3 perlfaq8 - System Interaction ($Revision: 1.14 $, $Date: 2002/11/10 17:35:47 $)
7 This section of the Perl FAQ covers questions involving operating
8 system interaction. Topics include interprocess communication (IPC),
9 control over the user-interface (keyboard, screen and pointing
10 devices), and most anything else not related to data manipulation.
12 Read the FAQs and documentation specific to the port of perl to your
13 operating system (eg, L<perlvms>, L<perlplan9>, ...). These should
14 contain more detailed information on the vagaries of your perl.
16 =head2 How do I find out which operating system I'm running under?
18 The $^O variable ($OSNAME if you use English) contains an indication of
19 the name of the operating system (not its release number) that your perl
22 =head2 How come exec() doesn't return?
24 Because that's what it does: it replaces your currently running
25 program with a different one. If you want to keep going (as is
26 probably the case if you're asking this question) use system()
29 =head2 How do I do fancy stuff with the keyboard/screen/mouse?
31 How you access/control keyboards, screens, and pointing devices
32 ("mice") is system-dependent. Try the following modules:
38 Term::Cap Standard perl distribution
40 Term::ReadLine::Gnu CPAN
41 Term::ReadLine::Perl CPAN
46 Term::Cap Standard perl distribution
56 Some of these specific cases are shown below.
58 =head2 How do I print something out in color?
60 In general, you don't, because you don't know whether
61 the recipient has a color-aware display device. If you
62 know that they have an ANSI terminal that understands
63 color, you can use the Term::ANSIColor module from CPAN:
66 print color("red"), "Stop!\n", color("reset");
67 print color("green"), "Go!\n", color("reset");
71 use Term::ANSIColor qw(:constants);
72 print RED, "Stop!\n", RESET;
73 print GREEN, "Go!\n", RESET;
75 =head2 How do I read just one key without waiting for a return key?
77 Controlling input buffering is a remarkably system-dependent matter.
78 On many systems, you can just use the B<stty> command as shown in
79 L<perlfunc/getc>, but as you see, that's already getting you into
82 open(TTY, "+</dev/tty") or die "no tty: $!";
83 system "stty cbreak </dev/tty >/dev/tty 2>&1";
84 $key = getc(TTY); # perhaps this works
86 sysread(TTY, $key, 1); # probably this does
87 system "stty -cbreak </dev/tty >/dev/tty 2>&1";
89 The Term::ReadKey module from CPAN offers an easy-to-use interface that
90 should be more efficient than shelling out to B<stty> for each key.
91 It even includes limited support for Windows.
98 However, using the code requires that you have a working C compiler
99 and can use it to build and install a CPAN module. Here's a solution
100 using the standard POSIX module, which is already on your systems
101 (assuming your system supports POSIX).
106 And here's the HotKey module, which hides the somewhat mystifying calls
107 to manipulate the POSIX termios structures.
113 @EXPORT = qw(cbreak cooked readkey);
116 use POSIX qw(:termios_h);
117 my ($term, $oterm, $echo, $noecho, $fd_stdin);
119 $fd_stdin = fileno(STDIN);
120 $term = POSIX::Termios->new();
121 $term->getattr($fd_stdin);
122 $oterm = $term->getlflag();
124 $echo = ECHO | ECHOK | ICANON;
125 $noecho = $oterm & ~$echo;
128 $term->setlflag($noecho); # ok, so i don't want echo either
129 $term->setcc(VTIME, 1);
130 $term->setattr($fd_stdin, TCSANOW);
134 $term->setlflag($oterm);
135 $term->setcc(VTIME, 0);
136 $term->setattr($fd_stdin, TCSANOW);
142 sysread(STDIN, $key, 1);
151 =head2 How do I check whether input is ready on the keyboard?
153 The easiest way to do this is to read a key in nonblocking mode with the
154 Term::ReadKey module from CPAN, passing it an argument of -1 to indicate
161 if (defined ($char = ReadKey(-1)) ) {
162 # input was waiting and it was $char
164 # no input was waiting
167 ReadMode('normal'); # restore normal tty settings
169 =head2 How do I clear the screen?
171 If you only have do so infrequently, use C<system>:
175 If you have to do this a lot, save the clear string
176 so you can print it 100 times without calling a program
179 $clear_string = `clear`;
182 If you're planning on doing other screen manipulations, like cursor
183 positions, etc, you might wish to use Term::Cap module:
186 $terminal = Term::Cap->Tgetent( {OSPEED => 9600} );
187 $clear_string = $terminal->Tputs('cl');
189 =head2 How do I get the screen size?
