Commit | Line | Data |
---|---|---|
73804d4b RS |
1 | @c -*-texinfo-*- |
2 | @c This is part of the GNU Emacs Lisp Reference Manual. | |
fd897522 | 3 | @c Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1998, 1999 |
177c0ea7 | 4 | @c Free Software Foundation, Inc. |
73804d4b RS |
5 | @c See the file elisp.texi for copying conditions. |
6 | @setfilename ../info/processes | |
969fe9b5 | 7 | @node Processes, Display, Abbrevs, Top |
73804d4b RS |
8 | @chapter Processes |
9 | @cindex child process | |
10 | @cindex parent process | |
11 | @cindex subprocess | |
12 | @cindex process | |
13 | ||
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}. | |
19 | ||
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 | |
29 | send input to it. | |
30 | ||
31 | @defun processp object | |
32 | This function returns @code{t} if @var{object} is a process, | |
33 | @code{nil} otherwise. | |
34 | @end defun | |
35 | ||
36 | @menu | |
37 | * Subprocess Creation:: Functions that start subprocesses. | |
a9f0a989 | 38 | * Shell Arguments:: Quoting an argument to pass it to a shell. |
73804d4b RS |
39 | * Synchronous Processes:: Details of using synchronous subprocesses. |
40 | * Asynchronous Processes:: Starting up an asynchronous subprocess. | |
41 | * Deleting Processes:: Eliminating an asynchronous subprocess. | |
42 | * Process Information:: Accessing run-status and other attributes. | |
43 | * Input to Processes:: Sending input to an asynchronous subprocess. | |
44 | * Signals to Processes:: Stopping, continuing or interrupting | |
45 | an asynchronous subprocess. | |
46 | * Output from Processes:: Collecting output from an asynchronous subprocess. | |
47 | * Sentinels:: Sentinels run when process run-status changes. | |
edc590bb | 48 | * Query Before Exit:: Whether to query if exiting will kill a process. |
73804d4b | 49 | * Transaction Queues:: Transaction-based communication with subprocesses. |
bfe721d1 | 50 | * Network:: Opening network connections. |
4bb24054 RS |
51 | * Network Servers:: Network servers let Emacs accept net connections. |
52 | * Datagrams:: | |
53 | * Low-Level Network:: Lower-level but more general function | |
54 | to create connections and servers. | |
73804d4b RS |
55 | @end menu |
56 | ||
57 | @node Subprocess Creation | |
58 | @section Functions that Create Subprocesses | |
59 | ||
60 | There are three functions that create a new subprocess in which to run | |
61 | a program. One of them, @code{start-process}, creates an asynchronous | |
62 | process and returns a process object (@pxref{Asynchronous Processes}). | |
63 | The other two, @code{call-process} and @code{call-process-region}, | |
64 | create a synchronous process and do not return a process object | |
65 | (@pxref{Synchronous Processes}). | |
66 | ||
8241495d | 67 | Synchronous and asynchronous processes are explained in the following |
73804d4b RS |
68 | sections. Since the three functions are all called in a similar |
69 | fashion, their common arguments are described here. | |
70 | ||
71 | @cindex execute program | |
72 | @cindex @code{PATH} environment variable | |
73 | @cindex @code{HOME} environment variable | |
74 | In all cases, the function's @var{program} argument specifies the | |
75 | program to be run. An error is signaled if the file is not found or | |
76 | cannot be executed. If the file name is relative, the variable | |
77 | @code{exec-path} contains a list of directories to search. Emacs | |
78 | initializes @code{exec-path} when it starts up, based on the value of | |
79 | the environment variable @code{PATH}. The standard file name | |
80 | constructs, @samp{~}, @samp{.}, and @samp{..}, are interpreted as usual | |
81 | in @code{exec-path}, but environment variable substitutions | |
82 | (@samp{$HOME}, etc.) are not recognized; use | |
83 | @code{substitute-in-file-name} to perform them (@pxref{File Name | |
84 | Expansion}). | |
85 | ||
cd14e1c9 RS |
86 | Executing a program can also try adding suffixes to the specified |
87 | name: | |
88 | ||
89 | @defvar exec-suffixes | |
90 | This variable is a list of suffixes (strings) to try adding to the | |
91 | specified program file name. The list should include @code{""} if you | |
92 | want the name to be tried exactly as specified. The default value is | |
93 | system-dependent. | |
94 | @end defvar | |
95 | ||
73804d4b RS |
96 | Each of the subprocess-creating functions has a @var{buffer-or-name} |
97 | argument which specifies where the standard output from the program will | |
1911e6e5 RS |
98 | go. It should be a buffer or a buffer name; if it is a buffer name, |
99 | that will create the buffer if it does not already exist. It can also | |
100 | be @code{nil}, which says to discard the output unless a filter function | |
101 | handles it. (@xref{Filter Functions}, and @ref{Read and Print}.) | |
102 | Normally, you should avoid having multiple processes send output to the | |
103 | same buffer because their output would be intermixed randomly. | |
73804d4b RS |
104 | |
105 | @cindex program arguments | |
106 | All three of the subprocess-creating functions have a @code{&rest} | |
107 | argument, @var{args}. The @var{args} must all be strings, and they are | |
108 | supplied to @var{program} as separate command line arguments. Wildcard | |
f9f59935 RS |
109 | characters and other shell constructs have no special meanings in these |
110 | strings, since the whole strings are passed directly to the specified | |
111 | program. | |
73804d4b | 112 | |
b22f3a19 | 113 | @strong{Please note:} The argument @var{program} contains only the |
73804d4b RS |
114 | name of the program; it may not contain any command-line arguments. You |
115 | must use @var{args} to provide those. | |
116 | ||
117 | The subprocess gets its current directory from the value of | |
118 | @code{default-directory} (@pxref{File Name Expansion}). | |
119 | ||
120 | @cindex environment variables, subprocesses | |
1911e6e5 | 121 | The subprocess inherits its environment from Emacs, but you can |
73804d4b RS |
122 | specify overrides for it with @code{process-environment}. @xref{System |
123 | Environment}. | |
124 | ||
177c0ea7 | 125 | @defvar exec-directory |
a9042418 | 126 | @pindex movemail |
8241495d | 127 | The value of this variable is a string, the name of a directory that |
1911e6e5 | 128 | contains programs that come with GNU Emacs, programs intended for Emacs |
a9042418 RS |
129 | to invoke. The program @code{movemail} is an example of such a program; |
130 | Rmail uses it to fetch new mail from an inbox. | |
73804d4b RS |
131 | @end defvar |
132 | ||
133 | @defopt exec-path | |
134 | The value of this variable is a list of directories to search for | |
135 | programs to run in subprocesses. Each element is either the name of a | |
136 | directory (i.e., a string), or @code{nil}, which stands for the default | |
137 | directory (which is the value of @code{default-directory}). | |
138 | @cindex program directories | |
139 | ||
140 | The value of @code{exec-path} is used by @code{call-process} and | |
141 | @code{start-process} when the @var{program} argument is not an absolute | |
142 | file name. | |
143 | @end defopt | |
144 | ||
969fe9b5 RS |
145 | @node Shell Arguments |
146 | @section Shell Arguments | |
147 | ||
148 | Lisp programs sometimes need to run a shell and give it a command | |
8241495d | 149 | that contains file names that were specified by the user. These |
969fe9b5 RS |
150 | programs ought to be able to support any valid file name. But the shell |
151 | gives special treatment to certain characters, and if these characters | |
152 | occur in the file name, they will confuse the shell. To handle these | |
153 | characters, use the function @code{shell-quote-argument}: | |
154 | ||
155 | @defun shell-quote-argument argument | |
156 | This function returns a string which represents, in shell syntax, | |
157 | an argument whose actual contents are @var{argument}. It should | |
158 | work reliably to concatenate the return value into a shell command | |
159 | and then pass it to a shell for execution. | |
160 | ||
161 | Precisely what this function does depends on your operating system. The | |
8241495d RS |
162 | function is designed to work with the syntax of your system's standard |
163 | shell; if you use an unusual shell, you will need to redefine this | |
164 | function. | |
969fe9b5 RS |
165 | |
166 | @example | |
167 | ;; @r{This example shows the behavior on GNU and Unix systems.} | |
168 | (shell-quote-argument "foo > bar") | |
169 | @result{} "foo\\ \\>\\ bar" | |
8241495d RS |
170 | |
171 | ;; @r{This example shows the behavior on MS-DOS and MS-Windows systems.} | |
172 | (shell-quote-argument "foo > bar") | |
173 | @result{} "\"foo > bar\"" | |
969fe9b5 RS |
174 | @end example |
175 | ||
176 | Here's an example of using @code{shell-quote-argument} to construct | |
177 | a shell command: | |
178 | ||
179 | @example | |
180 | (concat "diff -c " | |
181 | (shell-quote-argument oldfile) | |
182 | " " | |
183 | (shell-quote-argument newfile)) | |
184 | @end example | |
185 | @end defun | |
186 | ||
73804d4b RS |
187 | @node Synchronous Processes |
188 | @section Creating a Synchronous Process | |
189 | @cindex synchronous subprocess | |
190 | ||
191 | After a @dfn{synchronous process} is created, Emacs waits for the | |
8241495d RS |
192 | process to terminate before continuing. Starting Dired on GNU or |
193 | Unix@footnote{On other systems, Emacs uses a Lisp emulation of | |
194 | @code{ls}; see @ref{Contents of Directories}.} is an example of this: it | |
195 | runs @code{ls} in a synchronous process, then modifies the output | |
196 | slightly. Because the process is synchronous, the entire directory | |
197 | listing arrives in the buffer before Emacs tries to do anything with it. | |
73804d4b RS |
198 | |
199 | While Emacs waits for the synchronous subprocess to terminate, the | |
200 | user can quit by typing @kbd{C-g}. The first @kbd{C-g} tries to kill | |
201 | the subprocess with a @code{SIGINT} signal; but it waits until the | |
202 | subprocess actually terminates before quitting. If during that time the | |
203 | user types another @kbd{C-g}, that kills the subprocess instantly with | |
8241495d RS |
204 | @code{SIGKILL} and quits immediately (except on MS-DOS, where killing |
205 | other processes doesn't work). @xref{Quitting}. | |
73804d4b | 206 | |
969fe9b5 RS |
207 | The synchronous subprocess functions return an indication of how the |
208 | process terminated. | |
f9f59935 RS |
209 | |
210 | The output from a synchronous subprocess is generally decoded using a | |
211 | coding system, much like text read from a file. The input sent to a | |
212 | subprocess by @code{call-process-region} is encoded using a coding | |
213 | system, much like text written into a file. @xref{Coding Systems}. | |
73804d4b | 214 | |
22697dac | 215 | @defun call-process program &optional infile destination display &rest args |
73804d4b RS |
216 | This function calls @var{program} in a separate process and waits for |
217 | it to finish. | |
218 | ||
219 | The standard input for the process comes from file @var{infile} if | |
8241495d | 220 | @var{infile} is not @code{nil}, and from the null device otherwise. |
22697dac KH |
221 | The argument @var{destination} says where to put the process output. |
222 | Here are the possibilities: | |
223 | ||
224 | @table @asis | |
225 | @item a buffer | |
226 | Insert the output in that buffer, before point. This includes both the | |
227 | standard output stream and the standard error stream of the process. | |
228 | ||
229 | @item a string | |
969fe9b5 | 230 | Insert the output in a buffer with that name, before point. |
22697dac KH |
231 | |
232 | @item @code{t} | |
233 | Insert the output in the current buffer, before point. | |
234 | ||
235 | @item @code{nil} | |
236 | Discard the output. | |
237 | ||
238 | @item 0 | |
8241495d | 239 | Discard the output, and return @code{nil} immediately without waiting |
22697dac KH |
240 | for the subprocess to finish. |
241 | ||
242 | In this case, the process is not truly synchronous, since it can run in | |
243 | parallel with Emacs; but you can think of it as synchronous in that | |
244 | Emacs is essentially finished with the subprocess as soon as this | |
245 | function returns. | |
246 | ||
8241495d RS |
247 | MS-DOS doesn't support asynchronous subprocesses, so this option doesn't |
248 | work there. | |
249 | ||
