1PERLHACK(1) Perl Programmers Reference Guide PERLHACK(1)
2
6 perlhack - How to hack on Perl
7
9 This document explains how Perl development works. It includes details
10 about the Perl 5 Porters email list, the Perl repository, the Perl bug
11 tracker, patch guidelines, and commentary on Perl development
12 philosophy.
13
15 If you just want to submit a single small patch like a pod fix, a test
16 for a bug, comment fixes, etc., it's easy! Here's how:
17 · Check out the source repository
19 The perl source is in a git repository. You can clone the
21 repository with the following command:
22 % git clone https://github.com/Perl/perl5.git perl
24 · Ensure you're following the latest advice
26 In case the advice in this guide has been updated recently, read
28 the latest version directly from the perl source:
29 % perldoc pod/perlhack.pod
31 · Create a branch for your change
33 Create a branch based on blead to commit your change to, which will
35 later be used to send it to the Perl issue tracker.
36 % git checkout -b mychange
38 · Make your change
40 Hack, hack, hack. Keep in mind that Perl runs on many different
42 platforms, with different operating systems that have different
43 capabilities, different filesystem organizations, and even
44 different character sets. perlhacktips gives advice on this.
45 · Test your change
47 You can run all the tests with the following commands:
49 % ./Configure -des -Dusedevel
51 % make test
52 Keep hacking until the tests pass.
54 · Commit your change
56 Committing your work will save the change on your local system:
58 % git commit -a -m 'Commit message goes here'
60 Make sure the commit message describes your change in a single
62 sentence. For example, "Fixed spelling errors in perlhack.pod".
63 · Send your change to the Perl issue tracker
65 The next step is to submit your patch to the Perl core ticket
67 system.
68 Create a GitHub fork of the perl5 repository and add it as a
70 remote, if you haven't already, as described in the GitHub
71 documentation at
72 <https://help.github.com/en/articles/working-with-forks>.
73 % git remote add fork git@github.com:MyUser/perl5.git
75 Then, push your new branch to your fork.
77 % git push -u fork mychange
79 Finally, create a Pull Request on GitHub from your branch to blead
81 as described in the GitHub documentation at
82 <https://help.github.com/en/articles/creating-a-pull-request-from-a-fork>.
83 · Thank you
85 The porters appreciate the time you spent helping to make Perl
87 better. Thank you!
88 · Acknowledgement
90 All contributors are credited (by name and email address) in the
92 AUTHORS file, which is part of the perl distribution, as well as
93 the Git commit history.
94 If you donXt want to be included in the AUTHORS file, just let us
96 know. Otherwise we will take your submission of a patch as
97 permission to credit you in the AUTHORS file.
98 · Next time
100 The next time you wish to make a patch, you need to start from the
102 latest perl in a pristine state. Check you don't have any local
103 changes or added files in your perl check-out which you wish to
104 keep, then run these commands:
105 % git checkout blead
107 % git pull
108 % git reset --hard origin/blead
109 % git clean -dxf
110
112 If you want to report a bug in Perl, you must use the perlbug command
113 line tool. This tool will ensure that your bug report includes all the
114 relevant system and configuration information.
115 To browse existing Perl bugs and patches, you can use the web interface
117 at <http://rt.perl.org/>.
118 Please check the archive of the perl5-porters list (see below) and/or
120 the bug tracking system before submitting a bug report. Often, you'll
121 find that the bug has been reported already.
122 You can log in to the bug tracking system and comment on existing bug
124 reports. If you have additional information regarding an existing bug,
125 please add it. This will help the porters fix the bug.
126
128 The perl5-porters (p5p) mailing list is where the Perl standard
129 distribution is maintained and developed. The people who maintain Perl
130 are also referred to as the "Perl 5 Porters", "p5p" or just the
131 "porters".
132 A searchable archive of the list is available at
134 <http://markmail.org/search/?q=perl5-porters>. There is also an
135 archive at <http://archive.develooper.com/perl5-porters@perl.org/>.
136 perl-changes mailing list
138 The perl5-changes mailing list receives a copy of each patch that gets
139 submitted to the maintenance and development branches of the perl
140 repository. See <http://lists.perl.org/list/perl5-changes.html> for
141 subscription and archive information.
142 #p5p on IRC
144 Many porters are also active on the <irc://irc.perl.org/#p5p> channel.
145 Feel free to join the channel and ask questions about hacking on the
146 Perl core.
147
149 All of Perl's source code is kept centrally in a Git repository at
150 github.com. The repository contains many Perl revisions from Perl 1
151 onwards and all the revisions from Perforce, the previous version
152 control system.
153 For much more detail on using git with the Perl repository, please see
155 perlgit.
156 Read access via Git
158 You will need a copy of Git for your computer. You can fetch a copy of
159 the repository using the git protocol:
160 % git clone git://github.com/Perl/perl5.git perl
162 This clones the repository and makes a local copy in the perl
164 directory.
