lib/Test/Simple.pm

package Test::Simple;

use 5.004;

use strict 'vars';
use vars qw($VERSION);
$VERSION = '0.54';
$VERSION = eval $VERSION;    # make the alpha version come out as a number


use Test::Builder;
my $Test = Test::Builder->new;

sub import {
    my $self = shift;
    my $caller = caller;
    *{$caller.'::ok'} = \&ok;

    $Test->exported_to($caller);
    $Test->plan(@_);
}


=head1 NAME

Test::Simple - Basic utilities for writing tests.

=head1 SYNOPSIS

  use Test::Simple tests => 1;

  ok( $foo eq $bar, 'foo is bar' );


=head1 DESCRIPTION

** If you are unfamiliar with testing B<read Test::Tutorial> first! **

This is an extremely simple, extremely basic module for writing tests
suitable for CPAN modules and other pursuits.  If you wish to do more
complicated testing, use the Test::More module (a drop-in replacement
for this one).

The basic unit of Perl testing is the ok.  For each thing you want to
test your program will print out an "ok" or "not ok" to indicate pass
or fail.  You do this with the ok() function (see below).

The only other constraint is you must pre-declare how many tests you
plan to run.  This is in case something goes horribly wrong during the
test and your test program aborts, or skips a test or whatever.  You
do this like so:

    use Test::Simple tests => 23;

You must have a plan.


=over 4

=item B<ok>

  ok( $foo eq $bar, $name );
  ok( $foo eq $bar );

ok() is given an expression (in this case C<$foo eq $bar>).  If it's
true, the test passed.  If it's false, it didn't.  That's about it.

ok() prints out either "ok" or "not ok" along with a test number (it
keeps track of that for you).

  # This produces "ok 1 - Hell not yet frozen over" (or not ok)
  ok( get_temperature($hell) > 0, 'Hell not yet frozen over' );

If you provide a $name, that will be printed along with the "ok/not
ok" to make it easier to find your test when if fails (just search for
the name).  It also makes it easier for the next guy to understand
what your test is for.  It's highly recommended you use test names.

All tests are run in scalar context.  So this:

    ok( @stuff, 'I have some stuff' );

will do what you mean (fail if stuff is empty)

=cut

sub ok ($;$) {
    $Test->ok(@_);
}


=back

Test::Simple will start by printing number of tests run in the form
"1..M" (so "1..5" means you're going to run 5 tests).  This strange
format lets Test::Harness know how many tests you plan on running in
case something goes horribly wrong.

If all your tests passed, Test::Simple will exit with zero (which is
normal).  If anything failed it will exit with how many failed.  If
you run less (or more) tests than you planned, the missing (or extras)
will be considered failures.  If no tests were ever run Test::Simple
will throw a warning and exit with 255.  If the test died, even after
having successfully completed all its tests, it will still be
considered a failure and will exit with 255.

So the exit codes are...

    0                   all tests successful
    255                 test died
    any other number    how many failed (including missing or extras)

If you fail more than 254 tests, it will be reported as 254.

This module is by no means trying to be a complete testing system.
It's just to get you started.  Once you're off the ground its
recommended you look at L<Test::More>.


=head1 EXAMPLE

Here's an example of a simple .t file for the fictional Film module.

    use Test::Simple tests => 5;

    use Film;  # What you're testing.

    my $btaste = Film->new({ Title    => 'Bad Taste',
                             Director => 'Peter Jackson',
                             Rating   => 'R',
                             NumExplodingSheep => 1
                           });
    ok( defined($btaste) and ref $btaste eq 'Film',     'new() works' );

    ok( $btaste->Title      eq 'Bad Taste',     'Title() get'    );
    ok( $btaste->Director   eq 'Peter Jackson', 'Director() get' );
    ok( $btaste->Rating     eq 'R',             'Rating() get'   );
    ok( $btaste->NumExplodingSheep == 1,        'NumExplodingSheep() get' );

It will produce output like this:

    1..5
    ok 1 - new() works
    ok 2 - Title() get
    ok 3 - Director() get
    not ok 4 - Rating() get
    #    Failed test (t/film.t at line 14)
    ok 5 - NumExplodingSheep() get
    # Looks like you failed 1 tests of 5

Indicating the Film::Rating() method is broken.


=head1 CAVEATS

Test::Simple will only report a maximum of 254 failures in its exit
code.  If this is a problem, you probably have a huge test script.
Split it into multiple files.  (Otherwise blame the Unix folks for
using an unsigned short integer as the exit status).

Because VMS's exit codes are much, much different than the rest of the
universe, and perl does horrible mangling to them that gets in my way,
it works like this on VMS.