1911e6e5 | 250 | @item @code{(@var{real-destination} @var{error-destination})} |
22697dac KH |
251 | Keep the standard output stream separate from the standard error stream; |
252 | deal with the ordinary output as specified by @var{real-destination}, | |
253 | and dispose of the error output according to @var{error-destination}. | |
969fe9b5 RS |
254 | If @var{error-destination} is @code{nil}, that means to discard the |
255 | error output, @code{t} means mix it with the ordinary output, and a | |
256 | string specifies a file name to redirect error output into. | |
22697dac KH |
257 | |
258 | You can't directly specify a buffer to put the error output in; that is | |
259 | too difficult to implement. But you can achieve this result by sending | |
260 | the error output to a temporary file and then inserting the file into a | |
261 | buffer. | |
262 | @end table | |
73804d4b RS |
263 | |
264 | If @var{display} is non-@code{nil}, then @code{call-process} redisplays | |
a9f0a989 RS |
265 | the buffer as output is inserted. (However, if the coding system chosen |
266 | for decoding output is @code{undecided}, meaning deduce the encoding | |
267 | from the actual data, then redisplay sometimes cannot continue once | |
8241495d RS |
268 | non-@sc{ascii} characters are encountered. There are fundamental |
269 | reasons why it is hard to fix this; see @ref{Output from Processes}.) | |
270 | ||
271 | Otherwise the function @code{call-process} does no redisplay, and the | |
272 | results become visible on the screen only when Emacs redisplays that | |
273 | buffer in the normal course of events. | |
73804d4b RS |
274 | |
275 | The remaining arguments, @var{args}, are strings that specify command | |
276 | line arguments for the program. | |
277 | ||
278 | The value returned by @code{call-process} (unless you told it not to | |
279 | wait) indicates the reason for process termination. A number gives the | |
280 | exit status of the subprocess; 0 means success, and any other value | |
281 | means failure. If the process terminated with a signal, | |
282 | @code{call-process} returns a string describing the signal. | |
283 | ||
284 | In the examples below, the buffer @samp{foo} is current. | |
285 | ||
286 | @smallexample | |
287 | @group | |
288 | (call-process "pwd" nil t) | |
8241495d | 289 | @result{} 0 |
73804d4b RS |
290 | |
291 | ---------- Buffer: foo ---------- | |
292 | /usr/user/lewis/manual | |
293 | ---------- Buffer: foo ---------- | |
294 | @end group | |
295 | ||
296 | @group | |
297 | (call-process "grep" nil "bar" nil "lewis" "/etc/passwd") | |
8241495d | 298 | @result{} 0 |
73804d4b RS |
299 | |
300 | ---------- Buffer: bar ---------- | |
301 | lewis:5LTsHm66CSWKg:398:21:Bil Lewis:/user/lewis:/bin/csh | |
302 | ||
303 | ---------- Buffer: bar ---------- | |
304 | @end group | |
305 | @end smallexample | |
306 | ||
969fe9b5 RS |
307 | Here is a good example of the use of @code{call-process}, which used to |
308 | be found in the definition of @code{insert-directory}: | |
73804d4b RS |
309 | |
310 | @smallexample | |
311 | @group | |
969fe9b5 | 312 | (call-process insert-directory-program nil t nil @var{switches} |
73804d4b RS |
313 | (if full-directory-p |
314 | (concat (file-name-as-directory file) ".") | |
315 | file)) | |
316 | @end group | |
317 | @end smallexample | |
318 | @end defun | |
319 | ||
22697dac | 320 | @defun call-process-region start end program &optional delete destination display &rest args |
8241495d | 321 | This function sends the text from @var{start} to @var{end} as |
73804d4b | 322 | standard input to a process running @var{program}. It deletes the text |
f9f59935 RS |
323 | sent if @var{delete} is non-@code{nil}; this is useful when |
324 | @var{destination} is @code{t}, to insert the output in the current | |
325 | buffer in place of the input. | |
73804d4b | 326 | |
22697dac | 327 | The arguments @var{destination} and @var{display} control what to do |
73804d4b RS |
328 | with the output from the subprocess, and whether to update the display |
329 | as it comes in. For details, see the description of | |
22697dac | 330 | @code{call-process}, above. If @var{destination} is the integer 0, |
73804d4b | 331 | @code{call-process-region} discards the output and returns @code{nil} |
8241495d | 332 | immediately, without waiting for the subprocess to finish (this only |
4f939ab8 | 333 | works if asynchronous subprocesses are supported). |
73804d4b RS |
334 | |
335 | The remaining arguments, @var{args}, are strings that specify command | |
336 | line arguments for the program. | |
337 | ||
338 | The return value of @code{call-process-region} is just like that of | |
339 | @code{call-process}: @code{nil} if you told it to return without | |
340 | waiting; otherwise, a number or string which indicates how the | |
341 | subprocess terminated. | |
342 | ||
343 | In the following example, we use @code{call-process-region} to run the | |
344 | @code{cat} utility, with standard input being the first five characters | |
345 | in buffer @samp{foo} (the word @samp{input}). @code{cat} copies its | |
346 | standard input into its standard output. Since the argument | |
22697dac | 347 | @var{destination} is @code{t}, this output is inserted in the current |
73804d4b RS |
348 | buffer. |
349 | ||
350 | @smallexample | |
351 | @group | |
352 | ---------- Buffer: foo ---------- | |
353 | input@point{} | |
354 | ---------- Buffer: foo ---------- | |
355 | @end group | |
356 | ||
357 | @group | |
358 | (call-process-region 1 6 "cat" nil t) | |
8241495d | 359 | @result{} 0 |
73804d4b RS |
360 | |
361 | ---------- Buffer: foo ---------- | |
362 | inputinput@point{} | |
363 | ---------- Buffer: foo ---------- | |
364 | @end group | |
365 | @end smallexample | |
366 | ||
367 | The @code{shell-command-on-region} command uses | |
368 | @code{call-process-region} like this: | |
369 | ||
370 | @smallexample | |
371 | @group | |
177c0ea7 JB |
372 | (call-process-region |
373 | start end | |
73804d4b RS |
374 | shell-file-name ; @r{Name of program.} |
375 | nil ; @r{Do not delete region.} | |
376 | buffer ; @r{Send output to @code{buffer}.} | |
377 | nil ; @r{No redisplay during output.} | |
378 | "-c" command) ; @r{Arguments for the shell.} | |
379 | @end group | |
380 | @end smallexample | |
381 | @end defun | |
382 | ||
4bb24054 RS |
383 | @defun call-process-shell-command command &optional infile destination display &rest args |
384 | This function executes the shell command @var{command} synchronously | |
385 | in separate process. The final arguments @var{args} are additional | |
386 | arguments to add at the end of @var{command}. The other arguments | |
387 | are handled as in @code{call-process}. | |
388 | @end defun | |
389 | ||
f9f59935 RS |
390 | @defun shell-command-to-string command |
391 | This function executes @var{command} (a string) as a shell command, | |
392 | then returns the command's output as a string. | |
393 | @end defun | |
394 | ||
73804d4b RS |
395 | @node Asynchronous Processes |
396 | @section Creating an Asynchronous Process | |
397 | @cindex asynchronous subprocess | |
398 | ||
1911e6e5 RS |
399 | After an @dfn{asynchronous process} is created, Emacs and the subprocess |
400 | both continue running immediately. The process thereafter runs | |
a9f0a989 | 401 | in parallel with Emacs, and the two can communicate with each other |
8241495d | 402 | using the functions described in the following sections. However, |
a9f0a989 RS |
403 | communication is only partially asynchronous: Emacs sends data to the |
404 | process only when certain functions are called, and Emacs accepts data | |
405 | from the process only when Emacs is waiting for input or for a time | |
406 | delay. | |
407 | ||
408 | Here we describe how to create an asynchronous process. | |
73804d4b RS |
409 | |
410 | @defun start-process name buffer-or-name program &rest args | |
411 | This function creates a new asynchronous subprocess and starts the | |
412 | program @var{program} running in it. It returns a process object that | |
413 | stands for the new subprocess in Lisp. The argument @var{name} | |
414 | specifies the name for the process object; if a process with this name | |
f9f59935 RS |
415 | already exists, then @var{name} is modified (by appending @samp{<1>}, |
416 | etc.) to be unique. The buffer @var{buffer-or-name} is the buffer to | |
73804d4b RS |
417 | associate with the process. |
418 | ||
419 | The remaining arguments, @var{args}, are strings that specify command | |
420 | line arguments for the program. | |
421 | ||
422 | In the example below, the first process is started and runs (rather, | |
423 | sleeps) for 100 seconds. Meanwhile, the second process is started, and | |
424 | given the name @samp{my-process<1>} for the sake of uniqueness. It | |
425 | inserts the directory listing at the end of the buffer @samp{foo}, | |
426 | before the first process finishes. Then it finishes, and a message to | |
427 | that effect is inserted in the buffer. Much later, the first process | |
428 | finishes, and another message is inserted in the buffer for it. | |
429 | ||
430 | @smallexample | |
431 | @group | |
432 | (start-process "my-process" "foo" "sleep" "100") | |
433 | @result{} #<process my-process> | |
434 | @end group | |
435 | ||
436 | @group | |
437 | (start-process "my-process" "foo" "ls" "-l" "/user/lewis/bin") | |
438 | @result{} #<process my-process<1>> | |
439 | ||
440 | ---------- Buffer: foo ---------- | |
441 | total 2 | |
442 | lrwxrwxrwx 1 lewis 14 Jul 22 10:12 gnuemacs --> /emacs | |
443 | -rwxrwxrwx 1 lewis 19 Jul 30 21:02 lemon | |
444 | ||
445 | Process my-process<1> finished | |
446 | ||
447 | Process my-process finished | |
448 | ---------- Buffer: foo ---------- | |
449 | @end group | |
450 | @end smallexample | |
451 | @end defun | |
452 | ||
453 | @defun start-process-shell-command name buffer-or-name command &rest command-args | |
454 | This function is like @code{start-process} except that it uses a shell | |
455 | to execute the specified command. The argument @var{command} is a shell | |
456 | command name, and @var{command-args} are the arguments for the shell | |
f9f59935 RS |
457 | command. The variable @code{shell-file-name} specifies which shell to |
458 | use. | |
969fe9b5 RS |
459 | |
460 | The point of running a program through the shell, rather than directly | |
461 | with @code{start-process}, is so that you can employ shell features such | |
462 | as wildcards in the arguments. It follows that if you include an | |
8241495d | 463 | arbitrary user-specified arguments in the command, you should quote it |
969fe9b5 | 464 | with @code{shell-quote-argument} first, so that any special shell |
8241495d RS |
465 | characters do @emph{not} have their special shell meanings. @xref{Shell |
466 | Arguments}. | |
73804d4b RS |
467 | @end defun |
468 | ||
469 | @defvar process-connection-type | |
470 | @cindex pipes | |
471 | @cindex @sc{pty}s | |
472 | This variable controls the type of device used to communicate with | |
bfe721d1 KH |
473 | asynchronous subprocesses. If it is non-@code{nil}, then @sc{pty}s are |
474 | used, when available. Otherwise, pipes are used. | |
73804d4b RS |
475 | |
476 | @sc{pty}s are usually preferable for processes visible to the user, as | |
477 | in Shell mode, because they allow job control (@kbd{C-c}, @kbd{C-z}, | |
f9f59935 RS |
478 | etc.) to work between the process and its children, whereas pipes do |
479 | not. For subprocesses used for internal purposes by programs, it is | |
480 | often better to use a pipe, because they are more efficient. In | |
481 | addition, the total number of @sc{pty}s is limited on many systems and | |
482 | it is good not to waste them. | |
73804d4b | 483 | |
edc590bb | 484 | The value of @code{process-connection-type} takes effect when |
73804d4b RS |
485 | @code{start-process} is called. So you can specify how to communicate |
486 | with one subprocess by binding the variable around the call to | |
487 | @code{start-process}. | |
488 | ||
489 | @smallexample | |
490 | @group | |
491 | (let ((process-connection-type nil)) ; @r{Use a pipe.} | |
492 | (start-process @dots{})) | |
493 | @end group | |
494 | @end smallexample | |
22697dac KH |
495 | |
496 | To determine whether a given subprocess actually got a pipe or a | |
497 | @sc{pty}, use the function @code{process-tty-name} (@pxref{Process | |
498 | Information}). | |
73804d4b RS |
499 | @end defvar |
500 | ||
501 | @node Deleting Processes | |
