Commit | Line | Data |
a0d0e21e |
1 | =head1 NAME |
2 | |
184e9718 |
3 | perlipc - Perl interprocess communication (signals, fifos, pipes, safe subprocesses, sockets, and semaphores) |
a0d0e21e |
4 | |
5 | =head1 DESCRIPTION |
6 | |
4633a7c4 |
7 | The basic IPC facilities of Perl are built out of the good old Unix |
8 | signals, named pipes, pipe opens, the Berkeley socket routines, and SysV |
9 | IPC calls. Each is used in slightly different situations. |
10 | |
11 | =head1 Signals |
12 | |
13 | Perl uses a simple signal handling model: the %SIG hash contains names or |
14 | references of user-installed signal handlers. These handlers will be called |
15 | with an argument which is the name of the signal that triggered it. A |
16 | signal may be generated intentionally from a particular keyboard sequence like |
a2eb9003 |
17 | control-C or control-Z, sent to you from another process, or |
4633a7c4 |
18 | triggered automatically by the kernel when special events transpire, like |
54310121 |
19 | a child process exiting, your process running out of stack space, or |
4633a7c4 |
20 | hitting file size limit. |
21 | |
22 | For example, to trap an interrupt signal, set up a handler like this. |
7b05b7e3 |
23 | Do as little as you possibly can in your handler; notice how all we do is |
24 | set a global variable and then raise an exception. That's because on most |
25 | systems, libraries are not re-entrant; particularly, memory allocation and |
26 | I/O routines are not. That means that doing nearly I<anything> in your |
27 | handler could in theory trigger a memory fault and subsequent core dump. |
4633a7c4 |
28 | |
29 | sub catch_zap { |
30 | my $signame = shift; |
31 | $shucks++; |
32 | die "Somebody sent me a SIG$signame"; |
54310121 |
33 | } |
4633a7c4 |
34 | $SIG{INT} = 'catch_zap'; # could fail in modules |
35 | $SIG{INT} = \&catch_zap; # best strategy |
36 | |
37 | The names of the signals are the ones listed out by C<kill -l> on your |
38 | system, or you can retrieve them from the Config module. Set up an |
39 | @signame list indexed by number to get the name and a %signo table |
40 | indexed by name to get the number: |
41 | |
42 | use Config; |
43 | defined $Config{sig_name} || die "No sigs?"; |
44 | foreach $name (split(' ', $Config{sig_name})) { |
45 | $signo{$name} = $i; |
46 | $signame[$i] = $name; |
47 | $i++; |
54310121 |
48 | } |
4633a7c4 |
49 | |
6a3992aa |
50 | So to check whether signal 17 and SIGALRM were the same, do just this: |
4633a7c4 |
51 | |
52 | print "signal #17 = $signame[17]\n"; |
54310121 |
53 | if ($signo{ALRM}) { |
4633a7c4 |
54 | print "SIGALRM is $signo{ALRM}\n"; |
54310121 |
55 | } |
4633a7c4 |
56 | |
57 | You may also choose to assign the strings C<'IGNORE'> or C<'DEFAULT'> as |
58 | the handler, in which case Perl will try to discard the signal or do the |
f648820c |
59 | default thing. |
60 | |
19799a22 |
61 | On most Unix platforms, the C<CHLD> (sometimes also known as C<CLD>) signal |
f648820c |
62 | has special behavior with respect to a value of C<'IGNORE'>. |
63 | Setting C<$SIG{CHLD}> to C<'IGNORE'> on such a platform has the effect of |
64 | not creating zombie processes when the parent process fails to C<wait()> |
65 | on its child processes (i.e. child processes are automatically reaped). |
66 | Calling C<wait()> with C<$SIG{CHLD}> set to C<'IGNORE'> usually returns |
67 | C<-1> on such platforms. |
68 | |
69 | Some signals can be neither trapped nor ignored, such as |
4633a7c4 |
70 | the KILL and STOP (but not the TSTP) signals. One strategy for |
71 | temporarily ignoring signals is to use a local() statement, which will be |
72 | automatically restored once your block is exited. (Remember that local() |
73 | values are "inherited" by functions called from within that block.) |
74 | |
75 | sub precious { |
76 | local $SIG{INT} = 'IGNORE'; |
77 | &more_functions; |
54310121 |
78 | } |
4633a7c4 |
79 | sub more_functions { |
80 | # interrupts still ignored, for now... |
54310121 |
81 | } |
4633a7c4 |
82 | |
83 | Sending a signal to a negative process ID means that you send the signal |
fb73857a |
84 | to the entire Unix process-group. This code sends a hang-up signal to all |
85 | processes in the current process group (and sets $SIG{HUP} to IGNORE so |
86 | it doesn't kill itself): |
4633a7c4 |
87 | |
88 | { |
89 | local $SIG{HUP} = 'IGNORE'; |
90 | kill HUP => -$$; |
91 | # snazzy writing of: kill('HUP', -$$) |
92 | } |
a0d0e21e |
93 | |
4633a7c4 |
94 | Another interesting signal to send is signal number zero. This doesn't |
95 | actually affect another process, but instead checks whether it's alive |
54310121 |
96 | or has changed its UID. |
a0d0e21e |
97 | |
4633a7c4 |
98 | unless (kill 0 => $kid_pid) { |
99 | warn "something wicked happened to $kid_pid"; |
54310121 |
100 | } |
a0d0e21e |
101 | |
4633a7c4 |
102 | You might also want to employ anonymous functions for simple signal |
103 | handlers: |
a0d0e21e |
104 | |
4633a7c4 |
105 | $SIG{INT} = sub { die "\nOutta here!\n" }; |
a0d0e21e |
106 | |
4633a7c4 |
107 | But that will be problematic for the more complicated handlers that need |
54310121 |
108 | to reinstall themselves. Because Perl's signal mechanism is currently |
184e9718 |
109 | based on the signal(3) function from the C library, you may sometimes be so |
4633a7c4 |
110 | misfortunate as to run on systems where that function is "broken", that |
111 | is, it behaves in the old unreliable SysV way rather than the newer, more |
112 | reasonable BSD and POSIX fashion. So you'll see defensive people writing |
113 | signal handlers like this: |
a0d0e21e |
114 | |
54310121 |
115 | sub REAPER { |
4633a7c4 |
116 | $waitedpid = wait; |
6a3992aa |
117 | # loathe sysV: it makes us not only reinstate |
118 | # the handler, but place it after the wait |
54310121 |
119 | $SIG{CHLD} = \&REAPER; |
4633a7c4 |
120 | } |
121 | $SIG{CHLD} = \&REAPER; |
122 | # now do something that forks... |
123 | |
124 | or even the more elaborate: |
125 | |
6a3992aa |
126 | use POSIX ":sys_wait_h"; |
54310121 |
127 | sub REAPER { |
4633a7c4 |
128 | my $child; |
1450d070 |
129 | while (($child = waitpid(-1,WNOHANG)) > 0) { |
4633a7c4 |
130 | $Kid_Status{$child} = $?; |
54310121 |
131 | } |
6a3992aa |
132 | $SIG{CHLD} = \&REAPER; # still loathe sysV |
4633a7c4 |
133 | } |
134 | $SIG{CHLD} = \&REAPER; |
135 | # do something that forks... |
136 | |
137 | Signal handling is also used for timeouts in Unix, While safely |
138 | protected within an C<eval{}> block, you set a signal handler to trap |
139 | alarm signals and then schedule to have one delivered to you in some |
140 | number of seconds. Then try your blocking operation, clearing the alarm |
141 | when it's done but not before you've exited your C<eval{}> block. If it |
142 | goes off, you'll use die() to jump out of the block, much as you might |
143 | using longjmp() or throw() in other languages. |
144 | |
145 | Here's an example: |
146 | |
54310121 |
147 | eval { |
4633a7c4 |
148 | local $SIG{ALRM} = sub { die "alarm clock restart" }; |
54310121 |
149 | alarm 10; |
4633a7c4 |
150 | flock(FH, 2); # blocking write lock |
54310121 |
151 | alarm 0; |
4633a7c4 |
152 | }; |
153 | if ($@ and $@ !~ /alarm clock restart/) { die } |
154 | |
8a4f6ac2 |
155 | If the operation being timed out is system() or qx(), this technique |
156 | is liable to generate zombies. If this matters to you, you'll |
157 | need to do your own fork() and exec(), and kill the errant child process. |
158 | |
4633a7c4 |
159 | For more complex signal handling, you might see the standard POSIX |
160 | module. Lamentably, this is almost entirely undocumented, but |
161 | the F<t/lib/posix.t> file from the Perl source distribution has some |
162 | examples in it. |
163 | |
164 | =head1 Named Pipes |
165 | |
166 | A named pipe (often referred to as a FIFO) is an old Unix IPC |
167 | mechanism for processes communicating on the same machine. It works |
54310121 |
168 | just like a regular, connected anonymous pipes, except that the |
4633a7c4 |
169 | processes rendezvous using a filename and don't have to be related. |
170 | |
171 | To create a named pipe, use the Unix command mknod(1) or on some |
172 | systems, mkfifo(1). These may not be in your normal path. |
173 | |
