Debian Codegraph @

upstream/debian-unstable/man3/TAP::Harness.3perl

raw
1 .\" -*- mode: troff; coding: utf-8 -*-
2 .\" Automatically generated by Pod::Man 5.0102 (Pod::Simple 3.45)
3 .\"
4 .\" Standard preamble:
5 .\" ========================================================================
6 .de Sp \" Vertical space (when we can't use .PP)
7 .if t .sp .5v
8 .if n .sp
9 ..
10 .de Vb \" Begin verbatim text
11 .ft CW
12 .nf
13 .ne \\$1
14 ..
15 .de Ve \" End verbatim text
16 .ft R
17 .fi
18 ..
19 .\" \*(C` and \*(C' are quotes in nroff, nothing in troff, for use with C<>.
20 .ie n \{\
21 . ds C` ""
22 . ds C' ""
23 'br\}
24 .el\{\
25 . ds C`
26 . ds C'
27 'br\}
28 .\"
29 .\" Escape single quotes in literal strings from groff's Unicode transform.
30 .ie \n(.g .ds Aq \(aq
31 .el .ds Aq '
32 .\"
33 .\" If the F register is >0, we'll generate index entries on stderr for
34 .\" titles (.TH), headers (.SH), subsections (.SS), items (.Ip), and index
35 .\" entries marked with X<> in POD. Of course, you'll have to process the
36 .\" output yourself in some meaningful fashion.
37 .\"
38 .\" Avoid warning from groff about undefined register 'F'.
39 .de IX
40 ..
41 .nr rF 0
42 .if \n(.g .if rF .nr rF 1
43 .if (\n(rF:(\n(.g==0)) \{\
44 . if \nF \{\
45 . de IX
46 . tm Index:\\$1\t\\n%\t"\\$2"
47 ..
48 . if !\nF==2 \{\
49 . nr % 0
50 . nr F 2
51 . \}
52 . \}
53 .\}
54 .rr rF
55 .\" ========================================================================
56 .\"
57 .IX Title "TAP::Harness 3perl"
58 .TH TAP::Harness 3perl 2025-11-16 "perl v5.40.1" "Perl Programmers Reference Guide"
59 .\" For nroff, turn off justification. Always turn off hyphenation; it makes
60 .\" way too many mistakes in technical documents.
61 .if n .ad l
62 .nh
63 .SH NAME
64 TAP::Harness \- Run test scripts with statistics
65 .SH VERSION
66 .IX Header "VERSION"
67 Version 3.48
68 .SH DESCRIPTION
69 .IX Header "DESCRIPTION"
70 This is a simple test harness which allows tests to be run and results
71 automatically aggregated and output to STDOUT.
72 .SH SYNOPSIS
73 .IX Header "SYNOPSIS"
74 .Vb 3
75 \& use TAP::Harness;
76 \& my $harness = TAP::Harness\->new( \e%args );
77 \& $harness\->runtests(@tests);
78 .Ve
79 .SH METHODS
80 .IX Header "METHODS"
81 .SS "Class Methods"
82 .IX Subsection "Class Methods"
83 \fR\f(CI\*(C`new\*(C'\fR\fI\fR
84 .IX Subsection "new"
85 .PP
86 .Vb 5
87 \& my %args = (
88 \& verbosity => 1,
89 \& lib => [ \*(Aqlib\*(Aq, \*(Aqblib/lib\*(Aq, \*(Aqblib/arch\*(Aq ],
90 \& )
91 \& my $harness = TAP::Harness\->new( \e%args );
92 .Ve
93 .PP
94 The constructor returns a new \f(CW\*(C`TAP::Harness\*(C'\fR object. It accepts an
95 optional hashref whose allowed keys are:
96 .IP \(bu 4
97 \&\f(CW\*(C`verbosity\*(C'\fR
98 .Sp
99 Set the verbosity level:
100 .Sp
101 .Vb 6
102 \& 1 verbose Print individual test results to STDOUT.