502 | @section Deleting Processes | |
503 | @cindex deleting processes | |
504 | ||
505 | @dfn{Deleting a process} disconnects Emacs immediately from the | |
5517ea8a RS |
506 | subprocess. Processes are deleted automatically after they terminate, |
507 | but not necessarily right away. You can delete a process explicitly | |
508 | at any time. If you delete a terminated process explicitly before it | |
509 | is deleted automatically, no harm results. Deletion of a running | |
edc590bb RS |
510 | process sends a signal to terminate it (and its child processes if |
511 | any), and calls the process sentinel if it has one. | |
5517ea8a RS |
512 | |
513 | @code{get-buffer-process} and @code{process-list} do not remember a | |
514 | deleted process, but the process object itself continues to exist as | |
515 | long as other Lisp objects point to it. All the Lisp primitives that | |
516 | work on process objects accept deleted processes, but those that do | |
517 | I/O or send signals will report an error. The process mark continues | |
518 | to point to the same place as before, usually into a buffer where | |
519 | output from the process was being inserted. | |
73804d4b | 520 | |
1911e6e5 | 521 | @defopt delete-exited-processes |
73804d4b RS |
522 | This variable controls automatic deletion of processes that have |
523 | terminated (due to calling @code{exit} or to a signal). If it is | |
524 | @code{nil}, then they continue to exist until the user runs | |
525 | @code{list-processes}. Otherwise, they are deleted immediately after | |
526 | they exit. | |
1911e6e5 | 527 | @end defopt |
73804d4b RS |
528 | |
529 | @defun delete-process name | |
5517ea8a RS |
530 | This function deletes the process associated with @var{name}, killing |
531 | it with a @code{SIGKILL} signal. The argument @var{name} may be a | |
532 | process, the name of a process, a buffer, or the name of a buffer. | |
533 | Calling @code{delete-process} on a running process terminates it, | |
534 | updates the process status, and runs the sentinel (if any) immediately. | |
535 | If the process has already terminated, calling @code{delete-process} | |
536 | has no effect on its status, or on the running of its sentinel (which | |
537 | will happen sooner or later). | |
73804d4b RS |
538 | |
539 | @smallexample | |
540 | @group | |
541 | (delete-process "*shell*") | |
542 | @result{} nil | |
543 | @end group | |
544 | @end smallexample | |
545 | @end defun | |
546 | ||
73804d4b RS |
547 | @node Process Information |
548 | @section Process Information | |
549 | ||
550 | Several functions return information about processes. | |
551 | @code{list-processes} is provided for interactive use. | |
552 | ||
4bb24054 | 553 | @deffn Command list-processes &optional query-only |
73804d4b RS |
554 | This command displays a listing of all living processes. In addition, |
555 | it finally deletes any process whose status was @samp{Exited} or | |
556 | @samp{Signaled}. It returns @code{nil}. | |
4bb24054 RS |
557 | |
558 | If @var{query-only} is non-@code{nil} then it lists only processes | |
559 | whose query flag is non-@code{nil}. @xref{Query Before Exit}. | |
73804d4b RS |
560 | @end deffn |
561 | ||
562 | @defun process-list | |
563 | This function returns a list of all processes that have not been deleted. | |
564 | ||
565 | @smallexample | |
566 | @group | |
567 | (process-list) | |
568 | @result{} (#<process display-time> #<process shell>) | |
569 | @end group | |
570 | @end smallexample | |
571 | @end defun | |
572 | ||
573 | @defun get-process name | |
574 | This function returns the process named @var{name}, or @code{nil} if | |
575 | there is none. An error is signaled if @var{name} is not a string. | |
576 | ||
577 | @smallexample | |
578 | @group | |
579 | (get-process "shell") | |
580 | @result{} #<process shell> | |
581 | @end group | |
582 | @end smallexample | |
583 | @end defun | |
584 | ||
585 | @defun process-command process | |
586 | This function returns the command that was executed to start | |
587 | @var{process}. This is a list of strings, the first string being the | |
588 | program executed and the rest of the strings being the arguments that | |
589 | were given to the program. | |
590 | ||
591 | @smallexample | |
592 | @group | |
593 | (process-command (get-process "shell")) | |
594 | @result{} ("/bin/csh" "-i") | |
595 | @end group | |
596 | @end smallexample | |
597 | @end defun | |
598 | ||
599 | @defun process-id process | |
600 | This function returns the @sc{pid} of @var{process}. This is an | |
78608595 | 601 | integer that distinguishes the process @var{process} from all other |
73804d4b RS |
602 | processes running on the same computer at the current time. The |
603 | @sc{pid} of a process is chosen by the operating system kernel when the | |
604 | process is started and remains constant as long as the process exists. | |
605 | @end defun | |
606 | ||
607 | @defun process-name process | |
608 | This function returns the name of @var{process}. | |
609 | @end defun | |
610 | ||
611 | @defun process-status process-name | |
612 | This function returns the status of @var{process-name} as a symbol. | |
613 | The argument @var{process-name} must be a process, a buffer, a | |
614 | process name (string) or a buffer name (string). | |
615 | ||
616 | The possible values for an actual subprocess are: | |
617 | ||
618 | @table @code | |
619 | @item run | |
620 | for a process that is running. | |
621 | @item stop | |
622 | for a process that is stopped but continuable. | |
623 | @item exit | |
624 | for a process that has exited. | |
625 | @item signal | |
626 | for a process that has received a fatal signal. | |
627 | @item open | |
628 | for a network connection that is open. | |
629 | @item closed | |
630 | for a network connection that is closed. Once a connection | |
631 | is closed, you cannot reopen it, though you might be able to open | |
632 | a new connection to the same place. | |
4bb24054 RS |
633 | @item connect |
634 | for a non-blocking connection that is waiting to complete. | |
635 | @item failed | |
636 | for a non-blocking connection that has failed to complete. | |
637 | @item listen | |
638 | for a network server that is listening. | |
73804d4b RS |
639 | @item nil |
640 | if @var{process-name} is not the name of an existing process. | |
641 | @end table | |
642 | ||
643 | @smallexample | |
644 | @group | |
645 | (process-status "shell") | |
646 | @result{} run | |
647 | @end group | |
648 | @group | |
649 | (process-status (get-buffer "*shell*")) | |
650 | @result{} run | |
651 | @end group | |
652 | @group | |
653 | x | |
654 | @result{} #<process xx<1>> | |
655 | (process-status x) | |
656 | @result{} exit | |
657 | @end group | |
658 | @end smallexample | |
659 | ||
660 | For a network connection, @code{process-status} returns one of the symbols | |
661 | @code{open} or @code{closed}. The latter means that the other side | |
662 | closed the connection, or Emacs did @code{delete-process}. | |
73804d4b RS |
663 | @end defun |
664 | ||
665 | @defun process-exit-status process | |
666 | This function returns the exit status of @var{process} or the signal | |
667 | number that killed it. (Use the result of @code{process-status} to | |
668 | determine which of those it is.) If @var{process} has not yet | |
669 | terminated, the value is 0. | |
670 | @end defun | |
671 | ||
22697dac KH |
672 | @defun process-tty-name process |
673 | This function returns the terminal name that @var{process} is using for | |
674 | its communication with Emacs---or @code{nil} if it is using pipes | |
675 | instead of a terminal (see @code{process-connection-type} in | |
676 | @ref{Asynchronous Processes}). | |
677 | @end defun | |
678 | ||
f9f59935 RS |
679 | @defun process-coding-system process |
680 | This function returns a cons cell describing the coding systems in use | |
681 | for decoding output from @var{process} and for encoding input to | |
682 | @var{process} (@pxref{Coding Systems}). The value has this form: | |
683 | ||
684 | @example | |
969fe9b5 | 685 | (@var{coding-system-for-decoding} . @var{coding-system-for-encoding}) |
f9f59935 RS |
686 | @end example |
687 | @end defun | |
688 | ||
f9f59935 RS |
689 | @defun set-process-coding-system process decoding-system encoding-system |
690 | This function specifies the coding systems to use for subsequent output | |
691 | from and input to @var{process}. It will use @var{decoding-system} to | |
692 | decode subprocess output, and @var{encoding-system} to encode subprocess | |
693 | input. | |
4bb24054 RS |
694 | @end defun |
695 | ||
696 | Every process also has a property list that you can use to store | |
697 | miscellaneous values associated with the process. | |
698 | ||
699 | @defun process-get process propname | |
700 | This function returns the value of the @var{propname} property | |
701 | of @var{process}. | |
702 | @end defun | |
703 | ||
704 | @defun process-put process propname value | |
705 | This function sets the value of the @var{propname} property | |
706 | of @var{process} to @var{value}. | |
707 | @end defun | |
708 | ||
709 | @defun process-plist process | |
710 | This function returns the process plist of @var{process}. | |
711 | @end defun | |
712 | ||
713 | @defun set-process-plist process plist | |
714 | This function sets the process plist of @var{process} to @var{plist}. | |
f9f59935 RS |
715 | @end defun |
716 | ||
73804d4b RS |
717 | @node Input to Processes |
718 | @section Sending Input to Processes | |
719 | @cindex process input | |
720 | ||
721 | Asynchronous subprocesses receive input when it is sent to them by | |
722 | Emacs, which is done with the functions in this section. You must | |
723 | specify the process to send input to, and the input data to send. The | |
724 | data appears on the ``standard input'' of the subprocess. | |
725 | ||
726 | Some operating systems have limited space for buffered input in a | |
727 | @sc{pty}. On these systems, Emacs sends an @sc{eof} periodically amidst | |
728 | the other characters, to force them through. For most programs, | |
729 | these @sc{eof}s do no harm. | |
730 | ||
f9f59935 | 731 | Subprocess input is normally encoded using a coding system before the |
a9f0a989 RS |
732 | subprocess receives it, much like text written into a file. You can use |
733 | @code{set-process-coding-system} to specify which coding system to use | |
734 | (@pxref{Process Information}). Otherwise, the coding system comes from | |
735 | @code{coding-system-for-write}, if that is non-@code{nil}; or else from | |
736 | the defaulting mechanism (@pxref{Default Coding Systems}). | |
f9f59935 | 737 | |
b6954afd RS |
738 | Sometimes the system is unable to accept input for that process, |
739 | because the input buffer is full. When this happens, the send functions | |
740 | wait a short while, accepting output from subprocesses, and then try | |
741 | again. This gives the subprocess a chance to read more of its pending | |
742 | input and make space in the buffer. It also allows filters, sentinels | |
743 | and timers to run---so take account of that in writing your code. | |
744 | ||
73804d4b RS |
745 | @defun process-send-string process-name string |
746 | This function sends @var{process-name} the contents of @var{string} as | |
747 | standard input. The argument @var{process-name} must be a process or | |
748 | the name of a process. If it is @code{nil}, the current buffer's | |
749 | process is used. | |
750 | ||
751 | The function returns @code{nil}. | |
752 | ||
753 | @smallexample | |
754 | @group | |
755 | (process-send-string "shell<1>" "ls\n") | |
756 | @result{} nil | |
757 | @end group | |
758 | ||
759 | ||
760 | @group | |
761 | ---------- Buffer: *shell* ---------- | |
762 | ... | |
763 | introduction.texi syntax-tables.texi~ | |
764 | introduction.texi~ text.texi | |
765 | introduction.txt text.texi~ | |
766 | ... | |
767 | ---------- Buffer: *shell* ---------- | |
768 | @end group | |
769 | @end smallexample | |
770 | @end defun | |
771 | ||
8241495d | 772 | @defun process-send-region process-name start end |
73804d4b RS |
773 | This function sends the text in the region defined by @var{start} and |
774 | @var{end} as standard input to @var{process-name}, which is a process or | |
775 | a process name. (If it is @code{nil}, the current buffer's process is | |
776 | used.) | |
777 | ||
778 | An error is signaled unless both @var{start} and @var{end} are | |
