 output data is read in very small blocks, potentially resulting in
1166 very poor performance. This behavior can be remedied to some extent
1167 by setting the variable @var{process-adaptive-read-buffering} to a
1168 non-@code{nil} value (the default), as it will automatically delay reading
1169 from such processes, thus allowing them to produce more output before
1170 Emacs tries to read it.
1173 It is impossible to separate the standard output and standard error
1174 streams of the subprocess, because Emacs normally spawns the subprocess
1175 inside a pseudo-TTY, and a pseudo-TTY has only one output channel. If
1176 you want to keep the output to those streams separate, you should
1177 redirect one of them to a file---for example, by using an appropriate
1181 * Process Buffers:: If no filter, output is put in a buffer.
1182 * Filter Functions:: Filter functions accept output from the process.
1183 * Decoding Output:: Filters can get unibyte or multibyte strings.
1184 * Accepting Output:: How to wait until process output arrives.
1187 @node Process Buffers
1188 @subsection Process Buffers
1190 A process can (and usually does) have an @dfn{associated buffer},
1191 which is an ordinary Emacs buffer that is used for two purposes: storing
1192 the output from the process, and deciding when to kill the process. You
1193 can also use the buffer to identify a process to operate on, since in
1194 normal practice only one process is associated with any given buffer.
1195 Many applications of processes also use the buffer for editing input to
1196 be sent to the process, but this is not built into Emacs Lisp.
1198 Unless the process has a filter function (@pxref{Filter Functions}),
1199 its output is inserted in the associated buffer. The position to insert
1200 the output is determined by the @code{process-mark}, which is then
1201 updated to point to the end of the text just inserted. Usually, but not
1202 always, the @code{process-mark} is at the end of the buffer.
1204 @findex process-kill-buffer-query-function
1205 Killing the associated buffer of a process also kills the process.
1206 Emacs asks for confirmation first, if the process's
1207 @code{process-query-on-exit-flag} is non-@code{nil} (@pxref{Query
1208 Before Exit}). This confirmation is done by the function
1209 @code{process-kill-buffer-query-function}, which is run from
1210 @code{kill-buffer-query-functions} (@pxref{Killing Buffers}).
1212 @defun process-buffer process
1213 This function returns the associated buffer of the process
1218 (process-buffer (get-process "shell"))
1219 @result{} #<buffer *shell*>
1224 @defun process-mark process
1225 This function returns the process marker for @var{process}, which is the
1226 marker that says where to insert output from the process.
1228 If @var{process} does not have a buffer, @code{process-mark} returns a
1229 marker that points nowhere.
1231 Insertion of process output in a buffer uses this marker to decide where
1232 to insert, and updates it to point after the inserted text. That is why
1233 successive batches of output are inserted consecutively.
1235 Filter functions normally should use this marker in the same fashion
1236 as is done by direct insertion of output in the buffer. A good
1237 example of a filter function that uses @code{process-mark} is found at
1238 the end of the following section.
1240 When the user is expected to enter input in the process buffer for
1241 transmission to the process, the process marker separates the new input
1242 from previous output.
1245 @defun set-process-buffer process buffer
1246 This function sets the buffer associated with @var{process} to
1247 @var{buffer}. If @var{buffer} is @code{nil}, the process becomes
1248 associated with no buffer.
1251 @defun get-buffer-process buffer-or-name
1252 This function returns a nondeleted process associated with the buffer
1253 specified by @var{buffer-or-name}. If there are several processes
1254 associated with it, this function chooses one (currently, the one most
1255 recently created, but don't count on that). Deletion of a process
1256 (see @code{delete-process}) makes it ineligible for this function to
1259 It is usually a bad idea to have more than one process associated with
1264 (get-buffer-process "*shell*")
1265 @result{} #<process shell>
1269 Killing the process's buffer deletes the process, which kills the
1270 subprocess with a @code{SIGHUP} signal (@pxref{Signals to Processes}).
1273 @node Filter Functions
1274 @subsection Process Filter Functions
1275 @cindex filter function
1276 @cindex process filter
1278 A process @dfn{filter function} is a function that receives the
1279 standard output from the associated process. If a process has a filter,
1280 then @emph{all} output from that process is passed to the filter. The
1281 process buffer is used directly for output from the process only when
1284 The filter function can only be called when Emacs is waiting for
1285 something, because process output arrives only at such times. Emacs
1286 waits when reading terminal input, in @code{sit-for} and
1287 @code{sleep-for} (@pxref{Waiting}), and in @code{accept-process-output}
1288 (@pxref{Accepting Output}).
1290 A filter function must accept two arguments: the associated process
1291 and a string, which is output just received from it. The function is
1292 then free to do whatever it chooses with the output.
1294 Quitting is normally inhibited within a filter function---otherwise,
1295 the effect of typing @kbd{C-g} at command level or to quit a user
1296 command would be unpredictable. If you want to permit quitting inside
1297 a filter function, bind @code{inhibit-quit} to @code{nil}. In most
1298 cases, the right way to do this is with the macro
1299 @code{with-local-quit}. @xref{Quitting}.
1301 If an error happens during execution of a filter function, it is
1302 caught automatically, so that it doesn't stop the execution of whatever
1303 program was running when the filter function was started. However, if
1304 @code{debug-on-error} is non-@code{nil}, the error-catching is turned
1305 off. This makes it possible to use the Lisp debugger to debug the
1306 filter function. @xref{Debugger}.
1308 Many filter functions sometimes or always insert the text in the
1309 process's buffer, mimicking the actions of Emacs when there is no
1310 filter. Such filter functions need to use @code{set-buffer} in order to
1311 be sure to insert in that buffer. To avoid setting the current buffer
1312 semipermanently, these filter functions must save and restore the
1313 current buffer. They should also check whether the buffer is still
1314 alive, update the process marker, and in some cases update the value
1315 of point. Here is how to do these things:
1319 (defun ordinary-insertion-filter (proc string)
1320 (when (buffer-live-p (process-buffer proc))
1321 (with-current-buffer (process-buffer proc)
1322 (let ((moving (= (point) (process-mark proc))))
1326 ;; @r{Insert the text, advancing the process marker.}
1327 (goto-char (process-mark proc))
1329 (set-marker (process-mark proc) (point)))
1330 (if moving (goto-char (process-mark proc)))))))
1335 The reason to use @code{with-current-buffer}, rather than using
1336 @code{save-excursion} to save and restore the current buffer, is so as
1337 to preserve the change in point made by the second call to
1340 To make the filter force the process buffer to be visible whenever new
1341 text arrives, insert the following line just before the
1342 @code{with-current-buffer} construct:
1345 (display-buffer (process-buffer proc))
1348 To force point to the end of the new output, no matter where it was
1349 previously, eliminate the variable @code{moving} and call
1350 @code{goto-char} unconditionally.
1352 In earlier Emacs versions, every filter function that did regular
1353 expression searching or matching had to explicitly save and restore the
1354 match data. Now Emacs does this automatically for filter functions;
1355 they never need to do it explicitly. @xref{Match Data}.
1357 The output to the function may come in chunks of any size. A program
1358 that produces the same output twice in a row may send it as one batch of
1359 200 characters one time, and five batches of 40 characters the next. If
1360 the filter looks for certain text strings in the subprocess output, make
1361 sure to handle the case where one of these strings is split across two
1362 or more batches of output; one way to do this is to insert the
1363 received text into a temporary buffer, which can then be searched.
1365 @defun set-process-filter process filter
1366 This function gives @var{process} the filter function @var{filter}. If
1367 @var{filter} is @code{nil}, it gives the process no filter.
1370 @defun process-filter process
1371 This function returns the filter function of @var{process}, or @code{nil}
1375 Here is an example of use of a filter function:
1379 (defun keep-output (process output)
1380 (setq kept (cons output kept)))
1381 @result{} keep-output
1388 (set-process-filter (get-process "shell") 'keep-output)
1389 @result{} keep-output
1392 (process-send-string "shell" "ls ~/other\n")
1395 @result{} ("lewis@@slug[8] % "
1398 "FINAL-W87-SHORT.MSS backup.otl kolstad.mss~
1399 address.txt backup.psf kolstad.psf
1400 backup.bib~ david.mss resume-Dec-86.mss~
1401 backup.err david.psf resume-Dec.psf
1402 backup.mss dland syllabus.mss
1404 "#backups.mss# backup.mss~ kolstad.mss
1409 @ignore @c The code in this example doesn't show the right way to do things.