103 \& 0 normal
104 \& \-1 quiet Suppress some test output (mostly failures
105 \& while tests are running).
106 \& \-2 really quiet Suppress everything but the tests summary.
107 \& \-3 silent Suppress everything.
108 .Ve
109 .IP \(bu 4
110 \&\f(CW\*(C`timer\*(C'\fR
111 .Sp
112 Append run time for each test to output. Uses Time::HiRes if
113 available.
114 .IP \(bu 4
115 \&\f(CW\*(C`failures\*(C'\fR
116 .Sp
117 Show test failures (this is a no-op if \f(CW\*(C`verbose\*(C'\fR is selected).
118 .IP \(bu 4
119 \&\f(CW\*(C`comments\*(C'\fR
120 .Sp
121 Show test comments (this is a no-op if \f(CW\*(C`verbose\*(C'\fR is selected).
122 .IP \(bu 4
123 \&\f(CW\*(C`show_count\*(C'\fR
124 .Sp
125 Update the running test count during testing.
126 .IP \(bu 4
127 \&\f(CW\*(C`normalize\*(C'\fR
128 .Sp
129 Set to a true value to normalize the TAP that is emitted in verbose modes.
130 .IP \(bu 4
131 \&\f(CW\*(C`lib\*(C'\fR
132 .Sp
133 Accepts a scalar value or array ref of scalar values indicating which
134 paths to allowed libraries should be included if Perl tests are
135 executed. Naturally, this only makes sense in the context of tests
136 written in Perl.
137 .IP \(bu 4
138 \&\f(CW\*(C`switches\*(C'\fR
139 .Sp
140 Accepts a scalar value or array ref of scalar values indicating which
141 switches should be included if Perl tests are executed. Naturally, this
142 only makes sense in the context of tests written in Perl.
143 .IP \(bu 4
144 \&\f(CW\*(C`test_args\*(C'\fR
145 .Sp
146 A reference to an \f(CW@INC\fR style array of arguments to be passed to each
147 test program.
148 .Sp
149 .Vb 1
150 \& test_args => [\*(Aqfoo\*(Aq, \*(Aqbar\*(Aq],
151 .Ve
152 .Sp
153 if you want to pass different arguments to each test then you should
154 pass a hash of arrays, keyed by the alias for each test:
155 .Sp
156 .Vb 4
157 \& test_args => {
158 \& my_test => [\*(Aqfoo\*(Aq, \*(Aqbar\*(Aq],
159 \& other_test => [\*(Aqbaz\*(Aq],
160 \& }
161 .Ve
162 .IP \(bu 4
163 \&\f(CW\*(C`color\*(C'\fR
164 .Sp
165 Attempt to produce color output.
166 .IP \(bu 4
167 \&\f(CW\*(C`exec\*(C'\fR
168 .Sp
169 Typically, Perl tests are run through this. However, anything which
170 spits out TAP is fine. You can use this argument to specify the name of
171 the program (and optional switches) to run your tests with:
172 .Sp
173 .Vb 1
174 \& exec => [\*(Aq/usr/bin/ruby\*(Aq, \*(Aq\-w\*(Aq]
175 .Ve
176 .Sp
177 You can also pass a subroutine reference in order to determine and
178 return the proper program to run based on a given test script. The
179 subroutine reference should expect the TAP::Harness object itself as the
180 first argument, and the file name as the second argument. It should
181 return an array reference containing the command to be run and including
182 the test file name. It can also simply return \f(CW\*(C`undef\*(C'\fR, in which case
183 TAP::Harness will fall back on executing the test script in Perl:
184 .Sp
185 .Vb 2
186 \& exec => sub {
187 \& my ( $harness, $test_file ) = @_;
188 \&
189 \& # Let Perl tests run.
190 \& return undef if $test_file =~ /[.]t$/;
191 \& return [ qw( /usr/bin/ruby \-w ), $test_file ]
192 \& if $test_file =~ /[.]rb$/;
193 \& }
194 .Ve
195 .Sp
196 If the subroutine returns a scalar with a newline or a filehandle, it
197 will be interpreted as raw TAP or as a TAP stream, respectively.