165 If you cannot use the git protocol for firewall reasons, you can also
167 clone via http:
168 % git clone https://github.com/Perl/perl5.git perl
170 Read access via the web
172 You may access the repository over the web. This allows you to browse
173 the tree, see recent commits, subscribe to repository notifications,
174 search for particular commits and more. You may access it at
175 <https://github.com/Perl/perl5>.
176 Read access via rsync
178 You can also choose to use rsync to get a copy of the current source
179 tree for the bleadperl branch and all maintenance branches:
180 % rsync -avz rsync://perl5.git.perl.org/perl-current .
182 % rsync -avz rsync://perl5.git.perl.org/perl-5.12.x .
183 % rsync -avz rsync://perl5.git.perl.org/perl-5.10.x .
184 % rsync -avz rsync://perl5.git.perl.org/perl-5.8.x .
185 % rsync -avz rsync://perl5.git.perl.org/perl-5.6.x .
186 % rsync -avz rsync://perl5.git.perl.org/perl-5.005xx .
187 (Add the "--delete" option to remove leftover files.)
189 To get a full list of the available sync points:
191 % rsync perl5.git.perl.org::
193 Write access via git
195 If you have a commit bit, please see perlgit for more details on using
196 git.
197
199 If you're planning to do more extensive work than a single small fix,
200 we encourage you to read the documentation below. This will help you
201 focus your work and make your patches easier to incorporate into the
202 Perl source.
203 Submitting patches
205 If you have a small patch to submit, please submit it via the GitHub
206 Pull Request workflow. You may also send patches to the p5p list.
207 Patches are reviewed and discussed on GitHub or the p5p list. Simple,
209 uncontroversial patches will usually be applied without any discussion.
210 When the patch is applied, the ticket will be updated and you will
211 receive email.
212 In other cases, the patch will need more work or discussion. You are
214 encouraged to participate in the discussion and advocate for your
215 patch. Sometimes your patch may get lost in the shuffle. It's
216 appropriate to send a reminder email to p5p if no action has been taken
217 in a month. Please remember that the Perl 5 developers are all
218 volunteers, and be polite.
219 Changes are always applied directly to the main development branch,
221 called "blead". Some patches may be backported to a maintenance
222 branch. If you think your patch is appropriate for the maintenance
223 branch (see "MAINTENANCE BRANCHES" in perlpolicy), please explain why
224 when you submit it.
225 Getting your patch accepted
227 If you are submitting a code patch there are several things that you
228 can do to help the Perl 5 Porters accept your patch.
229 Patch style
231 Using the GitHub Pull Request workflow, your patch will automatically
233 be available in a suitable format. If you wish to submit a patch to
234 the p5p list for review, make sure to create it appropriately.
235 If you used git to check out the Perl source, then using "git
237 format-patch" will produce a patch in a style suitable for Perl. The
238 "format-patch" command produces one patch file for each commit you
239 made. If you prefer to send a single patch for all commits, you can
240 use "git diff".
241 % git checkout blead
243 % git pull
244 % git diff blead my-branch-name
245 This produces a patch based on the difference between blead and your
247 current branch. It's important to make sure that blead is up to date
248 before producing the diff, that's why we call "git pull" first.
249 We strongly recommend that you use git if possible. It will make your
251 life easier, and ours as well.
252 However, if you're not using git, you can still produce a suitable
254 patch. You'll need a pristine copy of the Perl source to diff against.
255 The porters prefer unified diffs. Using GNU "diff", you can produce a
256 diff like this:
257 % diff -Npurd perl.pristine perl.mine
259 Make sure that you "make realclean" in your copy of Perl to remove any
261 build artifacts, or you may get a confusing result.
262 Commit message
264 As you craft each patch you intend to submit to the Perl core, it's
266 important to write a good commit message. This is especially important
267 if your submission will consist of a series of commits.
268 The first line of the commit message should be a short description
270 without a period. It should be no longer than the subject line of an
271 email, 50 characters being a good rule of thumb.
272 A lot of Git tools (Gitweb, GitHub, git log --pretty=oneline, ...) will
274 only display the first line (cut off at 50 characters) when presenting
275 commit summaries.
276 The commit message should include a description of the problem that the
278 patch corrects or new functionality that the patch adds.
279 As a general rule of thumb, your commit message should help a
281 programmer who knows the Perl core quickly understand what you were
282 trying to do, how you were trying to do it, and why the change matters
283 to Perl.
284 · Why
286 Your commit message should describe why the change you are making
288 is important. When someone looks at your change in six months or
289 six years, your intent should be clear.
290 If you're deprecating a feature with the intent of later
292 simplifying another bit of code, say so. If you're fixing a
293 performance problem or adding a new feature to support some other
294 bit of the core, mention that.
295 · What
297 Your commit message should describe what part of the Perl core
299 you're changing and what you expect your patch to do.
300 · How
302 While it's not necessary for documentation changes, new tests or
304 trivial patches, it's often worth explaining how your change works.