174 | # system return val is backwards, so && not || |
175 | # |
176 | $ENV{PATH} .= ":/etc:/usr/etc"; |
54310121 |
177 | if ( system('mknod', $path, 'p') |
4633a7c4 |
178 | && system('mkfifo', $path) ) |
179 | { |
5a964f20 |
180 | die "mk{nod,fifo} $path failed"; |
54310121 |
181 | } |
4633a7c4 |
182 | |
183 | |
184 | A fifo is convenient when you want to connect a process to an unrelated |
185 | one. When you open a fifo, the program will block until there's something |
54310121 |
186 | on the other end. |
4633a7c4 |
187 | |
188 | For example, let's say you'd like to have your F<.signature> file be a |
189 | named pipe that has a Perl program on the other end. Now every time any |
6a3992aa |
190 | program (like a mailer, news reader, finger program, etc.) tries to read |
4633a7c4 |
191 | from that file, the reading program will block and your program will |
6a3992aa |
192 | supply the new signature. We'll use the pipe-checking file test B<-p> |
4633a7c4 |
193 | to find out whether anyone (or anything) has accidentally removed our fifo. |
194 | |
195 | chdir; # go home |
196 | $FIFO = '.signature'; |
197 | $ENV{PATH} .= ":/etc:/usr/games"; |
198 | |
199 | while (1) { |
200 | unless (-p $FIFO) { |
201 | unlink $FIFO; |
54310121 |
202 | system('mknod', $FIFO, 'p') |
4633a7c4 |
203 | && die "can't mknod $FIFO: $!"; |
54310121 |
204 | } |
4633a7c4 |
205 | |
206 | # next line blocks until there's a reader |
207 | open (FIFO, "> $FIFO") || die "can't write $FIFO: $!"; |
208 | print FIFO "John Smith (smith\@host.org)\n", `fortune -s`; |
209 | close FIFO; |
6a3992aa |
210 | sleep 2; # to avoid dup signals |
4633a7c4 |
211 | } |
a0d0e21e |
212 | |
5a964f20 |
213 | =head2 WARNING |
214 | |
215 | By installing Perl code to deal with signals, you're exposing yourself |
216 | to danger from two things. First, few system library functions are |
217 | re-entrant. If the signal interrupts while Perl is executing one function |
218 | (like malloc(3) or printf(3)), and your signal handler then calls the |
219 | same function again, you could get unpredictable behavior--often, a |
220 | core dump. Second, Perl isn't itself re-entrant at the lowest levels. |
221 | If the signal interrupts Perl while Perl is changing its own internal |
222 | data structures, similarly unpredictable behaviour may result. |
223 | |
224 | There are two things you can do, knowing this: be paranoid or be |
225 | pragmatic. The paranoid approach is to do as little as possible in your |
226 | signal handler. Set an existing integer variable that already has a |
227 | value, and return. This doesn't help you if you're in a slow system call, |
228 | which will just restart. That means you have to C<die> to longjump(3) out |
229 | of the handler. Even this is a little cavalier for the true paranoiac, |
230 | who avoids C<die> in a handler because the system I<is> out to get you. |
231 | The pragmatic approach is to say ``I know the risks, but prefer the |
232 | convenience'', and to do anything you want in your signal handler, |
233 | prepared to clean up core dumps now and again. |
234 | |
235 | To forbid signal handlers altogether would bars you from |
236 | many interesting programs, including virtually everything in this manpage, |
237 | since you could no longer even write SIGCHLD handlers. Their dodginess |
238 | is expected to be addresses in the 5.005 release. |
239 | |
a0d0e21e |
240 | |
4633a7c4 |
241 | =head1 Using open() for IPC |
242 | |
243 | Perl's basic open() statement can also be used for unidirectional interprocess |
244 | communication by either appending or prepending a pipe symbol to the second |
a2eb9003 |
245 | argument to open(). Here's how to start something up in a child process you |
4633a7c4 |
246 | intend to write to: |
247 | |
54310121 |
248 | open(SPOOLER, "| cat -v | lpr -h 2>/dev/null") |
4633a7c4 |
249 | || die "can't fork: $!"; |
250 | local $SIG{PIPE} = sub { die "spooler pipe broke" }; |
251 | print SPOOLER "stuff\n"; |
252 | close SPOOLER || die "bad spool: $! $?"; |
253 | |
254 | And here's how to start up a child process you intend to read from: |
255 | |
256 | open(STATUS, "netstat -an 2>&1 |") |
257 | || die "can't fork: $!"; |
258 | while (<STATUS>) { |
259 | next if /^(tcp|udp)/; |
260 | print; |
54310121 |
261 | } |
a2eb9003 |
262 | close STATUS || die "bad netstat: $! $?"; |
4633a7c4 |
263 | |
264 | If one can be sure that a particular program is a Perl script that is |
265 | expecting filenames in @ARGV, the clever programmer can write something |
266 | like this: |
267 | |
5a964f20 |
268 | % program f1 "cmd1|" - f2 "cmd2|" f3 < tmpfile |
4633a7c4 |
269 | |
270 | and irrespective of which shell it's called from, the Perl program will |
271 | read from the file F<f1>, the process F<cmd1>, standard input (F<tmpfile> |
272 | in this case), the F<f2> file, the F<cmd2> command, and finally the F<f3> |
273 | file. Pretty nifty, eh? |
274 | |
54310121 |
275 | You might notice that you could use backticks for much the |
4633a7c4 |
276 | same effect as opening a pipe for reading: |
277 | |
278 | print grep { !/^(tcp|udp)/ } `netstat -an 2>&1`; |
279 | die "bad netstat" if $?; |
280 | |
281 | While this is true on the surface, it's much more efficient to process the |
282 | file one line or record at a time because then you don't have to read the |
19799a22 |
283 | whole thing into memory at once. It also gives you finer control of the |
4633a7c4 |
284 | whole process, letting you to kill off the child process early if you'd |
285 | like. |
286 | |
287 | Be careful to check both the open() and the close() return values. If |
288 | you're I<writing> to a pipe, you should also trap SIGPIPE. Otherwise, |
289 | think of what happens when you start up a pipe to a command that doesn't |
290 | exist: the open() will in all likelihood succeed (it only reflects the |
291 | fork()'s success), but then your output will fail--spectacularly. Perl |
292 | can't know whether the command worked because your command is actually |
293 | running in a separate process whose exec() might have failed. Therefore, |
6a3992aa |
294 | while readers of bogus commands return just a quick end of file, writers |
4633a7c4 |
295 | to bogus command will trigger a signal they'd better be prepared to |
296 | handle. Consider: |
297 | |
5a964f20 |
298 | open(FH, "|bogus") or die "can't fork: $!"; |
299 | print FH "bang\n" or die "can't write: $!"; |
300 | close FH or die "can't close: $!"; |
301 | |
302 | That won't blow up until the close, and it will blow up with a SIGPIPE. |
303 | To catch it, you could use this: |
304 | |
305 | $SIG{PIPE} = 'IGNORE'; |
306 | open(FH, "|bogus") or die "can't fork: $!"; |
307 | print FH "bang\n" or die "can't write: $!"; |
308 | close FH or die "can't close: status=$?"; |
4633a7c4 |
309 | |
68dc0745 |
310 | =head2 Filehandles |
311 | |
5a964f20 |
312 | Both the main process and any child processes it forks share the same |
313 | STDIN, STDOUT, and STDERR filehandles. If both processes try to access |
45bc9206 |
314 | them at once, strange things can happen. You may also want to close |
5a964f20 |
315 | or reopen the filehandles for the child. You can get around this by |
316 | opening your pipe with open(), but on some systems this means that the |
317 | child process cannot outlive the parent. |
68dc0745 |
318 | |
319 | =head2 Background Processes |
320 | |
321 | You can run a command in the background with: |
322 | |
7b05b7e3 |
323 | system("cmd &"); |
68dc0745 |
324 | |
325 | The command's STDOUT and STDERR (and possibly STDIN, depending on your |
326 | shell) will be the same as the parent's. You won't need to catch |
327 | SIGCHLD because of the double-fork taking place (see below for more |
328 | details). |
329 | |
330 | =head2 Complete Dissociation of Child from Parent |
331 | |
332 | In some cases (starting server processes, for instance) you'll want to |
893af57a |
333 | completely dissociate the child process from the parent. This is |
334 | often called daemonization. A well behaved daemon will also chdir() |
335 | to the root directory (so it doesn't prevent unmounting the filesystem |
336 | containing the directory from which it was launched) and redirect its |
337 | standard file descriptors from and to F</dev/null> (so that random |