198 .IP \(bu 4
199 \&\f(CW\*(C`merge\*(C'\fR
200 .Sp
201 If \f(CW\*(C`merge\*(C'\fR is true the harness will create parsers that merge STDOUT
202 and STDERR together for any processes they start.
203 .IP \(bu 4
204 \&\f(CW\*(C`sources\*(C'\fR
205 .Sp
206 \&\fINEW to 3.18\fR.
207 .Sp
208 If set, \f(CW\*(C`sources\*(C'\fR must be a hashref containing the names of the
209 TAP::Parser::SourceHandlers to load and/or configure. The values are a
210 hash of configuration that will be accessible to the source handlers via
211 "config_for" in TAP::Parser::Source.
212 .Sp
213 For example:
214 .Sp
215 .Vb 5
216 \& sources => {
217 \& Perl => { exec => \*(Aq/path/to/custom/perl\*(Aq },
218 \& File => { extensions => [ \*(Aq.tap\*(Aq, \*(Aq.txt\*(Aq ] },
219 \& MyCustom => { some => \*(Aqconfig\*(Aq },
220 \& }
221 .Ve
222 .Sp
223 The \f(CW\*(C`sources\*(C'\fR parameter affects how \f(CW\*(C`source\*(C'\fR, \f(CW\*(C`tap\*(C'\fR and \f(CW\*(C`exec\*(C'\fR parameters
224 are handled.
225 .Sp
226 For more details, see the \f(CW\*(C`sources\*(C'\fR parameter in "new" in TAP::Parser,
227 TAP::Parser::Source, and TAP::Parser::IteratorFactory.
228 .IP \(bu 4
229 \&\f(CW\*(C`aggregator_class\*(C'\fR
230 .Sp
231 The name of the class to use to aggregate test results. The default is
232 TAP::Parser::Aggregator.
233 .IP \(bu 4
234 \&\f(CW\*(C`version\*(C'\fR
235 .Sp
236 \&\fINEW to 3.22\fR.
237 .Sp
238 Assume this TAP version for TAP::Parser instead of default TAP
239 version 12.
240 .IP \(bu 4
241 \&\f(CW\*(C`formatter_class\*(C'\fR
242 .Sp
243 The name of the class to use to format output. The default is
244 TAP::Formatter::Console, or TAP::Formatter::File if the output
245 isn't a TTY.
246 .IP \(bu 4
247 \&\f(CW\*(C`multiplexer_class\*(C'\fR
248 .Sp
249 The name of the class to use to multiplex tests during parallel testing.
250 The default is TAP::Parser::Multiplexer.
251 .IP \(bu 4
252 \&\f(CW\*(C`parser_class\*(C'\fR
253 .Sp
254 The name of the class to use to parse TAP. The default is
255 TAP::Parser.
256 .IP \(bu 4
257 \&\f(CW\*(C`scheduler_class\*(C'\fR
258 .Sp
259 The name of the class to use to schedule test execution. The default is
260 TAP::Parser::Scheduler.
261 .IP \(bu 4
262 \&\f(CW\*(C`formatter\*(C'\fR
263 .Sp
264 If set \f(CW\*(C`formatter\*(C'\fR must be an object that is capable of formatting the
265 TAP output. See TAP::Formatter::Console for an example.
266 .IP \(bu 4
267 \&\f(CW\*(C`errors\*(C'\fR
268 .Sp
269 If parse errors are found in the TAP output, a note of this will be
270 made in the summary report. To see all of the parse errors, set this
271 argument to true:
272 .Sp
273 .Vb 1
274 \& errors => 1
275 .Ve
276 .IP \(bu 4
277 \&\f(CW\*(C`directives\*(C'\fR
278 .Sp
279 If set to a true value, only test results with directives will be
280 displayed. This overrides other settings such as \f(CW\*(C`verbose\*(C'\fR or
281 \&\f(CW\*(C`failures\*(C'\fR.