305 Even if it's clear to you today, it may not be clear to a porter
306 next month or next year.
307 A commit message isn't intended to take the place of comments in your
309 code. Commit messages should describe the change you made, while code
310 comments should describe the current state of the code.
311 If you've just implemented a new feature, complete with doc, tests and
313 well-commented code, a brief commit message will often suffice. If,
314 however, you've just changed a single character deep in the parser or
315 lexer, you might need to write a small novel to ensure that future
316 readers understand what you did and why you did it.
317 Comments, Comments, Comments
319 Be sure to adequately comment your code. While commenting every line
321 is unnecessary, anything that takes advantage of side effects of
322 operators, that creates changes that will be felt outside of the
323 function being patched, or that others may find confusing should be
324 documented. If you are going to err, it is better to err on the side
325 of adding too many comments than too few.
326 The best comments explain why the code does what it does, not what it
328 does.
329 Style
331 In general, please follow the particular style of the code you are
333 patching.
334 In particular, follow these general guidelines for patching Perl
336 sources:
337 · 4-wide indents for code, 2-wide indents for nested CPP "#define"s,
339 with 8-wide tabstops.
340 · Use spaces for indentation, not tab characters.
342 The codebase is a mixture of tabs and spaces for indentation, and
344 we are moving to spaces only. Converting lines you're patching
345 from 8-wide tabs to spaces will help this migration.
346 · Try hard not to exceed 79-columns
348 · ANSI C prototypes
350 · Uncuddled elses and "K&R" style for indenting control constructs
352 · No C++ style (//) comments
354 · Mark places that need to be revisited with XXX (and revisit often!)
356 · Opening brace lines up with "if" when conditional spans multiple
358 lines; should be at end-of-line otherwise
359 · In function definitions, name starts in column 0 (return value-type
361 is on previous line)
362 · Single space after keywords that are followed by parens, no space
364 between function name and following paren
365 · Avoid assignments in conditionals, but if they're unavoidable, use
367 extra paren, e.g. "if (a && (b = c)) ..."
368 · "return foo;" rather than "return(foo);"
370 · "if (!foo) ..." rather than "if (foo == FALSE) ..." etc.
372 · Do not declare variables using "register". It may be
374 counterproductive with modern compilers, and is deprecated in C++,
375 under which the Perl source is regularly compiled.
376 · In-line functions that are in headers that are accessible to XS
378 code need to be able to compile without warnings with commonly used
379 extra compilation flags, such as gcc's "-Wswitch-default" which
380 warns whenever a switch statement does not have a "default" case.
381 The use of these extra flags is to catch potential problems in
382 legal C code, and is often used by Perl aggregators, such as Linux
383 distributors.
384 Test suite
386 If your patch changes code (rather than just changing documentation),
388 you should also include one or more test cases which illustrate the bug
389 you're fixing or validate the new functionality you're adding. In
390 general, you should update an existing test file rather than create a
391 new one.
392 Your test suite additions should generally follow these guidelines
394 (courtesy of Gurusamy Sarathy <gsar@activestate.com>):
395 · Know what you're testing. Read the docs, and the source.
397 · Tend to fail, not succeed.
399 · Interpret results strictly.
401 · Use unrelated features (this will flush out bizarre interactions).
403 · Use non-standard idioms (otherwise you are not testing TIMTOWTDI).
405 · Avoid using hardcoded test numbers whenever possible (the
407 EXPECTED/GOT found in t/op/tie.t is much more maintainable, and
408 gives better failure reports).
409 · Give meaningful error messages when a test fails.
411 · Avoid using qx// and system() unless you are testing for them. If
413 you do use them, make sure that you cover _all_ perl platforms.
414 · Unlink any temporary files you create.
416 · Promote unforeseen warnings to errors with $SIG{__WARN__}.
418 · Be sure to use the libraries and modules shipped with the version
420 being tested, not those that were already installed.
421 · Add comments to the code explaining what you are testing for.
423 · Make updating the '1..42' string unnecessary. Or make sure that
425 you update it.
426 · Test _all_ behaviors of a given operator, library, or function.
428 Test all optional arguments.
430 Test return values in various contexts (boolean, scalar, list,
432 lvalue).
433 Use both global and lexical variables.
435 Don't forget the exceptional, pathological cases.
437 Patching a core module
439 This works just like patching anything else, with one extra
440 consideration.
441 Modules in the cpan/ directory of the source tree are maintained
443 outside of the Perl core. When the author updates the module, the
444 updates are simply copied into the core. See that module's
445 documentation or its listing on <http://search.cpan.org/> for more
446 information on reporting bugs and submitting patches.
447 In most cases, patches to modules in cpan/ should be sent upstream and
449 should not be applied to the Perl core individually. If a patch to a
450 file in cpan/ absolutely cannot wait for the fix to be made upstream,
451 released to CPAN and copied to blead, you must add (or update) a
452 "CUSTOMIZED" entry in the "Porting/Maintainers.pl" file to flag that a
453 local modification has been made. See "Porting/Maintainers.pl" for
454 more details.