338 | output doesn't wind up on the user's terminal). |
339 | |
340 | use POSIX 'setsid'; |
341 | |
342 | sub daemonize { |
343 | chdir '/' or die "Can't chdir to /: $!"; |
344 | open STDIN, '/dev/null' or die "Can't read /dev/null: $!"; |
345 | open STDOUT, '>/dev/null' |
346 | or die "Can't write to /dev/null: $!"; |
347 | defined(my $pid = fork) or die "Can't fork: $!"; |
348 | exit if $pid; |
349 | setsid or die "Can't start a new session: $!"; |
350 | open STDERR, '>&STDOUT' or die "Can't dup stdout: $!"; |
351 | } |
5a964f20 |
352 | |
893af57a |
353 | The fork() has to come before the setsid() to ensure that you aren't a |
354 | process group leader (the setsid() will fail if you are). If your |
355 | system doesn't have the setsid() function, open F</dev/tty> and use the |
356 | C<TIOCNOTTY> ioctl() on it instead. See L<tty(4)> for details. |
5a964f20 |
357 | |
893af57a |
358 | Non-Unix users should check their Your_OS::Process module for other |
359 | solutions. |
68dc0745 |
360 | |
4633a7c4 |
361 | =head2 Safe Pipe Opens |
362 | |
363 | Another interesting approach to IPC is making your single program go |
364 | multiprocess and communicate between (or even amongst) yourselves. The |
365 | open() function will accept a file argument of either C<"-|"> or C<"|-"> |
366 | to do a very interesting thing: it forks a child connected to the |
367 | filehandle you've opened. The child is running the same program as the |
368 | parent. This is useful for safely opening a file when running under an |
369 | assumed UID or GID, for example. If you open a pipe I<to> minus, you can |
370 | write to the filehandle you opened and your kid will find it in his |
371 | STDIN. If you open a pipe I<from> minus, you can read from the filehandle |
372 | you opened whatever your kid writes to his STDOUT. |
373 | |
374 | use English; |
375 | my $sleep_count = 0; |
376 | |
54310121 |
377 | do { |
c07a80fd |
378 | $pid = open(KID_TO_WRITE, "|-"); |
4633a7c4 |
379 | unless (defined $pid) { |
380 | warn "cannot fork: $!"; |
381 | die "bailing out" if $sleep_count++ > 6; |
382 | sleep 10; |
54310121 |
383 | } |
4633a7c4 |
384 | } until defined $pid; |
385 | |
386 | if ($pid) { # parent |
c07a80fd |
387 | print KID_TO_WRITE @some_data; |
388 | close(KID_TO_WRITE) || warn "kid exited $?"; |
4633a7c4 |
389 | } else { # child |
390 | ($EUID, $EGID) = ($UID, $GID); # suid progs only |
54310121 |
391 | open (FILE, "> /safe/file") |
4633a7c4 |
392 | || die "can't open /safe/file: $!"; |
393 | while (<STDIN>) { |
394 | print FILE; # child's STDIN is parent's KID |
54310121 |
395 | } |
4633a7c4 |
396 | exit; # don't forget this |
54310121 |
397 | } |
4633a7c4 |
398 | |
399 | Another common use for this construct is when you need to execute |
400 | something without the shell's interference. With system(), it's |
54310121 |
401 | straightforward, but you can't use a pipe open or backticks safely. |
4633a7c4 |
402 | That's because there's no way to stop the shell from getting its hands on |
403 | your arguments. Instead, use lower-level control to call exec() directly. |
404 | |
54310121 |
405 | Here's a safe backtick or pipe open for read: |
4633a7c4 |
406 | |
407 | # add error processing as above |
c07a80fd |
408 | $pid = open(KID_TO_READ, "-|"); |
4633a7c4 |
409 | |
410 | if ($pid) { # parent |
c07a80fd |
411 | while (<KID_TO_READ>) { |
4633a7c4 |
412 | # do something interesting |
54310121 |
413 | } |
c07a80fd |
414 | close(KID_TO_READ) || warn "kid exited $?"; |
4633a7c4 |
415 | |
416 | } else { # child |
417 | ($EUID, $EGID) = ($UID, $GID); # suid only |
418 | exec($program, @options, @args) |
419 | || die "can't exec program: $!"; |
420 | # NOTREACHED |
54310121 |
421 | } |
4633a7c4 |
422 | |
423 | |
424 | And here's a safe pipe open for writing: |
425 | |
426 | # add error processing as above |
c07a80fd |
427 | $pid = open(KID_TO_WRITE, "|-"); |
4633a7c4 |
428 | $SIG{ALRM} = sub { die "whoops, $program pipe broke" }; |
429 | |
430 | if ($pid) { # parent |
431 | for (@data) { |
c07a80fd |
432 | print KID_TO_WRITE; |
54310121 |
433 | } |
c07a80fd |
434 | close(KID_TO_WRITE) || warn "kid exited $?"; |
4633a7c4 |
435 | |
436 | } else { # child |
437 | ($EUID, $EGID) = ($UID, $GID); |
438 | exec($program, @options, @args) |
439 | || die "can't exec program: $!"; |
440 | # NOTREACHED |
54310121 |
441 | } |
4633a7c4 |
442 | |
443 | Note that these operations are full Unix forks, which means they may not be |
444 | correctly implemented on alien systems. Additionally, these are not true |
54310121 |
445 | multithreading. If you'd like to learn more about threading, see the |
184e9718 |
446 | F<modules> file mentioned below in the SEE ALSO section. |
4633a7c4 |
447 | |
7b05b7e3 |
448 | =head2 Bidirectional Communication with Another Process |
4633a7c4 |
449 | |
450 | While this works reasonably well for unidirectional communication, what |
451 | about bidirectional communication? The obvious thing you'd like to do |
452 | doesn't actually work: |
453 | |
c07a80fd |
454 | open(PROG_FOR_READING_AND_WRITING, "| some program |") |
4633a7c4 |
455 | |
9f1b1f2d |
456 | and if you forget to use the C<use warnings> pragma or the B<-w> flag, |
457 | then you'll miss out entirely on the diagnostic message: |
4633a7c4 |
458 | |
459 | Can't do bidirectional pipe at -e line 1. |
460 | |
461 | If you really want to, you can use the standard open2() library function |
7b05b7e3 |
462 | to catch both ends. There's also an open3() for tridirectional I/O so you |
4633a7c4 |
463 | can also catch your child's STDERR, but doing so would then require an |
464 | awkward select() loop and wouldn't allow you to use normal Perl input |
465 | operations. |
466 | |
467 | If you look at its source, you'll see that open2() uses low-level |
5a964f20 |
468 | primitives like Unix pipe() and exec() calls to create all the connections. |
4633a7c4 |
469 | While it might have been slightly more efficient by using socketpair(), it |
470 | would have then been even less portable than it already is. The open2() |
471 | and open3() functions are unlikely to work anywhere except on a Unix |
472 | system or some other one purporting to be POSIX compliant. |
473 | |
474 | Here's an example of using open2(): |
475 | |
476 | use FileHandle; |
477 | use IPC::Open2; |
5a964f20 |
478 | $pid = open2(*Reader, *Writer, "cat -u -n" ); |
4633a7c4 |
479 | print Writer "stuff\n"; |
480 | $got = <Reader>; |
481 | |
6a3992aa |
482 | The problem with this is that Unix buffering is really going to |
483 | ruin your day. Even though your C<Writer> filehandle is auto-flushed, |
4633a7c4 |
484 | and the process on the other end will get your data in a timely manner, |
6a3992aa |
485 | you can't usually do anything to force it to give it back to you |
54310121 |
486 | in a similarly quick fashion. In this case, we could, because we |
4633a7c4 |
487 | gave I<cat> a B<-u> flag to make it unbuffered. But very few Unix |
488 | commands are designed to operate over pipes, so this seldom works |
54310121 |
489 | unless you yourself wrote the program on the other end of the |
4633a7c4 |
490 | double-ended pipe. |
491 | |
54310121 |
492 | A solution to this is the nonstandard F<Comm.pl> library. It uses |
4633a7c4 |
493 | pseudo-ttys to make your program behave more reasonably: |
494 | |
495 | require 'Comm.pl'; |
496 | $ph = open_proc('cat -n'); |
497 | for (1..10) { |
498 | print $ph "a line\n"; |
499 | print "got back ", scalar <$ph>; |
500 | } |
a0d0e21e |
501 | |
4633a7c4 |
502 | This way you don't have to have control over the source code of the |
54310121 |
503 | program you're using. The F<Comm> library also has expect() |
504 | and interact() functions. Find the library (and we hope its |
4633a7c4 |
505 | successor F<IPC::Chat>) at your nearest CPAN archive as detailed |
184e9718 |
506 | in the SEE ALSO section below. |
a0d0e21e |
507 | |
c8db1d39 |
508 | The newer Expect.pm module from CPAN also addresses this kind of thing. |
509 | This module requires two other modules from CPAN: IO::Pty and IO::Stty. |
510 | It sets up a pseudo-terminal to interact with programs that insist on |
511 | using talking to the terminal device driver. If your system is |
512 | amongst those supported, this may be your best bet. |
513 | |