282 .IP \(bu 4
283 \&\f(CW\*(C`ignore_exit\*(C'\fR
284 .Sp
285 If set to a true value instruct \f(CW\*(C`TAP::Parser\*(C'\fR to ignore exit and wait
286 status from test scripts.
287 .IP \(bu 4
288 \&\f(CW\*(C`jobs\*(C'\fR
289 .Sp
290 The maximum number of parallel tests to run at any time. Which tests
291 can be run in parallel is controlled by \f(CW\*(C`rules\*(C'\fR. The default is to
292 run only one test at a time.
293 .IP \(bu 4
294 \&\f(CW\*(C`rules\*(C'\fR
295 .Sp
296 A reference to a hash of rules that control which tests may be executed in
297 parallel. If no rules are declared and CPAN::Meta::YAML is available,
298 \&\f(CW\*(C`TAP::Harness\*(C'\fR attempts to load rules from a YAML file specified by the
299 \&\f(CW\*(C`rulesfile\*(C'\fR parameter. If no rules file exists, the default is for all
300 tests to be eligible to be run in parallel.
301 .Sp
302 Here some simple examples. For the full details of the data structure
303 and the related glob-style pattern matching, see
304 "Rules data structure" in TAP::Parser::Scheduler.
305 .Sp
306 .Vb 4
307 \& # Run all tests in sequence, except those starting with "p"
308 \& $harness\->rules({
309 \& par => \*(Aqt/p*.t\*(Aq
310 \& });
311 \&
312 \& # Equivalent YAML file
313 \& \-\-\-
314 \& par: t/p*.t
315 \&
316 \& # Run all tests in parallel, except those starting with "p"
317 \& $harness\->rules({
318 \& seq => [
319 \& { seq => \*(Aqt/p*.t\*(Aq },
320 \& { par => \*(Aq**\*(Aq },
321 \& ],
322 \& });
323 \&
324 \& # Equivalent YAML file
325 \& \-\-\-
326 \& seq:
327 \& \- seq: t/p*.t
328 \& \- par: **
329 \&
330 \& # Run some startup tests in sequence, then some parallel tests than some
331 \& # teardown tests in sequence.
332 \& $harness\->rules({
333 \& seq => [
334 \& { seq => \*(Aqt/startup/*.t\*(Aq },
335 \& { par => [\*(Aqt/a/*.t\*(Aq,\*(Aqt/b/*.t\*(Aq,\*(Aqt/c/*.t\*(Aq], }
336 \& { seq => \*(Aqt/shutdown/*.t\*(Aq },
337 \& ],
338 \&
339 \& });
340 \&
341 \& # Equivalent YAML file
342 \& \-\-\-
343 \& seq:
344 \& \- seq: t/startup/*.t
345 \& \- par:
346 \& \- t/a/*.t
347 \& \- t/b/*.t
348 \& \- t/c/*.t
349 \& \- seq: t/shutdown/*.t
350 .Ve
351 .Sp
352 This is an experimental feature and the interface may change.
353 .IP \(bu 4
354 \&\f(CW\*(C`rulesfiles\*(C'\fR
355 .Sp
356 This specifies where to find a YAML file of test scheduling rules. If not
357 provided, it looks for a default file to use. It first checks for a file given
358 in the \f(CW\*(C`HARNESS_RULESFILE\*(C'\fR environment variable, then it checks for
359 \&\fItestrules.yml\fR and then \fIt/testrules.yml\fR.