455 In contrast, modules in the dist/ directory are maintained in the core.
457 Updating perldelta
459 For changes significant enough to warrant a pod/perldelta.pod entry,
460 the porters will greatly appreciate it if you submit a delta entry
461 along with your actual change. Significant changes include, but are
462 not limited to:
463 · Adding, deprecating, or removing core features
465 · Adding, deprecating, removing, or upgrading core or dual-life
467 modules
468 · Adding new core tests
470 · Fixing security issues and user-visible bugs in the core
472 · Changes that might break existing code, either on the perl or C
474 level
475 · Significant performance improvements
477 · Adding, removing, or significantly changing documentation in the
479 pod/ directory
480 · Important platform-specific changes
482 Please make sure you add the perldelta entry to the right section
484 within pod/perldelta.pod. More information on how to write good
485 perldelta entries is available in the "Style" section of
486 Porting/how_to_write_a_perldelta.pod.
487 What makes for a good patch?
489 New features and extensions to the language can be contentious. There
490 is no specific set of criteria which determine what features get added,
491 but here are some questions to consider when developing a patch:
492 Does the concept match the general goals of Perl?
494 Our goals include, but are not limited to:
496 1. Keep it fast, simple, and useful.
498 2. Keep features/concepts as orthogonal as possible.
500 3. No arbitrary limits (platforms, data sizes, cultures).
502 4. Keep it open and exciting to use/patch/advocate Perl everywhere.
504 5. Either assimilate new technologies, or build bridges to them.
506 Where is the implementation?
508 All the talk in the world is useless without an implementation. In
510 almost every case, the person or people who argue for a new feature
511 will be expected to be the ones who implement it. Porters capable of
512 coding new features have their own agendas, and are not available to
513 implement your (possibly good) idea.
514 Backwards compatibility
516 It's a cardinal sin to break existing Perl programs. New warnings can
518 be contentious--some say that a program that emits warnings is not
519 broken, while others say it is. Adding keywords has the potential to
520 break programs, changing the meaning of existing token sequences or
521 functions might break programs.
522 The Perl 5 core includes mechanisms to help porters make backwards
524 incompatible changes more compatible such as the feature and deprecate
525 modules. Please use them when appropriate.
526 Could it be a module instead?
528 Perl 5 has extension mechanisms, modules and XS, specifically to avoid
530 the need to keep changing the Perl interpreter. You can write modules
531 that export functions, you can give those functions prototypes so they
532 can be called like built-in functions, you can even write XS code to
533 mess with the runtime data structures of the Perl interpreter if you
534 want to implement really complicated things.
535 Whenever possible, new features should be prototyped in a CPAN module
537 before they will be considered for the core.
538 Is the feature generic enough?
540 Is this something that only the submitter wants added to the language,
542 or is it broadly useful? Sometimes, instead of adding a feature with a
543 tight focus, the porters might decide to wait until someone implements
544 the more generalized feature.
545 Does it potentially introduce new bugs?
547 Radical rewrites of large chunks of the Perl interpreter have the
549 potential to introduce new bugs.
550 How big is it?
552 The smaller and more localized the change, the better. Similarly, a
554 series of small patches is greatly preferred over a single large patch.
555 Does it preclude other desirable features?
557 A patch is likely to be rejected if it closes off future avenues of
559 development. For instance, a patch that placed a true and final
560 interpretation on prototypes is likely to be rejected because there are
561 still options for the future of prototypes that haven't been addressed.
562 Is the implementation robust?
564 Good patches (tight code, complete, correct) stand more chance of going
566 in. Sloppy or incorrect patches might be placed on the back burner
567 until the pumpking has time to fix, or might be discarded altogether
568 without further notice.
569 Is the implementation generic enough to be portable?
571 The worst patches make use of system-specific features. It's highly
573 unlikely that non-portable additions to the Perl language will be
574 accepted.
575 Is the implementation tested?
577 Patches which change behaviour (fixing bugs or introducing new
579 features) must include regression tests to verify that everything works
580 as expected.
581 Without tests provided by the original author, how can anyone else
583 changing perl in the future be sure that they haven't unwittingly
584 broken the behaviour the patch implements? And without tests, how can
585 the patch's author be confident that his/her hard work put into the
586 patch won't be accidentally thrown away by someone in the future?
587 Is there enough documentation?
589 Patches without documentation are probably ill-thought out or
591 incomplete. No features can be added or changed without documentation,
592 so submitting a patch for the appropriate pod docs as well as the
593 source code is important.
594 Is there another way to do it?
596 Larry said "Although the Perl Slogan is There's More Than One Way to Do
598 It, I hesitate to make 10 ways to do something". This is a tricky
599 heuristic to navigate, though--one man's essential addition is another
600 man's pointless cruft.