    0     SS$_NORMAL        all tests successful
    4     SS$_ABORT         something went wrong

Unfortunately, I can't differentiate any further.


=head1 NOTES

Test::Simple is B<explicitly> tested all the way back to perl 5.004.

Test::Simple is thread-safe in perl 5.8.0 and up.

=head1 HISTORY

This module was conceived while talking with Tony Bowden in his
kitchen one night about the problems I was having writing some really
complicated feature into the new Testing module.  He observed that the
main problem is not dealing with these edge cases but that people hate
to write tests B<at all>.  What was needed was a dead simple module
that took all the hard work out of testing and was really, really easy
to learn.  Paul Johnson simultaneously had this idea (unfortunately,
he wasn't in Tony's kitchen).  This is it.


=head1 SEE ALSO

=over 4

=item L<Test::More>

More testing functions!  Once you outgrow Test::Simple, look at
Test::More.  Test::Simple is 100% forward compatible with Test::More
(i.e. you can just use Test::More instead of Test::Simple in your
programs and things will still work).

=item L<Test>

The original Perl testing module.

=item L<Test::Unit>

Elaborate unit testing.

=item L<Test::Inline>, L<SelfTest>

Embed tests in your code!

=item L<Test::Harness>

Interprets the output of your test program.

=back


=head1 AUTHORS

Idea by Tony Bowden and Paul Johnson, code by Michael G Schwern
E<lt>schwern@pobox.comE<gt>, wardrobe by Calvin Klein.


=head1 COPYRIGHT

Copyright 2001, 2002, 2004 by Michael G Schwern E<lt>schwern@pobox.comE<gt>.