360 .IP \(bu 4
361 \&\f(CW\*(C`stdout\*(C'\fR
362 .Sp
363 A filehandle for catching standard output.
364 .IP \(bu 4
365 \&\f(CW\*(C`trap\*(C'\fR
366 .Sp
367 Attempt to print summary information if run is interrupted by
368 SIGINT (Ctrl-C).
369 .PP
370 Any keys for which the value is \f(CW\*(C`undef\*(C'\fR will be ignored.
371 .SS "Instance Methods"
372 .IX Subsection "Instance Methods"
373 \fR\f(CI\*(C`runtests\*(C'\fR\fI\fR
374 .IX Subsection "runtests"
375 .PP
376 .Vb 1
377 \& $harness\->runtests(@tests);
378 .Ve
379 .PP
380 Accepts an array of \f(CW@tests\fR to be run. This should generally be the
381 names of test files, but this is not required. Each element in \f(CW@tests\fR
382 will be passed to \f(CWTAP::Parser::new()\fR as a \f(CW\*(C`source\*(C'\fR. See
383 TAP::Parser for more information.
384 .PP
385 It is possible to provide aliases that will be displayed in place of the
386 test name by supplying the test as a reference to an array containing
387 \&\f(CW\*(C`[ $test, $alias ]\*(C'\fR:
388 .PP
389 .Vb 2
390 \& $harness\->runtests( [ \*(Aqt/foo.t\*(Aq, \*(AqFoo Once\*(Aq ],
391 \& [ \*(Aqt/foo.t\*(Aq, \*(AqFoo Twice\*(Aq ] );
392 .Ve
393 .PP
394 Normally it is an error to attempt to run the same test twice. Aliases
395 allow you to overcome this limitation by giving each run of the test a
396 unique name.
397 .PP
398 Tests will be run in the order found.
399 .PP
400 If the environment variable \f(CW\*(C`PERL_TEST_HARNESS_DUMP_TAP\*(C'\fR is defined it
401 should name a directory into which a copy of the raw TAP for each test
402 will be written. TAP is written to files named for each test.
403 Subdirectories will be created as needed.
404 .PP
405 Returns a TAP::Parser::Aggregator containing the test results.
406 .PP
407 \fR\f(CI\*(C`summary\*(C'\fR\fI\fR
408 .IX Subsection "summary"
409 .PP
410 .Vb 1
411 \& $harness\->summary( $aggregator );
412 .Ve
413 .PP
414 Output the summary for a TAP::Parser::Aggregator.
415 .PP
416 \fR\f(CI\*(C`aggregate_tests\*(C'\fR\fI\fR
417 .IX Subsection "aggregate_tests"
418 .PP
419 .Vb 1
420 \& $harness\->aggregate_tests( $aggregate, @tests );
421 .Ve
422 .PP
423 Run the named tests and display a summary of result. Tests will be run
424 in the order found.
425 .PP
426 Test results will be added to the supplied TAP::Parser::Aggregator.
427 \&\f(CW\*(C`aggregate_tests\*(C'\fR may be called multiple times to run several sets of
428 tests. Multiple \f(CW\*(C`Test::Harness\*(C'\fR instances may be used to pass results
429 to a single aggregator so that different parts of a complex test suite
430 may be run using different \f(CW\*(C`TAP::Harness\*(C'\fR settings. This is useful, for
431 example, in the case where some tests should run in parallel but others
432 are unsuitable for parallel execution.
433 .PP
434 .Vb 8
435 \& my $formatter = TAP::Formatter::Console\->new;
436 \& my $ser_harness = TAP::Harness\->new( { formatter => $formatter } );
437 \& my $par_harness = TAP::Harness\->new(
438 \& { formatter => $formatter,
439 \& jobs => 9
440 \& }
441 \& );
442 \& my $aggregator = TAP::Parser::Aggregator\->new;
443 \&
444 \& $aggregator\->start();
445 \& $ser_harness\->aggregate_tests( $aggregator, @ser_tests );
446 \& $par_harness\->aggregate_tests( $aggregator, @par_tests );
447 \& $aggregator\->stop();
448 \& $formatter\->summary($aggregator);
449 .Ve
450 .PP
451 Note that for simpler testing requirements it will often be possible to
452 replace the above code with a single call to \f(CW\*(C`runtests\*(C'\fR.
453 .PP
454 Each element of the \f(CW@tests\fR array is either:
455 .IP \(bu 4
456 the source name of a test to run
457 .IP \(bu 4
458 a reference to a [ source name, display name ] array
459 .PP
460 In the case of a perl test suite, typically \fIsource names\fR are simply the file
461 names of the test scripts to run.