601 Does it create too much work?
603 Work for the pumpking, work for Perl programmers, work for module
605 authors, ... Perl is supposed to be easy.
606 Patches speak louder than words
608 Working code is always preferred to pie-in-the-sky ideas. A patch to
610 add a feature stands a much higher chance of making it to the language
611 than does a random feature request, no matter how fervently argued the
612 request might be. This ties into "Will it be useful?", as the fact
613 that someone took the time to make the patch demonstrates a strong
614 desire for the feature.
615
617 The core uses the same testing style as the rest of Perl, a simple
618 "ok/not ok" run through Test::Harness, but there are a few special
619 considerations.
620 There are three ways to write a test in the core: Test::More, t/test.pl
622 and ad hoc "print $test ? "ok 42\n" : "not ok 42\n"". The decision of
623 which to use depends on what part of the test suite you're working on.
624 This is a measure to prevent a high-level failure (such as Config.pm
625 breaking) from causing basic functionality tests to fail.
626 The t/test.pl library provides some of the features of Test::More, but
628 avoids loading most modules and uses as few core features as possible.
629 If you write your own test, use the Test Anything Protocol
631 <http://testanything.org>.
632 · t/base, t/comp and t/opbasic
634 Since we don't know if "require" works, or even subroutines, use ad
636 hoc tests for these three. Step carefully to avoid using the
637 feature being tested. Tests in t/opbasic, for instance, have been
638 placed there rather than in t/op because they test functionality
639 which t/test.pl presumes has already been demonstrated to work.
640 · t/cmd, t/run, t/io and t/op
642 Now that basic require() and subroutines are tested, you can use
644 the t/test.pl library.
645 You can also use certain libraries like Config conditionally, but
647 be sure to skip the test gracefully if it's not there.
648 · Everything else
650 Now that the core of Perl is tested, Test::More can and should be
652 used. You can also use the full suite of core modules in the
653 tests.
654 When you say "make test", Perl uses the t/TEST program to run the test
656 suite (except under Win32 where it uses t/harness instead). All tests
657 are run from the t/ directory, not the directory which contains the
658 test. This causes some problems with the tests in lib/, so here's some
659 opportunity for some patching.
660 You must be triply conscious of cross-platform concerns. This usually
662 boils down to using File::Spec, avoiding things like "fork()" and
663 "system()" unless absolutely necessary, and not assuming that a given
664 character has a particular ordinal value (code point) or that its UTF-8
665 representation is composed of particular bytes.
666 There are several functions available to specify characters and code
668 points portably in tests. The always-preloaded functions
669 "utf8::unicode_to_native()" and its inverse "utf8::native_to_unicode()"
670 take code points and translate appropriately. The file
671 t/charset_tools.pl has several functions that can be useful. It has
672 versions of the previous two functions that take strings as inputs --
673 not single numeric code points: "uni_to_native()" and
674 "native_to_uni()". If you must look at the individual bytes comprising
675 a UTF-8 encoded string, "byte_utf8a_to_utf8n()" takes as input a string
676 of those bytes encoded for an ASCII platform, and returns the
677 equivalent string in the native platform. For example,
678 "byte_utf8a_to_utf8n("\xC2\xA0")" returns the byte sequence on the
679 current platform that form the UTF-8 for "U+00A0", since "\xC2\xA0" are
680 the UTF-8 bytes on an ASCII platform for that code point. This
681 function returns "\xC2\xA0" on an ASCII platform, and "\x80\x41" on an
682 EBCDIC 1047 one.
683 But easiest is, if the character is specifiable as a literal, like "A"
685 or "%", to use that; if not so specificable, you can use use "\N{}" ,
686 if the side effects aren't troublesome. Simply specify all your
687 characters in hex, using "\N{U+ZZ}" instead of "\xZZ". "\N{}" is the
688 Unicode name, and so it always gives you the Unicode character.
689 "\N{U+41}" is the character whose Unicode code point is 0x41, hence is
690 'A' on all platforms. The side effects are:
691 · These select Unicode rules. That means that in double-quotish
693 strings, the string is always converted to UTF-8 to force a Unicode
694 interpretation (you can "utf8::downgrade()" afterwards to convert
695 back to non-UTF8, if possible). In regular expression patterns,
696 the conversion isn't done, but if the character set modifier would
697 otherwise be "/d", it is changed to "/u".
698 · If you use the form "\N{character name}", the charnames module gets
700 automatically loaded. This may not be suitable for the test level
701 you are doing.
702 If you are testing locales (see perllocale), there are helper functions
704 in t/loc_tools.pl to enable you to see what locales there are on the
705 current platform.
706 Special "make test" targets
708 There are various special make targets that can be used to test Perl
709 slightly differently than the standard "test" target. Not all them are
710 expected to give a 100% success rate. Many of them have several
711 aliases, and many of them are not available on certain operating
712 systems.
713 · test_porting
715 This runs some basic sanity tests on the source tree and helps
717 catch basic errors before you submit a patch.