191 If you have Term::ReadKey module installed from CPAN,
192 you can use it to fetch the width and height in characters
196 ($wchar, $hchar, $wpixels, $hpixels) = GetTerminalSize();
198 This is more portable than the raw C<ioctl>, but not as
201 require 'sys/ioctl.ph';
202 die "no TIOCGWINSZ " unless defined &TIOCGWINSZ;
203 open(TTY, "+</dev/tty") or die "No tty: $!";
204 unless (ioctl(TTY, &TIOCGWINSZ, $winsize='')) {
205 die sprintf "$0: ioctl TIOCGWINSZ (%08x: $!)\n", &TIOCGWINSZ;
207 ($row, $col, $xpixel, $ypixel) = unpack('S4', $winsize);
208 print "(row,col) = ($row,$col)";
209 print " (xpixel,ypixel) = ($xpixel,$ypixel)" if $xpixel || $ypixel;
212 =head2 How do I ask the user for a password?
214 (This question has nothing to do with the web. See a different
217 There's an example of this in L<perlfunc/crypt>). First, you put the
218 terminal into "no echo" mode, then just read the password normally.
219 You may do this with an old-style ioctl() function, POSIX terminal
220 control (see L<POSIX> or its documentation the Camel Book), or a call
221 to the B<stty> program, with varying degrees of portability.
223 You can also do this for most systems using the Term::ReadKey module
224 from CPAN, which is easier to use and in theory more portable.
229 $password = ReadLine(0);
231 =head2 How do I read and write the serial port?
233 This depends on which operating system your program is running on. In
234 the case of Unix, the serial ports will be accessible through files in
235 /dev; on other systems, device names will doubtless differ.
236 Several problem areas common to all device interaction are the
243 Your system may use lockfiles to control multiple access. Make sure
244 you follow the correct protocol. Unpredictable behavior can result
245 from multiple processes reading from one device.
249 If you expect to use both read and write operations on the device,
250 you'll have to open it for update (see L<perlfunc/"open"> for
251 details). You may wish to open it without running the risk of
252 blocking by using sysopen() and C<O_RDWR|O_NDELAY|O_NOCTTY> from the
253 Fcntl module (part of the standard perl distribution). See
254 L<perlfunc/"sysopen"> for more on this approach.
258 Some devices will be expecting a "\r" at the end of each line rather
259 than a "\n". In some ports of perl, "\r" and "\n" are different from
260 their usual (Unix) ASCII values of "\012" and "\015". You may have to
261 give the numeric values you want directly, using octal ("\015"), hex
262 ("0x0D"), or as a control-character specification ("\cM").
264 print DEV "atv1\012"; # wrong, for some devices
265 print DEV "atv1\015"; # right, for some devices
267 Even though with normal text files a "\n" will do the trick, there is
268 still no unified scheme for terminating a line that is portable
269 between Unix, DOS/Win, and Macintosh, except to terminate I<ALL> line
270 ends with "\015\012", and strip what you don't need from the output.
271 This applies especially to socket I/O and autoflushing, discussed
274 =item flushing output
276 If you expect characters to get to your device when you print() them,
277 you'll want to autoflush that filehandle. You can use select()
278 and the C<$|> variable to control autoflushing (see L<perlvar/$|>
279 and L<perlfunc/select>, or L<perlfaq5>, ``How do I flush/unbuffer an
280 output filehandle? Why must I do this?''):
286 You'll also see code that does this without a temporary variable, as in
288 select((select(DEV), $| = 1)[0]);
290 Or if you don't mind pulling in a few thousand lines
291 of code just because you're afraid of a little $| variable:
296 As mentioned in the previous item, this still doesn't work when using
297 socket I/O between Unix and Macintosh. You'll need to hard code your
298 line terminators, in that case.
300 =item non-blocking input
302 If you are doing a blocking read() or sysread(), you'll have to
303 arrange for an alarm handler to provide a timeout (see
304 L<perlfunc/alarm>). If you have a non-blocking open, you'll likely
305 have a non-blocking read, which means you may have to use a 4-arg
306 select() to determine whether I/O is ready on that device (see
307 L<perlfunc/"select">.
311 While trying to read from his caller-id box, the notorious Jamie Zawinski
312 <jwz@netscape.com>, after much gnashing of teeth and fighting with sysread,
313 sysopen, POSIX's tcgetattr business, and various other functions that
314 go bump in the night, finally came up with this:
318 my $stty = `/bin/stty -g`;
319 open2( \*MODEM_IN, \*MODEM_OUT, "cu -l$modem_device -s2400 2>&1");
320 # starting cu hoses /dev/tty's stty settings, even when it has
321 # been opened on a pipe...
322 system("/bin/stty $stty");
325 if ( !m/^Connected/ ) {
326 print STDERR "$0: cu printed `$_' instead of `Connected'\n";
330 =head2 How do I decode encrypted password files?