5a964f20 |
514 | =head2 Bidirectional Communication with Yourself |
515 | |
516 | If you want, you may make low-level pipe() and fork() |
517 | to stitch this together by hand. This example only |
518 | talks to itself, but you could reopen the appropriate |
519 | handles to STDIN and STDOUT and call other processes. |
520 | |
521 | #!/usr/bin/perl -w |
522 | # pipe1 - bidirectional communication using two pipe pairs |
523 | # designed for the socketpair-challenged |
524 | use IO::Handle; # thousands of lines just for autoflush :-( |
525 | pipe(PARENT_RDR, CHILD_WTR); # XXX: failure? |
526 | pipe(CHILD_RDR, PARENT_WTR); # XXX: failure? |
527 | CHILD_WTR->autoflush(1); |
528 | PARENT_WTR->autoflush(1); |
529 | |
530 | if ($pid = fork) { |
531 | close PARENT_RDR; close PARENT_WTR; |
532 | print CHILD_WTR "Parent Pid $$ is sending this\n"; |
533 | chomp($line = <CHILD_RDR>); |
534 | print "Parent Pid $$ just read this: `$line'\n"; |
535 | close CHILD_RDR; close CHILD_WTR; |
536 | waitpid($pid,0); |
537 | } else { |
538 | die "cannot fork: $!" unless defined $pid; |
539 | close CHILD_RDR; close CHILD_WTR; |
540 | chomp($line = <PARENT_RDR>); |
541 | print "Child Pid $$ just read this: `$line'\n"; |
542 | print PARENT_WTR "Child Pid $$ is sending this\n"; |
543 | close PARENT_RDR; close PARENT_WTR; |
544 | exit; |
545 | } |
546 | |
547 | But you don't actually have to make two pipe calls. If you |
548 | have the socketpair() system call, it will do this all for you. |
549 | |
550 | #!/usr/bin/perl -w |
551 | # pipe2 - bidirectional communication using socketpair |
552 | # "the best ones always go both ways" |
553 | |
554 | use Socket; |
555 | use IO::Handle; # thousands of lines just for autoflush :-( |
556 | # We say AF_UNIX because although *_LOCAL is the |
557 | # POSIX 1003.1g form of the constant, many machines |
558 | # still don't have it. |
559 | socketpair(CHILD, PARENT, AF_UNIX, SOCK_STREAM, PF_UNSPEC) |
560 | or die "socketpair: $!"; |
561 | |
562 | CHILD->autoflush(1); |
563 | PARENT->autoflush(1); |
564 | |
565 | if ($pid = fork) { |
566 | close PARENT; |
567 | print CHILD "Parent Pid $$ is sending this\n"; |
568 | chomp($line = <CHILD>); |
569 | print "Parent Pid $$ just read this: `$line'\n"; |
570 | close CHILD; |
571 | waitpid($pid,0); |
572 | } else { |
573 | die "cannot fork: $!" unless defined $pid; |
574 | close CHILD; |
575 | chomp($line = <PARENT>); |
576 | print "Child Pid $$ just read this: `$line'\n"; |
577 | print PARENT "Child Pid $$ is sending this\n"; |
578 | close PARENT; |
579 | exit; |
580 | } |
581 | |
4633a7c4 |
582 | =head1 Sockets: Client/Server Communication |
a0d0e21e |
583 | |
6a3992aa |
584 | While not limited to Unix-derived operating systems (e.g., WinSock on PCs |
4633a7c4 |
585 | provides socket support, as do some VMS libraries), you may not have |
184e9718 |
586 | sockets on your system, in which case this section probably isn't going to do |
6a3992aa |
587 | you much good. With sockets, you can do both virtual circuits (i.e., TCP |
588 | streams) and datagrams (i.e., UDP packets). You may be able to do even more |
4633a7c4 |
589 | depending on your system. |
590 | |
591 | The Perl function calls for dealing with sockets have the same names as |
592 | the corresponding system calls in C, but their arguments tend to differ |
593 | for two reasons: first, Perl filehandles work differently than C file |
594 | descriptors. Second, Perl already knows the length of its strings, so you |
595 | don't need to pass that information. |
a0d0e21e |
596 | |
4633a7c4 |
597 | One of the major problems with old socket code in Perl was that it used |
598 | hard-coded values for some of the constants, which severely hurt |
599 | portability. If you ever see code that does anything like explicitly |
600 | setting C<$AF_INET = 2>, you know you're in for big trouble: An |
601 | immeasurably superior approach is to use the C<Socket> module, which more |
602 | reliably grants access to various constants and functions you'll need. |
a0d0e21e |
603 | |
68dc0745 |
604 | If you're not writing a server/client for an existing protocol like |
605 | NNTP or SMTP, you should give some thought to how your server will |
606 | know when the client has finished talking, and vice-versa. Most |
607 | protocols are based on one-line messages and responses (so one party |
4a6725af |
608 | knows the other has finished when a "\n" is received) or multi-line |
68dc0745 |
609 | messages and responses that end with a period on an empty line |
610 | ("\n.\n" terminates a message/response). |
611 | |
5a964f20 |
612 | =head2 Internet Line Terminators |
613 | |
614 | The Internet line terminator is "\015\012". Under ASCII variants of |
615 | Unix, that could usually be written as "\r\n", but under other systems, |
616 | "\r\n" might at times be "\015\015\012", "\012\012\015", or something |
617 | completely different. The standards specify writing "\015\012" to be |
618 | conformant (be strict in what you provide), but they also recommend |
619 | accepting a lone "\012" on input (but be lenient in what you require). |
620 | We haven't always been very good about that in the code in this manpage, |
621 | but unless you're on a Mac, you'll probably be ok. |
622 | |
4633a7c4 |
623 | =head2 Internet TCP Clients and Servers |
a0d0e21e |
624 | |
4633a7c4 |
625 | Use Internet-domain sockets when you want to do client-server |
626 | communication that might extend to machines outside of your own system. |
627 | |
628 | Here's a sample TCP client using Internet-domain sockets: |
629 | |
630 | #!/usr/bin/perl -w |
4633a7c4 |
631 | use strict; |
632 | use Socket; |
633 | my ($remote,$port, $iaddr, $paddr, $proto, $line); |
634 | |
635 | $remote = shift || 'localhost'; |
636 | $port = shift || 2345; # random port |
637 | if ($port =~ /\D/) { $port = getservbyname($port, 'tcp') } |
638 | die "No port" unless $port; |
639 | $iaddr = inet_aton($remote) || die "no host: $remote"; |
640 | $paddr = sockaddr_in($port, $iaddr); |
641 | |
642 | $proto = getprotobyname('tcp'); |
643 | socket(SOCK, PF_INET, SOCK_STREAM, $proto) || die "socket: $!"; |
644 | connect(SOCK, $paddr) || die "connect: $!"; |
54310121 |
645 | while (defined($line = <SOCK>)) { |
4633a7c4 |
646 | print $line; |
54310121 |
647 | } |
4633a7c4 |
648 | |
649 | close (SOCK) || die "close: $!"; |
650 | exit; |
651 | |
652 | And here's a corresponding server to go along with it. We'll |
653 | leave the address as INADDR_ANY so that the kernel can choose |
54310121 |
654 | the appropriate interface on multihomed hosts. If you want sit |
c07a80fd |
655 | on a particular interface (like the external side of a gateway |
656 | or firewall machine), you should fill this in with your real address |
657 | instead. |
658 | |
659 | #!/usr/bin/perl -Tw |
c07a80fd |
660 | use strict; |
661 | BEGIN { $ENV{PATH} = '/usr/ucb:/bin' } |
662 | use Socket; |
663 | use Carp; |
5a964f20 |
664 | $EOL = "\015\012"; |
c07a80fd |
665 | |
54310121 |
666 | sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" } |
c07a80fd |
667 | |
668 | my $port = shift || 2345; |
669 | my $proto = getprotobyname('tcp'); |
6a3992aa |
670 | $port = $1 if $port =~ /(\d+)/; # untaint port number |
671 | |
c07a80fd |
672 | socket(Server, PF_INET, SOCK_STREAM, $proto) || die "socket: $!"; |
54310121 |
673 | setsockopt(Server, SOL_SOCKET, SO_REUSEADDR, |
c07a80fd |
674 | pack("l", 1)) || die "setsockopt: $!"; |
675 | bind(Server, sockaddr_in($port, INADDR_ANY)) || die "bind: $!"; |
676 | listen(Server,SOMAXCONN) || die "listen: $!"; |
677 | |
678 | logmsg "server started on port $port"; |
679 | |
680 | my $paddr; |
681 | |
682 | $SIG{CHLD} = \&REAPER; |
683 | |
684 | for ( ; $paddr = accept(Client,Server); close Client) { |
685 | my($port,$iaddr) = sockaddr_in($paddr); |
686 | my $name = gethostbyaddr($iaddr,AF_INET); |
687 | |
54310121 |
688 | logmsg "connection from $name [", |
689 | inet_ntoa($iaddr), "] |
c07a80fd |
690 | at port $port"; |
691 | |
54310121 |
692 | print Client "Hello there, $name, it's now ", |
5a964f20 |
693 | scalar localtime, $EOL; |
54310121 |
694 | } |
c07a80fd |
695 | |
54310121 |
696 | And here's a multithreaded version. It's multithreaded in that |
697 | like most typical servers, it spawns (forks) a slave server to |