1410 Here is another, more realistic example, which demonstrates how to use
1411 the process mark to do insertion in the same fashion as is done when
1412 there is no filter function:
1416 ;; @r{Insert input in the buffer specified by @code{my-shell-buffer}}
1417 ;; @r{and make sure that buffer is shown in some window.}
1418 (defun my-process-filter (proc str)
1419 (let ((cur (selected-window))
1421 (pop-to-buffer my-shell-buffer)
1424 (goto-char (point-max))
1426 (set-marker (process-mark proc) (point-max))
1427 (select-window cur)))
1432 @node Decoding Output
1433 @subsection Decoding Process Output
1434 @cindex decode process output
1436 When Emacs writes process output directly into a multibyte buffer,
1437 it decodes the output according to the process output coding system.
1438 If the coding system is @code{raw-text} or @code{no-conversion}, Emacs
1439 converts the unibyte output to multibyte using
1440 @code{string-to-multibyte}, and inserts the resulting multibyte text.
1442 You can use @code{set-process-coding-system} to specify which coding
1443 system to use (@pxref{Process Information}). Otherwise, the coding
1444 system comes from @code{coding-system-for-read}, if that is
1445 non-@code{nil}; or else from the defaulting mechanism (@pxref{Default
1446 Coding Systems}). If the text output by a process contains null
1447 bytes, Emacs by default uses @code{no-conversion} for it; see
1448 @ref{Lisp and Coding Systems, inhibit-null-byte-detection}, for how to
1449 control this behavior.
1451 @strong{Warning:} Coding systems such as @code{undecided} which
1452 determine the coding system from the data do not work entirely
1453 reliably with asynchronous subprocess output. This is because Emacs
1454 has to process asynchronous subprocess output in batches, as it
1455 arrives. Emacs must try to detect the proper coding system from one
1456 batch at a time, and this does not always work. Therefore, if at all
1457 possible, specify a coding system that determines both the character
1458 code conversion and the end of line conversion---that is, one like
1459 @code{latin-1-unix}, rather than @code{undecided} or @code{latin-1}.
1461 @c Let's keep the index entries that were there for
1462 @c set-process-filter-multibyte and process-filter-multibyte-p,
1463 @cindex filter multibyte flag, of process
1464 @cindex process filter multibyte flag
1465 When Emacs calls a process filter function, it provides the process
1466 output as a multibyte string or as a unibyte string according to the
1467 process's filter coding system. Emacs
1468 decodes the output according to the process output coding system,
1469 which usually produces a multibyte string, except for coding systems
1470 such as @code{binary} and @code{raw-text}
1472 @node Accepting Output
1473 @subsection Accepting Output from Processes
1474 @cindex accept input from processes
1476 Output from asynchronous subprocesses normally arrives only while
1477 Emacs is waiting for some sort of external event, such as elapsed time
1478 or terminal input. Occasionally it is useful in a Lisp program to
1479 explicitly permit output to arrive at a specific point, or even to wait
1480 until output arrives from a process.
1482 @defun accept-process-output &optional process seconds millisec just-this-one
1483 This function allows Emacs to read pending output from processes. The
1484 output is inserted in the associated buffers or given to their filter
1485 functions. If @var{process} is non-@code{nil} then this function does
1486 not return until some output has been received from @var{process}.
1489 The arguments @var{seconds} and @var{millisec} let you specify timeout
1490 periods. The former specifies a period measured in seconds and the
1491 latter specifies one measured in milliseconds. The two time periods
1492 thus specified are added together, and @code{accept-process-output}
1493 returns after that much time, whether or not there has been any
1496 The argument @var{millisec} is semi-obsolete nowadays because
1497 @var{seconds} can be a floating point number to specify waiting a
1498 fractional number of seconds. If @var{seconds} is 0, the function
1499 accepts whatever output is pending but does not wait.
1501 @c Emacs 22.1 feature
1502 If @var{process} is a process, and the argument @var{just-this-one} is
1503 non-@code{nil}, only output from that process is handled, suspending output
1504 from other processes until some output has been received from that
1505 process or the timeout expires. If @var{just-this-one} is an integer,
1506 also inhibit running timers. This feature is generally not
1507 recommended, but may be necessary for specific applications, such as
1510 The function @code{accept-process-output} returns non-@code{nil} if it
1511 did get some output, or @code{nil} if the timeout expired before output
1516 @section Sentinels: Detecting Process Status Changes
1517 @cindex process sentinel
1518 @cindex sentinel (of process)
1520 A @dfn{process sentinel} is a function that is called whenever the
1521 associated process changes status for any reason, including signals
1522 (whether sent by Emacs or caused by the process's own actions) that
1523 terminate, stop, or continue the process. The process sentinel is
1524 also called if the process exits. The sentinel receives two
1525 arguments: the process for which the event occurred, and a string
1526 describing the type of event.
1528 The string describing the event looks like one of the following:
1532 @code{"finished\n"}.
1535 @code{"exited abnormally with code @var{exitcode}\n"}.
1538 @code{"@var{name-of-signal}\n"}.
1541 @code{"@var{name-of-signal} (core dumped)\n"}.
1544 A sentinel runs only while Emacs is waiting (e.g., for terminal
1545 input, or for time to elapse, or for process output). This avoids the
1546 timing errors that could result from running them at random places in
1547 the middle of other Lisp programs. A program can wait, so that
1548 sentinels will run, by calling @code{sit-for} or @code{sleep-for}
1549 (@pxref{Waiting}), or @code{accept-process-output} (@pxref{Accepting
1550 Output}). Emacs also allows sentinels to run when the command loop is
1551 reading input. @code{delete-process} calls the sentinel when it
1552 terminates a running process.
1554 Emacs does not keep a queue of multiple reasons to call the sentinel
1555 of one process; it records just the current status and the fact that
1556 there has been a change. Therefore two changes in status, coming in
1557 quick succession, can call the sentinel just once. However, process
1558 termination will always run the sentinel exactly once. This is
1559 because the process status can't change again after termination.
1561 Emacs explicitly checks for output from the process before running
1562 the process sentinel. Once the sentinel runs due to process
1563 termination, no further output can arrive from the process.
1565 A sentinel that writes the output into the buffer of the process
1566 should check whether the buffer is still alive. If it tries to insert
1567 into a dead buffer, it will get an error. If the buffer is dead,
1568 @code{(buffer-name (process-buffer @var{process}))} returns @code{nil}.
1570 Quitting is normally inhibited within a sentinel---otherwise, the
1571 effect of typing @kbd{C-g} at command level or to quit a user command
1572 would be unpredictable. If you want to permit quitting inside a
1573 sentinel, bind @code{inhibit-quit} to @code{nil}. In most cases, the
1574 right way to do this is with the macro @code{with-local-quit}.
1577 If an error happens during execution of a sentinel, it is caught
1578 automatically, so that it doesn't stop the execution of whatever
1579 programs was running when the sentinel was started. However, if
1580 @code{debug-on-error} is non-@code{nil}, the error-catching is turned
1581 off. This makes it possible to use the Lisp debugger to debug the
1582 sentinel. @xref{Debugger}.
1584 While a sentinel is running, the process sentinel is temporarily
1585 set to @code{nil} so that the sentinel won't run recursively.
1586 For this reason it is not possible for a sentinel to specify
1589 In earlier Emacs versions, every sentinel that did regular expression
1590 searching or matching had to explicitly save and restore the match data.
1591 Now Emacs does this automatically for sentinels; they never need to do
1592 it explicitly. @xref{Match Data}.
1594 @defun set-process-sentinel process sentinel
1595 This function associates @var{sentinel} with @var{process}. If
1596 @var{sentinel} is @code{nil}, then the process will have no sentinel.
1597 The default behavior when there is no sentinel is to insert a message in
1598 the process's buffer when the process status changes.
1600 Changes in process sentinel take effect immediately---if the sentinel
1601 is slated to be run but has not been called yet, and you specify a new
1602 sentinel, the eventual call to the sentinel will use the new one.
