xref: /openbsd-src/gnu/usr.bin/perl/cpan/Test-Harness/bin/prove (revision f2da64fbbbf1b03f09f390ab01267c93dfd77c4c)
1#!/usr/bin/perl -w
2
3BEGIN { pop @INC if $INC[-1] eq '.' }
4use strict;
5use warnings;
6use App::Prove;
7
8my $app = App::Prove->new;
9$app->process_args(@ARGV);
10exit( $app->run ? 0 : 1 );
11
12__END__
13
14=head1 NAME
15
16prove - Run tests through a TAP harness.
17
18=head1 USAGE
19
20 prove [options] [files or directories]
21
22=head1 OPTIONS
23
24Boolean options:
25
26 -v,  --verbose         Print all test lines.
27 -l,  --lib             Add 'lib' to the path for your tests (-Ilib).
28 -b,  --blib            Add 'blib/lib' and 'blib/arch' to the path for
29                        your tests
30 -s,  --shuffle         Run the tests in random order.
31 -c,  --color           Colored test output (default).
32      --nocolor         Do not color test output.
33      --count           Show the X/Y test count when not verbose
34                        (default)
35      --nocount         Disable the X/Y test count.
36 -D   --dry             Dry run. Show test that would have run.
37 -f,  --failures        Show failed tests.
38 -o,  --comments        Show comments.
39      --ignore-exit     Ignore exit status from test scripts.
40 -m,  --merge           Merge test scripts' STDERR with their STDOUT.
41 -r,  --recurse         Recursively descend into directories.
42      --reverse         Run the tests in reverse order.
43 -q,  --quiet           Suppress some test output while running tests.
44 -Q,  --QUIET           Only print summary results.
45 -p,  --parse           Show full list of TAP parse errors, if any.
46      --directives      Only show results with TODO or SKIP directives.
47      --timer           Print elapsed time after each test.
48      --trap            Trap Ctrl-C and print summary on interrupt.
49      --normalize       Normalize TAP output in verbose output
50 -T                     Enable tainting checks.
51 -t                     Enable tainting warnings.
52 -W                     Enable fatal warnings.
53 -w                     Enable warnings.
54 -h,  --help            Display this help
55 -?,                    Display this help
56 -H,  --man             Longer manpage for prove
57      --norc            Don't process default .proverc
58
59Options that take arguments:
60
61 -I                     Library paths to include.
62 -P                     Load plugin (searches App::Prove::Plugin::*.)
63 -M                     Load a module.
64 -e,  --exec            Interpreter to run the tests ('' for compiled
65                        tests.)
66      --ext             Set the extension for tests (default '.t')
67      --harness         Define test harness to use.  See TAP::Harness.
68      --formatter       Result formatter to use. See FORMATTERS.
69      --source          Load and/or configure a SourceHandler. See
70                        SOURCE HANDLERS.
71 -a,  --archive out.tgz Store the resulting TAP in an archive file.
72 -j,  --jobs N          Run N test jobs in parallel (try 9.)
73      --state=opts      Control prove's persistent state.
74      --rc=rcfile       Process options from rcfile
75      --rules           Rules for parallel vs sequential processing.
76
77=head1 NOTES
78
79=head2 .proverc
80
81If F<~/.proverc> or F<./.proverc> exist they will be read and any
82options they contain processed before the command line options. Options
83in F<.proverc> are specified in the same way as command line options:
84
85    # .proverc
86    --state=hot,fast,save
87    -j9
88
89Additional option files may be specified with the C<--rc> option.
90Default option file processing is disabled by the C<--norc> option.
91
92Under Windows and VMS the option file is named F<_proverc> rather than
93F<.proverc> and is sought only in the current directory.
94
95=head2 Reading from C<STDIN>
96
97If you have a list of tests (or URLs, or anything else you want to test) in a
98file, you can add them to your tests by using a '-':
99
100 prove - < my_list_of_things_to_test.txt
101
102See the C<README> in the C<examples> directory of this distribution.
103
104=head2 Default Test Directory
105
106If no files or directories are supplied, C<prove> looks for all files
107matching the pattern C<t/*.t>.
108
109=head2 Colored Test Output
110
111Colored test output using L<TAP::Formatter::Color> is the default, but
112if output is not to a terminal, color is disabled. You can override this by
113adding the C<--color> switch.
114
115Color support requires L<Term::ANSIColor> on Unix-like platforms and
116L<Win32::Console> on windows. If the necessary module is not installed
117colored output will not be available.
118
119=head2 Exit Code
120
121If the tests fail C<prove> will exit with non-zero status.
