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 Perlbug
11 bug 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 git://perl5.git.perl.org/perl.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 · Make your change
33 Hack, hack, hack. Keep in mind that Perl runs on many different
35 platforms, with different operating systems that have different
36 capabilities, different filesystem organizations, and even
37 different character sets. perlhacktips gives advice on this.
38 · Test your change
40 You can run all the tests with the following commands:
42 % ./Configure -des -Dusedevel
44 % make test
45 Keep hacking until the tests pass.
47 · Commit your change
49 Committing your work will save the change on your local system:
51 % git commit -a -m 'Commit message goes here'
53 Make sure the commit message describes your change in a single
55 sentence. For example, "Fixed spelling errors in perlhack.pod".
56 · Send your change to perlbug
58 The next step is to submit your patch to the Perl core ticket
60 system via email.
61 If your changes are in a single git commit, run the following
63 commands to generate the patch file and attach it to your bug
64 report:
65 % git format-patch -1
67 % ./perl -Ilib utils/perlbug -p 0001-*.patch
68 The perlbug program will ask you a few questions about your email
70 address and the patch you're submitting. Once you've answered them
71 it will submit your patch via email.
72 If your changes are in multiple commits, generate a patch file for
74 each one and provide them to perlbug's "-p" option separated by
75 commas:
76 % git format-patch -3
78 % ./perl -Ilib utils/perlbug -p 0001-fix1.patch,0002-fix2.patch,\
79 > 0003-fix3.patch
80 When prompted, pick a subject that summarizes your changes.
82 · Thank you
84 The porters appreciate the time you spent helping to make Perl
86 better. Thank you!
87 · Next time
89 The next time you wish to make a patch, you need to start from the
91 latest perl in a pristine state. Check you don't have any local
92 changes or added files in your perl check-out which you wish to
93 keep, then run these commands:
94 % git pull
96 % git reset --hard origin/blead
97 % git clean -dxf
98
100 If you want to report a bug in Perl, you must use the perlbug command
101 line tool. This tool will ensure that your bug report includes all the
102 relevant system and configuration information.
103 To browse existing Perl bugs and patches, you can use the web interface
105 at <http://rt.perl.org/>.
106 Please check the archive of the perl5-porters list (see below) and/or
108 the bug tracking system before submitting a bug report. Often, you'll
109 find that the bug has been reported already.
110 You can log in to the bug tracking system and comment on existing bug
112 reports. If you have additional information regarding an existing bug,
113 please add it. This will help the porters fix the bug.
114
116 The perl5-porters (p5p) mailing list is where the Perl standard
117 distribution is maintained and developed. The people who maintain Perl
118 are also referred to as the "Perl 5 Porters", "p5p" or just the
119 "porters".
120 A searchable archive of the list is available at
122 <http://markmail.org/search/?q=perl5-porters>. There is also an
123 archive at <http://archive.develooper.com/perl5-porters@perl.org/>.
124 perl-changes mailing list
126 The perl5-changes mailing list receives a copy of each patch that gets
127 submitted to the maintenance and development branches of the perl
128 repository. See <http://lists.perl.org/list/perl5-changes.html> for
129 subscription and archive information.
130 #p5p on IRC
132 Many porters are also active on the <irc://irc.perl.org/#p5p> channel.
133 Feel free to join the channel and ask questions about hacking on the
134 Perl core.
135
137 All of Perl's source code is kept centrally in a Git repository at
138 perl5.git.perl.org. The repository contains many Perl revisions from
139 Perl 1 onwards and all the revisions from Perforce, the previous
140 version control system.
141 For much more detail on using git with the Perl repository, please see
143 perlgit.
144 Read access via Git
146 You will need a copy of Git for your computer. You can fetch a copy of
147 the repository using the git protocol:
148 % git clone git://perl5.git.perl.org/perl.git perl
150 This clones the repository and makes a local copy in the perl
152 directory.
