Test::Command - Test routines for external commands

Version 0.08

Test the exit status, signal, STDOUT or STDERR of an external command.

   use Test::Command tests => 11;

   ## testing exit status

   my $cmd = 'true';

   exit_is_num($cmd, 0);
   exit_cmp_ok($cmd, '<', 10);

   $cmd = 'false';

   exit_isnt_num($cmd, 0);

   ## testing terminating signal 

   $cmd = 'true';

   signal_is_num($cmd, 0);

   ## testing STDOUT

   $cmd         = [qw/ echo out /];  ## run as "system @$cmd"
   my $file_exp = 'echo_stdout.exp';

   stdout_is_eq($cmd, "out\n");
   stdout_isnt_eq($cmd, "out");
   stdout_is_file($cmd, $file_exp);

   ## testing STDERR

   $cmd = 'echo err >&2';

   stderr_like($cmd, /err/);
   stderr_unlike($cmd, /rre/);
   stderr_cmp_ok($cmd, 'eq', "err\n");

   ## run-once-test-many-OO-style
   ## the first test lazily runs command
   ## the second test uses cached results

   my $echo_test = Test::Command->new( cmd => 'echo out' );

   $echo_test->exit_is_num(0);
   $echo_test->signal_is_num(0);
   $echo_test->stdout_is_eq("out\n");

   ## force a re-run of the command

   $echo_test->run;

Test::Command intends to bridge the gap between the well tested functions and objects you choose and their usage in your programs. By examining the exit status, terminating signal, STDOUT and STDERR of your program you can determine if it is behaving as expected.

This includes testing the various combinations and permutations of options and arguments as well as the interactions between the various functions and objects that make up your program.

The various test functions below can accept either a command string or an array reference for the first argument. If the command is expressed as a string it is passed to system as is. If the command is expressed as an array reference it is dereferenced and passed to system as a list. See 'perldoc -f system' for how these may differ.

The final argument for the test functions, $name, is optional. By default the $name is a concatenation of the test function name, the command string and the expected value. This construction is generally sufficient for identifying a failing test, but you may always specify your own $name if desired.

Any of the test functions can be used as instance methods on a Test::Command object. This is done by dropping the initial $cmd argument and instead using arrow notation.

All of the following exit_is_num calls are equivalent.

   exit_is_num('true', 0);
   exit_is_num('true', 0, 'exit_is_num: true, 0');
   exit_is_num(['true'], 0);
   exit_is_num(['true'], 0, 'exit_is_num: true, 0');

   my $cmd = Test::Command->new( cmd => 'true' );

   exit_is_num($cmd, 0);
   exit_is_num($cmd, 0, 'exit_is_num: true, 0');
   $cmd->exit_is_num(0);
   $cmd->exit_is_num(0, 'exit_is_num: true, 0');

   $cmd = Test::Command->new( cmd => ['true'] );

   exit_is_num($cmd, 0);
   exit_is_num($cmd, 0, 'exit_is_num: true, 0');
   $cmd->exit_is_num(0);
   $cmd->exit_is_num(0, 'exit_is_num: true, 0');

All of the test functions mentioned below are exported by default.

   my $test_cmd_obj = Test::Command->new( cmd => $cmd )

This constructor creates and returns a Test::Command object. Use this to test multiple aspects of a single command execution while avoiding repeatedly running commands which are slow or resource intensive.

The cmd parameter can accept either a string or an array reference for its value. The value is dereferenced if necessary and passed directly to the system builtin.

   $test_cmd_obj->run;

This instance method forces the execution of the command specified by the invocant.

You only need to call this when you wish to re-run a command since the first test method invoked will lazily execute the command if necessary. However, if the state of your inputs has changed and you wish to re-run the command, you may do so by invoking this method at any point between your tests.

The test routines below compare against the exit status of the executed command right shifted by 8 (that is, $? >> 8).

   exit_is_num($cmd, $exp_num, $name)

If the exit status of the command is numerically equal to the expected number, this passes. Otherwise it fails.

   exit_isnt_num($cmd, $unexp_num, $name)

If the exit status of the command is not numerically equal to the given number, this passes. Otherwise it fails.

   exit_cmp_ok($cmd, $op, $operand, $name)

If the exit status of the command is compared with the given operand using the given operator, and that operation returns true, this passes. Otherwise it fails.

   exit_is_defined($cmd, $name)

If the exit status of the command is defined, this passes. Otherwise it fails. A defined exit status indicates that the command exited normally by calling exit() or running off the end of the program.

   exit_is_undef($cmd, $name)

If the exit status of the command is not defined, this passes. Otherwise it fails. An undefined exit status indicates that the command likely exited due to a signal.