1606 (defun msg-me (process event)
1608 (format "Process: %s had the event `%s'" process event)))
1609 (set-process-sentinel (get-process "shell") 'msg-me)
1613 (kill-process (get-process "shell"))
1614 @print{} Process: #<process shell> had the event `killed'
1615 @result{} #<process shell>
1620 @defun process-sentinel process
1621 This function returns the sentinel of @var{process}, or @code{nil} if it
1625 @defun waiting-for-user-input-p
1626 While a sentinel or filter function is running, this function returns
1627 non-@code{nil} if Emacs was waiting for keyboard input from the user at
1628 the time the sentinel or filter function was called, @code{nil} if it
1632 @node Query Before Exit
1633 @section Querying Before Exit
1635 When Emacs exits, it terminates all its subprocesses by sending them
1636 the @code{SIGHUP} signal. Because subprocesses may be doing
1637 valuable work, Emacs normally asks the user to confirm that it is ok
1638 to terminate them. Each process has a query flag which, if
1639 non-@code{nil}, says that Emacs should ask for confirmation before
1640 exiting and thus killing that process. The default for the query flag
1641 is @code{t}, meaning @emph{do} query.
1643 @defun process-query-on-exit-flag process
1644 This returns the query flag of @var{process}.
1647 @defun set-process-query-on-exit-flag process flag
1648 This function sets the query flag of @var{process} to @var{flag}. It
1651 Here is an example of using @code{set-process-query-on-exit-flag} on a
1652 shell process to avoid querying:
1656 (set-process-query-on-exit-flag (get-process "shell") nil)
1662 @node System Processes
1663 @section Accessing Other Processes
1664 @cindex system processes
1666 In addition to accessing and manipulating processes that are
1667 subprocesses of the current Emacs session, Emacs Lisp programs can
1668 also access other processes running on the same machine. We call
1669 these @dfn{system processes}, to distinguish between them and Emacs
1672 Emacs provides several primitives for accessing system processes.
1673 Not all platforms support these primitives; on those which don't,
1674 these primitives return @code{nil}.
1676 @defun list-system-processes
1677 This function returns a list of all the processes running on the
1678 system. Each process is identified by its @acronym{PID}, a numerical
1679 process ID that is assigned by the OS and distinguishes the process
1680 from all the other processes running on the same machine at the same
1684 @defun process-attributes pid
1685 This function returns an alist of attributes for the process specified
1686 by its process ID @var{pid}. Each association in the alist is of the
1687 form @code{(@var{key} . @var{value})}, where @var{key} designates the
1688 attribute and @var{value} is the value of that attribute. The various
1689 attribute @var{key}'s that this function can return are listed below.
1690 Not all platforms support all of these attributes; if an attribute is
1691 not supported, its association will not appear in the returned alist.
1692 Values that are numbers can be either integer or floating-point,
1693 depending on the magnitude of the value.
1697 The effective user ID of the user who invoked the process. The
1698 corresponding @var{value} is a number. If the process was invoked by
1699 the same user who runs the current Emacs session, the value is
1700 identical to what @code{user-uid} returns (@pxref{User
1704 User name corresponding to the process's effective user ID, a string.
1707 The group ID of the effective user ID, a number.
1710 Group name corresponding to the effective user's group ID, a string.
1713 The name of the command that runs in the process. This is a string
1714 that usually specifies the name of the executable file of the process,
1715 without the leading directories. However, some special system
1716 processes can report strings that do not correspond to an executable
1720 The state code of the process. This is a short string that encodes
1721 the scheduling state of the process. Here's a list of the most
1722 frequently seen codes:
1726 uninterruptible sleep (usually I/O)
1730 interruptible sleep (waiting for some event)
1732 stopped, e.g., by a job control signal
1734 ``zombie'': a process that terminated, but was not reaped by its parent
1738 For the full list of the possible states, see the manual page of the
1739 @command{ps} command.
1742 The process ID of the parent process, a number.
1745 The process group ID of the process, a number.
1748 The session ID of the process. This is a number that is the process
1749 ID of the process's @dfn{session leader}.
1752 A string that is the name of the process's controlling terminal. On
1753 Unix and GNU systems, this is normally the file name of the
1754 corresponding terminal device, such as @file{/dev/pts65}.
1757 The numerical process group ID of the foreground process group that
1758 uses the process's terminal.
1761 The number of minor page faults caused by the process since its
1762 beginning. (Minor page faults are those that don't involve reading
1766 The number of major page faults caused by the process since its
1767 beginning. (Major page faults require a disk to be read, and are thus
1768 more expensive than minor page faults.)
1772 Like @code{minflt} and @code{majflt}, but include the number of page
1773 faults for all the child processes of the given process.
1776 Time spent by the process in the user context, for running the
1777 application's code. The corresponding @var{value} is in the
1778 @w{@code{(@var{high} @var{low} @var{microsec})}} format, the same
1779 format used by functions @code{current-time} (@pxref{Time of Day,
1780 current-time}) and @code{file-attributes} (@pxref{File Attributes}).
1783 Time spent by the process in the system (kernel) context, for
1784 processing system calls. The corresponding @var{value} is in the same
1785 format as for @code{utime}.
1788 The sum of @code{utime} and @code{stime}. The corresponding
1789 @var{value} is in the same format as for @code{utime}.
1794 Like @code{utime}, @code{stime}, and @code{time}, but include the
1795 times of all the child processes of the given process.
1798 The numerical priority of the process.
1801 The @dfn{nice value} of the process, a number. (Processes with smaller
1802 nice values get scheduled more favorably.)
1805 The number of threads in the process.
1808 The time when the process was started, in the same
1809 @w{@code{(@var{high} @var{low} @var{microsec})}} format used by
1810 @code{current-time} and @code{file-attributes}.
1813 The time elapsed since the process started, in the @w{@code{(@var{high}
1814 @var{low} @var{microsec})}} format.
1817 The virtual memory size of the process, measured in kilobytes.
1820 The size of the process's @dfn{resident set}, the number of kilobytes
1821 occupied by the process in the machine's physical memory.
1824 The percentage of the CPU time used by the process since it started.
1825 The corresponding @var{value} is a floating-point number between 0 and
1829 The percentage of the total physical memory installed on the machine
1830 used by the process's resident set. The value is a floating-point
1831 number between 0 and 100.
1834 The command-line with which the process was invoked. This is a string
1835 in which individual command-line arguments are separated by blanks;
1836 whitespace characters that are embedded in the arguments are quoted as
1837 appropriate for the system's shell: escaped by backslash characters on
1838 GNU and Unix, and enclosed in double quote characters on Windows.
1839 Thus, this command-line string can be directly used in primitives such
1840 as @code{shell-command}.
1846 @node Transaction Queues
1847 @section Transaction Queues
1848 @cindex transaction queue
1850 You can use a @dfn{transaction queue} to communicate with a subprocess
1851 using transactions. First use @code{tq-create} to create a transaction
1852 queue communicating with a specified process. Then you can call
1853 @code{tq-enqueue} to send a transaction.
1855 @defun tq-create process
1856 This function creates and returns a transaction queue communicating with
1857 @var{process}. The argument @var{process} should be a subprocess
1858 capable of sending and receiving streams of bytes. It may be a child
1859 process, or it may be a TCP connection to a server, possibly on another
1863 @defun tq-enqueue queue question regexp closure fn &optional delay-question
1864 This function sends a transaction to queue @var{queue}. Specifying the
1865 queue has the effect of specifying the subprocess to talk to.
1867 The argument @var{question} is the outgoing message that starts the
1868 transaction. The argument @var{fn} is the function to call when the
1869 corresponding answer comes back; it is called with two arguments:
1870 @var{closure}, and the answer received.
1872 The argument @var{regexp} is a regular expression that should match
1873 text at the end of the entire answer, but nothing before; that's how
1874 @code{tq-enqueue} determines where the answer ends.
1876 If the argument @var{delay-question} is non-@code{nil}, delay sending
1877 this question until the process has finished replying to any previous
1878 questions. This produces more reliable results with some processes.
1880 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 to other processes on the same machine or other machines.