c07a80fd |
698 | handle the client request so that the master server can quickly |
699 | go back to service a new client. |
4633a7c4 |
700 | |
701 | #!/usr/bin/perl -Tw |
4633a7c4 |
702 | use strict; |
703 | BEGIN { $ENV{PATH} = '/usr/ucb:/bin' } |
a0d0e21e |
704 | use Socket; |
4633a7c4 |
705 | use Carp; |
5a964f20 |
706 | $EOL = "\015\012"; |
a0d0e21e |
707 | |
4633a7c4 |
708 | sub spawn; # forward declaration |
54310121 |
709 | sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" } |
a0d0e21e |
710 | |
4633a7c4 |
711 | my $port = shift || 2345; |
712 | my $proto = getprotobyname('tcp'); |
80aa6872 |
713 | $port = $1 if $port =~ /(\d+)/; # untaint port number |
54310121 |
714 | |
c07a80fd |
715 | socket(Server, PF_INET, SOCK_STREAM, $proto) || die "socket: $!"; |
54310121 |
716 | setsockopt(Server, SOL_SOCKET, SO_REUSEADDR, |
c07a80fd |
717 | pack("l", 1)) || die "setsockopt: $!"; |
718 | bind(Server, sockaddr_in($port, INADDR_ANY)) || die "bind: $!"; |
719 | listen(Server,SOMAXCONN) || die "listen: $!"; |
a0d0e21e |
720 | |
4633a7c4 |
721 | logmsg "server started on port $port"; |
a0d0e21e |
722 | |
4633a7c4 |
723 | my $waitedpid = 0; |
724 | my $paddr; |
a0d0e21e |
725 | |
54310121 |
726 | sub REAPER { |
4633a7c4 |
727 | $waitedpid = wait; |
6a3992aa |
728 | $SIG{CHLD} = \&REAPER; # loathe sysV |
4633a7c4 |
729 | logmsg "reaped $waitedpid" . ($? ? " with exit $?" : ''); |
730 | } |
731 | |
732 | $SIG{CHLD} = \&REAPER; |
733 | |
54310121 |
734 | for ( $waitedpid = 0; |
735 | ($paddr = accept(Client,Server)) || $waitedpid; |
736 | $waitedpid = 0, close Client) |
4633a7c4 |
737 | { |
6a3992aa |
738 | next if $waitedpid and not $paddr; |
4633a7c4 |
739 | my($port,$iaddr) = sockaddr_in($paddr); |
740 | my $name = gethostbyaddr($iaddr,AF_INET); |
741 | |
54310121 |
742 | logmsg "connection from $name [", |
743 | inet_ntoa($iaddr), "] |
4633a7c4 |
744 | at port $port"; |
a0d0e21e |
745 | |
54310121 |
746 | spawn sub { |
5a964f20 |
747 | print "Hello there, $name, it's now ", scalar localtime, $EOL; |
748 | exec '/usr/games/fortune' # XXX: `wrong' line terminators |
4633a7c4 |
749 | or confess "can't exec fortune: $!"; |
750 | }; |
a0d0e21e |
751 | |
54310121 |
752 | } |
a0d0e21e |
753 | |
4633a7c4 |
754 | sub spawn { |
755 | my $coderef = shift; |
a0d0e21e |
756 | |
54310121 |
757 | unless (@_ == 0 && $coderef && ref($coderef) eq 'CODE') { |
4633a7c4 |
758 | confess "usage: spawn CODEREF"; |
a0d0e21e |
759 | } |
4633a7c4 |
760 | |
761 | my $pid; |
762 | if (!defined($pid = fork)) { |
763 | logmsg "cannot fork: $!"; |
764 | return; |
765 | } elsif ($pid) { |
766 | logmsg "begat $pid"; |
6a3992aa |
767 | return; # I'm the parent |
4633a7c4 |
768 | } |
6a3992aa |
769 | # else I'm the child -- go spawn |
4633a7c4 |
770 | |
c07a80fd |
771 | open(STDIN, "<&Client") || die "can't dup client to stdin"; |
772 | open(STDOUT, ">&Client") || die "can't dup client to stdout"; |
4633a7c4 |
773 | ## open(STDERR, ">&STDOUT") || die "can't dup stdout to stderr"; |
774 | exit &$coderef(); |
54310121 |
775 | } |
4633a7c4 |
776 | |
777 | This server takes the trouble to clone off a child version via fork() for |
778 | each incoming request. That way it can handle many requests at once, |
779 | which you might not always want. Even if you don't fork(), the listen() |
780 | will allow that many pending connections. Forking servers have to be |
781 | particularly careful about cleaning up their dead children (called |
782 | "zombies" in Unix parlance), because otherwise you'll quickly fill up your |
783 | process table. |
784 | |
785 | We suggest that you use the B<-T> flag to use taint checking (see L<perlsec>) |
786 | even if we aren't running setuid or setgid. This is always a good idea |
787 | for servers and other programs run on behalf of someone else (like CGI |
788 | scripts), because it lessens the chances that people from the outside will |
789 | be able to compromise your system. |
790 | |
791 | Let's look at another TCP client. This one connects to the TCP "time" |
792 | service on a number of different machines and shows how far their clocks |
793 | differ from the system on which it's being run: |
794 | |
795 | #!/usr/bin/perl -w |
4633a7c4 |
796 | use strict; |
797 | use Socket; |
798 | |
799 | my $SECS_of_70_YEARS = 2208988800; |
54310121 |
800 | sub ctime { scalar localtime(shift) } |
4633a7c4 |
801 | |
54310121 |
802 | my $iaddr = gethostbyname('localhost'); |
803 | my $proto = getprotobyname('tcp'); |
804 | my $port = getservbyname('time', 'tcp'); |
4633a7c4 |
805 | my $paddr = sockaddr_in(0, $iaddr); |
806 | my($host); |
807 | |
808 | $| = 1; |
809 | printf "%-24s %8s %s\n", "localhost", 0, ctime(time()); |
810 | |
811 | foreach $host (@ARGV) { |
812 | printf "%-24s ", $host; |
813 | my $hisiaddr = inet_aton($host) || die "unknown host"; |
814 | my $hispaddr = sockaddr_in($port, $hisiaddr); |
815 | socket(SOCKET, PF_INET, SOCK_STREAM, $proto) || die "socket: $!"; |
816 | connect(SOCKET, $hispaddr) || die "bind: $!"; |
817 | my $rtime = ' '; |
818 | read(SOCKET, $rtime, 4); |
819 | close(SOCKET); |
820 | my $histime = unpack("N", $rtime) - $SECS_of_70_YEARS ; |
821 | printf "%8d %s\n", $histime - time, ctime($histime); |
a0d0e21e |
822 | } |
823 | |
4633a7c4 |
824 | =head2 Unix-Domain TCP Clients and Servers |
825 | |
a2eb9003 |
826 | That's fine for Internet-domain clients and servers, but what about local |
4633a7c4 |
827 | communications? While you can use the same setup, sometimes you don't |
828 | want to. Unix-domain sockets are local to the current host, and are often |
54310121 |
829 | used internally to implement pipes. Unlike Internet domain sockets, Unix |
4633a7c4 |
830 | domain sockets can show up in the file system with an ls(1) listing. |
831 | |
5a964f20 |
832 | % ls -l /dev/log |
4633a7c4 |
833 | srw-rw-rw- 1 root 0 Oct 31 07:23 /dev/log |
a0d0e21e |
834 | |
4633a7c4 |
835 | You can test for these with Perl's B<-S> file test: |
836 | |
837 | unless ( -S '/dev/log' ) { |
838 | die "something's wicked with the print system"; |
54310121 |
839 | } |
4633a7c4 |
840 | |
841 | Here's a sample Unix-domain client: |
842 | |
843 | #!/usr/bin/perl -w |
4633a7c4 |
844 | use Socket; |
845 | use strict; |
846 | my ($rendezvous, $line); |
847 | |
848 | $rendezvous = shift || '/tmp/catsock'; |
849 | socket(SOCK, PF_UNIX, SOCK_STREAM, 0) || die "socket: $!"; |
9607fc9c |
850 | connect(SOCK, sockaddr_un($rendezvous)) || die "connect: $!"; |
54310121 |
851 | while (defined($line = <SOCK>)) { |
4633a7c4 |
852 | print $line; |
54310121 |
853 | } |
4633a7c4 |
854 | exit; |
855 | |
5a964f20 |
856 | And here's a corresponding server. You don't have to worry about silly |
857 | network terminators here because Unix domain sockets are guaranteed |
858 | to be on the localhost, and thus everything works right. |
4633a7c4 |
859 | |
860 | #!/usr/bin/perl -Tw |
4633a7c4 |
861 | use strict; |
862 | use Socket; |
863 | use Carp; |
864 | |
865 | BEGIN { $ENV{PATH} = '/usr/ucb:/bin' } |
5a964f20 |
866 | sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" } |
4633a7c4 |
867 | |
868 | my $NAME = '/tmp/catsock'; |
869 | my $uaddr = sockaddr_un($NAME); |
870 | my $proto = getprotobyname('tcp'); |
871 | |
c07a80fd |
872 | socket(Server,PF_UNIX,SOCK_STREAM,0) || die "socket: $!"; |
4633a7c4 |
873 | unlink($NAME); |
c07a80fd |
874 | bind (Server, $uaddr) || die "bind: $!"; |
875 | listen(Server,SOMAXCONN) || die "listen: $!"; |
4633a7c4 |
876 | |
877 | logmsg "server started on $NAME"; |
878 | |
5a964f20 |
879 | my $waitedpid; |
880 | |
881 | sub REAPER { |
882 | $waitedpid = wait; |
883 | $SIG{CHLD} = \&REAPER; # loathe sysV |
884 | logmsg "reaped $waitedpid" . ($? ? " with exit $?" : ''); |
885 | } |
886 | |
4633a7c4 |
887 | $SIG{CHLD} = \&REAPER; |
888 | |
5a964f20 |
889 | |
54310121 |
890 | for ( $waitedpid = 0; |
891 | accept(Client,Server) || $waitedpid; |
892 | $waitedpid = 0, close Client) |
4633a7c4 |
893 | { |
894 | next if $waitedpid; |
895 | logmsg "connection on $NAME"; |
54310121 |
896 | spawn sub { |
4633a7c4 |
897 | print "Hello there, it's now ", scalar localtime, "\n"; |
898 | exec '/usr/games/fortune' or die "can't exec fortune: $!"; |
899 | }; |
54310121 |
900 | } |
4633a7c4 |
901 | |