718 · minitest
720 Run miniperl on t/base, t/comp, t/cmd, t/run, t/io, t/op, t/uni and
722 t/mro tests.
723 · test.valgrind check.valgrind
725 (Only in Linux) Run all the tests using the memory leak + naughty
727 memory access tool "valgrind". The log files will be named
728 testname.valgrind.
729 · test_harness
731 Run the test suite with the t/harness controlling program, instead
733 of t/TEST. t/harness is more sophisticated, and uses the
734 Test::Harness module, thus using this test target supposes that
735 perl mostly works. The main advantage for our purposes is that it
736 prints a detailed summary of failed tests at the end. Also, unlike
737 t/TEST, it doesn't redirect stderr to stdout.
738 Note that under Win32 t/harness is always used instead of t/TEST,
740 so there is no special "test_harness" target.
741 Under Win32's "test" target you may use the TEST_SWITCHES and
743 TEST_FILES environment variables to control the behaviour of
744 t/harness. This means you can say
745 nmake test TEST_FILES="op/*.t"
747 nmake test TEST_SWITCHES="-torture" TEST_FILES="op/*.t"
748 · test-notty test_notty
750 Sets PERL_SKIP_TTY_TEST to true before running normal test.
752 Parallel tests
754 The core distribution can now run its regression tests in parallel on
755 Unix-like platforms. Instead of running "make test", set "TEST_JOBS"
756 in your environment to the number of tests to run in parallel, and run
757 "make test_harness". On a Bourne-like shell, this can be done as
758 TEST_JOBS=3 make test_harness # Run 3 tests in parallel
760 An environment variable is used, rather than parallel make itself,
762 because TAP::Harness needs to be able to schedule individual non-
763 conflicting test scripts itself, and there is no standard interface to
764 "make" utilities to interact with their job schedulers.
765 Note that currently some test scripts may fail when run in parallel
767 (most notably dist/IO/t/io_dir.t). If necessary, run just the failing
768 scripts again sequentially and see if the failures go away.
769 Running tests by hand
771 You can run part of the test suite by hand by using one of the
772 following commands from the t/ directory:
773 ./perl -I../lib TEST list-of-.t-files
775 or
777 ./perl -I../lib harness list-of-.t-files
779 (If you don't specify test scripts, the whole test suite will be run.)
781 Using t/harness for testing
783 If you use "harness" for testing, you have several command line options
784 available to you. The arguments are as follows, and are in the order
785 that they must appear if used together.
786 harness -v -torture -re=pattern LIST OF FILES TO TEST
788 harness -v -torture -re LIST OF PATTERNS TO MATCH
789 If "LIST OF FILES TO TEST" is omitted, the file list is obtained from
791 the manifest. The file list may include shell wildcards which will be
792 expanded out.
793 · -v
795 Run the tests under verbose mode so you can see what tests were
797 run, and debug output.
798 · -torture
800 Run the torture tests as well as the normal set.
802 · -re=PATTERN
804 Filter the file list so that all the test files run match PATTERN.
806 Note that this form is distinct from the -re LIST OF PATTERNS form
807 below in that it allows the file list to be provided as well.
808 · -re LIST OF PATTERNS
810 Filter the file list so that all the test files run match
812 /(LIST|OF|PATTERNS)/. Note that with this form the patterns are
813 joined by '|' and you cannot supply a list of files, instead the
814 test files are obtained from the MANIFEST.
815 You can run an individual test by a command similar to
817 ./perl -I../lib path/to/foo.t
819 except that the harnesses set up some environment variables that may
821 affect the execution of the test:
822 · PERL_CORE=1
824 indicates that we're running this test as part of the perl core
826 test suite. This is useful for modules that have a dual life on
827 CPAN.
828 · PERL_DESTRUCT_LEVEL=2
830 is set to 2 if it isn't set already (see "PERL_DESTRUCT_LEVEL" in
832 perlhacktips).
833 · PERL
835 (used only by t/TEST) if set, overrides the path to the perl
837 executable that should be used to run the tests (the default being
838 ./perl).
839 · PERL_SKIP_TTY_TEST
841 if set, tells to skip the tests that need a terminal. It's
843 actually set automatically by the Makefile, but can also be forced
844 artificially by running 'make test_notty'.
845 Other environment variables that may influence tests
847 · PERL_TEST_Net_Ping
849 Setting this variable runs all the Net::Ping modules tests,
851 otherwise some tests that interact with the outside world are
852 skipped. See perl58delta.
853 · PERL_TEST_NOVREXX
855 Setting this variable skips the vrexx.t tests for OS2::REXX.
857 · PERL_TEST_NUMCONVERTS
859 This sets a variable in op/numconvert.t.
861 · PERL_TEST_MEMORY
863 Setting this variable includes the tests in t/bigmem/. This should
865 be set to the number of gigabytes of memory available for testing,
866 eg. "PERL_TEST_MEMORY=4" indicates that tests that require 4GiB of
867 available memory can be run safely.