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

See F<http://www.perl.com/perl/misc/Artistic.html>

=cut

1;
Revision:	1.1
Committed:	Thu Jun 30 14:27:15 2005 UTC (21 years, 3 months ago) by root
Branch:	MAIN
CVS Tags:	PERL-5-8-7, HEAD
Branch point for:	PERL
Log Message:	* empty log message *
#	Content
1	package Test::Simple;
2
3	use 5.004;
4
5	use strict 'vars';
6	use vars qw($VERSION);
7	$VERSION = '0.54';
8	$VERSION = eval $VERSION; # make the alpha version come out as a number
9
10
11	use Test::Builder;
12	my $Test = Test::Builder->new;
13
14	sub import {
15	my $self = shift;
16	my $caller = caller;
17	*{$caller.'::ok'} = \&ok;
18
19	$Test->exported_to($caller);
20	$Test->plan(@_);
21	}
22
23
24	=head1 NAME
25
26	Test::Simple - Basic utilities for writing tests.
27
28	=head1 SYNOPSIS
29
30	use Test::Simple tests => 1;
31
32	ok( $foo eq $bar, 'foo is bar' );
33
34
35	=head1 DESCRIPTION
36
37	If you are unfamiliar with testing B<read Test::Tutorial> first!
38
39	This is an extremely simple, extremely basic module for writing tests
40	suitable for CPAN modules and other pursuits. If you wish to do more
41	complicated testing, use the Test::More module (a drop-in replacement
42	for this one).
43
44	The basic unit of Perl testing is the ok. For each thing you want to
45	test your program will print out an "ok" or "not ok" to indicate pass
46	or fail. You do this with the ok() function (see below).
47
48	The only other constraint is you must pre-declare how many tests you
49	plan to run. This is in case something goes horribly wrong during the
50	test and your test program aborts, or skips a test or whatever. You
51	do this like so:
52
53	use Test::Simple tests => 23;
54
55	You must have a plan.
56
57
58	=over 4
59
60	=item B<ok>
61
62	ok( $foo eq $bar, $name );
63	ok( $foo eq $bar );
64
65	ok() is given an expression (in this case C<$foo eq $bar>). If it's
66	true, the test passed. If it's false, it didn't. That's about it.
67
68	ok() prints out either "ok" or "not ok" along with a test number (it
69	keeps track of that for you).
70
71	# This produces "ok 1 - Hell not yet frozen over" (or not ok)
72	ok( get_temperature($hell) > 0, 'Hell not yet frozen over' );
73
74	If you provide a $name, that will be printed along with the "ok/not
75	ok" to make it easier to find your test when if fails (just search for
76	the name). It also makes it easier for the next guy to understand
77	what your test is for. It's highly recommended you use test names.
78
79	All tests are run in scalar context. So this:
80
81	ok( @stuff, 'I have some stuff' );
82
83	will do what you mean (fail if stuff is empty)
84
85	=cut
86
87	sub ok ($;$) {
88	$Test->ok(@_);
89	}
90
91
92	=back
93
94	Test::Simple will start by printing number of tests run in the form
95	"1..M" (so "1..5" means you're going to run 5 tests). This strange
96	format lets Test::Harness know how many tests you plan on running in
97	case something goes horribly wrong.
98
99	If all your tests passed, Test::Simple will exit with zero (which is
100	normal). If anything failed it will exit with how many failed. If
101	you run less (or more) tests than you planned, the missing (or extras)
102	will be considered failures. If no tests were ever run Test::Simple
103	will throw a warning and exit with 255. If the test died, even after
104	having successfully completed all its tests, it will still be
105	considered a failure and will exit with 255.
106
107	So the exit codes are...
108
109	0 all tests successful
110	255 test died
111	any other number how many failed (including missing or extras)
112
113	If you fail more than 254 tests, it will be reported as 254.
114
115	This module is by no means trying to be a complete testing system.
116	It's just to get you started. Once you're off the ground its
117	recommended you look at L<Test::More>.
118
119
120	=head1 EXAMPLE
121
122	Here's an example of a simple .t file for the fictional Film module.
123
124	use Test::Simple tests => 5;
125
126	use Film; # What you're testing.
127
128	my $btaste = Film->new({ Title => 'Bad Taste',
129	Director => 'Peter Jackson',
130	Rating => 'R',
131	NumExplodingSheep => 1
132	});
133	ok( defined($btaste) and ref $btaste eq 'Film', 'new() works' );
134
135	ok( $btaste->Title eq 'Bad Taste', 'Title() get' );
136	ok( $btaste->Director eq 'Peter Jackson', 'Director() get' );
137	ok( $btaste->Rating eq 'R', 'Rating() get' );
138	ok( $btaste->NumExplodingSheep == 1, 'NumExplodingSheep() get' );
139
140	It will produce output like this:
141
142	1..5
143	ok 1 - new() works
144	ok 2 - Title() get
145	ok 3 - Director() get
146	not ok 4 - Rating() get
147	# Failed test (t/film.t at line 14)
148	ok 5 - NumExplodingSheep() get
149	# Looks like you failed 1 tests of 5
150
151	Indicating the Film::Rating() method is broken.
152
153
154	=head1 CAVEATS
155
156	Test::Simple will only report a maximum of 254 failures in its exit
157	code. If this is a problem, you probably have a huge test script.
158	Split it into multiple files. (Otherwise blame the Unix folks for
159	using an unsigned short integer as the exit status).
160
161	Because VMS's exit codes are much, much different than the rest of the
162	universe, and perl does horrible mangling to them that gets in my way,
163	it works like this on VMS.
164
165	0 SS$_NORMAL all tests successful
166	4 SS$_ABORT something went wrong
167
168	Unfortunately, I can't differentiate any further.
169
170
171	=head1 NOTES
172
173	Test::Simple is B<explicitly> tested all the way back to perl 5.004.
174
175	Test::Simple is thread-safe in perl 5.8.0 and up.
176
177	=head1 HISTORY
178
179	This module was conceived while talking with Tony Bowden in his
180	kitchen one night about the problems I was having writing some really
181	complicated feature into the new Testing module. He observed that the
182	main problem is not dealing with these edge cases but that people hate
183	to write tests B<at all>. What was needed was a dead simple module
184	that took all the hard work out of testing and was really, really easy
185	to learn. Paul Johnson simultaneously had this idea (unfortunately,
186	he wasn't in Tony's kitchen). This is it.
187
188
189	=head1 SEE ALSO
190
191	=over 4
192
193	=item L<Test::More>
194
195	More testing functions! Once you outgrow Test::Simple, look at
196	Test::More. Test::Simple is 100% forward compatible with Test::More
197	(i.e. you can just use Test::More instead of Test::Simple in your
198	programs and things will still work).
199
200	=item L<Test>
201
202	The original Perl testing module.
203
204	=item L<Test::Unit>
205
206	Elaborate unit testing.
207
208	=item L<Test::Inline>, L<SelfTest>
209
210	Embed tests in your code!
211
212	=item L<Test::Harness>
213
214	Interprets the output of your test program.
215
216	=back
217
218
219	=head1 AUTHORS
220
221	Idea by Tony Bowden and Paul Johnson, code by Michael G Schwern
222	E<lt>schwern@pobox.comE<gt>, wardrobe by Calvin Klein.
223
224
225	=head1 COPYRIGHT
226
227	Copyright 2001, 2002, 2004 by Michael G Schwern E<lt>schwern@pobox.comE<gt>.
228
229	This program is free software; you can redistribute it and/or
230	modify it under the same terms as Perl itself.
231
232	See F<http://www.perl.com/perl/misc/Artistic.html>
233
234	=cut
235
236	1;