3 use Test
::Harness
1.1601 ();
5 our($VERSION, @ISA, @EXPORT, @EXPORT_OK, $ntest, $TestLevel); #public-ish
6 our($TESTOUT, $ONFAIL, %todo, %history, $planned, @FAILDETAIL); #private-ish
10 @EXPORT=qw(&plan &ok &skip);
11 @EXPORT_OK=qw($ntest $TESTOUT);
13 $TestLevel = 0; # how many extra stack frames to skip
17 $TESTOUT = *STDOUT{IO};
19 # Use of this variable is strongly discouraged. It is set mainly to
20 # help test coverage analyzers know which test is running.
21 $ENV{REGRESSION_TEST} = $0;
24 croak "Test::plan(%args): odd number of arguments" if @_ & 1;
25 croak "Test::plan(): should not be called more than once" if $planned;
27 for (my $x=0; $x < @_; $x+=2) {
28 my ($k,$v) = @_[$x,$x+1];
29 if ($k =~ /^test(s)?$/) { $max = $v; }
30 elsif ($k eq 'todo' or
31 $k eq 'failok') { for (@$v) { $todo{$_}=1; }; }
32 elsif ($k eq 'onfail') {
33 ref $v eq 'CODE' or croak "Test::plan(onfail => $v): must be CODE";
36 else { carp "Test::plan(): skipping unrecognized directive '$k'" }
38 my @todo = sort { $a <=> $b } keys %todo;
40 print $TESTOUT "1..$max todo ".join(' ', @todo).";\n";
42 print $TESTOUT "1..$max\n";
49 (ref $v or '') eq 'CODE' ? $v->() : $v;
53 croak "ok: plan before you test!" if !$planned;
54 my ($pkg,$file,$line) = caller($TestLevel);
55 my $repetition = ++$history{"$file:$line"};
56 my $context = ("$file at line $line".
57 ($repetition > 1 ? " fail \#$repetition" : ''));
59 my $result = to_value(shift);
64 $expected = to_value(shift);
66 if (!defined $expected) {
67 $ok = !defined $result;
68 } elsif (!defined $result) {
70 } elsif ((ref($expected)||'') eq 'Regexp') {
71 $ok = $result =~ /$expected/;
72 } elsif (($regex) = ($expected =~ m,^ / (.+) / $,sx) or
73 ($ignore, $regex) = ($expected =~ m,^ m([^\w\s]) (.+) \1 $,sx)) {
74 $ok = $result =~ /$regex/;
76 $ok = $result eq $expected;
79 my $todo = $todo{$ntest};
81 $context .= ' TODO?!' if $todo;
82 print $TESTOUT "ok $ntest # ($context)\n";
84 # Issuing two separate print()s causes severe trouble with
85 # Test::Harness on VMS. The "not "'s for failed tests occur
86 # on a separate line and would not get counted as failures.
87 #print $TESTOUT "not " if !$ok;
88 #print $TESTOUT "ok $ntest\n";
89 # Replace with a single print() as a workaround:
91 $okline = "not " if !$ok;
92 $okline .= "ok $ntest\n";
93 print $TESTOUT $okline;
96 my $detail = { 'repetition' => $repetition, 'package' => $pkg,
97 'result' => $result, 'todo' => $todo };
98 $$detail{expected} = $expected if defined $expected;
99 $diag = $$detail{diagnostic} = to_value(shift) if @_;
100 $context .= ' *TODO*' if $todo;
101 if (!defined $expected) {
103 print $TESTOUT "# Failed test $ntest in $context\n";
105 print $TESTOUT "# Failed test $ntest in $context: $diag\n";
108 my $prefix = "Test $ntest";
109 print $TESTOUT "# $prefix got: ".