153 If you cannot use the git protocol for firewall reasons, you can also
155 clone via http, though this is much slower:
156 % git clone http://perl5.git.perl.org/perl.git perl
158 Read access via the web
160 You may access the repository over the web. This allows you to browse
161 the tree, see recent commits, subscribe to RSS feeds for the changes,
162 search for particular commits and more. You may access it at
163 <http://perl5.git.perl.org/perl.git>. A mirror of the repository is
164 found at <https://github.com/Perl/perl5>.
165 Read access via rsync
167 You can also choose to use rsync to get a copy of the current source
168 tree for the bleadperl branch and all maintenance branches:
169 % rsync -avz rsync://perl5.git.perl.org/perl-current .
171 % rsync -avz rsync://perl5.git.perl.org/perl-5.12.x .
172 % rsync -avz rsync://perl5.git.perl.org/perl-5.10.x .
173 % rsync -avz rsync://perl5.git.perl.org/perl-5.8.x .
174 % rsync -avz rsync://perl5.git.perl.org/perl-5.6.x .
175 % rsync -avz rsync://perl5.git.perl.org/perl-5.005xx .
176 (Add the "--delete" option to remove leftover files.)
178 To get a full list of the available sync points:
180 % rsync perl5.git.perl.org::
182 Write access via git
184 If you have a commit bit, please see perlgit for more details on using
185 git.
186
188 If you're planning to do more extensive work than a single small fix,
189 we encourage you to read the documentation below. This will help you
190 focus your work and make your patches easier to incorporate into the
191 Perl source.
192 Submitting patches
194 If you have a small patch to submit, please submit it via perlbug. You
195 can also send email directly to perlbug@perl.org. Please note that
196 messages sent to perlbug may be held in a moderation queue, so you
197 won't receive a response immediately.
198 You'll know your submission has been processed when you receive an
200 email from our ticket tracking system. This email will give you a
201 ticket number. Once your patch has made it to the ticket tracking
202 system, it will also be sent to the perl5-porters@perl.org list.
203 Patches are reviewed and discussed on the p5p list. Simple,
205 uncontroversial patches will usually be applied without any discussion.
206 When the patch is applied, the ticket will be updated and you will
207 receive email. In addition, an email will be sent to the p5p list.
208 In other cases, the patch will need more work or discussion. That will
210 happen on the p5p list.
211 You are encouraged to participate in the discussion and advocate for
213 your patch. Sometimes your patch may get lost in the shuffle. It's
214 appropriate to send a reminder email to p5p if no action has been taken
215 in a month. Please remember that the Perl 5 developers are all
216 volunteers, and be polite.
217 Changes are always applied directly to the main development branch,
219 called "blead". Some patches may be backported to a maintenance
220 branch. If you think your patch is appropriate for the maintenance
221 branch (see "MAINTENANCE BRANCHES" in perlpolicy), please explain why
222 when you submit it.
223 Getting your patch accepted
225 If you are submitting a code patch there are several things that you
226 can do to help the Perl 5 Porters accept your patch.
227 Patch style
229 If you used git to check out the Perl source, then using "git
231 format-patch" will produce a patch in a style suitable for Perl. The
232 "format-patch" command produces one patch file for each commit you
233 made. If you prefer to send a single patch for all commits, you can
234 use "git diff".
235 % git checkout blead
237 % git pull
238 % git diff blead my-branch-name
239 This produces a patch based on the difference between blead and your
241 current branch. It's important to make sure that blead is up to date
242 before producing the diff, that's why we call "git pull" first.
243 We strongly recommend that you use git if possible. It will make your
245 life easier, and ours as well.
246 However, if you're not using git, you can still produce a suitable
248 patch. You'll need a pristine copy of the Perl source to diff against.
249 The porters prefer unified diffs. Using GNU "diff", you can produce a
250 diff like this:
251 % diff -Npurd perl.pristine perl.mine
253 Make sure that you "make realclean" in your copy of Perl to remove any
255 build artifacts, or you may get a confusing result.
256 Commit message
258 As you craft each patch you intend to submit to the Perl core, it's
260 important to write a good commit message. This is especially important
261 if your submission will consist of a series of commits.