902 | As you see, it's remarkably similar to the Internet domain TCP server, so |
903 | much so, in fact, that we've omitted several duplicate functions--spawn(), |
904 | logmsg(), ctime(), and REAPER()--which are exactly the same as in the |
905 | other server. |
906 | |
907 | So why would you ever want to use a Unix domain socket instead of a |
908 | simpler named pipe? Because a named pipe doesn't give you sessions. You |
909 | can't tell one process's data from another's. With socket programming, |
910 | you get a separate session for each client: that's why accept() takes two |
911 | arguments. |
912 | |
913 | For example, let's say that you have a long running database server daemon |
914 | that you want folks from the World Wide Web to be able to access, but only |
915 | if they go through a CGI interface. You'd have a small, simple CGI |
916 | program that does whatever checks and logging you feel like, and then acts |
917 | as a Unix-domain client and connects to your private server. |
918 | |
7b05b7e3 |
919 | =head1 TCP Clients with IO::Socket |
920 | |
921 | For those preferring a higher-level interface to socket programming, the |
922 | IO::Socket module provides an object-oriented approach. IO::Socket is |
923 | included as part of the standard Perl distribution as of the 5.004 |
924 | release. If you're running an earlier version of Perl, just fetch |
925 | IO::Socket from CPAN, where you'll also find find modules providing easy |
926 | interfaces to the following systems: DNS, FTP, Ident (RFC 931), NIS and |
927 | NISPlus, NNTP, Ping, POP3, SMTP, SNMP, SSLeay, Telnet, and Time--just |
928 | to name a few. |
929 | |
930 | =head2 A Simple Client |
931 | |
932 | Here's a client that creates a TCP connection to the "daytime" |
933 | service at port 13 of the host name "localhost" and prints out everything |
934 | that the server there cares to provide. |
935 | |
936 | #!/usr/bin/perl -w |
937 | use IO::Socket; |
938 | $remote = IO::Socket::INET->new( |
939 | Proto => "tcp", |
940 | PeerAddr => "localhost", |
941 | PeerPort => "daytime(13)", |
942 | ) |
943 | or die "cannot connect to daytime port at localhost"; |
944 | while ( <$remote> ) { print } |
945 | |
946 | When you run this program, you should get something back that |
947 | looks like this: |
948 | |
949 | Wed May 14 08:40:46 MDT 1997 |
950 | |
951 | Here are what those parameters to the C<new> constructor mean: |
952 | |
953 | =over |
954 | |
955 | =item C<Proto> |
956 | |
957 | This is which protocol to use. In this case, the socket handle returned |
958 | will be connected to a TCP socket, because we want a stream-oriented |
959 | connection, that is, one that acts pretty much like a plain old file. |
960 | Not all sockets are this of this type. For example, the UDP protocol |
961 | can be used to make a datagram socket, used for message-passing. |
962 | |
963 | =item C<PeerAddr> |
964 | |
965 | This is the name or Internet address of the remote host the server is |
966 | running on. We could have specified a longer name like C<"www.perl.com">, |
967 | or an address like C<"204.148.40.9">. For demonstration purposes, we've |
968 | used the special hostname C<"localhost">, which should always mean the |
969 | current machine you're running on. The corresponding Internet address |
970 | for localhost is C<"127.1">, if you'd rather use that. |
971 | |
972 | =item C<PeerPort> |
973 | |
974 | This is the service name or port number we'd like to connect to. |
975 | We could have gotten away with using just C<"daytime"> on systems with a |
976 | well-configured system services file,[FOOTNOTE: The system services file |
977 | is in I</etc/services> under Unix] but just in case, we've specified the |
978 | port number (13) in parentheses. Using just the number would also have |
979 | worked, but constant numbers make careful programmers nervous. |
980 | |
981 | =back |
982 | |
983 | Notice how the return value from the C<new> constructor is used as |
984 | a filehandle in the C<while> loop? That's what's called an indirect |
985 | filehandle, a scalar variable containing a filehandle. You can use |
986 | it the same way you would a normal filehandle. For example, you |
987 | can read one line from it this way: |
988 | |
989 | $line = <$handle>; |
990 | |
991 | all remaining lines from is this way: |
992 | |
993 | @lines = <$handle>; |
994 | |
995 | and send a line of data to it this way: |
996 | |
997 | print $handle "some data\n"; |
998 | |
999 | =head2 A Webget Client |
1000 | |
1001 | Here's a simple client that takes a remote host to fetch a document |
1002 | from, and then a list of documents to get from that host. This is a |
1003 | more interesting client than the previous one because it first sends |
1004 | something to the server before fetching the server's response. |
1005 | |
1006 | #!/usr/bin/perl -w |
1007 | use IO::Socket; |
1008 | unless (@ARGV > 1) { die "usage: $0 host document ..." } |
1009 | $host = shift(@ARGV); |
5a964f20 |
1010 | $EOL = "\015\012"; |
1011 | $BLANK = $EOL x 2; |
7b05b7e3 |
1012 | foreach $document ( @ARGV ) { |
1013 | $remote = IO::Socket::INET->new( Proto => "tcp", |
1014 | PeerAddr => $host, |
1015 | PeerPort => "http(80)", |
1016 | ); |
1017 | unless ($remote) { die "cannot connect to http daemon on $host" } |
1018 | $remote->autoflush(1); |
5a964f20 |
1019 | print $remote "GET $document HTTP/1.0" . $BLANK; |
7b05b7e3 |
1020 | while ( <$remote> ) { print } |
1021 | close $remote; |
1022 | } |
1023 | |
1024 | The web server handing the "http" service, which is assumed to be at |
1025 | its standard port, number 80. If your the web server you're trying to |
1026 | connect to is at a different port (like 1080 or 8080), you should specify |
c47ff5f1 |
1027 | as the named-parameter pair, C<< PeerPort => 8080 >>. The C<autoflush> |
7b05b7e3 |
1028 | method is used on the socket because otherwise the system would buffer |
1029 | up the output we sent it. (If you're on a Mac, you'll also need to |
1030 | change every C<"\n"> in your code that sends data over the network to |
1031 | be a C<"\015\012"> instead.) |
1032 | |
1033 | Connecting to the server is only the first part of the process: once you |
1034 | have the connection, you have to use the server's language. Each server |
1035 | on the network has its own little command language that it expects as |
1036 | input. The string that we send to the server starting with "GET" is in |
1037 | HTTP syntax. In this case, we simply request each specified document. |
1038 | Yes, we really are making a new connection for each document, even though |
1039 | it's the same host. That's the way you always used to have to speak HTTP. |
1040 | Recent versions of web browsers may request that the remote server leave |
1041 | the connection open a little while, but the server doesn't have to honor |
1042 | such a request. |
1043 | |
1044 | Here's an example of running that program, which we'll call I<webget>: |
1045 | |
5a964f20 |
1046 | % webget www.perl.com /guanaco.html |
7b05b7e3 |
1047 | HTTP/1.1 404 File Not Found |
1048 | Date: Thu, 08 May 1997 18:02:32 GMT |
1049 | Server: Apache/1.2b6 |
1050 | Connection: close |
1051 | Content-type: text/html |
1052 | |
1053 | <HEAD><TITLE>404 File Not Found</TITLE></HEAD> |
1054 | <BODY><H1>File Not Found</H1> |
1055 | The requested URL /guanaco.html was not found on this server.<P> |
1056 | </BODY> |
1057 | |
1058 | Ok, so that's not very interesting, because it didn't find that |
1059 | particular document. But a long response wouldn't have fit on this page. |
1060 | |
1061 | For a more fully-featured version of this program, you should look to |
1062 | the I<lwp-request> program included with the LWP modules from CPAN. |
1063 | |
1064 | =head2 Interactive Client with IO::Socket |
1065 | |
1066 | Well, that's all fine if you want to send one command and get one answer, |
1067 | but what about setting up something fully interactive, somewhat like |
1068 | the way I<telnet> works? That way you can type a line, get the answer, |
1069 | type a line, get the answer, etc. |
1070 | |
1071 | This client is more complicated than the two we've done so far, but if |
1072 | you're on a system that supports the powerful C<fork> call, the solution |
1073 | isn't that rough. Once you've made the connection to whatever service |