110 (defined $result? "'$result'":'<UNDEF>')." ($context)\n";
111 $prefix = ' ' x (length($prefix) - 5);
112 if ((ref($expected)||'') eq 'Regexp') {
113 $expected = 'qr/'.$expected.'/'
115 $expected = "'$expected'";
118 print $TESTOUT "# $prefix Expected: $expected\n";
120 print $TESTOUT "# $prefix Expected: $expected ($diag)\n";
123 push @FAILDETAIL, $detail;
131 my $whyskip = to_value(shift);
133 $whyskip = 'skip' if $whyskip =~ m/^\d+$/;
134 print $TESTOUT "ok $ntest # $whyskip\n";
138 local($TestLevel) = $TestLevel+1; #ignore this stack frame
144 $ONFAIL->(\@FAILDETAIL) if @FAILDETAIL && $ONFAIL;
152 Test - provides a simple framework for writing test scripts
159 # use a BEGIN block so we print our plan before MyModule is loaded
160 BEGIN { plan tests => 14, todo => [3,4] }
162 # load your module...
168 ok(0); # ok, expected failure (see todo list, above)
169 ok(1); # surprise success!
171 ok(0,1); # failure: '0' ne '1'
172 ok('broke','fixed'); # failure: 'broke' ne 'fixed'
173 ok('fixed','fixed'); # success: 'fixed' eq 'fixed'
174 ok('fixed',qr/x/); # success: 'fixed' =~ qr/x/
176 ok(sub { 1+1 }, 2); # success: '2' eq '2'
177 ok(sub { 1+1 }, 3); # failure: '2' ne '3'
178 ok(0, int(rand(2)); # (just kidding :-)
181 ok @list, 3, "\@list=".join(',',@list); #extra diagnostics
182 ok 'segmentation fault', '/(?i)success/'; #regex match
184 skip($feature_is_missing, ...); #do platform specific test
188 L<Test::Harness|Test::Harness> expects to see particular output when it
189 executes tests. This module aims to make writing proper test scripts just
190 a little bit easier (and less error prone :-).
198 These tests are expected to succeed. If they don't something's
201 =item * SKIPPED TESTS
203 Skip is for tests that might or might not be possible to run depending
204 on the availability of platform specific features. The first argument
205 should evaluate to true (think "yes, please skip") if the required
206 feature is not available. After the first argument, skip works
207 exactly the same way as do normal tests.
211 TODO tests are designed for maintaining an B<executable TODO list>.
212 These tests are expected NOT to succeed. If a TODO test does succeed,
213 the feature in question should not be on the TODO list, now should it?
215 Packages should NOT be released with succeeding TODO tests. As soon
216 as a TODO test starts working, it should be promoted to a normal test
217 and the newly working feature should be documented in the release
224 Both C<ok> and C<skip> return true if their test succeeds and false
225 otherwise in a scalar context.
229 BEGIN { plan test => 4, onfail => sub { warn "CALL 911!" } }
231 While test failures should be enough, extra diagnostics can be
232 triggered at the end of a test run. C<onfail> is passed an array ref
233 of hash refs that describe each test failure. Each hash will contain
234 at least the following fields: C<package>, C<repetition>, and
235 C<result>. (The file, line, and test number are not included because
236 their correspondence to a particular test is tenuous.) If the test
237 had an expected value or a diagnostic string, these will also be
240 The B<optional> C<onfail> hook might be used simply to print out the
241 version of your package and/or how to report problems. It might also
242 be used to generate extremely sophisticated diagnostics for a
243 particularly bizarre test failure. However it's not a panacea. Core
244 dumps or other unrecoverable errors prevent the C<onfail> hook from
245 running. (It is run inside an C<END> block.) Besides, C<onfail> is
246 probably over-kill in most cases. (Your test code should be simpler
247 than the code it is testing, yes?)
251 L<Test::Harness> and, perhaps, test coverage analysis tools.
255 Copyright (c) 1998-1999 Joshua Nathaniel Pritikin. All rights reserved.
257 This package is free software and is provided "as is" without express
258 or implied warranty. It may be used, redistributed and/or modified
259 under the terms of the Perl Artistic License (see
260 http://www.perl.com/perl/misc/Artistic.html)