The test routines below compare against the lower 8 bits of the exit status of the executed command.

   signal_is_num($cmd, $exp_num, $name)

If the terminating signal of the command is numerically equal to the expected number, this passes. Otherwise it fails.

   signal_isnt_num($cmd, $unexp_num, $name)

If the terminating signal of the command is not numerically equal to the given number, this passes. Otherwise it fails.

   signal_cmp_ok($cmd, $op, $operand, $name)

If the terminating signal of the command is compared with the given operand using the given operator, and that operation returns true, this passes. Otherwise it fails.

   signal_is_defined($cmd, $name)

If the terminating signal of the command is defined, this passes. Otherwise it fails. A defined signal indicates that the command likely exited due to a signal.

   signal_is_undef($cmd, $name)

If the terminating signal of the command is not defined, this passes. Otherwise it fails. An undefined signal indicates that the command exited normally by calling exit() or running off the end of the program.

Except where specified, the test routines below treat STDOUT as a single slurped string.

   stdout_is_eq($cmd, $exp_string, $name)

If the STDOUT of the command is equal (compared using eq) to the expected string, then this passes. Otherwise it fails.

   stdout_isnt_eq($cmd, $unexp_string, $name)

If the STDOUT of the command is not equal (compared using eq) to the given string, this passes. Otherwise it fails.

   stdout_is_num($cmd, $exp_num, $name)

If the STDOUT of the command is equal (compared using ==) to the expected number, then this passes. Otherwise it fails.

   stdout_isnt_num($cmd, $unexp_num, $name)

If the STDOUT of the command is not equal (compared using ==) to the given number, this passes. Otherwise it fails.

   stdout_like($cmd, $exp_regex, $name)

If the STDOUT of the command matches the expected regular expression, this passes. Otherwise it fails.

   stdout_unlike($cmd, $unexp_regex, $name)

If the STDOUT of the command does not match the given regular expression, this passes. Otherwise it fails.

   stdout_cmp_ok($cmd, $op, $operand, $name)

If the STDOUT of the command is compared with the given operand using the given operator, and that operation returns true, this passes. Otherwise it fails.

   stdout_is_file($cmd, $exp_file, $name)

If the STDOUT of the command is equal (compared using eq) to the contents of the given file, then this passes. Otherwise it fails. Note that this comparison is performed line by line, rather than slurping the entire file.

Except where specified, the test routines below treat STDERR as a single slurped string.

   stderr_is_eq($cmd, $exp_string, $name)

If the STDERR of the command is equal (compared using eq) to the expected string, then this passes. Otherwise it fails.

   stderr_isnt_eq($cmd, $unexp_string, $name)

If the STDERR of the command is not equal (compared using eq) to the given string, this passes. Otherwise it fails.

   stderr_is_num($cmd, $exp_num, $name)

If the STDERR of the command is equal (compared using ==) to the expected number, then this passes. Otherwise it fails.

   stderr_isnt_num($cmd, $unexp_num, $name)

If the STDERR of the command is not equal (compared using ==) to the given number, this passes. Otherwise it fails.

   stderr_like($cmd, $exp_regex, $name)

If the STDERR of the command matches the expected regular expression, this passes. Otherwise it fails.

   stderr_unlike($cmd, $unexp_regex, $name)

If the STDERR of the command does not match the given regular expression, this passes. Otherwise it fails.

   stderr_cmp_ok($cmd, $op, $operand, $name)

If the STDERR of the command is compared with the given operand using the given operator, and that operation returns true, this passes. Otherwise it fails.

   stderr_is_file($cmd, $exp_file, $name)

If the STDERR of the command is equal (compared using eq) to the contents of the given file, then this passes. Otherwise it fails. Note that this comparison is performed line by line, rather than slurping the entire file.

Daniel B. Boorstein, <danboo at cpan.org>

Please report any bugs or feature requests to bug-test-command at rt.cpan.org, or through the web interface at http://rt.cpan.org/NoAuth/ReportBug.html?Queue=Test-Command. I will be notified, and then you'll automatically be notified of progress on your bug as I make changes.

You can find documentation for this module with the perldoc command.

    perldoc Test::Command

You can also look for information at:

* AnnoCPAN: Annotated CPAN documentation

http://annocpan.org/dist/Test-Command

* CPAN Ratings

http://cpanratings.perl.org/d/Test-Command

* RT: CPAN's request tracker

http://rt.cpan.org/NoAuth/Bugs.html?Dist=Test-Command

* Search CPAN

http://search.cpan.org/dist/Test-Command

Test::Builder by Michael Schwern allowed me to focus on the specifics related to testing system commands by making it easy to produce proper test output.

Copyright 2007 Daniel B. Boorstein, all rights reserved.

This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

* create a tool that produces test scripts given a list of commands to run

* optionally save the temp files with STDOUT and STDERR for user debugging

* if user defines all options and sample arguments to basic command

* create tool to enumerate all possible means of calling program

* allow testing with randomized/permuted/collapsed opts and args

* potential test functions:

* time_lt($cmd, $seconds)

* time_gt($cmd, $seconds)

* stdout_line_custom($cmd, \&code)

* stderr_line_custom($cmd, \&code)

Test::Builder provides the testing methods used in this module.

Test::Builder::Module is the superclass of this module.