262 The first line of the commit message should be a short description
264 without a period. It should be no longer than the subject line of an
265 email, 50 characters being a good rule of thumb.
266 A lot of Git tools (Gitweb, GitHub, git log --pretty=oneline, ...) will
268 only display the first line (cut off at 50 characters) when presenting
269 commit summaries.
270 The commit message should include a description of the problem that the
272 patch corrects or new functionality that the patch adds.
273 As a general rule of thumb, your commit message should help a
275 programmer who knows the Perl core quickly understand what you were
276 trying to do, how you were trying to do it, and why the change matters
277 to Perl.
278 · Why
280 Your commit message should describe why the change you are making
282 is important. When someone looks at your change in six months or
283 six years, your intent should be clear.
284 If you're deprecating a feature with the intent of later
286 simplifying another bit of code, say so. If you're fixing a
287 performance problem or adding a new feature to support some other
288 bit of the core, mention that.
289 · What
291 Your commit message should describe what part of the Perl core
293 you're changing and what you expect your patch to do.
294 · How
296 While it's not necessary for documentation changes, new tests or
298 trivial patches, it's often worth explaining how your change works.
299 Even if it's clear to you today, it may not be clear to a porter
300 next month or next year.
301 A commit message isn't intended to take the place of comments in your
303 code. Commit messages should describe the change you made, while code
304 comments should describe the current state of the code.
305 If you've just implemented a new feature, complete with doc, tests and
307 well-commented code, a brief commit message will often suffice. If,
308 however, you've just changed a single character deep in the parser or
309 lexer, you might need to write a small novel to ensure that future
310 readers understand what you did and why you did it.
311 Comments, Comments, Comments
313 Be sure to adequately comment your code. While commenting every line
315 is unnecessary, anything that takes advantage of side effects of
316 operators, that creates changes that will be felt outside of the
317 function being patched, or that others may find confusing should be
318 documented. If you are going to err, it is better to err on the side
319 of adding too many comments than too few.
320 The best comments explain why the code does what it does, not what it
322 does.
323 Style
325 In general, please follow the particular style of the code you are
327 patching.
328 In particular, follow these general guidelines for patching Perl
330 sources:
331 · 4-wide indents for code, 2-wide indents for nested CPP "#define"s,
333 with 8-wide tabstops.
334 · Use spaces for indentation, not tab characters.
336 The codebase is a mixture of tabs and spaces for indentation, and
338 we are moving to spaces only. Converting lines you're patching
339 from 8-wide tabs to spaces will help this migration.
340 · Try hard not to exceed 79-columns
342 · ANSI C prototypes
344 · Uncuddled elses and "K&R" style for indenting control constructs
346 · No C++ style (//) comments
348 · Mark places that need to be revisited with XXX (and revisit often!)
350 · Opening brace lines up with "if" when conditional spans multiple
352 lines; should be at end-of-line otherwise
353 · In function definitions, name starts in column 0 (return value-type
355 is on previous line)
356 · Single space after keywords that are followed by parens, no space
358 between function name and following paren
359 · Avoid assignments in conditionals, but if they're unavoidable, use
361 extra paren, e.g. "if (a && (b = c)) ..."
362 · "return foo;" rather than "return(foo);"
364 · "if (!foo) ..." rather than "if (foo == FALSE) ..." etc.
366 · Do not declare variables using "register". It may be
368 counterproductive with modern compilers, and is deprecated in C++,
369 under which the Perl source is regularly compiled.
370 · In-line functions that are in headers that are accessible to XS
372 code need to be able to compile without warnings with commonly used
373 extra compilation flags, such as gcc's "-Wswitch-default" which
374 warns whenever a switch statement does not have a "default" case.
375 The use of these extra flags is to catch potential problems in
376 legal C code, and is often used by Perl aggregators, such as Linux
377 distributors.
378 Test suite
380 If your patch changes code (rather than just changing documentation),
382 you should also include one or more test cases which illustrate the bug
383 you're fixing or validate the new functionality you're adding. In
384 general, you should update an existing test file rather than create a
385 new one.