462 .PP
463 When you supply a separate display name it becomes possible to run a
464 test more than once; the display name is effectively the alias by which
465 the test is known inside the harness. The harness doesn't care if it
466 runs the same test more than once when each invocation uses a
467 different name.
468 .PP
469 \fR\f(CI\*(C`make_scheduler\*(C'\fR\fI\fR
470 .IX Subsection "make_scheduler"
471 .PP
472 Called by the harness when it needs to create a
473 TAP::Parser::Scheduler. Override in a subclass to provide an
474 alternative scheduler. \f(CW\*(C`make_scheduler\*(C'\fR is passed the list of tests
475 that was passed to \f(CW\*(C`aggregate_tests\*(C'\fR.
476 .PP
477 \fR\f(CI\*(C`jobs\*(C'\fR\fI\fR
478 .IX Subsection "jobs"
479 .PP
480 Gets or sets the number of concurrent test runs the harness is
481 handling. By default, this value is 1 \-\- for parallel testing, this
482 should be set higher.
483 .PP
484 \fR\f(CI\*(C`make_parser\*(C'\fR\fI\fR
485 .IX Subsection "make_parser"
486 .PP
487 Make a new parser and display formatter session. Typically used and/or
488 overridden in subclasses.
489 .PP
490 .Vb 1
491 \& my ( $parser, $session ) = $harness\->make_parser;
492 .Ve
493 .PP
494 \fR\f(CI\*(C`finish_parser\*(C'\fR\fI\fR
495 .IX Subsection "finish_parser"
496 .PP
497 Terminate use of a parser. Typically used and/or overridden in
498 subclasses. The parser isn't destroyed as a result of this.
499 .SH CONFIGURING
500 .IX Header "CONFIGURING"
501 \&\f(CW\*(C`TAP::Harness\*(C'\fR is designed to be easy to configure.
502 .SS Plugins
503 .IX Subsection "Plugins"
504 \&\f(CW\*(C`TAP::Parser\*(C'\fR plugins let you change the way TAP is \fIinput\fR to and \fIoutput\fR
505 from the parser.
506 .PP
507 TAP::Parser::SourceHandlers handle TAP \fIinput\fR. You can configure them
508 and load custom handlers using the \f(CW\*(C`sources\*(C'\fR parameter to "new".
509 .PP
510 TAP::Formatters handle TAP \fIoutput\fR. You can load custom formatters by
511 using the \f(CW\*(C`formatter_class\*(C'\fR parameter to "new". To configure a formatter,
512 you currently need to instantiate it outside of TAP::Harness and pass it in
513 with the \f(CW\*(C`formatter\*(C'\fR parameter to "new". This \fImay\fR be addressed by adding
514 a \fIformatters\fR parameter to "new" in the future.
515 .ie n .SS """Module::Build"""
516 .el .SS \f(CWModule::Build\fP
517 .IX Subsection "Module::Build"
518 Module::Build version \f(CW0.30\fR supports \f(CW\*(C`TAP::Harness\*(C'\fR.
519 .PP
520 To load \f(CW\*(C`TAP::Harness\*(C'\fR plugins, you'll need to use the \f(CW\*(C`tap_harness_args\*(C'\fR
521 parameter to \f(CW\*(C`new\*(C'\fR, typically from your \f(CW\*(C`Build.PL\*(C'\fR. For example:
522 .PP
523 .Vb 10
524 \& Module::Build\->new(
525 \& module_name => \*(AqMyApp\*(Aq,
526 \& test_file_exts => [qw(.t .tap .txt)],
527 \& use_tap_harness => 1,
528 \& tap_harness_args => {
529 \& sources => {
530 \& MyCustom => {},
531 \& File => {
532 \& extensions => [\*(Aq.tap\*(Aq, \*(Aq.txt\*(Aq],
533 \& },
534 \& },
535 \& formatter_class => \*(AqTAP::Formatter::HTML\*(Aq,
536 \& },
537 \& build_requires => {
538 \& \*(AqModule::Build\*(Aq => \*(Aq0.30\*(Aq,
539 \& \*(AqTAP::Harness\*(Aq => \*(Aq3.18\*(Aq,
540 \& },
541 \& )\->create_build_script;
542 .Ve
543 .PP
544 See "new"
545 .ie n .SS """ExtUtils::MakeMaker"""
546 .el .SS \f(CWExtUtils::MakeMaker\fP
547 .IX Subsection "ExtUtils::MakeMaker"