332 You spend lots and lots of money on dedicated hardware, but this is
333 bound to get you talked about.
335 Seriously, you can't if they are Unix password files--the Unix
336 password system employs one-way encryption. It's more like hashing than
337 encryption. The best you can check is whether something else hashes to
338 the same string. You can't turn a hash back into the original string.
340 can forcibly (and intelligently) try to guess passwords, but don't
341 (can't) guarantee quick success.
343 If you're worried about users selecting bad passwords, you should
344 proactively check when they try to change their password (by modifying
345 passwd(1), for example).
347 =head2 How do I start a process in the background?
349 Several modules can start other processes that do not block
350 your Perl program. You can use IPC::Open3, Parallel::Jobs,
351 IPC::Run, and some of the POE modules. See CPAN for more
358 or you could use fork as documented in L<perlfunc/"fork">, with
359 further examples in L<perlipc>. Some things to be aware of, if you're
360 on a Unix-like system:
364 =item STDIN, STDOUT, and STDERR are shared
366 Both the main process and the backgrounded one (the "child" process)
367 share the same STDIN, STDOUT and STDERR filehandles. If both try to
368 access them at once, strange things can happen. You may want to close
369 or reopen these for the child. You can get around this with
370 C<open>ing a pipe (see L<perlfunc/"open">) but on some systems this
371 means that the child process cannot outlive the parent.
375 You'll have to catch the SIGCHLD signal, and possibly SIGPIPE too.
376 SIGCHLD is sent when the backgrounded process finishes. SIGPIPE is
377 sent when you write to a filehandle whose child process has closed (an
378 untrapped SIGPIPE can cause your program to silently die). This is
379 not an issue with C<system("cmd&")>.
383 You have to be prepared to "reap" the child process when it finishes.
385 $SIG{CHLD} = sub { wait };
387 $SIG{CHLD} = 'IGNORE';
389 You can also use a double fork. You immediately wait() for your
390 first child, and the init daemon will wait() for your grandchild once
393 unless ($pid = fork) {
395 exec "what you really wanna do";
403 See L<perlipc/"Signals"> for other examples of code to do this.
404 Zombies are not an issue with C<system("prog &")>.
408 =head2 How do I trap control characters/signals?
410 You don't actually "trap" a control character. Instead, that character
411 generates a signal which is sent to your terminal's currently
412 foregrounded process group, which you then trap in your process.
413 Signals are documented in L<perlipc/"Signals"> and the
414 section on ``Signals'' in the Camel.
416 Be warned that very few C libraries are re-entrant. Therefore, if you
417 attempt to print() in a handler that got invoked during another stdio
418 operation your internal structures will likely be in an
419 inconsistent state, and your program will dump core. You can
420 sometimes avoid this by using syswrite() instead of print().
422 Unless you're exceedingly careful, the only safe things to do inside a
423 signal handler are (1) set a variable and (2) exit. In the first case,
424 you should only set a variable in such a way that malloc() is not
425 called (eg, by setting a variable that already has a value).
429 $Interrupted = 0; # to ensure it has a value
432 syswrite(STDERR, "ouch\n", 5);
435 However, because syscalls restart by default, you'll find that if
436 you're in a "slow" call, such as <FH>, read(), connect(), or
437 wait(), that the only way to terminate them is by "longjumping" out;
438 that is, by raising an exception. See the time-out handler for a
439 blocking flock() in L<perlipc/"Signals"> or the section on ``Signals''
442 =head2 How do I modify the shadow password file on a Unix system?
444 If perl was installed correctly and your shadow library was written
445 properly, the getpw*() functions described in L<perlfunc> should in
446 theory provide (read-only) access to entries in the shadow password
447 file. To change the file, make a new shadow password file (the format
448 varies from system to system--see L<passwd(5)> for specifics) and use
449 pwd_mkdb(8) to install it (see L<pwd_mkdb(8)> for more details).
451 =head2 How do I set the time and date?
453 Assuming you're running under sufficient permissions, you should be
454 able to set the system-wide date and time by running the date(1)
455 program. (There is no way to set the time and date on a per-process
456 basis.) This mechanism will work for Unix, MS-DOS, Windows, and NT;
457 the VMS equivalent is C<set time>.
459 However, if all you want to do is change your time zone, you can
460 probably get away with setting an environment variable:
462 $ENV{TZ} = "MST7MDT"; # unixish
463 $ENV{'SYS$TIMEZONE_DIFFERENTIAL'}="-5" # vms
464 system "trn comp.lang.perl.misc";
466 =head2 How can I sleep() or alarm() for under a second?