868 See also the documentation for the Test and Test::Harness modules, for
870 more environment variables that affect testing.
871 Performance testing
873 The file t/perf/benchmarks contains snippets of perl code which are
874 intended to be benchmarked across a range of perls by the
875 Porting/bench.pl tool. If you fix or enhance a performance issue, you
876 may want to add a representative code sample to the file, then run
877 bench.pl against the previous and current perls to see what difference
878 it has made, and whether anything else has slowed down as a
879 consequence.
880 The file t/perf/opcount.t is designed to test whether a particular code
882 snippet has been compiled into an optree containing specified numbers
883 of particular op types. This is good for testing whether optimisations
884 which alter ops, such as converting an "aelem" op into an "aelemfast"
885 op, are really doing that.
886 The files t/perf/speed.t and t/re/speed.t are designed to test things
888 that run thousands of times slower if a particular optimisation is
889 broken (for example, the utf8 length cache on long utf8 strings). Add
890 a test that will take a fraction of a second normally, and minutes
891 otherwise, causing the test file to time out on failure.
892 Building perl at older commits
894 In the course of hacking on the Perl core distribution, you may have
895 occasion to configure, build and test perl at an old commit. Sometimes
896 "make" will fail during this process. If that happens, you may be able
897 to salvage the situation by using the Devel::PatchPerl library from
898 CPAN (not included in the core) to bring the source code at that commit
899 to a buildable state.
900 Here's a real world example, taken from work done to resolve perl
902 #72414 <https://rt.perl.org/Ticket/Display.html?id=72414>. Use of
903 Porting/bisect.pl had identified commit
904 "ba77e4cc9d1ceebf472c9c5c18b2377ee47062e6" as the commit in which a bug
905 was corrected. To confirm, a P5P developer wanted to configure and
906 build perl at commit "ba77e4c^" (presumably "bad") and then at
907 "ba77e4c" (presumably "good"). Normal configuration and build was
908 attempted:
909 $ sh ./Configure -des -Dusedevel
911 $ make test_prep
912 "make", however, failed with output (excerpted) like this:
914 cc -fstack-protector -L/usr/local/lib -o miniperl \
916 gv.o toke.o perly.o pad.o regcomp.o dump.o util.o \
917 mg.o reentr.o mro.o hv.o av.o run.o pp_hot.o sv.o \
918 pp.o scope.o pp_ctl.o pp_sys.o doop.o doio.o regexec.o \
919 utf8.o taint.o deb.o universal.o globals.o perlio.o \
920 perlapi.o numeric.o mathoms.o locale.o pp_pack.o pp_sort.o \
921 miniperlmain.o opmini.o perlmini.o
922 pp.o: In function `Perl_pp_pow':
923 pp.c:(.text+0x2db9): undefined reference to `pow'
924 ...
925 collect2: error: ld returned 1 exit status
926 makefile:348: recipe for target 'miniperl' failed
927 make: *** [miniperl] Error 1
928 Another P5P contributor recommended installation and use of
930 Devel::PatchPerl for this situation, first to determine the version of
931 perl at the commit in question, then to patch the source code at that
932 point to facilitate a build.
933 $ perl -MDevel::PatchPerl -e \
935 'print Devel::PatchPerl->determine_version("/path/to/sourcecode"), "\n";'
936 5.11.1
937 $ perl -MDevel::PatchPerl -e \
938 'Devel::PatchPerl->patch_source("5.11.1", "/path/to/sourcecode");'
939 Once the source was patched, "./Configure" and "make test_prep" were
941 called and completed successfully, enabling confirmation of the
942 findings in RT #72414.
943
945 To hack on the Perl guts, you'll need to read the following things:
946 · perlsource
948 An overview of the Perl source tree. This will help you find the
950 files you're looking for.
951 · perlinterp
953 An overview of the Perl interpreter source code and some details on
955 how Perl does what it does.
956 · perlhacktut
958 This document walks through the creation of a small patch to Perl's
960 C code. If you're just getting started with Perl core hacking,
961 this will help you understand how it works.
962 · perlhacktips
964 More details on hacking the Perl core. This document focuses on
966 lower level details such as how to write tests, compilation issues,
967 portability, debugging, etc.
968 If you plan on doing serious C hacking, make sure to read this.
970 · perlguts
972 This is of paramount importance, since it's the documentation of
974 what goes where in the Perl source. Read it over a couple of times
975 and it might start to make sense - don't worry if it doesn't yet,
976 because the best way to study it is to read it in conjunction with
977 poking at Perl source, and we'll do that later on.
978 Gisle Aas's "illustrated perlguts", also known as illguts, has very
980 helpful pictures:
981 <http://search.cpan.org/dist/illguts/>
983 · perlxstut and perlxs
985 A working knowledge of XSUB programming is incredibly useful for
987 core hacking; XSUBs use techniques drawn from the PP code, the
988 portion of the guts that actually executes a Perl program. It's a
989 lot gentler to learn those techniques from simple examples and
990 explanation than from the core itself.