779 | integers or markers that indicate positions in the current buffer. (It | |
780 | is unimportant which number is larger.) | |
8241495d | 781 | @end defun |
73804d4b RS |
782 | |
783 | @defun process-send-eof &optional process-name | |
784 | This function makes @var{process-name} see an end-of-file in its | |
785 | input. The @sc{eof} comes after any text already sent to it. | |
786 | ||
787 | If @var{process-name} is not supplied, or if it is @code{nil}, then | |
788 | this function sends the @sc{eof} to the current buffer's process. An | |
789 | error is signaled if the current buffer has no process. | |
790 | ||
791 | The function returns @var{process-name}. | |
792 | ||
793 | @smallexample | |
794 | @group | |
795 | (process-send-eof "shell") | |
796 | @result{} "shell" | |
797 | @end group | |
798 | @end smallexample | |
799 | @end defun | |
800 | ||
b6954afd RS |
801 | @defun process-running-child-p process |
802 | @tindex process-running-child-p process | |
803 | This function will tell you whether a subprocess has given control of | |
804 | its terminal to its own child process. The value is @code{t} if this is | |
805 | true, or if Emacs cannot tell; it is @code{nil} if Emacs can be certain | |
806 | that this is not so. | |
807 | @end defun | |
808 | ||
73804d4b RS |
809 | @node Signals to Processes |
810 | @section Sending Signals to Processes | |
811 | @cindex process signals | |
812 | @cindex sending signals | |
813 | @cindex signals | |
814 | ||
815 | @dfn{Sending a signal} to a subprocess is a way of interrupting its | |
816 | activities. There are several different signals, each with its own | |
817 | meaning. The set of signals and their names is defined by the operating | |
818 | system. For example, the signal @code{SIGINT} means that the user has | |
819 | typed @kbd{C-c}, or that some analogous thing has happened. | |
820 | ||
821 | Each signal has a standard effect on the subprocess. Most signals | |
822 | kill the subprocess, but some stop or resume execution instead. Most | |
823 | signals can optionally be handled by programs; if the program handles | |
824 | the signal, then we can say nothing in general about its effects. | |
825 | ||
826 | You can send signals explicitly by calling the functions in this | |
827 | section. Emacs also sends signals automatically at certain times: | |
828 | killing a buffer sends a @code{SIGHUP} signal to all its associated | |
829 | processes; killing Emacs sends a @code{SIGHUP} signal to all remaining | |
830 | processes. (@code{SIGHUP} is a signal that usually indicates that the | |
831 | user hung up the phone.) | |
832 | ||
833 | Each of the signal-sending functions takes two optional arguments: | |
834 | @var{process-name} and @var{current-group}. | |
835 | ||
836 | The argument @var{process-name} must be either a process, the name of | |
837 | one, or @code{nil}. If it is @code{nil}, the process defaults to the | |
838 | process associated with the current buffer. An error is signaled if | |
839 | @var{process-name} does not identify a process. | |
840 | ||
841 | The argument @var{current-group} is a flag that makes a difference | |
842 | when you are running a job-control shell as an Emacs subprocess. If it | |
843 | is non-@code{nil}, then the signal is sent to the current process-group | |
78608595 | 844 | of the terminal that Emacs uses to communicate with the subprocess. If |
73804d4b RS |
845 | the process is a job-control shell, this means the shell's current |
846 | subjob. If it is @code{nil}, the signal is sent to the process group of | |
847 | the immediate subprocess of Emacs. If the subprocess is a job-control | |
848 | shell, this is the shell itself. | |
849 | ||
850 | The flag @var{current-group} has no effect when a pipe is used to | |
851 | communicate with the subprocess, because the operating system does not | |
852 | support the distinction in the case of pipes. For the same reason, | |
853 | job-control shells won't work when a pipe is used. See | |
854 | @code{process-connection-type} in @ref{Asynchronous Processes}. | |
855 | ||
856 | @defun interrupt-process &optional process-name current-group | |
857 | This function interrupts the process @var{process-name} by sending the | |
858 | signal @code{SIGINT}. Outside of Emacs, typing the ``interrupt | |
859 | character'' (normally @kbd{C-c} on some systems, and @code{DEL} on | |
860 | others) sends this signal. When the argument @var{current-group} is | |
861 | non-@code{nil}, you can think of this function as ``typing @kbd{C-c}'' | |
862 | on the terminal by which Emacs talks to the subprocess. | |
863 | @end defun | |
864 | ||
865 | @defun kill-process &optional process-name current-group | |
866 | This function kills the process @var{process-name} by sending the | |
867 | signal @code{SIGKILL}. This signal kills the subprocess immediately, | |
868 | and cannot be handled by the subprocess. | |
869 | @end defun | |
870 | ||
871 | @defun quit-process &optional process-name current-group | |
872 | This function sends the signal @code{SIGQUIT} to the process | |
873 | @var{process-name}. This signal is the one sent by the ``quit | |
874 | character'' (usually @kbd{C-b} or @kbd{C-\}) when you are not inside | |
875 | Emacs. | |
876 | @end defun | |
877 | ||
878 | @defun stop-process &optional process-name current-group | |
879 | This function stops the process @var{process-name} by sending the | |
880 | signal @code{SIGTSTP}. Use @code{continue-process} to resume its | |
881 | execution. | |
882 | ||
969fe9b5 | 883 | Outside of Emacs, on systems with job control, the ``stop character'' |
f9f59935 RS |
884 | (usually @kbd{C-z}) normally sends this signal. When |
885 | @var{current-group} is non-@code{nil}, you can think of this function as | |
886 | ``typing @kbd{C-z}'' on the terminal Emacs uses to communicate with the | |
887 | subprocess. | |
73804d4b RS |
888 | @end defun |
889 | ||
890 | @defun continue-process &optional process-name current-group | |
891 | This function resumes execution of the process @var{process} by sending | |
892 | it the signal @code{SIGCONT}. This presumes that @var{process-name} was | |
893 | stopped previously. | |
894 | @end defun | |
895 | ||
896 | @c Emacs 19 feature | |
4bb24054 RS |
897 | @defun signal-process process signal |
898 | This function sends a signal to process @var{process}. The argument | |
899 | @var{signal} specifies which signal to send; it should be an integer. | |
900 | ||
901 | You can specify the target process by its process @sc{id}; that allows | |
902 | you to send signals to processes that are not children of Emacs. | |
73804d4b RS |
903 | @end defun |
904 | ||
905 | @node Output from Processes | |
906 | @section Receiving Output from Processes | |
907 | @cindex process output | |
908 | @cindex output from processes | |
909 | ||
910 | There are two ways to receive the output that a subprocess writes to | |
911 | its standard output stream. The output can be inserted in a buffer, | |
912 | which is called the associated buffer of the process, or a function | |
78608595 RS |
913 | called the @dfn{filter function} can be called to act on the output. If |
914 | the process has no buffer and no filter function, its output is | |
915 | discarded. | |
177c0ea7 | 916 | |
edc590bb RS |
917 | When a subprocess terminates, Emacs reads any pending output, |
918 | then stops reading output from that subprocess. Therefore, if the | |
919 | subprocess has children that are still live and still producing | |
920 | output, Emacs won't receive that output. | |
921 | ||
a9f0a989 RS |
922 | Output from a subprocess can arrive only while Emacs is waiting: when |
923 | reading terminal input, in @code{sit-for} and @code{sleep-for} | |
924 | (@pxref{Waiting}), and in @code{accept-process-output} (@pxref{Accepting | |
925 | Output}). This minimizes the problem of timing errors that usually | |
926 | plague parallel programming. For example, you can safely create a | |
927 | process and only then specify its buffer or filter function; no output | |
928 | can arrive before you finish, if the code in between does not call any | |
929 | primitive that waits. | |
930 | ||
8f3efb4e RS |
931 | It is impossible to separate the standard output and standard error |
932 | streams of the subprocess, because Emacs normally spawns the subprocess | |
933 | inside a pseudo-TTY, and a pseudo-TTY has only one output channel. If | |
934 | you want to keep the output to those streams separate, you should | |
935 | redirect one of them to a file--for example, by using an appropriate | |
936 | shell command. | |
937 | ||
73804d4b | 938 | @menu |
582ef186 RS |
939 | * Process Buffers:: If no filter, output is put in a buffer. |
940 | * Filter Functions:: Filter functions accept output from the process. | |
591a3500 | 941 | * Decoding Output:: Filters can get unibyte or multibyte strings. |
582ef186 | 942 | * Accepting Output:: How to wait until process output arrives. |
73804d4b RS |
943 | @end menu |
944 | ||
945 | @node Process Buffers | |
946 | @subsection Process Buffers | |
947 | ||
948 | A process can (and usually does) have an @dfn{associated buffer}, | |
949 | which is an ordinary Emacs buffer that is used for two purposes: storing | |
950 | the output from the process, and deciding when to kill the process. You | |
951 | can also use the buffer to identify a process to operate on, since in | |
952 | normal practice only one process is associated with any given buffer. | |
953 | Many applications of processes also use the buffer for editing input to | |
954 | be sent to the process, but this is not built into Emacs Lisp. | |
955 | ||
956 | Unless the process has a filter function (@pxref{Filter Functions}), | |
957 | its output is inserted in the associated buffer. The position to insert | |
78608595 RS |
958 | the output is determined by the @code{process-mark}, which is then |
959 | updated to point to the end of the text just inserted. Usually, but not | |
960 | always, the @code{process-mark} is at the end of the buffer. | |
73804d4b RS |
961 | |
962 | @defun process-buffer process | |
963 | This function returns the associated buffer of the process | |
964 | @var{process}. | |
965 | ||
966 | @smallexample | |
967 | @group | |
968 | (process-buffer (get-process "shell")) | |
969 | @result{} #<buffer *shell*> | |
970 | @end group | |
971 | @end smallexample | |
972 | @end defun | |
973 | ||
974 | @defun process-mark process | |
975 | This function returns the process marker for @var{process}, which is the | |
976 | marker that says where to insert output from the process. | |
977 | ||
978 | If @var{process} does not have a buffer, @code{process-mark} returns a | |
979 | marker that points nowhere. | |
980 | ||
981 | Insertion of process output in a buffer uses this marker to decide where | |
982 | to insert, and updates it to point after the inserted text. That is why | |
983 | successive batches of output are inserted consecutively. | |
984 | ||
985 | Filter functions normally should use this marker in the same fashion | |
986 | as is done by direct insertion of output in the buffer. A good | |
987 | example of a filter function that uses @code{process-mark} is found at | |
988 | the end of the following section. | |
989 | ||
990 | When the user is expected to enter input in the process buffer for | |
f9f59935 RS |
991 | transmission to the process, the process marker separates the new input |
992 | from previous output. | |
73804d4b RS |
993 | @end defun |
994 | ||
995 | @defun set-process-buffer process buffer | |
996 | This function sets the buffer associated with @var{process} to | |
997 | @var{buffer}. If @var{buffer} is @code{nil}, the process becomes | |
998 | associated with no buffer. | |
999 | @end defun | |
1000 | ||
1001 | @defun get-buffer-process buffer-or-name | |
5517ea8a RS |
1002 | This function returns a nondeleted process associated with the buffer |
1003 | specified by @var{buffer-or-name}. If there are several processes | |
1004 | associated with it, this function chooses one (currently, the one most | |
1005 | recently created, but don't count on that). Deletion of a process | |
1006 | (see @code{delete-process}) makes it ineligible for this function to | |
1007 | return. | |
1008 | ||
1009 | It is usually a bad idea to have more than one process associated with | |
1010 | the same buffer. | |
73804d4b RS |
1011 | |
1012 | @smallexample | |
1013 | @group | |
1014 | (get-buffer-process "*shell*") | |