1074 | you'd like to chat with, call C<fork> to clone your process. Each of |
1075 | these two identical process has a very simple job to do: the parent |
1076 | copies everything from the socket to standard output, while the child |
1077 | simultaneously copies everything from standard input to the socket. |
1078 | To accomplish the same thing using just one process would be I<much> |
1079 | harder, because it's easier to code two processes to do one thing than it |
1080 | is to code one process to do two things. (This keep-it-simple principle |
5a964f20 |
1081 | a cornerstones of the Unix philosophy, and good software engineering as |
1082 | well, which is probably why it's spread to other systems.) |
7b05b7e3 |
1083 | |
1084 | Here's the code: |
1085 | |
1086 | #!/usr/bin/perl -w |
1087 | use strict; |
1088 | use IO::Socket; |
1089 | my ($host, $port, $kidpid, $handle, $line); |
1090 | |
1091 | unless (@ARGV == 2) { die "usage: $0 host port" } |
1092 | ($host, $port) = @ARGV; |
1093 | |
1094 | # create a tcp connection to the specified host and port |
1095 | $handle = IO::Socket::INET->new(Proto => "tcp", |
1096 | PeerAddr => $host, |
1097 | PeerPort => $port) |
1098 | or die "can't connect to port $port on $host: $!"; |
1099 | |
1100 | $handle->autoflush(1); # so output gets there right away |
1101 | print STDERR "[Connected to $host:$port]\n"; |
1102 | |
1103 | # split the program into two processes, identical twins |
1104 | die "can't fork: $!" unless defined($kidpid = fork()); |
1105 | |
1106 | # the if{} block runs only in the parent process |
1107 | if ($kidpid) { |
1108 | # copy the socket to standard output |
1109 | while (defined ($line = <$handle>)) { |
1110 | print STDOUT $line; |
1111 | } |
1112 | kill("TERM", $kidpid); # send SIGTERM to child |
1113 | } |
1114 | # the else{} block runs only in the child process |
1115 | else { |
1116 | # copy standard input to the socket |
1117 | while (defined ($line = <STDIN>)) { |
1118 | print $handle $line; |
1119 | } |
1120 | } |
1121 | |
1122 | The C<kill> function in the parent's C<if> block is there to send a |
1123 | signal to our child process (current running in the C<else> block) |
1124 | as soon as the remote server has closed its end of the connection. |
1125 | |
7b05b7e3 |
1126 | If the remote server sends data a byte at time, and you need that |
1127 | data immediately without waiting for a newline (which might not happen), |
1128 | you may wish to replace the C<while> loop in the parent with the |
1129 | following: |
1130 | |
1131 | my $byte; |
1132 | while (sysread($handle, $byte, 1) == 1) { |
1133 | print STDOUT $byte; |
1134 | } |
1135 | |
1136 | Making a system call for each byte you want to read is not very efficient |
1137 | (to put it mildly) but is the simplest to explain and works reasonably |
1138 | well. |
1139 | |
1140 | =head1 TCP Servers with IO::Socket |
1141 | |
5a964f20 |
1142 | As always, setting up a server is little bit more involved than running a client. |
7b05b7e3 |
1143 | The model is that the server creates a special kind of socket that |
1144 | does nothing but listen on a particular port for incoming connections. |
c47ff5f1 |
1145 | It does this by calling the C<< IO::Socket::INET->new() >> method with |
7b05b7e3 |
1146 | slightly different arguments than the client did. |
1147 | |
1148 | =over |
1149 | |
1150 | =item Proto |
1151 | |
1152 | This is which protocol to use. Like our clients, we'll |
1153 | still specify C<"tcp"> here. |
1154 | |
1155 | =item LocalPort |
1156 | |
1157 | We specify a local |
1158 | port in the C<LocalPort> argument, which we didn't do for the client. |
1159 | This is service name or port number for which you want to be the |
1160 | server. (Under Unix, ports under 1024 are restricted to the |
1161 | superuser.) In our sample, we'll use port 9000, but you can use |
1162 | any port that's not currently in use on your system. If you try |
1163 | to use one already in used, you'll get an "Address already in use" |
19799a22 |
1164 | message. Under Unix, the C<netstat -a> command will show |
7b05b7e3 |
1165 | which services current have servers. |
1166 | |
1167 | =item Listen |
1168 | |
1169 | The C<Listen> parameter is set to the maximum number of |
1170 | pending connections we can accept until we turn away incoming clients. |
1171 | Think of it as a call-waiting queue for your telephone. |
1172 | The low-level Socket module has a special symbol for the system maximum, which |
1173 | is SOMAXCONN. |
1174 | |
1175 | =item Reuse |
1176 | |
1177 | The C<Reuse> parameter is needed so that we restart our server |
1178 | manually without waiting a few minutes to allow system buffers to |
1179 | clear out. |
1180 | |
1181 | =back |
1182 | |
1183 | Once the generic server socket has been created using the parameters |
1184 | listed above, the server then waits for a new client to connect |
1185 | to it. The server blocks in the C<accept> method, which eventually an |
1186 | bidirectional connection to the remote client. (Make sure to autoflush |
1187 | this handle to circumvent buffering.) |
1188 | |
1189 | To add to user-friendliness, our server prompts the user for commands. |
1190 | Most servers don't do this. Because of the prompt without a newline, |
1191 | you'll have to use the C<sysread> variant of the interactive client above. |
1192 | |
1193 | This server accepts one of five different commands, sending output |
1194 | back to the client. Note that unlike most network servers, this one |
1195 | only handles one incoming client at a time. Multithreaded servers are |
f83494b9 |
1196 | covered in Chapter 6 of the Camel. |
7b05b7e3 |
1197 | |
1198 | Here's the code. We'll |
1199 | |
1200 | #!/usr/bin/perl -w |
1201 | use IO::Socket; |
1202 | use Net::hostent; # for OO version of gethostbyaddr |
1203 | |
1204 | $PORT = 9000; # pick something not in use |
1205 | |
1206 | $server = IO::Socket::INET->new( Proto => 'tcp', |
1207 | LocalPort => $PORT, |
1208 | Listen => SOMAXCONN, |
1209 | Reuse => 1); |
1210 | |
1211 | die "can't setup server" unless $server; |
1212 | print "[Server $0 accepting clients]\n"; |
1213 | |
1214 | while ($client = $server->accept()) { |
1215 | $client->autoflush(1); |
1216 | print $client "Welcome to $0; type help for command list.\n"; |
1217 | $hostinfo = gethostbyaddr($client->peeraddr); |
1218 | printf "[Connect from %s]\n", $hostinfo->name || $client->peerhost; |
1219 | print $client "Command? "; |
1220 | while ( <$client>) { |
1221 | next unless /\S/; # blank line |
1222 | if (/quit|exit/i) { last; } |
1223 | elsif (/date|time/i) { printf $client "%s\n", scalar localtime; } |
1224 | elsif (/who/i ) { print $client `who 2>&1`; } |
1225 | elsif (/cookie/i ) { print $client `/usr/games/fortune 2>&1`; } |
1226 | elsif (/motd/i ) { print $client `cat /etc/motd 2>&1`; } |
1227 | else { |
1228 | print $client "Commands: quit date who cookie motd\n"; |
1229 | } |
1230 | } continue { |
1231 | print $client "Command? "; |
1232 | } |
1233 | close $client; |
1234 | } |
1235 | |
1236 | =head1 UDP: Message Passing |
4633a7c4 |
1237 | |
1238 | Another kind of client-server setup is one that uses not connections, but |
1239 | messages. UDP communications involve much lower overhead but also provide |
1240 | less reliability, as there are no promises that messages will arrive at |
1241 | all, let alone in order and unmangled. Still, UDP offers some advantages |
1242 | over TCP, including being able to "broadcast" or "multicast" to a whole |
1243 | bunch of destination hosts at once (usually on your local subnet). If you |
1244 | find yourself overly concerned about reliability and start building checks |
6a3992aa |
1245 | into your message system, then you probably should use just TCP to start |
4633a7c4 |
1246 | with. |
1247 | |
1248 | Here's a UDP program similar to the sample Internet TCP client given |
7b05b7e3 |
1249 | earlier. However, instead of checking one host at a time, the UDP version |
4633a7c4 |
1250 | will check many of them asynchronously by simulating a multicast and then |
1251 | using select() to do a timed-out wait for I/O. To do something similar |
1252 | with TCP, you'd have to use a different socket handle for each host. |
1253 | |
