1IPC::Cmd(3)           User Contributed Perl Documentation          IPC::Cmd(3)
2
3
4

NAME

6       IPC::Cmd - finding and running system commands made easy
7

SYNOPSIS

9           use IPC::Cmd qw[can_run run run_forked];
10
11           my $full_path = can_run('wget') or warn 'wget is not installed!';
12
13           ### commands can be arrayrefs or strings ###
14           my $cmd = "$full_path -b theregister.co.uk";
15           my $cmd = [$full_path, '-b', 'theregister.co.uk'];
16
17           ### in scalar context ###
18           my $buffer;
19           if( scalar run( command => $cmd,
20                           verbose => 0,
21                           buffer  => \$buffer,
22                           timeout => 20 )
23           ) {
24               print "fetched webpage successfully: $buffer\n";
25           }
26
27
28           ### in list context ###
29           my( $success, $error_message, $full_buf, $stdout_buf, $stderr_buf ) =
30                   run( command => $cmd, verbose => 0 );
31
32           if( $success ) {
33               print "this is what the command printed:\n";
34               print join "", @$full_buf;
35           }
36
37           ### run_forked example ###
38           my $result = run_forked("$full_path -q -O - theregister.co.uk", {'timeout' => 20});
39           if ($result->{'exit_code'} eq 0 && !$result->{'timeout'}) {
40               print "this is what wget returned:\n";
41               print $result->{'stdout'};
42           }
43
44           ### check for features
45           print "IPC::Open3 available: "  . IPC::Cmd->can_use_ipc_open3;
46           print "IPC::Run available: "    . IPC::Cmd->can_use_ipc_run;
47           print "Can capture buffer: "    . IPC::Cmd->can_capture_buffer;
48
49           ### don't have IPC::Cmd be verbose, ie don't print to stdout or
50           ### stderr when running commands -- default is '0'
51           $IPC::Cmd::VERBOSE = 0;
52

DESCRIPTION

54       IPC::Cmd allows you to run commands platform independently,
55       interactively if desired, but have them still work.
56
57       The "can_run" function can tell you if a certain binary is installed
58       and if so where, whereas the "run" function can actually execute any of
59       the commands you give it and give you a clear return value, as well as
60       adhere to your verbosity settings.
61

CLASS METHODS

63   $ipc_run_version = IPC::Cmd->can_use_ipc_run( [VERBOSE] )
64       Utility function that tells you if "IPC::Run" is available.  If the
65       "verbose" flag is passed, it will print diagnostic messages if IPC::Run
66       can not be found or loaded.
67
68   $ipc_open3_version = IPC::Cmd->can_use_ipc_open3( [VERBOSE] )
69       Utility function that tells you if "IPC::Open3" is available.  If the
70       verbose flag is passed, it will print diagnostic messages if
71       "IPC::Open3" can not be found or loaded.
72
73   $bool = IPC::Cmd->can_capture_buffer
74       Utility function that tells you if "IPC::Cmd" is capable of capturing
75       buffers in it's current configuration.
76
77   $bool = IPC::Cmd->can_use_run_forked
78       Utility function that tells you if "IPC::Cmd" is capable of providing
79       "run_forked" on the current platform.
80