1899 A network connection is handled by Lisp much like a subprocess, and is
1900 represented by a process object. However, the process you are
1901 communicating with is not a child of the Emacs process, so it has no
1902 process @acronym{ID}, and you can't kill it or send it signals. All you
1903 can do is send and receive data. @code{delete-process} closes the
1904 connection, but does not kill the program at the other end; that
1905 program must decide what to do about closure of the connection.
1907 Lisp programs can listen for connections by creating network
1908 servers. A network server is also represented by a kind of process
1909 object, but unlike a network connection, the network server never
1910 transfers data itself. When it receives a connection request, it
1911 creates a new network connection to represent the connection just
1912 made. (The network connection inherits certain information, including
1913 the process plist, from the server.) The network server then goes
1914 back to listening for more connection requests.
1916 Network connections and servers are created by calling
1917 @code{make-network-process} with an argument list consisting of
1918 keyword/argument pairs, for example @code{:server t} to create a
1919 server process, or @code{:type 'datagram} to create a datagram
1920 connection. @xref{Low-Level Network}, for details. You can also use
1921 the @code{open-network-stream} function described below.
1923 To distinguish the different types of processes, the
1924 @code{process-type} function returns the symbol @code{network} for a
1925 network connection or server, @code{serial} for a serial port
1926 connection, or @code{real} for a real subprocess.
1928 The @code{process-status} function returns @code{open},
1929 @code{closed}, @code{connect}, and @code{failed} for network
1930 connections. For a network server, the status is always
1931 @code{listen}. None of those values is possible for a real
1932 subprocess. @xref{Process Information}.
1934 You can stop and resume operation of a network process by calling
1935 @code{stop-process} and @code{continue-process}. For a server
1936 process, being stopped means not accepting new connections. (Up to 5
1937 connection requests will be queued for when you resume the server; you
1938 can increase this limit, unless it is imposed by the operating
1939 system.) For a network stream connection, being stopped means not
1940 processing input (any arriving input waits until you resume the
1941 connection). For a datagram connection, some number of packets may be
1942 queued but input may be lost. You can use the function
1943 @code{process-command} to determine whether a network connection or
1944 server is stopped; a non-@code{nil} value means yes.
1946 @cindex network connection, encrypted
1947 @cindex encrypted network connections
1948 @cindex TLS network connections
1949 @cindex STARTTLS network connections
1950 @defun open-network-stream name buffer-or-name host service &rest parameters
1951 This function opens a TCP connection, with optional encryption, and
1952 returns a process object that represents the connection.
1954 The @var{name} argument specifies the name for the process object. It
1955 is modified as necessary to make it unique.
1957 The @var{buffer-or-name} argument is the buffer to associate with the
1958 connection. Output from the connection is inserted in the buffer,
1959 unless you specify a filter function to handle the output. If
1960 @var{buffer-or-name} is @code{nil}, it means that the connection is not
1961 associated with any buffer.
1963 The arguments @var{host} and @var{service} specify where to connect to;
1964 @var{host} is the host name (a string), and @var{service} is the name of
1965 a defined network service (a string) or a port number (an integer).
1967 @c FIXME? Is this too lengthy for the printed manual?
1968 The remaining arguments @var{parameters} are keyword/argument pairs
1969 that are mainly relevant to encrypted connections:
1973 @item :nowait @var{boolean}
1974 If non-@code{nil}, try to make an asynchronous connection.
1976 @item :type @var{type}
1977 The type of connection. Options are:
1981 An ordinary, unencrypted connection.
1984 A TLS (``Transport Layer Security'') connection.
1987 Start with a plain connection, and if parameters @samp{:success}
1988 and @samp{:capability-command} are supplied, try to upgrade to an encrypted
1989 connection via STARTTLS. If that fails, retain the unencrypted connection.
1991 As for @code{nil}, but if STARTTLS fails drop the connection.
1996 @item :always-query-capabilities @var{boolean}
1997 If non-@code{nil}, always ask for the server's capabilities, even when
1998 doing a @samp{plain} connection.
2000 @item :capability-command @var{capability-command}
2001 Command string to query the host capabilities.
2003 @item :end-of-command @var{regexp}
2004 @itemx :end-of-capability @var{regexp}
2005 Regular expression matching the end of a command, or the end of the
2006 command @var{capability-command}. The latter defaults to the former.
2008 @item :starttls-function @var{function}
2009 Function of one argument (the response to @var{capability-command}),
2010 which returns either @code{nil}, or the command to activate STARTTLS
2013 @item :success @var{regexp}
2014 Regular expression matching a successful STARTTLS negotiation.
2016 @item :use-starttls-if-possible @var{boolean}
2017 If non-@code{nil}, do opportunistic STARTTLS upgrades even if Emacs
2018 doesn't have built-in TLS support.
2020 @item :client-certificate @var{list-or-t}
2021 Either a list of the form @code{(@var{key-file} @var{cert-file})},
2022 naming the certificate key file and certificate file itself, or
2023 @code{t}, meaning to query @code{auth-source} for this information
2024 (@pxref{Top,,auth-source, auth, Emacs auth-source Library}).
2025 Only used for TLS or STARTTLS.
2027 @item :return-list @var{cons-or-nil}
2028 The return value of this function. If omitted or @code{nil}, return a
2029 process object. Otherwise, a cons of the form @code{(@var{process-object}
2030 . @var{plist})}, where @var{plist} has keywords:
2033 @item :greeting @var{string-or-nil}
2034 If non-@code{nil}, the greeting string returned by the host.
2035 @item :capabilities @var{string-or-nil}
2036 If non-@code{nil}, the host's capability string.
2037 @item :type @var{symbol}
2038 The connection type: @samp{plain} or @samp{tls}.
2045 @node Network Servers
2046 @section Network Servers
2047 @cindex network servers
2049 You create a server by calling @code{make-network-process} with
2050 @code{:server t}. The server will listen for connection requests from
2051 clients. When it accepts a client connection request, that creates a
2052 new network connection, itself a process object, with the following
2057 The connection's process name is constructed by concatenating the
2058 server process's @var{name} with a client identification string. The
2059 client identification string for an IPv4 connection looks like
2060 @samp{<@var{a}.@var{b}.@var{c}.@var{d}:@var{p}>}. Otherwise, it is a
2061 unique number in brackets, as in @samp{<@var{nnn}>}. The number
2062 is unique for each connection in the Emacs session.
2065 If the server's filter is non-@code{nil}, the connection process does
2066 not get a separate process buffer; otherwise, Emacs creates a new
2067 buffer for the purpose. The buffer name is the server's buffer name
2068 or process name, concatenated with the client identification string.
2070 The server's process buffer value is never used directly, but the log
2071 function can retrieve it and use it to log connections by inserting
2075 The communication type and the process filter and sentinel are
2076 inherited from those of the server. The server never directly
2077 uses its filter and sentinel; their sole purpose is to initialize
2078 connections made to the server.
2081 The connection's process contact info is set according to the client's
2082 addressing information (typically an IP address and a port number).
2083 This information is associated with the @code{process-contact}
2084 keywords @code{:host}, @code{:service}, @code{:remote}.
2087 The connection's local address is set up according to the port
2088 number used for the connection.
2091 The client process's plist is initialized from the server's plist.
2098 A datagram connection communicates with individual packets rather
2099 than streams of data. Each call to @code{process-send} sends one
2100 datagram packet (@pxref{Input to Processes}), and each datagram
2101 received results in one call to the filter function.
2103 The datagram connection doesn't have to talk with the same remote
2104 peer all the time. It has a @dfn{remote peer address} which specifies
2105 where to send datagrams to. Each time an incoming datagram is passed
2106 to the filter function, the peer address is set to the address that
2107 datagram came from; that way, if the filter function sends a datagram,
2108 it will go back to that place. You can specify the remote peer
2109 address when you create the datagram connection using the
2110 @code{:remote} keyword. You can change it later on by calling
2111 @code{set-process-datagram-address}.
2113 @defun process-datagram-address process
2114 If @var{process} is a datagram connection or server, this function
2115 returns its remote peer address.
2118 @defun set-process-datagram-address process address
2119 If @var{process} is a datagram connection or server, this function
2120 sets its remote peer address to @var{address}.
2123 @node Low-Level Network
2124 @section Low-Level Network Access
2126 You can also create network connections by operating at a lower
2127 level than that of @code{open-network-stream}, using
2128 @code{make-network-process}.