1254 | #!/usr/bin/perl -w |
1255 | use strict; |
4633a7c4 |
1256 | use Socket; |
1257 | use Sys::Hostname; |
1258 | |
54310121 |
1259 | my ( $count, $hisiaddr, $hispaddr, $histime, |
1260 | $host, $iaddr, $paddr, $port, $proto, |
4633a7c4 |
1261 | $rin, $rout, $rtime, $SECS_of_70_YEARS); |
1262 | |
1263 | $SECS_of_70_YEARS = 2208988800; |
1264 | |
1265 | $iaddr = gethostbyname(hostname()); |
1266 | $proto = getprotobyname('udp'); |
1267 | $port = getservbyname('time', 'udp'); |
1268 | $paddr = sockaddr_in(0, $iaddr); # 0 means let kernel pick |
1269 | |
1270 | socket(SOCKET, PF_INET, SOCK_DGRAM, $proto) || die "socket: $!"; |
1271 | bind(SOCKET, $paddr) || die "bind: $!"; |
1272 | |
1273 | $| = 1; |
1274 | printf "%-12s %8s %s\n", "localhost", 0, scalar localtime time; |
1275 | $count = 0; |
1276 | for $host (@ARGV) { |
1277 | $count++; |
1278 | $hisiaddr = inet_aton($host) || die "unknown host"; |
1279 | $hispaddr = sockaddr_in($port, $hisiaddr); |
1280 | defined(send(SOCKET, 0, 0, $hispaddr)) || die "send $host: $!"; |
1281 | } |
1282 | |
1283 | $rin = ''; |
1284 | vec($rin, fileno(SOCKET), 1) = 1; |
1285 | |
1286 | # timeout after 10.0 seconds |
1287 | while ($count && select($rout = $rin, undef, undef, 10.0)) { |
1288 | $rtime = ''; |
1289 | ($hispaddr = recv(SOCKET, $rtime, 4, 0)) || die "recv: $!"; |
1290 | ($port, $hisiaddr) = sockaddr_in($hispaddr); |
1291 | $host = gethostbyaddr($hisiaddr, AF_INET); |
1292 | $histime = unpack("N", $rtime) - $SECS_of_70_YEARS ; |
1293 | printf "%-12s ", $host; |
1294 | printf "%8d %s\n", $histime - time, scalar localtime($histime); |
1295 | $count--; |
1296 | } |
1297 | |
1298 | =head1 SysV IPC |
1299 | |
1300 | While System V IPC isn't so widely used as sockets, it still has some |
1301 | interesting uses. You can't, however, effectively use SysV IPC or |
1302 | Berkeley mmap() to have shared memory so as to share a variable amongst |
1303 | several processes. That's because Perl would reallocate your string when |
1304 | you weren't wanting it to. |
1305 | |
54310121 |
1306 | Here's a small example showing shared memory usage. |
a0d0e21e |
1307 | |
0ade1984 |
1308 | use IPC::SysV qw(IPC_PRIVATE IPC_RMID S_IRWXU S_IRWXG S_IRWXO); |
1309 | |
a0d0e21e |
1310 | $size = 2000; |
0ade1984 |
1311 | $key = shmget(IPC_PRIVATE, $size, S_IRWXU|S_IRWXG|S_IRWXO) || die "$!"; |
1312 | print "shm key $key\n"; |
a0d0e21e |
1313 | |
1314 | $message = "Message #1"; |
0ade1984 |
1315 | shmwrite($key, $message, 0, 60) || die "$!"; |
1316 | print "wrote: '$message'\n"; |
1317 | shmread($key, $buff, 0, 60) || die "$!"; |
1318 | print "read : '$buff'\n"; |
a0d0e21e |
1319 | |
0ade1984 |
1320 | # the buffer of shmread is zero-character end-padded. |
1321 | substr($buff, index($buff, "\0")) = ''; |
1322 | print "un" unless $buff eq $message; |
1323 | print "swell\n"; |
a0d0e21e |
1324 | |
0ade1984 |
1325 | print "deleting shm $key\n"; |
1326 | shmctl($key, IPC_RMID, 0) || die "$!"; |
a0d0e21e |
1327 | |
1328 | Here's an example of a semaphore: |
1329 | |
0ade1984 |
1330 | use IPC::SysV qw(IPC_CREAT); |
1331 | |
a0d0e21e |
1332 | $IPC_KEY = 1234; |
0ade1984 |
1333 | $key = semget($IPC_KEY, 10, 0666 | IPC_CREAT ) || die "$!"; |
1334 | print "shm key $key\n"; |
a0d0e21e |
1335 | |
a2eb9003 |
1336 | Put this code in a separate file to be run in more than one process. |
a0d0e21e |
1337 | Call the file F<take>: |
1338 | |
1339 | # create a semaphore |
1340 | |
1341 | $IPC_KEY = 1234; |
1342 | $key = semget($IPC_KEY, 0 , 0 ); |
1343 | die if !defined($key); |
1344 | |
1345 | $semnum = 0; |
1346 | $semflag = 0; |
1347 | |
1348 | # 'take' semaphore |
1349 | # wait for semaphore to be zero |
1350 | $semop = 0; |
1351 | $opstring1 = pack("sss", $semnum, $semop, $semflag); |
1352 | |
1353 | # Increment the semaphore count |
1354 | $semop = 1; |
1355 | $opstring2 = pack("sss", $semnum, $semop, $semflag); |
1356 | $opstring = $opstring1 . $opstring2; |
1357 | |
1358 | semop($key,$opstring) || die "$!"; |
1359 | |
a2eb9003 |
1360 | Put this code in a separate file to be run in more than one process. |
a0d0e21e |
1361 | Call this file F<give>: |
1362 | |
4633a7c4 |
1363 | # 'give' the semaphore |
a0d0e21e |
1364 | # run this in the original process and you will see |
1365 | # that the second process continues |
1366 | |
1367 | $IPC_KEY = 1234; |
1368 | $key = semget($IPC_KEY, 0, 0); |
1369 | die if !defined($key); |
1370 | |
1371 | $semnum = 0; |
1372 | $semflag = 0; |
1373 | |
1374 | # Decrement the semaphore count |
1375 | $semop = -1; |
1376 | $opstring = pack("sss", $semnum, $semop, $semflag); |
1377 | |
1378 | semop($key,$opstring) || die "$!"; |
1379 | |
7b05b7e3 |
1380 | The SysV IPC code above was written long ago, and it's definitely |
0ade1984 |
1381 | clunky looking. For a more modern look, see the IPC::SysV module |
1382 | which is included with Perl starting from Perl 5.005. |
4633a7c4 |
1383 | |
1384 | =head1 NOTES |
1385 | |
5a964f20 |
1386 | Most of these routines quietly but politely return C<undef> when they |
1387 | fail instead of causing your program to die right then and there due to |
1388 | an uncaught exception. (Actually, some of the new I<Socket> conversion |
1389 | functions croak() on bad arguments.) It is therefore essential to |
1390 | check return values from these functions. Always begin your socket |
1391 | programs this way for optimal success, and don't forget to add B<-T> |
1392 | taint checking flag to the #! line for servers: |
4633a7c4 |
1393 | |
5a964f20 |
1394 | #!/usr/bin/perl -Tw |
4633a7c4 |
1395 | use strict; |
1396 | use sigtrap; |
1397 | use Socket; |
1398 | |
1399 | =head1 BUGS |
1400 | |
1401 | All these routines create system-specific portability problems. As noted |
1402 | elsewhere, Perl is at the mercy of your C libraries for much of its system |
1403 | behaviour. It's probably safest to assume broken SysV semantics for |
6a3992aa |
1404 | signals and to stick with simple TCP and UDP socket operations; e.g., don't |
a2eb9003 |
1405 | try to pass open file descriptors over a local UDP datagram socket if you |
4633a7c4 |
1406 | want your code to stand a chance of being portable. |
1407 | |
5a964f20 |
1408 | As mentioned in the signals section, because few vendors provide C |
1409 | libraries that are safely re-entrant, the prudent programmer will do |
1410 | little else within a handler beyond setting a numeric variable that |
1411 | already exists; or, if locked into a slow (restarting) system call, |
1412 | using die() to raise an exception and longjmp(3) out. In fact, even |
1413 | these may in some cases cause a core dump. It's probably best to avoid |
1414 | signals except where they are absolutely inevitable. This |
1415 | will be addressed in a future release of Perl. |
4633a7c4 |
1416 | |
1417 | =head1 AUTHOR |
1418 | |
1419 | Tom Christiansen, with occasional vestiges of Larry Wall's original |
7b05b7e3 |
1420 | version and suggestions from the Perl Porters. |
4633a7c4 |
1421 | |
1422 | =head1 SEE ALSO |
1423 | |
7b05b7e3 |
1424 | There's a lot more to networking than this, but this should get you |
1425 | started. |
1426 | |
5a964f20 |
1427 | For intrepid programmers, the indispensable textbook is I<Unix Network |
1428 | Programming> by W. Richard Stevens (published by Addison-Wesley). Note |
1429 | that most books on networking address networking from the perspective of |
1430 | a C programmer; translation to Perl is left as an exercise for the reader. |
7b05b7e3 |
1431 | |
1432 | The IO::Socket(3) manpage describes the object library, and the Socket(3) |
1433 | manpage describes the low-level interface to sockets. Besides the obvious |
1434 | functions in L<perlfunc>, you should also check out the F<modules> file |
1435 | at your nearest CPAN site. (See L<perlmodlib> or best yet, the F<Perl |
1436 | FAQ> for a description of what CPAN is and where to get it.) |
1437 | |
4633a7c4 |
1438 | Section 5 of the F<modules> file is devoted to "Networking, Device Control |
6a3992aa |
1439 | (modems), and Interprocess Communication", and contains numerous unbundled |
4633a7c4 |
1440 | modules numerous networking modules, Chat and Expect operations, CGI |
1441 | programming, DCE, FTP, IPC, NNTP, Proxy, Ptty, RPC, SNMP, SMTP, Telnet, |
1442 | Threads, and ToolTalk--just to name a few. |