1015 | @result{} #<process shell> | |
1016 | @end group | |
1017 | @end smallexample | |
1018 | ||
1019 | Killing the process's buffer deletes the process, which kills the | |
1020 | subprocess with a @code{SIGHUP} signal (@pxref{Signals to Processes}). | |
1021 | @end defun | |
1022 | ||
1023 | @node Filter Functions | |
1024 | @subsection Process Filter Functions | |
1025 | @cindex filter function | |
1026 | @cindex process filter | |
1027 | ||
1028 | A process @dfn{filter function} is a function that receives the | |
1029 | standard output from the associated process. If a process has a filter, | |
78608595 RS |
1030 | then @emph{all} output from that process is passed to the filter. The |
1031 | process buffer is used directly for output from the process only when | |
1032 | there is no filter. | |
73804d4b | 1033 | |
a9f0a989 RS |
1034 | The filter function can only be called when Emacs is waiting for |
1035 | something, because process output arrives only at such times. Emacs | |
1036 | waits when reading terminal input, in @code{sit-for} and | |
1037 | @code{sleep-for} (@pxref{Waiting}), and in @code{accept-process-output} | |
1038 | (@pxref{Accepting Output}). | |
1039 | ||
f9f59935 RS |
1040 | A filter function must accept two arguments: the associated process |
1041 | and a string, which is output just received from it. The function is | |
1042 | then free to do whatever it chooses with the output. | |
73804d4b | 1043 | |
73804d4b RS |
1044 | Quitting is normally inhibited within a filter function---otherwise, |
1045 | the effect of typing @kbd{C-g} at command level or to quit a user | |
1046 | command would be unpredictable. If you want to permit quitting inside a | |
1047 | filter function, bind @code{inhibit-quit} to @code{nil}. | |
1048 | @xref{Quitting}. | |
1049 | ||
22697dac KH |
1050 | If an error happens during execution of a filter function, it is |
1051 | caught automatically, so that it doesn't stop the execution of whatever | |
eaac2be1 | 1052 | program was running when the filter function was started. However, if |
22697dac KH |
1053 | @code{debug-on-error} is non-@code{nil}, the error-catching is turned |
1054 | off. This makes it possible to use the Lisp debugger to debug the | |
1055 | filter function. @xref{Debugger}. | |
1056 | ||
73804d4b RS |
1057 | Many filter functions sometimes or always insert the text in the |
1058 | process's buffer, mimicking the actions of Emacs when there is no | |
1059 | filter. Such filter functions need to use @code{set-buffer} in order to | |
1060 | be sure to insert in that buffer. To avoid setting the current buffer | |
f9f59935 RS |
1061 | semipermanently, these filter functions must save and restore the |
1062 | current buffer. They should also update the process marker, and in some | |
1063 | cases update the value of point. Here is how to do these things: | |
73804d4b RS |
1064 | |
1065 | @smallexample | |
1066 | @group | |
1067 | (defun ordinary-insertion-filter (proc string) | |
f9f59935 RS |
1068 | (with-current-buffer (process-buffer proc) |
1069 | (let ((moving (= (point) (process-mark proc)))) | |
73804d4b RS |
1070 | @end group |
1071 | @group | |
f9f59935 RS |
1072 | (save-excursion |
1073 | ;; @r{Insert the text, advancing the process marker.} | |
1074 | (goto-char (process-mark proc)) | |
1075 | (insert string) | |
1076 | (set-marker (process-mark proc) (point))) | |
1077 | (if moving (goto-char (process-mark proc)))))) | |
73804d4b RS |
1078 | @end group |
1079 | @end smallexample | |
1080 | ||
1081 | @noindent | |
f9f59935 RS |
1082 | The reason to use @code{with-current-buffer}, rather than using |
1083 | @code{save-excursion} to save and restore the current buffer, is so as | |
1084 | to preserve the change in point made by the second call to | |
1085 | @code{goto-char}. | |
73804d4b RS |
1086 | |
1087 | To make the filter force the process buffer to be visible whenever new | |
1088 | text arrives, insert the following line just before the | |
f9f59935 | 1089 | @code{with-current-buffer} construct: |
73804d4b RS |
1090 | |
1091 | @smallexample | |
1092 | (display-buffer (process-buffer proc)) | |
1093 | @end smallexample | |
1094 | ||
f9f59935 RS |
1095 | To force point to the end of the new output, no matter where it was |
1096 | previously, eliminate the variable @code{moving} and call | |
73804d4b RS |
1097 | @code{goto-char} unconditionally. |
1098 | ||
f9f59935 RS |
1099 | In earlier Emacs versions, every filter function that did regular |
1100 | expression searching or matching had to explicitly save and restore the | |
1101 | match data. Now Emacs does this automatically for filter functions; | |
1102 | they never need to do it explicitly. @xref{Match Data}. | |
73804d4b RS |
1103 | |
1104 | A filter function that writes the output into the buffer of the | |
bfe721d1 | 1105 | process should check whether the buffer is still alive. If it tries to |
f9f59935 RS |
1106 | insert into a dead buffer, it will get an error. The expression |
1107 | @code{(buffer-name (process-buffer @var{process}))} returns @code{nil} | |
1108 | if the buffer is dead. | |
73804d4b RS |
1109 | |
1110 | The output to the function may come in chunks of any size. A program | |
f9f59935 RS |
1111 | that produces the same output twice in a row may send it as one batch of |
1112 | 200 characters one time, and five batches of 40 characters the next. If | |
1113 | the filter looks for certain text strings in the subprocess output, make | |
1114 | sure to handle the case where one of these strings is split across two | |
1115 | or more batches of output. | |
73804d4b RS |
1116 | |
1117 | @defun set-process-filter process filter | |
1118 | This function gives @var{process} the filter function @var{filter}. If | |
1119 | @var{filter} is @code{nil}, it gives the process no filter. | |
1120 | @end defun | |
1121 | ||
1122 | @defun process-filter process | |
1123 | This function returns the filter function of @var{process}, or @code{nil} | |
1124 | if it has none. | |
1125 | @end defun | |
1126 | ||
1127 | Here is an example of use of a filter function: | |
1128 | ||
1129 | @smallexample | |
1130 | @group | |
1131 | (defun keep-output (process output) | |
1132 | (setq kept (cons output kept))) | |
1133 | @result{} keep-output | |
1134 | @end group | |
1135 | @group | |
1136 | (setq kept nil) | |
1137 | @result{} nil | |
1138 | @end group | |
1139 | @group | |
1140 | (set-process-filter (get-process "shell") 'keep-output) | |
1141 | @result{} keep-output | |
1142 | @end group | |
1143 | @group | |
1144 | (process-send-string "shell" "ls ~/other\n") | |
1145 | @result{} nil | |
1146 | kept | |
1147 | @result{} ("lewis@@slug[8] % " | |
1148 | @end group | |
1149 | @group | |
1150 | "FINAL-W87-SHORT.MSS backup.otl kolstad.mss~ | |
1151 | address.txt backup.psf kolstad.psf | |
1152 | backup.bib~ david.mss resume-Dec-86.mss~ | |
1153 | backup.err david.psf resume-Dec.psf | |
1154 | backup.mss dland syllabus.mss | |
1155 | " | |
1156 | "#backups.mss# backup.mss~ kolstad.mss | |
1157 | ") | |
1158 | @end group | |
1159 | @end smallexample | |
1160 | ||
1161 | @ignore @c The code in this example doesn't show the right way to do things. | |
1162 | Here is another, more realistic example, which demonstrates how to use | |
1163 | the process mark to do insertion in the same fashion as is done when | |
1164 | there is no filter function: | |
1165 | ||
1166 | @smallexample | |
1167 | @group | |
1168 | ;; @r{Insert input in the buffer specified by @code{my-shell-buffer}} | |
1169 | ;; @r{and make sure that buffer is shown in some window.} | |
1170 | (defun my-process-filter (proc str) | |
f9f59935 RS |
1171 | (let ((cur (selected-window)) |
1172 | (pop-up-windows t)) | |
1173 | (pop-to-buffer my-shell-buffer) | |
73804d4b RS |
1174 | @end group |
1175 | @group | |
f9f59935 RS |
1176 | (goto-char (point-max)) |
1177 | (insert str) | |
1178 | (set-marker (process-mark proc) (point-max)) | |
1179 | (select-window cur))) | |
73804d4b RS |
1180 | @end group |
1181 | @end smallexample | |
1182 | @end ignore | |
1183 | ||
582ef186 RS |
1184 | @node Decoding Output |
1185 | @subsection Decoding Process Output | |
1186 | ||
1187 | When Emacs writes process output directly into a multibyte buffer, | |
1188 | it decodes the output according to the process output coding system. | |
1189 | If the coding system is @code{raw-text} or @code{no-conversion}, Emacs | |
1190 | converts the unibyte output to multibyte using | |
1191 | @code{string-to-multibyte}, inserts the resulting multibyte text. | |
1192 | ||
1193 | You can use @code{set-process-coding-system} to specify which coding | |
1194 | system to use (@pxref{Process Information}). Otherwise, the coding | |
1195 | system comes from @code{coding-system-for-read}, if that is | |
1196 | non-@code{nil}; or else from the defaulting mechanism (@pxref{Default | |
1197 | Coding Systems}). | |
1198 | ||
1199 | @strong{Warning:} Coding systems such as @code{undecided} which | |
1200 | determine the coding system from the data do not work entirely | |
1201 | reliably with asynchronous subprocess output. This is because Emacs | |
1202 | has to process asynchronous subprocess output in batches, as it | |
1203 | arrives. Emacs must try to detect the proper coding system from one | |
1204 | batch at a time, and this does not always work. Therefore, if at all | |
1205 | possible, specify a coding system that determines both the character | |
1206 | code conversion and the end of line conversion---that is, one like | |
1207 | @code{latin-1-unix}, rather than @code{undecided} or @code{latin-1}. | |
1208 | ||
1209 | @cindex filter multibyte flag, of process | |
1210 | @cindex process filter multibyte flag | |
1211 | When Emacs calls a process filter function, it provides the process | |
1212 | output as a multibyte string or as a unibyte string according to the | |
1213 | process's filter multibyte flag. If the flag is non-@code{nil}, Emacs | |
1214 | decodes the output according to the process output coding system to | |
1215 | produce a multibyte string, and passes that to the process. If the | |
1216 | flag is @code{nil}, Emacs puts the output into a unibyte string, with | |
1217 | no decoding, and passes that. | |
1218 | ||
1219 | When you create a process, the filter multibyte flag takes its | |
1220 | initial value from @code{default-enable-multibyte-characters}. If you | |
1221 | want to change the flag later on, use | |
1222 | @code{set-process-filter-multibyte}. | |
1223 | ||
1224 | @defun set-process-filter-multibyte process multibyte | |
1225 | This function sets the filter multibyte flag of @var{process} | |
1226 | to @var{multibyte}. | |
1227 | @end defun | |
1228 | ||
1229 | @defun process-filter-multibyte-p process | |
1230 | This function returns the filter multibyte flag of @var{process}. | |
1231 | @end defun | |
1232 | ||
73804d4b RS |
1233 | @node Accepting Output |
1234 | @subsection Accepting Output from Processes | |
1235 | ||
1236 | Output from asynchronous subprocesses normally arrives only while | |
1237 | Emacs is waiting for some sort of external event, such as elapsed time | |
1238 | or terminal input. Occasionally it is useful in a Lisp program to | |
1239 | explicitly permit output to arrive at a specific point, or even to wait | |
1240 | until output arrives from a process. | |
1241 | ||
1242 | @defun accept-process-output &optional process seconds millisec | |
1243 | This function allows Emacs to read pending output from processes. The | |
1244 | output is inserted in the associated buffers or given to their filter | |
1245 | functions. If @var{process} is non-@code{nil} then this function does | |
1246 | not return until some output has been received from @var{process}. | |
1247 | ||
1248 | @c Emacs 19 feature | |
1249 | The arguments @var{seconds} and @var{millisec} let you specify timeout | |
1250 | periods. The former specifies a period measured in seconds and the | |
1251 | latter specifies one measured in milliseconds. The two time periods | |
1252 | thus specified are added together, and @code{accept-process-output} | |
1253 | returns after that much time whether or not there has been any | |
1254 | subprocess output. | |
1255 | ||
bfe721d1 KH |
1256 | The argument @var{seconds} need not be an integer. If it is a floating |
1257 | point number, this function waits for a fractional number of seconds. | |
1258 | Some systems support only a whole number of seconds; on these systems, | |