2131 * Proc: Network Processes. Using @code{make-network-process}.
2132 * Options: Network Options. Further control over network connections.
2133 * Features: Network Feature Testing.
2134 Determining which network features work on
2135 the machine you are using.
2138 @node Network Processes
2139 @subsection @code{make-network-process}
2141 The basic function for creating network connections and network
2142 servers is @code{make-network-process}. It can do either of those
2143 jobs, depending on the arguments you give it.
2145 @defun make-network-process &rest args
2146 This function creates a network connection or server and returns the
2147 process object that represents it. The arguments @var{args} are a
2148 list of keyword/argument pairs. Omitting a keyword is always
2149 equivalent to specifying it with value @code{nil}, except for
2150 @code{:coding}, @code{:filter-multibyte}, and @code{:reuseaddr}. Here
2151 are the meaningful keywords:
2154 @item :name @var{name}
2155 Use the string @var{name} as the process name. It is modified if
2156 necessary to make it unique.
2158 @item :type @var{type}
2159 Specify the communication type. A value of @code{nil} specifies a
2160 stream connection (the default); @code{datagram} specifies a datagram
2161 connection; @code{seqpacket} specifies a ``sequenced packet stream''
2162 connection. Both connections and servers can be of these types.
2164 @item :server @var{server-flag}
2165 If @var{server-flag} is non-@code{nil}, create a server. Otherwise,
2166 create a connection. For a stream type server, @var{server-flag} may
2167 be an integer which then specifies the length of the queue of pending
2168 connections to the server. The default queue length is 5.
2170 @item :host @var{host}
2171 Specify the host to connect to. @var{host} should be a host name or
2172 Internet address, as a string, or the symbol @code{local} to specify
2173 the local host. If you specify @var{host} for a server, it must
2174 specify a valid address for the local host, and only clients
2175 connecting to that address will be accepted.
2177 @item :service @var{service}
2178 @var{service} specifies a port number to connect to, or, for a server,
2179 the port number to listen on. It should be a service name that
2180 translates to a port number, or an integer specifying the port number
2181 directly. For a server, it can also be @code{t}, which means to let
2182 the system select an unused port number.
2184 @item :family @var{family}
2185 @var{family} specifies the address (and protocol) family for
2186 communication. @code{nil} means determine the proper address family
2187 automatically for the given @var{host} and @var{service}.
2188 @code{local} specifies a Unix socket, in which case @var{host} is
2189 ignored. @code{ipv4} and @code{ipv6} specify to use IPv4 and IPv6
2192 @item :local @var{local-address}
2193 For a server process, @var{local-address} is the address to listen on.
2194 It overrides @var{family}, @var{host} and @var{service}, and you
2195 may as well not specify them.
2197 @item :remote @var{remote-address}
2198 For a connection, @var{remote-address} is the address to connect to.
2199 It overrides @var{family}, @var{host} and @var{service}, and you
2200 may as well not specify them.
2202 For a datagram server, @var{remote-address} specifies the initial
2203 setting of the remote datagram address.
2205 The format of @var{local-address} or @var{remote-address} depends on
2210 An IPv4 address is represented as a five-element vector of four 8-bit
2211 integers and one 16-bit integer
2212 @code{[@var{a} @var{b} @var{c} @var{d} @var{p}]} corresponding to
2213 numeric IPv4 address @var{a}.@var{b}.@var{c}.@var{d} and port number
2217 An IPv6 address is represented as a nine-element vector of 16-bit
2218 integers @code{[@var{a} @var{b} @var{c} @var{d} @var{e} @var{f}
2219 @var{g} @var{h} @var{p}]} corresponding to numeric IPv6 address
2220 @var{a}:@var{b}:@var{c}:@var{d}:@var{e}:@var{f}:@var{g}:@var{h} and
2221 port number @var{p}.
2224 A local address is represented as a string which specifies the address
2225 in the local address space.
2228 An ``unsupported family'' address is represented by a cons
2229 @code{(@var{f} . @var{av})}, where @var{f} is the family number and
2230 @var{av} is a vector specifying the socket address using one element
2231 per address data byte. Do not rely on this format in portable code,
2232 as it may depend on implementation defined constants, data sizes, and
2233 data structure alignment.
2236 @item :nowait @var{bool}
2237 If @var{bool} is non-@code{nil} for a stream connection, return
2238 without waiting for the connection to complete. When the connection
2239 succeeds or fails, Emacs will call the sentinel function, with a
2240 second argument matching @code{"open"} (if successful) or
2241 @code{"failed"}. The default is to block, so that
2242 @code{make-network-process} does not return until the connection
2243 has succeeded or failed.
2245 @item :stop @var{stopped}
2246 Start the network connection or server in the `stopped' state if
2247 @var{stopped} is non-@code{nil}.
2249 @item :buffer @var{buffer}
2250 Use @var{buffer} as the process buffer.
2252 @item :coding @var{coding}
2253 Use @var{coding} as the coding system for this process. To specify
2254 different coding systems for decoding data from the connection and for
2255 encoding data sent to it, specify @code{(@var{decoding} .
2256 @var{encoding})} for @var{coding}.
2258 If you don't specify this keyword at all, the default
2259 is to determine the coding systems from the data.
2261 @item :noquery @var{query-flag}
2262 Initialize the process query flag to @var{query-flag}.
2263 @xref{Query Before Exit}.
2265 @item :filter @var{filter}
2266 Initialize the process filter to @var{filter}.
2268 @item :sentinel @var{sentinel}
2269 Initialize the process sentinel to @var{sentinel}.
2271 @item :log @var{log}
2272 Initialize the log function of a server process to @var{log}. The log
2273 function is called each time the server accepts a network connection
2274 from a client. The arguments passed to the log function are
2275 @var{server}, @var{connection}, and @var{message}, where @var{server}
2276 is the server process, @var{connection} is the new process for the
2277 connection, and @var{message} is a string describing what has
2280 @item :plist @var{plist}
2281 Initialize the process plist to @var{plist}.
2284 The original argument list, modified with the actual connection
2285 information, is available via the @code{process-contact} function.
2288 @node Network Options
2289 @subsection Network Options
2291 The following network options can be specified when you create a
2292 network process. Except for @code{:reuseaddr}, you can also set or
2293 modify these options later, using @code{set-network-process-option}.
2295 For a server process, the options specified with
2296 @code{make-network-process} are not inherited by the client
2297 connections, so you will need to set the necessary options for each
2298 child connection as it is created.
2301 @item :bindtodevice @var{device-name}
2302 If @var{device-name} is a non-empty string identifying a network
2303 interface name (see @code{network-interface-list}), only handle
2304 packets received on that interface. If @var{device-name} is @code{nil}
2305 (the default), handle packets received on any interface.
2307 Using this option may require special privileges on some systems.
2309 @item :broadcast @var{broadcast-flag}
2310 If @var{broadcast-flag} is non-@code{nil} for a datagram process, the
2311 process will receive datagram packet sent to a broadcast address, and
2312 be able to send packets to a broadcast address. Ignored for a stream
2315 @item :dontroute @var{dontroute-flag}
2316 If @var{dontroute-flag} is non-@code{nil}, the process can only send
2317 to hosts on the same network as the local host.
2319 @item :keepalive @var{keepalive-flag}
2320 If @var{keepalive-flag} is non-@code{nil} for a stream connection,
2321 enable exchange of low-level keep-alive messages.