FUNCTIONS

82   $path = can_run( PROGRAM );
83       "can_run" takes only one argument: the name of a binary you wish to
84       locate. "can_run" works much like the unix binary "which" or the bash
85       command "type", which scans through your path, looking for the
86       requested binary.
87
88       Unlike "which" and "type", this function is platform independent and
89       will also work on, for example, Win32.
90
91       If called in a scalar context it will return the full path to the
92       binary you asked for if it was found, or "undef" if it was not.
93
94       If called in a list context and the global variable $INSTANCES is a
95       true value, it will return a list of the full paths to instances of the
96       binary where found in "PATH", or an empty list if it was not found.
97
98   $ok | ($ok, $err, $full_buf, $stdout_buff, $stderr_buff) = run( command =>
99       COMMAND, [verbose => BOOL, buffer => \$SCALAR, timeout => DIGIT] );
100       "run" takes 4 arguments:
101
102       command
103           This is the command to execute. It may be either a string or an
104           array reference.  This is a required argument.
105
106           See "Caveats" for remarks on how commands are parsed and their
107           limitations.
108
109       verbose
110           This controls whether all output of a command should also be
111           printed to STDOUT/STDERR or should only be trapped in buffers
112           (NOTE: buffers require IPC::Run to be installed, or your system
113           able to work with IPC::Open3).
114
115           It will default to the global setting of $IPC::Cmd::VERBOSE, which
116           by default is 0.
117
118       buffer
119           This will hold all the output of a command. It needs to be a
120           reference to a scalar.  Note that this will hold both the STDOUT
121           and STDERR messages, and you have no way of telling which is which.
122           If you require this distinction, run the "run" command in list
123           context and inspect the individual buffers.
124
125           Of course, this requires that the underlying call supports buffers.
126           See the note on buffers above.
127
128       timeout
129           Sets the maximum time the command is allowed to run before
130           aborting, using the built-in alarm() call. If the timeout is
131           triggered, the "errorcode" in the return value will be set to an
132           object of the "IPC::Cmd::TimeOut" class. See the "error message"
133           section below for details.
134
135           Defaults to 0, meaning no timeout is set.
136
137       "run" will return a simple "true" or "false" when called in scalar
138       context.  In list context, you will be returned a list of the following
139       items:
140
141       success
142           A simple boolean indicating if the command executed without errors
143           or not.
144
145       error message
146           If the first element of the return value ("success") was 0, then
147           some error occurred. This second element is the error message the
148           command you requested exited with, if available. This is generally
149           a pretty printed value of $? or $@. See "perldoc perlvar" for
150           details on what they can contain.  If the error was a timeout, the
151           "error message" will be prefixed with the string
152           "IPC::Cmd::TimeOut", the timeout class.
153
154       full_buffer
155           This is an array reference containing all the output the command
156           generated.  Note that buffers are only available if you have
157           IPC::Run installed, or if your system is able to work with
158           IPC::Open3 -- see below).  Otherwise, this element will be "undef".
159
160       out_buffer
161           This is an array reference containing all the output sent to STDOUT
162           the command generated. The notes from "full_buffer" apply.
163
164       error_buffer
165           This is an arrayreference containing all the output sent to STDERR
166           the command generated. The notes from "full_buffer" apply.
167
168       See the "HOW IT WORKS" section below to see how "IPC::Cmd" decides what
169       modules or function calls to use when issuing a command.
170
171   $hashref = run_forked( COMMAND, { child_stdin => SCALAR, timeout => DIGIT,
172       stdout_handler => CODEREF, stderr_handler => CODEREF} );
173       "run_forked" is used to execute some program or a coderef, optionally
174       feed it with some input, get its return code and output (both stdout
175       and stderr into separate buffers).  In addition, it allows to terminate
176       the program if it takes too long to finish.
177
178       The important and distinguishing feature of run_forked is execution