468 If you want finer granularity than the 1 second that the sleep()
469 function provides, the easiest way is to use the select() function as
470 documented in L<perlfunc/"select">. Try the Time::HiRes and
471 the BSD::Itimer modules (available from CPAN, and starting from
472 Perl 5.8 Time::HiRes is part of the standard distribution).
474 =head2 How can I measure time under a second?
476 In general, you may not be able to. The Time::HiRes module (available
477 from CPAN, and starting from Perl 5.8 part of the standard distribution)
478 provides this functionality for some systems.
480 If your system supports both the syscall() function in Perl as well as
481 a system call like gettimeofday(2), then you may be able to do
484 require 'sys/syscall.ph';
488 $done = $start = pack($TIMEVAL_T, ());
490 syscall(&SYS_gettimeofday, $start, 0) != -1
491 or die "gettimeofday: $!";
493 ##########################
494 # DO YOUR OPERATION HERE #
495 ##########################
497 syscall( &SYS_gettimeofday, $done, 0) != -1
498 or die "gettimeofday: $!";
500 @start = unpack($TIMEVAL_T, $start);
501 @done = unpack($TIMEVAL_T, $done);
504 for ($done[1], $start[1]) { $_ /= 1_000_000 }
506 $delta_time = sprintf "%.4f", ($done[0] + $done[1] )
508 ($start[0] + $start[1] );
510 =head2 How can I do an atexit() or setjmp()/longjmp()? (Exception handling)
512 Release 5 of Perl added the END block, which can be used to simulate
513 atexit(). Each package's END block is called when the program or
514 thread ends (see L<perlmod> manpage for more details).
516 For example, you can use this to make sure your filter program
517 managed to finish its output without filling up the disk:
520 close(STDOUT) || die "stdout close failed: $!";
523 The END block isn't called when untrapped signals kill the program,
524 though, so if you use END blocks you should also use
526 use sigtrap qw(die normal-signals);
528 Perl's exception-handling mechanism is its eval() operator. You can
529 use eval() as setjmp and die() as longjmp. For details of this, see
530 the section on signals, especially the time-out handler for a blocking
531 flock() in L<perlipc/"Signals"> or the section on ``Signals'' in
534 If exception handling is all you're interested in, try the
535 exceptions.pl library (part of the standard perl distribution).
537 If you want the atexit() syntax (and an rmexit() as well), try the
538 AtExit module available from CPAN.
540 =head2 Why doesn't my sockets program work under System V (Solaris)? What does the error message "Protocol not supported" mean?
542 Some Sys-V based systems, notably Solaris 2.X, redefined some of the
543 standard socket constants. Since these were constant across all
544 architectures, they were often hardwired into perl code. The proper
545 way to deal with this is to "use Socket" to get the correct values.
547 Note that even though SunOS and Solaris are binary compatible, these
548 values are different. Go figure.
550 =head2 How can I call my system's unique C functions from Perl?
552 In most cases, you write an external module to do it--see the answer
553 to "Where can I learn about linking C with Perl? [h2xs, xsubpp]".
554 However, if the function is a system call, and your system supports
555 syscall(), you can use the syscall function (documented in
558 Remember to check the modules that came with your distribution, and
559 CPAN as well--someone may already have written a module to do it.
561 =head2 Where do I get the include files to do ioctl() or syscall()?
563 Historically, these would be generated by the h2ph tool, part of the
564 standard perl distribution. This program converts cpp(1) directives
565 in C header files to files containing subroutine definitions, like
566 &SYS_getitimer, which you can use as arguments to your functions.
567 It doesn't work perfectly, but it usually gets most of the job done.
568 Simple files like F<errno.h>, F<syscall.h>, and F<socket.h> were fine,
569 but the hard ones like F<ioctl.h> nearly always need to hand-edited.
570 Here's how to install the *.ph files:
576 If your system supports dynamic loading, for reasons of portability and
577 sanity you probably ought to use h2xs (also part of the standard perl
578 distribution). This tool converts C header files to Perl extensions.
579 See L<perlxstut> for how to get started with h2xs.
581 If your system doesn't support dynamic loading, you still probably
582 ought to use h2xs. See L<perlxstut> and L<ExtUtils::MakeMaker> for
583 more information (in brief, just use B<make perl> instead of a plain
584 B<make> to rebuild perl with a new static extension).
586 =head2 Why do setuid perl scripts complain about kernel problems?
588 Some operating systems have bugs in the kernel that make setuid
589 scripts inherently insecure. Perl gives you a number of options
590 (described in L<perlsec>) to work around such systems.
592 =head2 How can I open a pipe both to and from a command?