2323 @item :linger @var{linger-arg}
2324 If @var{linger-arg} is non-@code{nil}, wait for successful
2325 transmission of all queued packets on the connection before it is
2326 deleted (see @code{delete-process}). If @var{linger-arg} is an
2327 integer, it specifies the maximum time in seconds to wait for queued
2328 packets to be sent before closing the connection. Default is
2329 @code{nil} which means to discard unsent queued packets when the
2332 @item :oobinline @var{oobinline-flag}
2333 If @var{oobinline-flag} is non-@code{nil} for a stream connection,
2334 receive out-of-band data in the normal data stream. Otherwise, ignore
2337 @item :priority @var{priority}
2338 Set the priority for packets sent on this connection to the integer
2339 @var{priority}. The interpretation of this number is protocol
2340 specific, such as setting the TOS (type of service) field on IP
2341 packets sent on this connection. It may also have system dependent
2342 effects, such as selecting a specific output queue on the network
2345 @item :reuseaddr @var{reuseaddr-flag}
2346 If @var{reuseaddr-flag} is non-@code{nil} (the default) for a stream
2347 server process, allow this server to reuse a specific port number (see
2348 @code{:service}) unless another process on this host is already
2349 listening on that port. If @var{reuseaddr-flag} is @code{nil}, there
2350 may be a period of time after the last use of that port (by any
2351 process on the host), where it is not possible to make a new server on
2355 @defun set-network-process-option process option value &optional no-error
2356 This function sets or modifies a network option for network process
2357 @var{process}. See @code{make-network-process} for details of options
2358 @var{option} and their corresponding values @var{value}. If
2359 @var{no-error} is non-@code{nil}, this function returns @code{nil}
2360 instead of signaling an error if @var{option} is not a supported
2361 option. If the function successfully completes, it returns @code{t}.
2363 The current setting of an option is available via the
2364 @code{process-contact} function.
2367 @node Network Feature Testing
2368 @subsection Testing Availability of Network Features
2370 To test for the availability of a given network feature, use
2371 @code{featurep} like this:
2374 (featurep 'make-network-process '(@var{keyword} @var{value}))
2378 The result of the first form is @code{t} if it works to specify
2379 @var{keyword} with value @var{value} in @code{make-network-process}.
2380 The result of the second form is @code{t} if @var{keyword} is
2381 supported by @code{make-network-process}. Here are some of the
2382 @var{keyword}---@var{value} pairs you can test in
2387 Non-@code{nil} if non-blocking connect is supported.
2388 @item (:type datagram)
2389 Non-@code{nil} if datagrams are supported.
2390 @item (:family local)
2391 Non-@code{nil} if local (a.k.a.@: ``UNIX domain'') sockets are supported.
2392 @item (:family ipv6)
2393 Non-@code{nil} if IPv6 is supported.
2395 Non-@code{nil} if the system can select the port for a server.
2398 To test for the availability of a given network option, use
2399 @code{featurep} like this:
2402 (featurep 'make-network-process '@var{keyword})
2406 Here are some of the options you can test in this way.
2417 That particular network option is supported by
2418 @code{make-network-process} and @code{set-network-process-option}.
2422 @section Misc Network Facilities
2424 These additional functions are useful for creating and operating
2425 on network connections. Note that they are supported only on some
2428 @defun network-interface-list
2429 This function returns a list describing the network interfaces
2430 of the machine you are using. The value is an alist whose
2431 elements have the form @code{(@var{name} . @var{address})}.
2432 @var{address} has the same form as the @var{local-address}
2433 and @var{remote-address} arguments to @code{make-network-process}.
2436 @defun network-interface-info ifname
2437 This function returns information about the network interface named
2438 @var{ifname}. The value is a list of the form
2439 @code{(@var{addr} @var{bcast} @var{netmask} @var{hwaddr} @var{flags})}.
2443 The Internet protocol address.
2445 The broadcast address.
2449 The layer 2 address (Ethernet MAC address, for instance).
2451 The current flags of the interface.
2455 @defun format-network-address address &optional omit-port
2456 This function converts the Lisp representation of a network address to
2459 A five-element vector @code{[@var{a} @var{b} @var{c} @var{d} @var{p}]}
2460 represents an IPv4 address @var{a}.@var{b}.@var{c}.@var{d} and port
2461 number @var{p}. @code{format-network-address} converts that to the
2462 string @code{"@var{a}.@var{b}.@var{c}.@var{d}:@var{p}"}.
2464 A nine-element vector @code{[@var{a} @var{b} @var{c} @var{d} @var{e}
2465 @var{f} @var{g} @var{h} @var{p}]} represents an IPv6 address along
2466 with a port number. @code{format-network-address} converts that to
2468 @code{"[@var{a}:@var{b}:@var{c}:@var{d}:@var{e}:@var{f}:@var{g}:@var{h}]:@var{p}"}.
2470 If the vector does not include the port number, @var{p}, or if
2471 @var{omit-port} is non-@code{nil}, the result does not include the
2472 @code{:@var{p}} suffix.
2476 @section Communicating with Serial Ports
2477 @cindex @file{/dev/tty}
2479 @cindex serial connections
2481 Emacs can communicate with serial ports. For interactive use,
2482 @kbd{M-x serial-term} opens a terminal window. In a Lisp program,
2483 @code{make-serial-process} creates a process object.
2485 The serial port can be configured at run-time, without having to
2486 close and re-open it. The function @code{serial-process-configure}
2487 lets you change the speed, bytesize, and other parameters. In a
2488 terminal window created by @code{serial-term}, you can click on the
2489 mode line for configuration.
2491 A serial connection is represented by a process object which can be
2492 used similar to a subprocess or network process. You can send and
2493 receive data and configure the serial port. A serial process object
2494 has no process ID, you can't send signals to it, and the status codes
2495 are different from other types of processes.
2496 @code{delete-process} on the process object or @code{kill-buffer} on
2497 the process buffer close the connection, but this does not affect the
2498 device connected to the serial port.
2500 The function @code{process-type} returns the symbol @code{serial}
2501 for a process object representing a serial port connection.
2503 Serial ports are available on GNU/Linux, Unix, and Windows systems.
2505 @deffn Command serial-term port speed
2506 Start a terminal-emulator for a serial port in a new buffer.
2507 @var{port} is the name of the serial port to which to connect. For
2508 example, this could be @file{/dev/ttyS0} on Unix. On Windows, this
2509 could be @file{COM1}, or @file{\\.\COM10} (double the backslashes in
2512 @var{speed} is the speed of the serial port in bits per second. 9600
2513 is a common value. The buffer is in Term mode; see @ref{Term Mode,,,
2514 emacs, The GNU Emacs Manual}, for the commands to use in that buffer.
2515 You can change the speed and the configuration in the mode line menu.
2518 @defun make-serial-process &rest args
2519 This function creates a process and a buffer. Arguments are specified
2520 as keyword/argument pairs. Here's the list of the meaningful keywords:
2523 @item :port @var{port}@r{ (mandatory)}
2524 This is the name of the serial port. On Unix and GNU systems, this is
2525 a file name such as @file{/dev/ttyS0}. On Windows, this could be
2526 @file{COM1}, or @file{\\.\COM10} for ports higher than @file{COM9}
2527 (double the backslashes in Lisp strings).
2529 @item :speed @var{speed}@r{ (mandatory)}
2530 The speed of the serial port in bits per second. This function calls
2531 @code{serial-process-configure} to handle the speed.
2533 @item :name @var{name}
2534 The name of the process. If @var{name} is not given, @var{port} will
2535 serve as the process name as well.
2537 @item :buffer @var{buffer}
2538 The buffer to associate with the process. The value could be either a
2539 buffer or a string that names a buffer. Process output goes at the
2540 end of that buffer, unless you specify an output stream or filter
2541 function to handle the output. If @var{buffer} is not given, the
2542 process buffer's name is taken from the value of the @code{:name}
2545 @item :coding @var{coding}
2546 If @var{coding} is a symbol, it specifies the coding system used for
2547 both reading and writing for this process. If @var{coding} is a cons
2548 @code{(decoding . encoding)}, @var{decoding} is used for reading, and
2549 @var{encoding} is used for writing. If not specified, the default is
2550 to determine the coding systems from data itself.
2552 @item :noquery @var{query-flag}
2553 Initialize the process query flag to @var{query-flag}. @xref{Query
2554 Before Exit}. The flags defaults to @code{nil} if unspecified.
2556 @item :stop @var{bool}
2557 Start process in the @code{stopped} state if @var{bool} is
2558 non-@code{nil}. In the stopped state, a serial process does not
2559 accept incoming data, but you can send outgoing data. The stopped
2560 state is cleared by @code{continue-process} and set by
2561 @code{stop-process}.
2563 @item :filter @var{filter}
2564 Install @var{filter} as the process filter.
2566 @item :sentinel @var{sentinel}
2567 Install @var{sentinel} as the process sentinel.
2569 @item :plist @var{plist}
2570 Install @var{plist} as the initial plist of the process.
2577 These are handled by @code{serial-process-configure}, which is called
2578 by @code{make-serial-process}.