548 ExtUtils::MakeMaker does not support TAP::Harness out-of-the-box.
549 .ie n .SS """prove"""
550 .el .SS \f(CWprove\fP
551 .IX Subsection "prove"
552 prove supports \f(CW\*(C`TAP::Harness\*(C'\fR plugins, and has a plugin system of its
553 own. See "FORMATTERS" in prove, "SOURCE HANDLERS" in prove and App::Prove
554 for more details.
555 .SH "WRITING PLUGINS"
556 .IX Header "WRITING PLUGINS"
557 If you can't configure \f(CW\*(C`TAP::Harness\*(C'\fR to do what you want, and you can't find
558 an existing plugin, consider writing one.
559 .PP
560 The two primary use cases supported by TAP::Harness for plugins are \fIinput\fR
561 and \fIoutput\fR:
562 .IP "Customize how TAP gets into the parser" 2
563 .IX Item "Customize how TAP gets into the parser"
564 To do this, you can either extend an existing TAP::Parser::SourceHandler,
565 or write your own. It's a pretty simple API, and they can be loaded and
566 configured using the \f(CW\*(C`sources\*(C'\fR parameter to "new".
567 .IP "Customize how TAP results are output from the parser" 2
568 .IX Item "Customize how TAP results are output from the parser"
569 To do this, you can either extend an existing TAP::Formatter, or write your
570 own. Writing formatters are a bit more involved than writing a
571 \&\fISourceHandler\fR, as you'll need to understand the TAP::Parser API. A
572 good place to start is by understanding how "aggregate_tests" works.
573 .Sp
574 Custom formatters can be loaded configured using the \f(CW\*(C`formatter_class\*(C'\fR
575 parameter to "new".
576 .SH SUBCLASSING
577 .IX Header "SUBCLASSING"
578 If you can't configure \f(CW\*(C`TAP::Harness\*(C'\fR to do exactly what you want, and writing
579 a plugin isn't an option, consider extending it. It is designed to be (mostly)
580 easy to subclass, though the cases when sub-classing is necessary should be few
581 and far between.
582 .SS Methods
583 .IX Subsection "Methods"
584 The following methods are ones you may wish to override if you want to
585 subclass \f(CW\*(C`TAP::Harness\*(C'\fR.
586 .IP """new""" 4
587 .IX Item """new"""
588 .PD 0
589 .IP """runtests""" 4
590 .IX Item """runtests"""
591 .IP """summary""" 4
592 .IX Item """summary"""
593 .PD
594 .SH REPLACING
595 .IX Header "REPLACING"
596 If you like the \f(CW\*(C`prove\*(C'\fR utility and TAP::Parser but you want your
597 own harness, all you need to do is write one and provide \f(CW\*(C`new\*(C'\fR and
598 \&\f(CW\*(C`runtests\*(C'\fR methods. Then you can use the \f(CW\*(C`prove\*(C'\fR utility like so:
599 .PP
600 .Vb 1
601 \& prove \-\-harness My::Test::Harness
602 .Ve
603 .PP
604 Note that while \f(CW\*(C`prove\*(C'\fR accepts a list of tests (or things to be
605 tested), \f(CW\*(C`new\*(C'\fR has a fairly rich set of arguments. You'll probably want
606 to read over this code carefully to see how all of them are being used.
607 .SH "SEE ALSO"
608 .IX Header "SEE ALSO"
609 Test::Harness