179       timeout which at first seems to be quite a simple task but if you think
180       that the program which you're spawning might spawn some children itself
181       (which in their turn could do the same and so on) it turns out to be
182       not a simple issue.
183
184       "run_forked" is designed to survive and successfully terminate almost
185       any long running task, even a fork bomb in case your system has the
186       resources to survive during given timeout.
187
188       This is achieved by creating separate watchdog process which spawns the
189       specified program in a separate process session and supervises it:
190       optionally feeds it with input, stores its exit code, stdout and
191       stderr, terminates it in case it runs longer than specified.
192
193       Invocation requires the command to be executed or a coderef and
194       optionally a hashref of options:
195
196       "timeout"
197           Specify in seconds how long to run the command before it is killed
198           with SIG_KILL (9), which effectively terminates it and all of its
199           children (direct or indirect).
200
201       "child_stdin"
202           Specify some text that will be passed into the "STDIN" of the
203           executed program.
204
205       "stdout_handler"
206           Coderef of a subroutine to call when a portion of data is received
207           on STDOUT from the executing program.
208
209       "stderr_handler"
210           Coderef of a subroutine to call when a portion of data is received
211           on STDERR from the executing program.
212
213       "wait_loop_callback"
214           Coderef of a subroutine to call inside of the main waiting loop
215           (while "run_forked" waits for the external to finish or fail).  It
216           is useful to stop running external process before it ends by
217           itself, e.g.
218
219             my $r = run_forked("some external command", {
220                     'wait_loop_callback' => sub {
221                     if (condition) {
222                         kill(1, $$);
223                     }
224                     },
225                     'terminate_on_signal' => 'HUP',
226                     });
227
228           Combined with "stdout_handler" and "stderr_handler" allows
229           terminating external command based on its output. Could also be
230           used as a timer without engaging with alarm (signals).
231
232           Remember that this code could be called every millisecond
233           (depending on the output which external command generates), so try
234           to make it as lightweight as possible.
235
236       "discard_output"
237           Discards the buffering of the standard output and standard errors
238           for return by run_forked().  With this option you have to use the
239           std*_handlers to read what the command outputs.  Useful for
240           commands that send a lot of output.
241
242       "terminate_on_parent_sudden_death"
243           Enable this option if you wish all spawned processes to be killed
244           if the initially spawned process (the parent) is killed or dies
245           without waiting for child processes.
246
247       "run_forked" will return a HASHREF with the following keys:
248
249       "exit_code"
250           The exit code of the executed program.
251
252       "timeout"
253           The number of seconds the program ran for before being terminated,
254           or 0 if no timeout occurred.
255
256       "stdout"
257           Holds the standard output of the executed command (or empty string
258           if there was no STDOUT output or if "discard_output" was used; it's
259           always defined!)
260
261       "stderr"
262           Holds the standard error of the executed command (or empty string
263           if there was no STDERR output or if "discard_output" was used; it's
264           always defined!)
265
266       "merged"
267           Holds the standard output and error of the executed command merged
268           into one stream (or empty string if there was no output at all or
269           if "discard_output" was used; it's always defined!)
270
271       "err_msg"
272           Holds some explanation in the case of an error.
273
274   $q = QUOTE
275       Returns the character used for quoting strings on this platform. This
276       is usually a "'" (single quote) on most systems, but some systems use
277       different quotes. For example, "Win32" uses """ (double quote).
278
279       You can use it as follows:
280
281         use IPC::Cmd qw[run QUOTE];
282         my $cmd = q[echo ] . QUOTE . q[foo bar] . QUOTE;
283
284       This makes sure that "foo bar" is treated as a string, rather than two
285       separate arguments to the "echo" function.
286