1911e6e5 | 1259 | @var{seconds} is rounded down. |
bfe721d1 | 1260 | |
73804d4b RS |
1261 | Not all operating systems support waiting periods other than multiples |
1262 | of a second; on those that do not, you get an error if you specify | |
1263 | nonzero @var{millisec}. | |
1264 | ||
1265 | The function @code{accept-process-output} returns non-@code{nil} if it | |
1266 | did get some output, or @code{nil} if the timeout expired before output | |
1267 | arrived. | |
1268 | @end defun | |
1269 | ||
1270 | @node Sentinels | |
1271 | @section Sentinels: Detecting Process Status Changes | |
1272 | @cindex process sentinel | |
1273 | @cindex sentinel | |
1274 | ||
1275 | A @dfn{process sentinel} is a function that is called whenever the | |
1276 | associated process changes status for any reason, including signals | |
1277 | (whether sent by Emacs or caused by the process's own actions) that | |
5517ea8a RS |
1278 | terminate, stop, or continue the process. The process sentinel is |
1279 | also called if the process exits. The sentinel receives two | |
1280 | arguments: the process for which the event occurred, and a string | |
1281 | describing the type of event. | |
73804d4b RS |
1282 | |
1283 | The string describing the event looks like one of the following: | |
1284 | ||
1285 | @itemize @bullet | |
177c0ea7 | 1286 | @item |
73804d4b RS |
1287 | @code{"finished\n"}. |
1288 | ||
1289 | @item | |
1290 | @code{"exited abnormally with code @var{exitcode}\n"}. | |
1291 | ||
1292 | @item | |
1293 | @code{"@var{name-of-signal}\n"}. | |
1294 | ||
1295 | @item | |
1296 | @code{"@var{name-of-signal} (core dumped)\n"}. | |
1297 | @end itemize | |
1298 | ||
5517ea8a RS |
1299 | A sentinel runs only while Emacs is waiting (e.g., for terminal |
1300 | input, or for time to elapse, or for process output). This avoids the | |
1301 | timing errors that could result from running them at random places in | |
1302 | the middle of other Lisp programs. A program can wait, so that | |
1303 | sentinels will run, by calling @code{sit-for} or @code{sleep-for} | |
78608595 | 1304 | (@pxref{Waiting}), or @code{accept-process-output} (@pxref{Accepting |
a9f0a989 | 1305 | Output}). Emacs also allows sentinels to run when the command loop is |
5517ea8a RS |
1306 | reading input. @code{delete-process} calls the sentinel when it |
1307 | terminates a running process. | |
1308 | ||
1309 | Emacs does not keep a queue of multiple reasons to call the sentinel | |
1310 | of one process; it records just the current status and the fact that | |
1311 | there has been a change. Therefore two changes in status, coming in | |
1312 | quick succession, can call the sentinel just once. However, process | |
1313 | termination will always run the sentinel exactly once. This is | |
1314 | because the process status can't change again after termination. | |
73804d4b RS |
1315 | |
1316 | Quitting is normally inhibited within a sentinel---otherwise, the | |
1317 | effect of typing @kbd{C-g} at command level or to quit a user command | |
1318 | would be unpredictable. If you want to permit quitting inside a | |
1319 | sentinel, bind @code{inhibit-quit} to @code{nil}. @xref{Quitting}. | |
1320 | ||
1321 | A sentinel that writes the output into the buffer of the process | |
bfe721d1 | 1322 | should check whether the buffer is still alive. If it tries to insert |
73804d4b RS |
1323 | into a dead buffer, it will get an error. If the buffer is dead, |
1324 | @code{(buffer-name (process-buffer @var{process}))} returns @code{nil}. | |
1325 | ||
22697dac KH |
1326 | If an error happens during execution of a sentinel, it is caught |
1327 | automatically, so that it doesn't stop the execution of whatever | |
1328 | programs was running when the sentinel was started. However, if | |
1329 | @code{debug-on-error} is non-@code{nil}, the error-catching is turned | |
1330 | off. This makes it possible to use the Lisp debugger to debug the | |
1331 | sentinel. @xref{Debugger}. | |
1332 | ||
8a56d8ed RS |
1333 | While a sentinel is running, the process sentinel is temporarily |
1334 | set to @code{nil} so that the sentinel won't run recursively. | |
1335 | For this reason it is not possible for a sentinel to specify | |
1336 | a new sentinel. | |
1337 | ||
f9f59935 RS |
1338 | In earlier Emacs versions, every sentinel that did regular expression |
1339 | searching or matching had to explicitly save and restore the match data. | |
1340 | Now Emacs does this automatically for sentinels; they never need to do | |
1341 | it explicitly. @xref{Match Data}. | |
bfe721d1 | 1342 | |
73804d4b RS |
1343 | @defun set-process-sentinel process sentinel |
1344 | This function associates @var{sentinel} with @var{process}. If | |
1345 | @var{sentinel} is @code{nil}, then the process will have no sentinel. | |
1346 | The default behavior when there is no sentinel is to insert a message in | |
1347 | the process's buffer when the process status changes. | |
1348 | ||
5517ea8a RS |
1349 | Changes in process sentinel take effect immediately---if the sentinel |
1350 | is slated to be run but has not been called yet, and you specify a new | |
1351 | sentinel, the eventual call to the sentinel will use the new one. | |
1352 | ||
73804d4b RS |
1353 | @smallexample |
1354 | @group | |
1355 | (defun msg-me (process event) | |
1356 | (princ | |
1357 | (format "Process: %s had the event `%s'" process event))) | |
1358 | (set-process-sentinel (get-process "shell") 'msg-me) | |
1359 | @result{} msg-me | |
1360 | @end group | |
1361 | @group | |
1362 | (kill-process (get-process "shell")) | |
1363 | @print{} Process: #<process shell> had the event `killed' | |
1364 | @result{} #<process shell> | |
1365 | @end group | |
1366 | @end smallexample | |
1367 | @end defun | |
1368 | ||
1369 | @defun process-sentinel process | |
1370 | This function returns the sentinel of @var{process}, or @code{nil} if it | |
1371 | has none. | |
1372 | @end defun | |
1373 | ||
1374 | @defun waiting-for-user-input-p | |
1375 | While a sentinel or filter function is running, this function returns | |
1376 | non-@code{nil} if Emacs was waiting for keyboard input from the user at | |
1377 | the time the sentinel or filter function was called, @code{nil} if it | |
1378 | was not. | |
1379 | @end defun | |
1380 | ||
edc590bb | 1381 | @node Query Before Exit |
177c0ea7 | 1382 | @section Querying Before Exit |
edc590bb RS |
1383 | |
1384 | When Emacs exits, it terminates all its subprocesses by sending them | |
1385 | the @code{SIGHUP} signal. Because some subprocesses are doing | |
1386 | valuable work, Emacs normally asks the user to confirm that it is ok | |
1387 | to terminate them. Each process has a query flag which, if | |
1388 | non-@code{nil}, says that Emacs should ask for confirmation before | |
1389 | exiting and thus killing that process. The default for the query flag | |
1390 | is @code{t}, meaning @emph{do} query. | |
1391 | ||
1392 | @tindex process-query-on-exit-flag | |
1393 | @defun process-query-on-exit-flag process | |
1394 | This returns the query flag of @var{process}. | |
1395 | @end defun | |
1396 | ||
1397 | @tindex set-process-query-on-exit-flag | |
1398 | @defun set-process-query-on-exit-flag process flag | |
1399 | This function sets the query flag of @var{process} to @var{flag}. It | |
1400 | returns @var{flag}. | |
1401 | ||
1402 | @smallexample | |
1403 | @group | |
1404 | ;; @r{Don't query about the shell process} | |
1405 | (set-process-query-on-exit-flag (get-process "shell") nil) | |
1406 | @result{} t | |
1407 | @end group | |
1408 | @end smallexample | |
1409 | @end defun | |
1410 | ||
1411 | @defun process-kill-without-query process &optional do-query | |
1412 | This function clears the query flag of @var{process}, so that | |
1413 | Emacs will not query the user on account of that process. | |
1414 | ||
1415 | Actually, the function does more than that: it returns the old value of | |
1416 | the process's query flag, and sets the query flag to @var{do-query}. | |
1417 | Please don't use this function to do those things any more---please | |
1418 | use the newer, cleaner functions @code{process-query-on-exit-flag} and | |
1419 | @code{set-process-query-on-exit-flag} in all but the simplest cases. | |
1420 | The only way you should use @code{process-kill-without-query} nowadays | |
1421 | is like this: | |
1422 | ||
1423 | @smallexample | |
1424 | @group | |
1425 | ;; @r{Don't query about the shell process} | |
1426 | (process-kill-without-query (get-process "shell")) | |
1427 | @end group | |
1428 | @end smallexample | |
1429 | @end defun | |
1430 | ||
73804d4b RS |
1431 | @node Transaction Queues |
1432 | @section Transaction Queues | |
1433 | @cindex transaction queue | |
1434 | ||
f9f59935 RS |
1435 | You can use a @dfn{transaction queue} to communicate with a subprocess |
1436 | using transactions. First use @code{tq-create} to create a transaction | |
1437 | queue communicating with a specified process. Then you can call | |
1438 | @code{tq-enqueue} to send a transaction. | |
73804d4b RS |
1439 | |
1440 | @defun tq-create process | |
1441 | This function creates and returns a transaction queue communicating with | |
1442 | @var{process}. The argument @var{process} should be a subprocess | |
1443 | capable of sending and receiving streams of bytes. It may be a child | |
78608595 | 1444 | process, or it may be a TCP connection to a server, possibly on another |
73804d4b RS |
1445 | machine. |
1446 | @end defun | |
1447 | ||
1448 | @defun tq-enqueue queue question regexp closure fn | |
1449 | This function sends a transaction to queue @var{queue}. Specifying the | |
1450 | queue has the effect of specifying the subprocess to talk to. | |
1451 | ||
78608595 | 1452 | The argument @var{question} is the outgoing message that starts the |
73804d4b RS |
1453 | transaction. The argument @var{fn} is the function to call when the |
1454 | corresponding answer comes back; it is called with two arguments: | |
1455 | @var{closure}, and the answer received. | |
1456 | ||
6f6fd5ef RS |
1457 | The argument @var{regexp} is a regular expression that should match |
1458 | text at the end of the entire answer, but nothing before; that's how | |
1459 | @code{tq-enqueue} determines where the answer ends. | |
73804d4b RS |
1460 | |
1461 | The return value of @code{tq-enqueue} itself is not meaningful. | |
1462 | @end defun | |
1463 | ||
1464 | @defun tq-close queue | |
1465 | Shut down transaction queue @var{queue}, waiting for all pending transactions | |
1466 | to complete, and then terminate the connection or child process. | |
1467 | @end defun | |
1468 | ||
1469 | Transaction queues are implemented by means of a filter function. | |
1470 | @xref{Filter Functions}. | |
1471 | ||
bfe721d1 KH |
1472 | @node Network |
1473 | @section Network Connections | |
1474 | @cindex network connection | |
73804d4b | 1475 | @cindex TCP |
0bbffed7 | 1476 | @cindex UDP |
73804d4b | 1477 | |
0bbffed7 KS |
1478 | Emacs Lisp programs can open stream (TCP) and datagram (UDP) network |
1479 | connections to other processes on the same machine or other machines. | |
1480 | A network connection is handled by Lisp much like a subprocess, and is | |
4bb24054 RS |
1481 | represented by a process object. However, the process you are |
1482 | communicating with is not a child of the Emacs process, so it has no | |
1483 | process @sc{id}, and you can't kill it or send it signals. All you | |
1484 | can do is send and receive data. @code{delete-process} closes the | |
1485 | connection, but does not kill the program at the other end; that | |
1486 | program must decide what to do about closure of the connection. | |
1487 | ||
1488 | Lisp programs can listen for connections by creating network | |
1489 | servers. A network server is also represented by a kind of process | |
1490 | object, but unlike a network connection, the network server never | |
1491 | transfers data itself. When it receives a connection request, it | |
1492 | creates a new network connection to represent the connection just | |
1493 | made. (The network connection inherits certain information, including | |
1494 | the process plist, from the server.) The network server then goes | |
1495 | back to listening for more connection requests. | |
73804d4b RS |
1496 | |
1497 | You can distinguish process objects representing network connections | |
4bb24054 RS |
1498 | and servers from those representing subprocesses with the |