594 The IPC::Open2 module (part of the standard perl distribution) is an
595 easy-to-use approach that internally uses pipe(), fork(), and exec() to do
596 the job. Make sure you read the deadlock warnings in its documentation,
597 though (see L<IPC::Open2>). See
598 L<perlipc/"Bidirectional Communication with Another Process"> and
599 L<perlipc/"Bidirectional Communication with Yourself">
601 You may also use the IPC::Open3 module (part of the standard perl
602 distribution), but be warned that it has a different order of
603 arguments from IPC::Open2 (see L<IPC::Open3>).
605 =head2 Why can't I get the output of a command with system()?
607 You're confusing the purpose of system() and backticks (``). system()
608 runs a command and returns exit status information (as a 16 bit value:
609 the low 7 bits are the signal the process died from, if any, and
610 the high 8 bits are the actual exit value). Backticks (``) run a
611 command and return what it sent to STDOUT.
613 $exit_status = system("mail-users");
614 $output_string = `ls`;
616 =head2 How can I capture STDERR from an external command?
618 There are three basic ways of running external commands:
620 system $cmd; # using system()
621 $output = `$cmd`; # using backticks (``)
622 open (PIPE, "cmd |"); # using open()
624 With system(), both STDOUT and STDERR will go the same place as the
625 script's STDOUT and STDERR, unless the system() command redirects them.
626 Backticks and open() read B<only> the STDOUT of your command.
628 You can also use the open3() function from IPC::Open3. Benjamin
629 Goldberg provides some sample code:
631 To capture a program's STDOUT, but discard its STDERR:
635 use Symbol qw(gensym);
636 open(NULL, ">", File::Spec->devnull);
637 my $pid = open3(gensym, \*PH, ">&NULL", "cmd");
641 To capture a program's STDERR, but discard its STDOUT:
645 use Symbol qw(gensym);
646 open(NULL, ">", File::Spec->devnull);
647 my $pid = open3(gensym, ">&NULL", \*PH, "cmd");
651 To capture a program's STDERR, and let its STDOUT go to our own STDERR:
654 use Symbol qw(gensym);
655 my $pid = open3(gensym, ">&STDERR", \*PH, "cmd");
659 To read both a command's STDOUT and its STDERR separately, you can
660 redirect them to temp files, let the command run, then read the temp
664 use Symbol qw(gensym);
666 local *CATCHOUT = IO::File->new_tempfile;
667 local *CATCHERR = IO::File->new_tempfile;
668 my $pid = open3(gensym, ">&CATCHOUT", ">&CATCHERR", "cmd");
670 seek $_, 0, 0 for \*CATCHOUT, \*CATCHERR;
671 while( <CATCHOUT> ) {}
672 while( <CATCHERR> ) {}
674 But there's no real need for *both* to be tempfiles... the following
675 should work just as well, without deadlocking:
678 use Symbol qw(gensym);
680 local *CATCHERR = IO::File->new_tempfile;
681 my $pid = open3(gensym, \*CATCHOUT, ">&CATCHERR", "cmd");
682 while( <CATCHOUT> ) {}
685 while( <CATCHERR> ) {}
687 And it'll be faster, too, since we can begin processing the program's
688 stdout immediately, rather than waiting for the program to finish.
690 With any of these, you can change file descriptors before the call:
692 open(STDOUT, ">logfile");
695 or you can use Bourne shell file-descriptor redirection:
697 $output = `$cmd 2>some_file`;
698 open (PIPE, "cmd 2>some_file |");
700 You can also use file-descriptor redirection to make STDERR a
703 $output = `$cmd 2>&1`;
704 open (PIPE, "cmd 2>&1 |");
706 Note that you I<cannot> simply open STDERR to be a dup of STDOUT
707 in your Perl program and avoid calling the shell to do the redirection.