HOW IT WORKS

288       "run" will try to execute your command using the following logic:
289
290       •   If you have "IPC::Run" installed, and the variable
291           $IPC::Cmd::USE_IPC_RUN is set to true (See the "Global Variables"
292           section) use that to execute the command. You will have the full
293           output available in buffers, interactive commands are sure to work
294           and you are guaranteed to have your verbosity settings honored
295           cleanly.
296
297       •   Otherwise, if the variable $IPC::Cmd::USE_IPC_OPEN3 is set to true
298           (See the "Global Variables" section), try to execute the command
299           using IPC::Open3. Buffers will be available on all platforms,
300           interactive commands will still execute cleanly, and also your
301           verbosity settings will be adhered to nicely;
302
303       •   Otherwise, if you have the "verbose" argument set to true, we fall
304           back to a simple system() call. We cannot capture any buffers, but
305           interactive commands will still work.
306
307       •   Otherwise we will try and temporarily redirect STDERR and STDOUT,
308           do a system() call with your command and then re-open STDERR and
309           STDOUT.  This is the method of last resort and will still allow you
310           to execute your commands cleanly. However, no buffers will be
311           available.
312

Global Variables

314       The behaviour of IPC::Cmd can be altered by changing the following
315       global variables:
316
317   $IPC::Cmd::VERBOSE
318       This controls whether IPC::Cmd will print any output from the commands
319       to the screen or not. The default is 0.
320
321   $IPC::Cmd::USE_IPC_RUN
322       This variable controls whether IPC::Cmd will try to use IPC::Run when
323       available and suitable.
324
325   $IPC::Cmd::USE_IPC_OPEN3
326       This variable controls whether IPC::Cmd will try to use IPC::Open3 when
327       available and suitable. Defaults to true.
328
329   $IPC::Cmd::WARN
330       This variable controls whether run-time warnings should be issued, like
331       the failure to load an "IPC::*" module you explicitly requested.
332
333       Defaults to true. Turn this off at your own risk.
334
335   $IPC::Cmd::INSTANCES
336       This variable controls whether "can_run" will return all instances of
337       the binary it finds in the "PATH" when called in a list context.
338
339       Defaults to false, set to true to enable the described behaviour.
340
341   $IPC::Cmd::ALLOW_NULL_ARGS
342       This variable controls whether "run" will remove any empty/null
343       arguments it finds in command arguments.
344
345       Defaults to false, so it will remove null arguments. Set to true to
346       allow them.
347

Caveats

349       Whitespace and IPC::Open3 / system()
350           When using "IPC::Open3" or "system", if you provide a string as the
351           "command" argument, it is assumed to be appropriately escaped. You
352           can use the "QUOTE" constant to use as a portable quote character
353           (see above).  However, if you provide an array reference, special
354           rules apply:
355
356           If your command contains special characters (< > | &), it will be
357           internally stringified before executing the command, to avoid that
358           these special characters are escaped and passed as arguments
359           instead of retaining their special meaning.
360
361           However, if the command contained arguments that contained
362           whitespace, stringifying the command would lose the significance of
363           the whitespace.  Therefore, "IPC::Cmd" will quote any arguments
364           containing whitespace in your command if the command is passed as
365           an arrayref and contains special characters.
366
367       Whitespace and IPC::Run
368           When using "IPC::Run", if you provide a string as the "command"
369           argument, the string will be split on whitespace to determine the
370           individual elements of your command. Although this will usually
371           just Do What You Mean, it may break if you have files or commands
372           with whitespace in them.
373
374           If you do not wish this to happen, you should provide an array
375           reference, where all parts of your command are already separated
376           out.  Note however, if there are extra or spurious whitespaces in
377           these parts, the parser or underlying code may not interpret it
378           correctly, and cause an error.
379
380           Example: The following code
381
382               gzip -cdf foo.tar.gz | tar -xf -
383
384           should either be passed as
385
386               "gzip -cdf foo.tar.gz | tar -xf -"
387
388           or as
389
390               ['gzip', '-cdf', 'foo.tar.gz', '|', 'tar', '-xf', '-']
391
392           But take care not to pass it as, for example
393
394               ['gzip -cdf foo.tar.gz', '|', 'tar -xf -']
395
396           Since this will lead to issues as described above.
397
398       IO Redirect
399           Currently it is too complicated to parse your command for IO
400           redirections. For capturing STDOUT or STDERR there is a work around
401           however, since you can just inspect your buffers for the contents.
402
403       Interleaving STDOUT/STDERR
404           Neither IPC::Run nor IPC::Open3 can interleave STDOUT and STDERR.
405           For short bursts of output from a program, e.g. this sample,
406
407               for ( 1..4 ) {
408                   $_ % 2 ? print STDOUT $_ : print STDERR $_;
409               }
410
411           IPC::[Run|Open3] will first read all of STDOUT, then all of STDERR,
412           meaning the output looks like '13' on STDOUT and '24' on STDERR,
413           instead of
414
415               1
416               2
417               3
418               4
419
420           This has been recorded in rt.cpan.org as bug #37532: Unable to
421           interleave STDOUT and STDERR.
422

See Also

424       IPC::Run, IPC::Open3
425

ACKNOWLEDGEMENTS

427       Thanks to James Mastros and Martijn van der Streek for their help in
428       getting IPC::Open3 to behave nicely.
429
430       Thanks to Petya Kohts for the "run_forked" code.
431

BUG REPORTS

433       Please report bugs or other issues to <bug-ipc-cmd@rt.cpan.org>.
434

AUTHOR

436       Original author: Jos Boumans <kane@cpan.org>.  Current maintainer:
437       Chris Williams <bingos@cpan.org>.
438
440       This library is free software; you may redistribute and/or modify it
441       under the same terms as Perl itself.
442
443
444
445perl v5.36.0                      2023-01-20                       IPC::Cmd(3)
Impressum