1499 | @code{process-status} function. The possible status values for | |
1500 | network connections are @code{open}, @code{closed}, @code{connect}, | |
1501 | and @code{failed}. For a network server, the status is always | |
1502 | @code{listen}. None of those values is possible for a real | |
1503 | subprocess. @xref{Process Information}. | |
1504 | ||
1505 | You can stop and resume operation of a network processes by calling | |
1506 | @code{stop-process} and @code{continue-process}. For a server | |
1507 | process, being stopped means not accepting new connections. (Up to 5 | |
0bbffed7 KS |
1508 | connection requests will be queued for when you resume the server; you |
1509 | can increase this limit, unless it is imposed by the operating | |
1510 | systems.) For a network connection, being stopped means not | |
1511 | processing input (any arriving input waits until you resume the | |
1512 | connection). You can use the function @code{process-command} to | |
1513 | determine whether a network connection or server is stopped; a | |
1514 | non-@code{nil} value means yes. | |
73804d4b RS |
1515 | |
1516 | @defun open-network-stream name buffer-or-name host service | |
4bb24054 RS |
1517 | This function opens a TCP connection, and returns a process object |
1518 | that represents the connection. | |
73804d4b RS |
1519 | |
1520 | The @var{name} argument specifies the name for the process object. It | |
1521 | is modified as necessary to make it unique. | |
1522 | ||
1523 | The @var{buffer-or-name} argument is the buffer to associate with the | |
1524 | connection. Output from the connection is inserted in the buffer, | |
1525 | unless you specify a filter function to handle the output. If | |
1526 | @var{buffer-or-name} is @code{nil}, it means that the connection is not | |
1527 | associated with any buffer. | |
1528 | ||
1529 | The arguments @var{host} and @var{service} specify where to connect to; | |
1530 | @var{host} is the host name (a string), and @var{service} is the name of | |
1531 | a defined network service (a string) or a port number (an integer). | |
1532 | @end defun | |
ab5796a9 | 1533 | |
4bb24054 RS |
1534 | @defun open-network-stream-nowait name buffer-or-name host service &optional sentinel filter |
1535 | This function opens a TCP connection, like @code{open-network-stream}, | |
1536 | but it returns immediately without waiting for the request to be | |
1537 | accepted or rejected by the remote server. When the request is | |
1538 | subsequently accepted or rejected, the process's sentinel function | |
1539 | will be called with a string that starts with @code{"open"} (on | |
1540 | success) or @code{"failed"} (on error). | |
1541 | ||
1542 | Some systems do not support non-blocking connections; on those | |
1543 | systems, @code{open-network-stream-nowait} returns @code{nil} | |
1544 | and does nothing. | |
1545 | ||
1546 | The optional arguments @var{sentinel} and @var{filter} specify the | |
1547 | sentinel and filter functions for this network connection. It is | |
1548 | useful to specify them when opening the connection, because they will | |
1549 | be used later asynchronously. The other arguments mean the same as in | |
1550 | @code{open-network-stream}. | |
1551 | @end defun | |
1552 | ||
1553 | @defun process-contact process &optional key | |
1554 | This function returns information about how a network process was set | |
1555 | up. For a connection, when @var{key} is @code{nil}, it returns | |
1556 | @code{(@var{hostname} @var{service})} which specifies what you | |
1557 | connected to. | |
1558 | ||
1559 | If @var{key} is @code{t}, the value is the complete status information | |
1560 | for the connection or server; that is, the list of keywords and values | |
1561 | specified in @code{make-network-process}, except that some of the | |
1562 | values represent the current status instead of what you specified: | |
1563 | ||
1564 | @table @code | |
1565 | @item :buffer | |
1566 | The associated value is the process buffer. | |
1567 | @item :filter | |
1568 | The associated value is the process filter function. | |
1569 | @item :sentinel | |
0bbffed7 | 1570 | The associated value is the process sentinel function. |
4bb24054 RS |
1571 | @item :remote |
1572 | In a connection, this is the address in internal format of the remote peer. | |
1573 | @item :local | |
1574 | The local address, in internal format. | |
1575 | @item :service | |
1576 | In a server, if you specified @code{t} for @var{service}, | |
1577 | this value is the actual port number. | |
1578 | @end table | |
1579 | ||
1580 | @code{:local} and @code{:remote} are included even if they were not | |
1581 | specified explicitly in @code{make-network-process}. | |
1582 | ||
1583 | If @var{key} is a keyword, the function returns the value corresponding | |
1584 | to that keyword. | |
1585 | ||
1586 | For an ordinary child process, this function always returns @code{t}. | |
1587 | @end defun | |
1588 | ||
1589 | @node Network Servers | |
1590 | @section Network Servers | |
1591 | ||
1592 | You create a server by calling @code{make-network-process} with | |
1593 | @code{:server t}. The server will listen for connection requests from | |
1594 | clients. When it accepts a client connection request, that creates a | |
1595 | new network connection, itself a process object, with the following | |
1596 | parameters: | |
1597 | ||
1598 | @itemize @bullet | |
1599 | @item | |
1600 | The connection's process name is constructed by concatenating the | |
1601 | server process' @var{name} with a client identification string. The | |
1602 | client identification string for an IPv4 connection looks like | |
1603 | @samp{<@var{a}.@var{b}.@var{c}.@var{d}:@var{p}>}. Otherwise, it is a | |
0bbffed7 | 1604 | unique number in brackets, as in @samp{<@var{nnn}>}. The number |
4bb24054 RS |
1605 | is unique for each connection in the Emacs session. |
1606 | ||
1607 | @item | |
1608 | If the server's filter is non-@code{nil}, the connection process does | |
0bbffed7 | 1609 | not get a separate process buffer; otherwise, Emacs creates a new |
4bb24054 RS |
1610 | buffer for the purpose. The buffer name is the server's buffer name |
1611 | or process name, concatenated with the client identification string. | |
1612 | ||
1613 | The server's process buffer value is never used directly by Emacs, but | |
1614 | it is passed to the log function, which can log connections by | |
1615 | inserting text there. | |
1616 | ||
1617 | @item | |
1618 | The communication type and the process filter and sentinel are | |
1619 | inherited from those of the server. The server never directly | |
1620 | uses its filter and sentinel; their sole purpose is to initialize | |
1621 | connections made to the server. | |
1622 | ||
1623 | @item | |
1624 | The connection's process contact info is set according to the client's | |
1625 | addressing information (typically an IP address and a port number). | |
1626 | This information is associated with the @code{process-contact} | |
1627 | keywords @code{:host}, @code{:service}, @code{:remote}. | |
1628 | ||
1629 | @item | |
1630 | The connection's local address is set up according to the port | |
1631 | number used for the connection. | |
1632 | ||
1633 | @item | |
1634 | The client process' plist is initialized from the server's plist. | |
1635 | @end itemize | |
1636 | ||
1637 | @defun open-network-stream-server name buffer-or-name service &optional sentinel filter | |
1638 | Create a network server process for a TCP service. | |
1639 | It returns nil if server processes are not supported; otherwise, | |
1640 | it returns a subprocess-object to represent the server. | |
1641 | ||
1642 | When a client connects to the specified service, Emacs creates a new | |
1643 | subprocess to handle the new connection, and then calls its sentinel | |
1644 | function (which it has inherited from the server). | |
1645 | ||
1646 | The optional arguments @var{sentinel} and @var{filter} specify the | |
1647 | sentinel and filter functions for the server. It is useful to specify | |
1648 | them now, because they will be used later asynchronously when the | |
1649 | server receives a connection request. The three arguments @var{name}, | |
1650 | @var{buffer-or-name} and @var{service} mean the same thing as in | |
1651 | @code{open-network-stream}, but @var{service} can be @code{t} | |
1652 | meaning ask the system to allocate an unused port to listen on. | |
1653 | @end defun | |
1654 | ||
1655 | @node Datagrams | |
1656 | @section Datagrams | |
1657 | @cindex datagrams | |
1658 | ||
1659 | A datagram connection communicates with individual packets | |
1660 | rather than streams of data. Each call to @code{process-send} | |
1661 | sends one datagram packet, and each datagram received results | |
1662 | in one call to the filter function. | |
1663 | ||
1664 | The datagram connection doesn't have to talk with the same remote | |
1665 | peer all the time. It has a @dfn{remote peer address} which specifies | |
1666 | where to send datagrams to. Each time an incoming datagram is passed | |
1667 | to the filter function, the peer address is set to the address that | |
1668 | datagram came from; that way, if the filter function sends a datagram, | |
1669 | it will go back to that place. You can specify the remote peer | |
1670 | address when you create the datagram connection using the | |
1671 | @code{:remote} keyword. You can change it later on by calling | |
1672 | @code{set-process-datagram-address}. | |
1673 | ||
1674 | @defun process-datagram-address process | |
1675 | If @var{process} is a datagram connection or server, this function | |
1676 | returns its remote peer address. | |
1677 | @end defun | |
1678 | ||
1679 | @defun set-process-datagram-address process address | |
1680 | If @var{process} is a datagram connection or server, this function | |
1681 | sets its remote peer address to @var{address}. | |
1682 | @end defun | |
1683 | ||
1684 | @node Low-Level Network | |
1685 | @section Low-Level Network Access | |
1686 | ||
1687 | The basic function for creating network connections and network | |
1688 | servers is @code{make-network-process}. It can do either of those | |
1689 | jobs, depending on the arguments you give it. | |
1690 | ||
1691 | @defun make-network-process &rest args | |
1692 | This function creates a network connection or server and returns the | |
1693 | process object that represents it. The arguments @var{args} are a | |
1694 | list of keyword/argument pairs. Omitting a keyword is always | |
1695 | equivalent to specifying it with value @code{nil}, except for | |
0bbffed7 KS |
1696 | @code{:coding}, @code{:filter-multibyte}, and @code{:reuseaddr}. Here |
1697 | are the meaningful keywords: | |
4bb24054 RS |
1698 | |
1699 | @table @asis | |
1700 | @item :name name | |
1701 | Use the string @var{name} as the process name. It is modified if | |
1702 | necessary to make it unique. | |
1703 | ||
1704 | @item :type @var{type} | |
1705 | Specify the communication type. A value of @code{nil} specifies a | |
1706 | stream connection (the default); @code{datagram} specifies a datagram | |
1707 | connection. Both connections and servers can be of either type. | |
1708 | ||
1709 | @item :server @var{server-flag} | |
1710 | If @var{server-flag} is non-@code{nil}, create a server. Otherwise, | |
0bbffed7 KS |
1711 | create a connection. For a stream type server, @var{server-flag} may |
1712 | be an integer which then specifies the length of the queue of pending | |
1713 | connections to the server. The default queue length is 5. | |
4bb24054 RS |
1714 | |
1715 | @item :host @var{host} | |
1716 | Specify the host to connect to. @var{host} should be a host name or | |
1717 | internet address, as a string, or the symbol @code{local} to specify | |
1718 | the local host. If you specify @var{host} for a server, it must | |
1719 | specify a valid address for the local host, and only clients | |
1720 | connecting to that address will be accepted. | |
1721 | ||
1722 | @item :service @var{service} | |
1723 | @var{service} specifies a port number to connect to, or, for a server, | |
1724 | the port number to listen on. It should be a service name that | |
1725 | translates to a port number, or an integer specifying the port number | |
1726 | directly. For a server, it can also be @code{t}, which means to let | |
1727 | the system select an unused port number. | |
1728 | ||
1729 | @item :family @var{family} | |
1730 | @var{family} specifies the address (and protocol) family for | |