710 open(STDERR, ">&STDOUT");
711 $alloutput = `cmd args`; # stderr still escapes
713 This fails because the open() makes STDERR go to where STDOUT was
714 going at the time of the open(). The backticks then make STDOUT go to
715 a string, but don't change STDERR (which still goes to the old
718 Note that you I<must> use Bourne shell (sh(1)) redirection syntax in
719 backticks, not csh(1)! Details on why Perl's system() and backtick
720 and pipe opens all use the Bourne shell are in the
721 F<versus/csh.whynot> article in the "Far More Than You Ever Wanted To
722 Know" collection in http://www.cpan.org/misc/olddoc/FMTEYEWTK.tgz . To
723 capture a command's STDERR and STDOUT together:
725 $output = `cmd 2>&1`; # either with backticks
726 $pid = open(PH, "cmd 2>&1 |"); # or with an open pipe
727 while (<PH>) { } # plus a read
729 To capture a command's STDOUT but discard its STDERR:
731 $output = `cmd 2>/dev/null`; # either with backticks
732 $pid = open(PH, "cmd 2>/dev/null |"); # or with an open pipe
733 while (<PH>) { } # plus a read
735 To capture a command's STDERR but discard its STDOUT:
737 $output = `cmd 2>&1 1>/dev/null`; # either with backticks
738 $pid = open(PH, "cmd 2>&1 1>/dev/null |"); # or with an open pipe
739 while (<PH>) { } # plus a read
741 To exchange a command's STDOUT and STDERR in order to capture the STDERR
742 but leave its STDOUT to come out our old STDERR:
744 $output = `cmd 3>&1 1>&2 2>&3 3>&-`; # either with backticks
745 $pid = open(PH, "cmd 3>&1 1>&2 2>&3 3>&-|");# or with an open pipe
746 while (<PH>) { } # plus a read
748 To read both a command's STDOUT and its STDERR separately, it's easiest
749 and safest to redirect them separately to files, and then read from those
750 files when the program is done:
752 system("program args 1>/tmp/program.stdout 2>/tmp/program.stderr");
754 Ordering is important in all these examples. That's because the shell
755 processes file descriptor redirections in strictly left to right order.
757 system("prog args 1>tmpfile 2>&1");
758 system("prog args 2>&1 1>tmpfile");
760 The first command sends both standard out and standard error to the
761 temporary file. The second command sends only the old standard output
762 there, and the old standard error shows up on the old standard out.
764 =head2 Why doesn't open() return an error when a pipe open fails?
766 If the second argument to a piped open() contains shell
767 metacharacters, perl fork()s, then exec()s a shell to decode the
768 metacharacters and eventually run the desired program. If the program
769 couldn't be run, it's the shell that gets the message, not Perl. All
770 your Perl program can find out is whether the shell itself could be
771 successfully started. You can still capture the shell's STDERR and
772 check it for error messages. See L<"How can I capture STDERR from an
773 external command?"> elsewhere in this document, or use the
776 If there are no shell metacharacters in the argument of open(), Perl
777 runs the command directly, without using the shell, and can correctly
778 report whether the command started.
780 =head2 What's wrong with using backticks in a void context?
782 Strictly speaking, nothing. Stylistically speaking, it's not a good
783 way to write maintainable code. Perl has several operators for
784 running external commands. Backticks are one; they collect the output
785 from the command for use in your program. The C<system> function is
786 another; it doesn't do this.
788 Writing backticks in your program sends a clear message to the readers
789 of your code that you wanted to collect the output of the command.
790 Why send a clear message that isn't true?
796 You forgot to check C<$?> to see whether the program even ran
797 correctly. Even if you wrote
799 print `cat /etc/termcap`;
801 this code could and probably should be written as
803 system("cat /etc/termcap") == 0
804 or die "cat program failed!";
806 which will get the output quickly (as it is generated, instead of only
807 at the end) and also check the return value.
809 system() also provides direct control over whether shell wildcard
810 processing may take place, whereas backticks do not.
812 =head2 How can I call backticks without shell processing?
814 This is a bit tricky. You can't simply write the command
817 @ok = `grep @opts '$search_string' @filenames`;
819 As of Perl 5.8.0, you can use open() with multiple arguments.
820 Just like the list forms of system() and exec(), no shell
823 open( GREP, "-|", 'grep', @opts, $search_string, @filenames );
830 if (open(GREP, "-|")) {
837 exec 'grep', @opts, $search_string, @filenames;
840 Just as with system(), no shell escapes happen when you exec() a list.
841 Further examples of this can be found in L<perlipc/"Safe Pipe Opens">.
843 Note that if you're use Microsoft, no solution to this vexing issue
844 is even possible. Even if Perl were to emulate fork(), you'd still
845 be stuck, because Microsoft does not have a argc/argv-style API.
847 =head2 Why can't my script read from STDIN after I gave it EOF (^D on Unix, ^Z on MS-DOS)?
849 Some stdio's set error and eof flags that need clearing. The
850 POSIX module defines clearerr() that you can use. That is the
851 technically correct way to do it. Here are some less reliable
858 Try keeping around the seekpointer and go there, like this:
861 seek(LOG, $where, 0);
865 If that doesn't work, try seeking to a different part of the file and
870 If that doesn't work, try seeking to a different part of
871 the file, reading something, and then seeking back.
875 If that doesn't work, give up on your stdio package and use sysread.
879 =head2 How can I convert my shell script to perl?
881 Learn Perl and rewrite it. Seriously, there's no simple converter.
882 Things that are awkward to do in the shell are easy to do in Perl, and
883 this very awkwardness is what would make a shell->perl converter
884 nigh-on impossible to write. By rewriting it, you'll think about what
885 you're really trying to do, and hopefully will escape the shell's
886 pipeline datastream paradigm, which while convenient for some matters,
887 causes many inefficiencies.