991 · perlapi
993 The documentation for the Perl API explains what some of the
995 internal functions do, as well as the many macros used in the
996 source.
997 · Porting/pumpkin.pod
999 This is a collection of words of wisdom for a Perl porter; some of
1001 it is only useful to the pumpkin holder, but most of it applies to
1002 anyone wanting to go about Perl development.
1003
1005 The CPAN testers ( <http://testers.cpan.org/> ) are a group of
1006 volunteers who test CPAN modules on a variety of platforms.
1007 Perl Smokers ( <http://www.nntp.perl.org/group/perl.daily-build/> and
1009 <http://www.nntp.perl.org/group/perl.daily-build.reports/> )
1010 automatically test Perl source releases on platforms with various
1011 configurations.
1012 Both efforts welcome volunteers. In order to get involved in smoke
1014 testing of the perl itself visit
1015 <http://search.cpan.org/dist/Test-Smoke/>. In order to start smoke
1016 testing CPAN modules visit
1017 <http://search.cpan.org/dist/CPANPLUS-YACSmoke/> or
1018 <http://search.cpan.org/dist/minismokebox/> or
1019 <http://search.cpan.org/dist/CPAN-Reporter/>.
1020
1022 If you've read all the documentation in the document and the ones
1023 listed above, you're more than ready to hack on Perl.
1024 Here's some more recommendations
1026 · Subscribe to perl5-porters, follow the patches and try and
1028 understand them; don't be afraid to ask if there's a portion you're
1029 not clear on - who knows, you may unearth a bug in the patch...
1030 · Do read the README associated with your operating system, e.g.
1032 README.aix on the IBM AIX OS. Don't hesitate to supply patches to
1033 that README if you find anything missing or changed over a new OS
1034 release.
1035 · Find an area of Perl that seems interesting to you, and see if you
1037 can work out how it works. Scan through the source, and step over
1038 it in the debugger. Play, poke, investigate, fiddle! You'll
1039 probably get to understand not just your chosen area but a much
1040 wider range of perl's activity as well, and probably sooner than
1041 you'd think.
1042 "The Road goes ever on and on, down from the door where it began."
1044 If you can do these things, you've started on the long road to Perl
1045 porting. Thanks for wanting to help make Perl better - and happy
1046 hacking!
1047 Metaphoric Quotations
1049 If you recognized the quote about the Road above, you're in luck.
1050 Most software projects begin each file with a literal description of
1052 each file's purpose. Perl instead begins each with a literary allusion
1053 to that file's purpose.
1054 Like chapters in many books, all top-level Perl source files (along
1056 with a few others here and there) begin with an epigrammatic
1057 inscription that alludes, indirectly and metaphorically, to the
1058 material you're about to read.
1059 Quotations are taken from writings of J.R.R. Tolkien pertaining to his
1061 Legendarium, almost always from The Lord of the Rings. Chapters and
1062 page numbers are given using the following editions:
1063 · The Hobbit, by J.R.R. Tolkien. The hardcover, 70th-anniversary
1065 edition of 2007 was used, published in the UK by Harper Collins
1066 Publishers and in the US by the Houghton Mifflin Company.
1067 · The Lord of the Rings, by J.R.R. Tolkien. The hardcover,
1069 50th-anniversary edition of 2004 was used, published in the UK by
1070 Harper Collins Publishers and in the US by the Houghton Mifflin
1071 Company.
1072 · The Lays of Beleriand, by J.R.R. Tolkien and published posthumously
1074 by his son and literary executor, C.J.R. Tolkien, being the 3rd of
1075 the 12 volumes in Christopher's mammoth History of Middle Earth.
1076 Page numbers derive from the hardcover edition, first published in
1077 1983 by George Allen & Unwin; no page numbers changed for the
1078 special 3-volume omnibus edition of 2002 or the various trade-paper
1079 editions, all again now by Harper Collins or Houghton Mifflin.
1080 Other JRRT books fair game for quotes would thus include The Adventures
1082 of Tom Bombadil, The Silmarillion, Unfinished Tales, and The Tale of
1083 the Children of Hurin, all but the first posthumously assembled by
1084 CJRT. But The Lord of the Rings itself is perfectly fine and probably
1085 best to quote from, provided you can find a suitable quote there.
1086 So if you were to supply a new, complete, top-level source file to add
1088 to Perl, you should conform to this peculiar practice by yourself
1089 selecting an appropriate quotation from Tolkien, retaining the original
1090 spelling and punctuation and using the same format the rest of the
1091 quotes are in. Indirect and oblique is just fine; remember, it's a
1092 metaphor, so being meta is, after all, what it's for.
1093
1095 This document was originally written by Nathan Torkington, and is
1096 maintained by the perl5-porters mailing list.
1097perl v5.30.1 2019-11-29 PERLHACK(1)