122
123=head2 Arguments to Tests
124
125It is possible to supply arguments to tests. To do so separate them from
126prove's own arguments with the arisdottle, '::'. For example
127
128 prove -v t/mytest.t :: --url http://example.com
129
130would run F<t/mytest.t> with the options '--url http://example.com'.
131When running multiple tests they will each receive the same arguments.
132
133=head2 C<--exec>
134
135Normally you can just pass a list of Perl tests and the harness will know how
136to execute them.  However, if your tests are not written in Perl or if you
137want all tests invoked exactly the same way, use the C<-e>, or C<--exec>
138switch:
139
140 prove --exec '/usr/bin/ruby -w' t/
141 prove --exec '/usr/bin/perl -Tw -mstrict -Ilib' t/
142 prove --exec '/path/to/my/customer/exec'
143
144=head2 C<--merge>
145
146If you need to make sure your diagnostics are displayed in the correct
147order relative to test results you can use the C<--merge> option to
148merge the test scripts' STDERR into their STDOUT.
149
150This guarantees that STDOUT (where the test results appear) and STDERR
151(where the diagnostics appear) will stay in sync. The harness will
152display any diagnostics your tests emit on STDERR.
153
154Caveat: this is a bit of a kludge. In particular note that if anything
155that appears on STDERR looks like a test result the test harness will
156get confused. Use this option only if you understand the consequences
157and can live with the risk.
158
159=head2 C<--trap>
160
161The C<--trap> option will attempt to trap SIGINT (Ctrl-C) during a test
162run and display the test summary even if the run is interrupted
163
164=head2 C<--state>
165
166You can ask C<prove> to remember the state of previous test runs and
167select and/or order the tests to be run based on that saved state.
168
169The C<--state> switch requires an argument which must be a comma
170separated list of one or more of the following options.
171
172=over
173
174=item C<last>
175
176Run the same tests as the last time the state was saved. This makes it
177possible, for example, to recreate the ordering of a shuffled test.
178
179    # Run all tests in random order
180    $ prove -b --state=save --shuffle
181
182    # Run them again in the same order
183    $ prove -b --state=last
184
185=item C<failed>
186
187Run only the tests that failed on the last run.
188
189    # Run all tests
190    $ prove -b --state=save
191
192    # Run failures
193    $ prove -b --state=failed
194
195If you also specify the C<save> option newly passing tests will be
196excluded from subsequent runs.
197
198    # Repeat until no more failures
199    $ prove -b --state=failed,save
200
201=item C<passed>
202
203Run only the passed tests from last time. Useful to make sure that no
204new problems have been introduced.
205
206=item C<all>
207
208Run all tests in normal order. Multple options may be specified, so to
209run all tests with the failures from last time first:
210
211    $ prove -b --state=failed,all,save
212
213=item C<hot>
214
215Run the tests that most recently failed first. The last failure time of
216each test is stored. The C<hot> option causes tests to be run in most-recent-
217failure order.
218
219    $ prove -b --state=hot,save
220
221Tests that have never failed will not be selected. To run all tests with
222the most recently failed first use
223
224    $ prove -b --state=hot,all,save
225
226This combination of options may also be specified thus
227
228    $ prove -b --state=adrian
229
230=item C<todo>
231
232Run any tests with todos.
233
234=item C<slow>
235
236Run the tests in slowest to fastest order. This is useful in conjunction
237with the C<-j> parallel testing switch to ensure that your slowest tests
238start running first.
239
240    $ prove -b --state=slow -j9
241
242=item C<fast>
243
244Run test tests in fastest to slowest order.
245
246=item C<new>
247
248Run the tests in newest to oldest order based on the modification times
249of the test scripts.
250
251=item C<old>
252
253Run the tests in oldest to newest order.
254
255=item C<fresh>
256
257Run those test scripts that have been modified since the last test run.
258
259=item C<save>
260
261Save the state on exit. The state is stored in a file called F<.prove>
262(F<_prove> on Windows and VMS) in the current directory.
263
264=back
265
266The C<--state> switch may be used more than once.
267
268    $ prove -b --state=hot --state=all,save
269
270=head2 --rules
271
272The C<--rules> option is used to control which tests are run sequentially and
273which are run in parallel, if the C<--jobs> option is specified. The option may
274be specified multiple times, and the order matters.
275
276The most practical use is likely to specify that some tests are not
277"parallel-ready".  Since mentioning a file with --rules doesn't cause it to
278be selected to run as a test, you can "set and forget" some rules preferences in
279your .proverc file. Then you'll be able to take maximum advantage of the
280performance benefits of parallel testing, while some exceptions are still run
281in parallel.