889 =head2 Can I use perl to run a telnet or ftp session?
891 Try the Net::FTP, TCP::Client, and Net::Telnet modules (available from
892 CPAN). http://www.cpan.org/scripts/netstuff/telnet.emul.shar
893 will also help for emulating the telnet protocol, but Net::Telnet is
894 quite probably easier to use..
896 If all you want to do is pretend to be telnet but don't need
897 the initial telnet handshaking, then the standard dual-process
898 approach will suffice:
900 use IO::Socket; # new in 5.004
901 $handle = IO::Socket::INET->new('www.perl.com:80')
902 || die "can't connect to port 80 on www.perl.com: $!";
903 $handle->autoflush(1);
904 if (fork()) { # XXX: undef means failure
906 print while <STDIN>; # everything from stdin to socket
908 print while <$handle>; # everything from socket to stdout
913 =head2 How can I write expect in Perl?
915 Once upon a time, there was a library called chat2.pl (part of the
916 standard perl distribution), which never really got finished. If you
917 find it somewhere, I<don't use it>. These days, your best bet is to
918 look at the Expect module available from CPAN, which also requires two
919 other modules from CPAN, IO::Pty and IO::Stty.
921 =head2 Is there a way to hide perl's command line from programs such as "ps"?
923 First of all note that if you're doing this for security reasons (to
924 avoid people seeing passwords, for example) then you should rewrite
925 your program so that critical information is never given as an
926 argument. Hiding the arguments won't make your program completely
929 To actually alter the visible command line, you can assign to the
930 variable $0 as documented in L<perlvar>. This won't work on all
931 operating systems, though. Daemon programs like sendmail place their
934 $0 = "orcus [accepting connections]";
936 =head2 I {changed directory, modified my environment} in a perl script. How come the change disappeared when I exited the script? How do I get my changes to be visible?
942 In the strictest sense, it can't be done--the script executes as a
943 different process from the shell it was started from. Changes to a
944 process are not reflected in its parent--only in any children
945 created after the change. There is shell magic that may allow you to
946 fake it by eval()ing the script's output in your shell; check out the
947 comp.unix.questions FAQ for details.
951 =head2 How do I close a process's filehandle without waiting for it to complete?
953 Assuming your system supports such things, just send an appropriate signal
954 to the process (see L<perlfunc/"kill">). It's common to first send a TERM
955 signal, wait a little bit, and then send a KILL signal to finish it off.
957 =head2 How do I fork a daemon process?
959 If by daemon process you mean one that's detached (disassociated from
960 its tty), then the following process is reported to work on most
961 Unixish systems. Non-Unix users should check their Your_OS::Process
962 module for other solutions.
968 Open /dev/tty and use the TIOCNOTTY ioctl on it. See L<tty(4)>
969 for details. Or better yet, you can just use the POSIX::setsid()
970 function, so you don't have to worry about process groups.
974 Change directory to /
978 Reopen STDIN, STDOUT, and STDERR so they're not connected to the old
983 Background yourself like this:
989 The Proc::Daemon module, available from CPAN, provides a function to
990 perform these actions for you.
992 =head2 How do I find out if I'm running interactively or not?
994 Good question. Sometimes C<-t STDIN> and C<-t STDOUT> can give clues,
997 if (-t STDIN && -t STDOUT) {
1001 On POSIX systems, you can test whether your own process group matches
1002 the current process group of your controlling terminal as follows:
1004 use POSIX qw/getpgrp tcgetpgrp/;
1005 open(TTY, "/dev/tty") or die $!;
1006 $tpgrp = tcgetpgrp(fileno(*TTY));
1008 if ($tpgrp == $pgrp) {
1009 print "foreground\n";
1011 print "background\n";
1014 =head2 How do I timeout a slow event?
1016 Use the alarm() function, probably in conjunction with a signal
1017 handler, as documented in L<perlipc/"Signals"> and the section on
1018 ``Signals'' in the Camel. You may instead use the more flexible
1019 Sys::AlarmCall module available from CPAN.
1021 The alarm() function is not implemented on all versions of Windows.
1022 Check the documentation for your specific version of Perl.
1024 =head2 How do I set CPU limits?
1026 Use the BSD::Resource module from CPAN.
1028 =head2 How do I avoid zombies on a Unix system?
1030 Use the reaper code from L<perlipc/"Signals"> to call wait() when a
1031 SIGCHLD is received, or else use the double-fork technique described
1032 in L<perlfaq8/"How do I start a process in the background?">.