386 Your test suite additions should generally follow these guidelines
388 (courtesy of Gurusamy Sarathy <gsar@activestate.com>):
389 · Know what you're testing. Read the docs, and the source.
391 · Tend to fail, not succeed.
393 · Interpret results strictly.
395 · Use unrelated features (this will flush out bizarre interactions).
397 · Use non-standard idioms (otherwise you are not testing TIMTOWTDI).
399 · Avoid using hardcoded test numbers whenever possible (the
401 EXPECTED/GOT found in t/op/tie.t is much more maintainable, and
402 gives better failure reports).
403 · Give meaningful error messages when a test fails.
405 · Avoid using qx// and system() unless you are testing for them. If
407 you do use them, make sure that you cover _all_ perl platforms.
408 · Unlink any temporary files you create.
410 · Promote unforeseen warnings to errors with $SIG{__WARN__}.
412 · Be sure to use the libraries and modules shipped with the version
414 being tested, not those that were already installed.
415 · Add comments to the code explaining what you are testing for.
417 · Make updating the '1..42' string unnecessary. Or make sure that
419 you update it.
420 · Test _all_ behaviors of a given operator, library, or function.
422 Test all optional arguments.
424 Test return values in various contexts (boolean, scalar, list,
426 lvalue).
427 Use both global and lexical variables.
429 Don't forget the exceptional, pathological cases.
431 Patching a core module
433 This works just like patching anything else, with one extra
434 consideration.
435 Modules in the cpan/ directory of the source tree are maintained
437 outside of the Perl core. When the author updates the module, the
438 updates are simply copied into the core. See that module's
439 documentation or its listing on <http://search.cpan.org/> for more
440 information on reporting bugs and submitting patches.
441 In most cases, patches to modules in cpan/ should be sent upstream and
443 should not be applied to the Perl core individually. If a patch to a
444 file in cpan/ absolutely cannot wait for the fix to be made upstream,
445 released to CPAN and copied to blead, you must add (or update) a
446 "CUSTOMIZED" entry in the "Porting/Maintainers.pl" file to flag that a
447 local modification has been made. See "Porting/Maintainers.pl" for
448 more details.
449 In contrast, modules in the dist/ directory are maintained in the core.
451 Updating perldelta
453 For changes significant enough to warrant a pod/perldelta.pod entry,
454 the porters will greatly appreciate it if you submit a delta entry
455 along with your actual change. Significant changes include, but are
456 not limited to:
457 · Adding, deprecating, or removing core features
459 · Adding, deprecating, removing, or upgrading core or dual-life
461 modules
462 · Adding new core tests
464 · Fixing security issues and user-visible bugs in the core
466 · Changes that might break existing code, either on the perl or C
468 level
469 · Significant performance improvements
471 · Adding, removing, or significantly changing documentation in the
473 pod/ directory
474 · Important platform-specific changes
476 Please make sure you add the perldelta entry to the right section
478 within pod/perldelta.pod. More information on how to write good
479 perldelta entries is available in the "Style" section of
480 Porting/how_to_write_a_perldelta.pod.
481 What makes for a good patch?
483 New features and extensions to the language can be contentious. There
484 is no specific set of criteria which determine what features get added,
485 but here are some questions to consider when developing a patch:
486 Does the concept match the general goals of Perl?
488 Our goals include, but are not limited to:
490 1. Keep it fast, simple, and useful.
492 2. Keep features/concepts as orthogonal as possible.
494 3. No arbitrary limits (platforms, data sizes, cultures).
496 4. Keep it open and exciting to use/patch/advocate Perl everywhere.
498 5. Either assimilate new technologies, or build bridges to them.
500 Where is the implementation?
502 All the talk in the world is useless without an implementation. In
504 almost every case, the person or people who argue for a new feature
505 will be expected to be the ones who implement it. Porters capable of
506 coding new features have their own agendas, and are not available to
507 implement your (possibly good) idea.
508 Backwards compatibility
510 It's a cardinal sin to break existing Perl programs. New warnings can
512 be contentious--some say that a program that emits warnings is not
513 broken, while others say it is. Adding keywords has the potential to
514 break programs, changing the meaning of existing token sequences or
515 functions might break programs.