1731 | communication. @code{nil} stands for IPv4. @code{local} specifies a | |
1732 | Unix socket, in which case @var{host} is ignored. | |
1733 | ||
1734 | @item :local @var{local-address} | |
1735 | For a server process, @var{local-address} is the address to listen on. | |
1736 | It overrides @var{family}, @var{host} and @var{service}, and you | |
1737 | may as well not specify them. | |
1738 | ||
1739 | @item :remote @var{remote-address} | |
1740 | For a connection, @var{remote-address} is the address to connect to. | |
1741 | It overrides @var{family}, @var{host} and @var{service}, and you | |
1742 | may as well not specify them. | |
1743 | ||
1744 | For a datagram server, @var{remote-address} specifies the initial | |
1745 | setting of the remote datagram address. | |
1746 | ||
1747 | The format of @var{local-address} or @var{remote-address} depends on | |
1748 | the address family: | |
1749 | ||
1750 | @itemize - | |
1751 | @item | |
1752 | An IPv4 address is represented as a vector of integers @code{[@var{a} | |
1753 | @var{b} @var{c} @var{d} @var{p}]} corresponding to numeric IP address | |
1754 | @var{a}.@var{b}.@var{c}.@var{d} and port number @var{p}. | |
1755 | ||
1756 | @item | |
1757 | A local address is represented as a string which specifies the address | |
1758 | in the local address space. | |
1759 | ||
1760 | @item | |
1761 | An ``unsupported family'' address is represented by a cons | |
1762 | @code{(@var{f} . @var{av})}, where @var{f} is the family number and | |
1763 | @var{av} is a vector specifying the socket address using with one | |
1764 | element per address data byte. Do not rely on this format in portable | |
1765 | code, as it may depend on implementation defined constants, data | |
1766 | sizes, and data structure alignment. | |
1767 | @end itemize | |
1768 | ||
1769 | @item :nowait @var{bool} | |
1770 | If @var{bool} is non-@code{nil} for a stream connection, return | |
1771 | without waiting for the connection to complete. When the connection | |
1772 | succeeds or fails, Emacs will call the sentinel function, with a | |
1773 | second argument matching @code{"open"} (if successful) or | |
1774 | @code{"failed"}. The default is to block, so that | |
1775 | @code{make-network-process} does not return until the connection | |
1776 | has succeeded or failed. | |
1777 | ||
1778 | @item :stop @var{stopped} | |
1779 | Start the network connection or server in the `stopped' state if | |
1780 | @var{stopped} is non-@code{nil}. | |
1781 | ||
1782 | @item :buffer @var{buffer} | |
1783 | Use @var{buffer} as the process buffer. | |
1784 | ||
1785 | @item :coding @var{coding} | |
1786 | Use @var{coding} as the coding system for this process. To specify | |
1787 | different coding systems for decoding data from the connection and for | |
1788 | encoding data sent to it, specify @code{(@var{decoding} . | |
1789 | @var{encoding})} for @var{coding}. | |
1790 | ||
1791 | If you don't specify this keyword at all, the default | |
0bbffed7 | 1792 | is to determine the coding systems from the data. |
4bb24054 RS |
1793 | |
1794 | @item :noquery @var{query-flag} | |
0bbffed7 | 1795 | Initialize the process query flag to @var{query-flag}. @xref{Query Before Exit}. |
4bb24054 RS |
1796 | |
1797 | @item :filter @var{filter} | |
1798 | Initialize the process filter to @var{filter}. | |
1799 | ||
1800 | @item :filter-multibyte @var{bool} | |
1801 | If @var{bool} is non-@code{nil}, strings given to the process filter | |
1802 | are multibyte, otherwise they are unibyte. If you don't specify this | |
1803 | keyword at all, the default is that the strings are multibyte if | |
1804 | @code{default-enable-multibyte-characters} is non-@code{nil}. | |
1805 | ||
1806 | @item :sentinel @var{sentinel} | |
1807 | Initialize the process sentinel to @var{sentinel}. | |
1808 | ||
1809 | @item :log @var{log} | |
1810 | Initialize the log function of a server process to @var{log}. The log | |
1811 | function is called each time the server accepts a network connection | |
1812 | from a client. The arguments passed to the log function are | |
1813 | @var{server}, @var{connection}, and @var{message}, where @var{server} | |
1814 | is the server process, @var{connection} is the new process for the | |
1815 | connection, and @var{message} is a string describing what has | |
1816 | happened. | |
1817 | ||
1818 | @item :plist @var{plist} | |
1819 | Initialize the process plist to @var{plist}. | |
1820 | @end table | |
1821 | ||
0bbffed7 KS |
1822 | The following network options can be specified for the network |
1823 | process. Except for @code{:reuseaddr}, you can set or modify these | |
1824 | options later using @code{set-network-process-option}. | |
1825 | ||
1826 | For a server process, the options specified with | |
1827 | @code{make-network-process} are not inherited by the client | |
1828 | connections, so you will need to set the necessary options for each | |
1829 | child connection as they are created. | |
1830 | ||
41beda59 | 1831 | @table @asis |
0bbffed7 KS |
1832 | @item :bindtodevice @var{device-name} |
1833 | If @var{device-name} is a non-empty string identifying a network | |
1834 | interface name (see @code{network-interface-list}), only handle | |
1835 | packets received on that interface. If @var{device-name} is nil (the | |
1836 | default), handle packets received on any interface. | |
1837 | ||
1838 | Using this option may require special privileges on some systems. | |
1839 | ||
1840 | @item :broadcast @var{broadcast-flag} | |
1841 | If @var{broadcast-flag} is non-@code{nil} for a datagram process, the | |
1842 | process will receive datagram packet sent to a broadcast address, and | |
1843 | be able to send packets to a broadcast address. Ignored for a stream | |
1844 | connection. | |
1845 | ||
1846 | @item :dontroute @var{dontroute-flag} | |
1847 | If @var{dontroute-flag} is non-@code{nil}, the process can only send | |
1848 | to hosts on the same network as the local host. | |
1849 | ||
1850 | @item :keepalive @var{keepalive-flag} | |
1851 | If @var{keepalive-flag} is non-@code{nil} for a stream connection, | |
1852 | enable exchange of low-level keep-alive messa | |
1853 | ||
1854 | @item :linger @var{linger-arg} | |
1855 | If @var{linger-arg} is non-@code{nil}, wait for successful | |
1856 | transmission of all queued packets on the connection before it is | |
1857 | deleted (see @code{delete-process}). If @var{linger-arg} is an | |
1858 | integer, it specifies the maximum time in seconds to wait for queued | |
1859 | packets to be sent before closing the connection. Default is | |
1860 | @code{nil} which means to discard unsent queued packets when the | |
1861 | process is deleted. | |
1862 | ||
1863 | @item :oobinline @var{oobinline-flag} | |
1864 | If @var{oobinline-flag} is non-@code{nil} for a stream connection, | |
1865 | receive out-of-band data in the normal data stream. Otherwise, ignore | |
1866 | out-of-band data. | |
1867 | ||
1868 | @item :priority @var{priority} | |
1869 | Set the priority for packets sent on this connection to the integer | |
1870 | @var{priority}. The interpretation of this number is protocol | |
1871 | specific, such as setting the TOS (type of service) field on IP | |
1872 | packets sent on this connection. It may also have system dependent | |
1873 | effects, such as selecting a specific output queue on the network | |
1874 | interface. | |
1875 | ||
1876 | @item :reuseaddr @var{reuseaddr-flag} | |
1877 | If @var{reuseaddr-flag} is non-@code{nil} (the default) for a stream | |
1878 | server process, allow this server to reuse a specific port number (see | |
1879 | @code{:service}) unless another process on this host is already | |
1880 | listening on that port. If @var{reuseaddr-flag} is @code{nil}, there | |
1881 | may be a period of time after the last use of that port (by any | |
1882 | process on the host), where it is not possible to make a new server on | |
1883 | that port. | |
1884 | ||
1885 | @end table | |
1886 | ||
4bb24054 RS |
1887 | The original argument list, modified with the actual connection |
1888 | information, is available via the `process-contact' function. | |
1889 | @end defun | |
1890 | ||
0bbffed7 KS |
1891 | @defun set-network-process-option process option value |
1892 | This function sets or modifies a network option for network process | |
1893 | @var{process}. See @code{make-network-process} for details of options | |
1894 | @var{option} and their corresponding values @var{value}. | |
1895 | ||
1896 | The current setting of an option is available via the | |
1897 | `process-contact' function. | |
1898 | @end defun | |
1899 | ||
4bb24054 RS |
1900 | @defun network-interface-list |
1901 | This function returns a list describing the network interfaces | |
1902 | of the machine you are using. The value is an alist whose | |
1903 | elements have the form @code{(@var{name} . @var{address})}. | |
1904 | @var{address} has the same form as the @var{local-address} | |
1905 | and @var{remote-address} arguments to @code{make-network-process}. | |
1906 | @end defun | |
1907 | ||
1908 | @defun network-interface-info ifname | |
1909 | This function returns information about the network interface named | |
1910 | @var{ifname}. The value is a list of the form @code{(@var{addr} @var{bcast} @var{netmask} @var{hwaddr} @var{flags})}. | |
1911 | ||
1912 | @table @var | |
1913 | @item addr | |
1914 | The internet protocol address. | |
1915 | @item bcast | |
1916 | The broadcast address. | |
1917 | @item netmask | |
1918 | The network mask. | |
1919 | @item hwaddr | |
1920 | The layer 2 address (Ethernet MAC address, for instance). | |
1921 | @item flags | |
1922 | The current flags of the interface. | |
1923 | @end table | |
1924 | @end defun | |
1925 | ||
1926 | @defun format-network-address address &optional omit-port | |
1927 | This function converts the lisp representation of a network address to | |
1928 | a string. For example, a five-element vector @code{[@var{a} @var{b} | |
1929 | @var{c} @var{d} @var{p}]} represents an IP address | |
1930 | @var{a}.@var{b}.@var{c}.@var{d} and port number @var{p}. | |
1931 | @code{format-network-address} converts that to the string | |
1932 | @code{"@var{a}.@var{b}.@var{c}.@var{d}:@var{p}"}. | |
1933 | ||
1934 | If @var{omit-port} is non-@code{nil}, the value does not include | |
1935 | the port number. | |
1936 | @end defun | |
1937 | ||
1938 | To test for the availability of a given network feature, use | |
1939 | @code{featurep} like this: | |
1940 | ||
1941 | @example | |
1942 | (featurep 'make-network-process '(@var{keyword} @var{value})) | |
1943 | @end example | |
1944 | ||
1945 | @noindent | |
0bbffed7 KS |
1946 | The result of the first form is @code{t} if it works to specify |
1947 | @var{keyword} with value @var{value} in @code{make-network-process}. | |
1948 | The result of the second form is @code{t} if @var{keyword} is | |
1949 | supported by @code{make-network-process}. Here are some of the | |
1950 | @var{keyword}---@var{value} pairs you can test in | |
1951 | this way. | |
4bb24054 RS |
1952 | |
1953 | @table @code | |
1954 | @item (:nowait t) | |
1955 | Non-@code{nil} if non-blocking connect is supported. | |
1956 | @item (:type datagram) | |
1957 | Non-@code{nil} if datagrams are supported. | |
1958 | @item (:family local) | |
1959 | Non-@code{nil} if local (aka ``UNIX domain'') sockets are supported. | |
1960 | @item (:service t) | |
1961 | Non-@code{nil} if the system can select the port for a server. | |
0bbffed7 KS |
1962 | @end table |
1963 | ||
1964 | To test for the availability of a given network option, use | |
1965 | @code{featurep} like this: | |
1966 | ||
1967 | @example | |
1968 | (featurep 'make-network-process '@var{keyword}) | |
1969 | @end example | |
1970 | ||
1971 | Here are some of the option @var{keyword}s you can test in | |
1972 | this way. | |
1973 | ||
1974 | @table @code | |
1975 | @item :bindtodevice | |
1976 | @itemx :broadcast | |
1977 | @itemx :dontroute | |
1978 | @itemx :keepalive | |
1979 | @itemx :linger | |
1980 | @itemx :oobinline | |
1981 | @itemx :priority | |
1982 | @itemx :reuseaddr | |
1983 | That particular network option is supported by | |
1984 | @code{make-network-process} and @code{set-network-process-option}. | |
4bb24054 RS |
1985 | @end table |
1986 | ||
ab5796a9 MB |
1987 | @ignore |
1988 | arch-tag: ba9da253-e65f-4e7f-b727-08fba0a1df7a | |
1989 | @end ignore | |
4bb24054 | 1990 |