2581 The original argument list, possibly modified by later configuration,
2582 is available via the function @code{process-contact}.
2587 (make-serial-process :port "/dev/ttyS0" :speed 9600)
2591 @defun serial-process-configure &rest args
2592 @cindex baud, in serial connections
2593 @cindex bytesize, in serial connections
2594 @cindex parity, in serial connections
2595 @cindex stopbits, in serial connections
2596 @cindex flowcontrol, in serial connections
2598 This functions configures a serial port connection. Arguments are
2599 specified as keyword/argument pairs. Attributes that are not given
2600 are re-initialized from the process's current configuration (available
2601 via the function @code{process-contact}) or set to reasonable default
2602 values. The following arguments are defined:
2605 @item :process @var{process}
2606 @itemx :name @var{name}
2607 @itemx :buffer @var{buffer}
2608 @itemx :port @var{port}
2609 Any of these arguments can be given to identify the process that is to
2610 be configured. If none of these arguments is given, the current
2611 buffer's process is used.
2613 @item :speed @var{speed}
2614 The speed of the serial port in bits per second, a.k.a.@: @dfn{baud
2615 rate}. The value can be any number, but most serial ports work only
2616 at a few defined values between 1200 and 115200, with 9600 being the
2617 most common value. If @var{speed} is @code{nil}, the function ignores
2618 all other arguments and does not configure the port. This may be
2619 useful for special serial ports such as Bluetooth-to-serial converters
2620 which can only be configured through AT commands sent through the
2621 connection. The value of @code{nil} for @var{speed} is valid only for
2622 connections that were already opened by a previous call to
2623 @code{make-serial-process} or @code{serial-term}.
2625 @item :bytesize @var{bytesize}
2626 The number of bits per byte, which can be 7 or 8. If @var{bytesize}
2627 is not given or @code{nil}, it defaults to 8.
2629 @item :parity @var{parity}
2630 The value can be @code{nil} (don't use parity), the symbol
2631 @code{odd} (use odd parity), or the symbol @code{even} (use even
2632 parity). If @var{parity} is not given, it defaults to no parity.
2634 @item :stopbits @var{stopbits}
2635 The number of stopbits used to terminate a transmission
2636 of each byte. @var{stopbits} can be 1 or 2. If @var{stopbits} is not
2637 given or @code{nil}, it defaults to 1.
2639 @item :flowcontrol @var{flowcontrol}
2640 The type of flow control to use for this connection, which is either
2641 @code{nil} (don't use flow control), the symbol @code{hw} (use RTS/CTS
2642 hardware flow control), or the symbol @code{sw} (use XON/XOFF software
2643 flow control). If @var{flowcontrol} is not given, it defaults to no
2647 @code{serial-process-configure} is called by
2648 @code{make-serial-process} for the initial configuration of the serial
2653 @section Packing and Unpacking Byte Arrays
2654 @cindex byte packing and unpacking
2656 This section describes how to pack and unpack arrays of bytes,
2657 usually for binary network protocols. These functions convert byte arrays
2658 to alists, and vice versa. The byte array can be represented as a
2659 unibyte string or as a vector of integers, while the alist associates
2660 symbols either with fixed-size objects or with recursive sub-alists.
2663 @cindex deserializing
2666 Conversion from byte arrays to nested alists is also known as
2667 @dfn{deserializing} or @dfn{unpacking}, while going in the opposite
2668 direction is also known as @dfn{serializing} or @dfn{packing}.
2671 * Bindat Spec:: Describing data layout.
2672 * Bindat Functions:: Doing the unpacking and packing.
2673 * Bindat Examples:: Samples of what bindat.el can do for you!
2677 @subsection Describing Data Layout
2679 To control unpacking and packing, you write a @dfn{data layout
2680 specification}, a special nested list describing named and typed
2681 @dfn{fields}. This specification controls length of each field to be
2682 processed, and how to pack or unpack it. We normally keep bindat specs
2683 in variables whose names end in @samp{-bindat-spec}; that kind of name
2684 is automatically recognized as ``risky''.
2688 @cindex little endian
2689 @cindex network byte ordering
2690 A field's @dfn{type} describes the size (in bytes) of the object
2691 that the field represents and, in the case of multibyte fields, how
2692 the bytes are ordered within the field. The two possible orderings
2693 are ``big endian'' (also known as ``network byte ordering'') and
2694 ``little endian''. For instance, the number @code{#x23cd} (decimal
2695 9165) in big endian would be the two bytes @code{#x23} @code{#xcd};
2696 and in little endian, @code{#xcd} @code{#x23}. Here are the possible
2702 Unsigned byte, with length 1.
2707 Unsigned integer in network byte order, with length 2.
2710 Unsigned integer in network byte order, with length 3.
2715 Unsigned integer in network byte order, with length 4.
2716 Note: These values may be limited by Emacs's integer implementation limits.
2721 Unsigned integer in little endian order, with length 2, 3 and 4, respectively.
2724 String of length @var{len}.
2726 @item strz @var{len}
2727 Zero-terminated string, in a fixed-size field with length @var{len}.
2729 @item vec @var{len} [@var{type}]
2730 Vector of @var{len} elements of type @var{type}, or bytes if not
2731 @var{type} is specified.
2732 The @var{type} is any of the simple types above, or another vector
2733 specified as a list @code{(vec @var{len} [@var{type}])}.
2736 Four-byte vector representing an Internet address. For example:
2737 @code{[127 0 0 1]} for localhost.
2739 @item bits @var{len}
2740 List of set bits in @var{len} bytes. The bytes are taken in big
2741 endian order and the bits are numbered starting with @code{8 *
2742 @var{len} @minus{} 1} and ending with zero. For example: @code{bits
2743 2} unpacks @code{#x28} @code{#x1c} to @code{(2 3 4 11 13)} and
2744 @code{#x1c} @code{#x28} to @code{(3 5 10 11 12)}.
2746 @item (eval @var{form})
2747 @var{form} is a Lisp expression evaluated at the moment the field is
2748 unpacked or packed. The result of the evaluation should be one of the
2749 above-listed type specifications.
2752 For a fixed-size field, the length @var{len} is given as an integer
2753 specifying the number of bytes in the field.
2755 When the length of a field is not fixed, it typically depends on the
2756 value of a preceding field. In this case, the length @var{len} can be
2757 given either as a list @code{(@var{name} ...)} identifying a
2758 @dfn{field name} in the format specified for @code{bindat-get-field}
2759 below, or by an expression @code{(eval @var{form})} where @var{form}
2760 should evaluate to an integer, specifying the field length.
2762 A field specification generally has the form @code{([@var{name}]
2763 @var{handler})}. The square braces indicate that @var{name} is
2764 optional. (Don't use names that are symbols meaningful as type
2765 specifications (above) or handler specifications (below), since that
2766 would be ambiguous.) @var{name} can be a symbol or the expression
2767 @code{(eval @var{form})}, in which case @var{form} should evaluate to
2770 @var{handler} describes how to unpack or pack the field and can be one
2775 Unpack/pack this field according to the type specification @var{type}.
2777 @item eval @var{form}
2778 Evaluate @var{form}, a Lisp expression, for side-effect only. If the
2779 field name is specified, the value is bound to that field name.
2781 @item fill @var{len}
2782 Skip @var{len} bytes. In packing, this leaves them unchanged,
2783 which normally means they remain zero. In unpacking, this means
2786 @item align @var{len}
2787 Skip to the next multiple of @var{len} bytes.
2789 @item struct @var{spec-name}
2790 Process @var{spec-name} as a sub-specification. This describes a
2791 structure nested within another structure.
2793 @item union @var{form} (@var{tag} @var{spec})@dots{}
2794 @c ??? I don't see how one would actually use this.
2795 @c ??? what kind of expression would be useful for @var{form}?
2796 Evaluate @var{form}, a Lisp expression, find the first @var{tag}
2797 that matches it, and process its associated data layout specification
2798 @var{spec}. Matching can occur in one of three ways:
2802 If a @var{tag} has the form @code{(eval @var{expr})}, evaluate
2803 @var{expr} with the variable @code{tag} dynamically bound to the value
2804 of @var{form}. A non-@code{nil} result indicates a match.