1034 =head2 How do I use an SQL database?
1036 There are a number of excellent interfaces to SQL databases. See the
1037 DBD::* modules available from http://www.cpan.org/modules/by-module/DBD/ .
1038 A lot of information on this can be found at http://dbi.perl.org/
1040 =head2 How do I make a system() exit on control-C?
1042 You can't. You need to imitate the system() call (see L<perlipc> for
1043 sample code) and then have a signal handler for the INT signal that
1044 passes the signal on to the subprocess. Or you can check for it:
1047 if ($rc & 127) { die "signal death" }
1049 =head2 How do I open a file without blocking?
1051 If you're lucky enough to be using a system that supports
1052 non-blocking reads (most Unixish systems do), you need only to use the
1053 O_NDELAY or O_NONBLOCK flag from the Fcntl module in conjunction with
1057 sysopen(FH, "/tmp/somefile", O_WRONLY|O_NDELAY|O_CREAT, 0644)
1058 or die "can't open /tmp/somefile: $!":
1060 =head2 How do I install a module from CPAN?
1062 The easiest way is to have a module also named CPAN do it for you.
1063 This module comes with perl version 5.004 and later.
1065 $ perl -MCPAN -e shell
1067 cpan shell -- CPAN exploration and modules installation (v1.59_54)
1068 ReadLine support enabled
1070 cpan> install Some::Module
1072 To manually install the CPAN module, or any well-behaved CPAN module
1073 for that matter, follow these steps:
1079 Unpack the source into a temporary area.
1099 If your version of perl is compiled without dynamic loading, then you
1100 just need to replace step 3 (B<make>) with B<make perl> and you will
1101 get a new F<perl> binary with your extension linked in.
1103 See L<ExtUtils::MakeMaker> for more details on building extensions.
1104 See also the next question, ``What's the difference between require
1107 =head2 What's the difference between require and use?
1109 Perl offers several different ways to include code from one file into
1110 another. Here are the deltas between the various inclusion constructs:
1112 1) do $file is like eval `cat $file`, except the former
1113 1.1: searches @INC and updates %INC.
1114 1.2: bequeaths an *unrelated* lexical scope on the eval'ed code.
1116 2) require $file is like do $file, except the former
1117 2.1: checks for redundant loading, skipping already loaded files.
1118 2.2: raises an exception on failure to find, compile, or execute $file.
1120 3) require Module is like require "Module.pm", except the former
1121 3.1: translates each "::" into your system's directory separator.
1122 3.2: primes the parser to disambiguate class Module as an indirect object.
1124 4) use Module is like require Module, except the former
1125 4.1: loads the module at compile time, not run-time.
1126 4.2: imports symbols and semantics from that package to the current one.
1128 In general, you usually want C<use> and a proper Perl module.
1130 =head2 How do I keep my own module/library directory?
1132 When you build modules, use the PREFIX option when generating
1135 perl Makefile.PL PREFIX=/u/mydir/perl
1137 then either set the PERL5LIB environment variable before you run
1138 scripts that use the modules/libraries (see L<perlrun>) or say
1140 use lib '/u/mydir/perl';
1142 This is almost the same as
1145 unshift(@INC, '/u/mydir/perl');
1148 except that the lib module checks for machine-dependent subdirectories.
1149 See Perl's L<lib> for more information.
1151 =head2 How do I add the directory my program lives in to the module/library search path?
1154 use lib "$FindBin::Bin";
1155 use your_own_modules;
1157 =head2 How do I add a directory to my include path at runtime?
1159 Here are the suggested ways of modifying your include path:
1161 the PERLLIB environment variable
1162 the PERL5LIB environment variable
1163 the perl -Idir command line flag
1164 the use lib pragma, as in
1165 use lib "$ENV{HOME}/myown_perllib";
1167 The latter is particularly useful because it knows about machine
1168 dependent architectures. The lib.pm pragmatic module was first
1169 included with the 5.002 release of Perl.
1171 =head2 What is socket.ph and where do I get it?
1173 It's a perl4-style file defining values for system networking
1174 constants. Sometimes it is built using h2ph when Perl is installed,
1175 but other times it is not. Modern programs C<use Socket;> instead.
1177 =head1 AUTHOR AND COPYRIGHT
1179 Copyright (c) 1997-2002 Tom Christiansen and Nathan Torkington.
1180 All rights reserved.
1182 This documentation is free; you can redistribute it and/or modify it
1183 under the same terms as Perl itself.
1185 Irrespective of its distribution, all code examples in this file
1186 are hereby placed into the public domain. You are permitted and
1187 encouraged to use this code in your own programs for fun
1188 or for profit as you see fit. A simple comment in the code giving
1189 credit would be courteous but is not required.