282
283=head3 --rules examples
284
285    # All tests are allowed to run in parallel, except those starting with "p"
286    --rules='seq=t/p*.t' --rules='par=**'
287
288    # All tests must run in sequence except those starting with "p", which should be run parallel
289    --rules='par=t/p*.t'
290
291=head3 --rules resolution
292
293=over 4
294
295=item * By default, all tests are eligible to be run in parallel. Specifying any of your own rules removes this one.
296
297=item * "First match wins". The first rule that matches a test will be the one that applies.
298
299=item * Any test which does not match a rule will be run in sequence at the end of the run.
300
301=item * The existence of a rule does not imply selecting a test. You must still specify the tests to run.
302
303=item * Specifying a rule to allow tests to run in parallel does not make them run in parallel. You still need specify the number of parallel C<jobs> in your Harness object.
304
305=back
306
307=head3 --rules Glob-style pattern matching
308
309We implement our own glob-style pattern matching for --rules. Here are the
310supported patterns:
311
312    ** is any number of characters, including /, within a pathname
313    * is zero or more characters within a filename/directory name
314    ? is exactly one character within a filename/directory name
315    {foo,bar,baz} is any of foo, bar or baz.
316    \ is an escape character
317
318=head3 More advanced specifications for parallel vs sequence run rules
319
320If you need more advanced management of what runs in parallel vs in sequence, see
321the associated 'rules' documentation in L<TAP::Harness> and L<TAP::Parser::Scheduler>.
322If what's possible directly through C<prove> is not sufficient, you can write your own
323harness to access these features directly.
324
325=head2 @INC
326
327prove introduces a separation between "options passed to the perl which
328runs prove" and "options passed to the perl which runs tests"; this
329distinction is by design. Thus the perl which is running a test starts
330with the default C<@INC>. Additional library directories can be added
331via the C<PERL5LIB> environment variable, via -Ifoo in C<PERL5OPT> or
332via the C<-Ilib> option to F<prove>.
333
334=head2 Taint Mode
335
336Normally when a Perl program is run in taint mode the contents of the
337C<PERL5LIB> environment variable do not appear in C<@INC>.
338
339Because C<PERL5LIB> is often used during testing to add build
340directories to C<@INC> prove passes the names of any directories found
341in C<PERL5LIB> as -I switches. The net effect of this is that
342C<PERL5LIB> is honoured even when prove is run in taint mode.
343
344
345=head1 FORMATTERS
346
347You can load a custom L<TAP::Parser::Formatter>:
348
349  prove --formatter MyFormatter
350
351=head1 SOURCE HANDLERS
352
353You can load custom L<TAP::Parser::SourceHandler>s, to change the way the
354parser interprets particular I<sources> of TAP.
355
356  prove --source MyHandler --source YetAnother t
357
358If you want to provide config to the source you can use:
359
360  prove --source MyCustom \
361        --source Perl --perl-option 'foo=bar baz' --perl-option avg=0.278 \
362        --source File --file-option extensions=.txt --file-option extensions=.tmp t
363        --source pgTAP --pgtap-option pset=format=html --pgtap-option pset=border=2
364
365Each C<--$source-option> option must specify a key/value pair separated by an
366C<=>. If an option can take multiple values, just specify it multiple times,
367as with the C<extensions=> examples above. If the option should be a hash
368reference, specify the value as a second pair separated by a C<=>, as in the
369C<pset=> examples above (escape C<=> with a backslash).
370
371All C<--sources> are combined into a hash, and passed to L<TAP::Harness/new>'s
372C<sources> parameter.
373
374See L<TAP::Parser::IteratorFactory> for more details on how configuration is
375passed to I<SourceHandlers>.
376
377=head1 PLUGINS
378
379Plugins can be loaded using the C<< -PI<plugin> >> syntax, eg:
380
381  prove -PMyPlugin
382
383This will search for a module named C<App::Prove::Plugin::MyPlugin>, or failing
384that, C<MyPlugin>.  If the plugin can't be found, C<prove> will complain & exit.
385
386You can pass arguments to your plugin by appending C<=arg1,arg2,etc> to the
387plugin name:
388
389  prove -PMyPlugin=fou,du,fafa
390
391Please check individual plugin documentation for more details.
392
393=head2 Available Plugins
394
395For an up-to-date list of plugins available, please check CPAN:
396
397L<http://search.cpan.org/search?query=App%3A%3AProve+Plugin>
398
399=head2 Writing Plugins
400
401Please see L<App::Prove/PLUGINS>.
402
403=cut
404
405# vim:ts=4:sw=4:et:sta
406