2807 @var{tag} matches if it is @code{equal} to the value of @var{form}.
2810 @var{tag} matches unconditionally if it is @code{t}.
2813 @item repeat @var{count} @var{field-specs}@dots{}
2814 Process the @var{field-specs} recursively, in order, then repeat
2815 starting from the first one, processing all the specs @var{count}
2816 times overall. The @var{count} is given using the same formats as a
2817 field length---if an @code{eval} form is used, it is evaluated just once.
2818 For correct operation, each spec in @var{field-specs} must include a name.
2821 For the @code{(eval @var{form})} forms used in a bindat specification,
2822 the @var{form} can access and update these dynamically bound variables
2827 Value of the last field processed.
2830 The data as a byte array.
2833 Current index (within @code{bindat-raw}) for unpacking or packing.
2836 The alist containing the structured data that have been unpacked so
2837 far, or the entire structure being packed. You can use
2838 @code{bindat-get-field} to access specific fields of this structure.
2842 Inside a @code{repeat} block, these contain the maximum number of
2843 repetitions (as specified by the @var{count} parameter), and the
2844 current repetition number (counting from 0). Setting @code{count} to
2845 zero will terminate the inner-most repeat block after the current
2846 repetition has completed.
2849 @node Bindat Functions
2850 @subsection Functions to Unpack and Pack Bytes
2852 In the following documentation, @var{spec} refers to a data layout
2853 specification, @code{bindat-raw} to a byte array, and @var{struct} to an
2854 alist representing unpacked field data.
2856 @defun bindat-unpack spec bindat-raw &optional bindat-idx
2857 This function unpacks data from the unibyte string or byte
2858 array @code{bindat-raw}
2859 according to @var{spec}. Normally this starts unpacking at the
2860 beginning of the byte array, but if @var{bindat-idx} is non-@code{nil}, it
2861 specifies a zero-based starting position to use instead.
2863 The value is an alist or nested alist in which each element describes
2867 @defun bindat-get-field struct &rest name
2868 This function selects a field's data from the nested alist
2869 @var{struct}. Usually @var{struct} was returned by
2870 @code{bindat-unpack}. If @var{name} corresponds to just one argument,
2871 that means to extract a top-level field value. Multiple @var{name}
2872 arguments specify repeated lookup of sub-structures. An integer name
2873 acts as an array index.
2875 For example, if @var{name} is @code{(a b 2 c)}, that means to find
2876 field @code{c} in the third element of subfield @code{b} of field
2877 @code{a}. (This corresponds to @code{struct.a.b[2].c} in C.)
2880 Although packing and unpacking operations change the organization of
2881 data (in memory), they preserve the data's @dfn{total length}, which is
2882 the sum of all the fields' lengths, in bytes. This value is not
2883 generally inherent in either the specification or alist alone; instead,
2884 both pieces of information contribute to its calculation. Likewise, the
2885 length of a string or array being unpacked may be longer than the data's
2886 total length as described by the specification.
2888 @defun bindat-length spec struct
2889 This function returns the total length of the data in @var{struct},
2890 according to @var{spec}.
2893 @defun bindat-pack spec struct &optional bindat-raw bindat-idx
2894 This function returns a byte array packed according to @var{spec} from
2895 the data in the alist @var{struct}. Normally it creates and fills a
2896 new byte array starting at the beginning. However, if @var{bindat-raw}
2897 is non-@code{nil}, it specifies a pre-allocated unibyte string or vector to
2898 pack into. If @var{bindat-idx} is non-@code{nil}, it specifies the starting
2899 offset for packing into @code{bindat-raw}.
2901 When pre-allocating, you should make sure @code{(length @var{bindat-raw})}
2902 meets or exceeds the total length to avoid an out-of-range error.
2905 @defun bindat-ip-to-string ip
2906 Convert the Internet address vector @var{ip} to a string in the usual
2910 (bindat-ip-to-string [127 0 0 1])
2911 @result{} "127.0.0.1"
2915 @node Bindat Examples
2916 @subsection Examples of Byte Unpacking and Packing
2918 Here is a complete example of byte unpacking and packing:
2921 (defvar fcookie-index-spec
2929 (:offset repeat (:count)
2931 "Description of a fortune cookie index file's contents.")
2933 (defun fcookie (cookies &optional index)
2934 "Display a random fortune cookie from file COOKIES.
2935 Optional second arg INDEX specifies the associated index
2936 filename, which is by default constructed by appending
2937 \".dat\" to COOKIES. Display cookie text in possibly
2938 new buffer \"*Fortune Cookie: BASENAME*\" where BASENAME
2939 is COOKIES without the directory part."
2940 (interactive "fCookies file: ")
2941 (let* ((info (with-temp-buffer
2942 (insert-file-contents-literally
2943 (or index (concat cookies ".dat")))
2944 (bindat-unpack fcookie-index-spec
2946 (sel (random (bindat-get-field info :count)))
2947 (beg (cdar (bindat-get-field info :offset sel)))
2948 (end (or (cdar (bindat-get-field info
2950 (nth 7 (file-attributes cookies)))))
2953 (format "*Fortune Cookie: %s*"
2954 (file-name-nondirectory cookies))))
2956 (insert-file-contents-literally
2957 cookies nil beg (- end 3))))
2959 (defun fcookie-create-index (cookies &optional index delim)
2960 "Scan file COOKIES, and write out its index file.
2961 Optional second arg INDEX specifies the index filename,
2962 which is by default constructed by appending \".dat\" to
2963 COOKIES. Optional third arg DELIM specifies the unibyte
2964 character which, when found on a line of its own in
2965 COOKIES, indicates the border between entries."
2966 (interactive "fCookies file: ")
2967 (setq delim (or delim ?%))
2968 (let ((delim-line (format "\n%c\n" delim))
2971 min p q len offsets)
2972 (unless (= 3 (string-bytes delim-line))
2973 (error "Delimiter cannot be represented in one byte"))
2975 (insert-file-contents-literally cookies)
2976 (while (and (setq p (point))
2977 (search-forward delim-line (point-max) t)
2978 (setq len (- (point) 3 p)))
2979 (setq count (1+ count)
2981 min (min (or min max) len)
2982 offsets (cons (1- p) offsets))))
2984 (set-buffer-multibyte nil)
2994 (:offset . ,(mapcar (lambda (o)
2995 (list (cons :foo o)))
2996 (nreverse offsets))))))
2997 (let ((coding-system-for-write 'raw-text-unix))
2998 (write-file (or index (concat cookies ".dat")))))))
3001 Following is an example of defining and unpacking a complex structure.
3002 Consider the following C structures:
3006 unsigned long dest_ip;
3007 unsigned long src_ip;
3008 unsigned short dest_port;
3009 unsigned short src_port;
3014 unsigned char opcode;
3015 unsigned short length; /* In network byte order */
3016 unsigned char id[8]; /* null-terminated string */
3017 unsigned char data[/* (length + 3) & ~3 */];
3021 struct header header;
3022 unsigned long counters[2]; /* In little endian order */
3023 unsigned char items;
3024 unsigned char filler[3];
3025 struct data item[/* items */];
3030 The corresponding data layout specification:
3042 (length u16) ;; network byte order
3048 '((header struct header-spec)
3049 (counters vec 2 u32r) ;; little endian order
3052 (item repeat (items)
3053 (struct data-spec))))
3056 A binary data representation:
3060 [ 192 168 1 100 192 168 1 101 01 28 21 32
3061 160 134 1 0 5 1 0 0 2 0 0 0
3062 2 3 0 5 ?A ?B ?C ?D ?E ?F 0 0 1 2 3 4 5 0 0 0
3063 1 4 0 7 ?B ?C ?D ?E ?F ?G 0 0 6 7 8 9 10 11 12 0 ])
3066 The corresponding decoded structure:
3069 (setq decoded (bindat-unpack packet-spec binary-data))
3072 (dest-ip . [192 168 1 100])
3073 (src-ip . [192 168 1 101])
3076 (counters . [100000 261])
3078 (item ((data . [1 2 3 4 5])
3083 ((data . [6 7 8 9 10 11 12])
3090 Fetching data from this structure:
3093 (bindat-get-field decoded 'item 1 'id)