516 The Perl 5 core includes mechanisms to help porters make backwards
518 incompatible changes more compatible such as the feature and deprecate
519 modules. Please use them when appropriate.
520 Could it be a module instead?
522 Perl 5 has extension mechanisms, modules and XS, specifically to avoid
524 the need to keep changing the Perl interpreter. You can write modules
525 that export functions, you can give those functions prototypes so they
526 can be called like built-in functions, you can even write XS code to
527 mess with the runtime data structures of the Perl interpreter if you
528 want to implement really complicated things.
529 Whenever possible, new features should be prototyped in a CPAN module
531 before they will be considered for the core.
532 Is the feature generic enough?
534 Is this something that only the submitter wants added to the language,
536 or is it broadly useful? Sometimes, instead of adding a feature with a
537 tight focus, the porters might decide to wait until someone implements
538 the more generalized feature.
539 Does it potentially introduce new bugs?
541 Radical rewrites of large chunks of the Perl interpreter have the
543 potential to introduce new bugs.
544 How big is it?
546 The smaller and more localized the change, the better. Similarly, a
548 series of small patches is greatly preferred over a single large patch.
549 Does it preclude other desirable features?
551 A patch is likely to be rejected if it closes off future avenues of
553 development. For instance, a patch that placed a true and final
554 interpretation on prototypes is likely to be rejected because there are
555 still options for the future of prototypes that haven't been addressed.
556 Is the implementation robust?
558 Good patches (tight code, complete, correct) stand more chance of going
560 in. Sloppy or incorrect patches might be placed on the back burner
561 until the pumpking has time to fix, or might be discarded altogether
562 without further notice.
563 Is the implementation generic enough to be portable?
565 The worst patches make use of system-specific features. It's highly
567 unlikely that non-portable additions to the Perl language will be
568 accepted.
569 Is the implementation tested?
571 Patches which change behaviour (fixing bugs or introducing new
573 features) must include regression tests to verify that everything works
574 as expected.
575 Without tests provided by the original author, how can anyone else
577 changing perl in the future be sure that they haven't unwittingly
578 broken the behaviour the patch implements? And without tests, how can
579 the patch's author be confident that his/her hard work put into the
580 patch won't be accidentally thrown away by someone in the future?
581 Is there enough documentation?
583 Patches without documentation are probably ill-thought out or
585 incomplete. No features can be added or changed without documentation,
586 so submitting a patch for the appropriate pod docs as well as the
587 source code is important.
588 Is there another way to do it?
590 Larry said "Although the Perl Slogan is There's More Than One Way to Do
592 It, I hesitate to make 10 ways to do something". This is a tricky
593 heuristic to navigate, though--one man's essential addition is another
594 man's pointless cruft.
595 Does it create too much work?
597 Work for the pumpking, work for Perl programmers, work for module
599 authors, ... Perl is supposed to be easy.
600 Patches speak louder than words
602 Working code is always preferred to pie-in-the-sky ideas. A patch to
604 add a feature stands a much higher chance of making it to the language
605 than does a random feature request, no matter how fervently argued the
606 request might be. This ties into "Will it be useful?", as the fact
607 that someone took the time to make the patch demonstrates a strong
608 desire for the feature.
609
611 The core uses the same testing style as the rest of Perl, a simple
612 "ok/not ok" run through Test::Harness, but there are a few special
613 considerations.
614 There are three ways to write a test in the core: Test::More, t/test.pl
616 and ad hoc "print $test ? "ok 42\n" : "not ok 42\n"". The decision of
617 which to use depends on what part of the test suite you're working on.
618 This is a measure to prevent a high-level failure (such as Config.pm
619 breaking) from causing basic functionality tests to fail.
620 The t/test.pl library provides some of the features of Test::More, but
622 avoids loading most modules and uses as few core features as possible.
623 If you write your own test, use the Test Anything Protocol
625 <http://testanything.org>.
626 · t/base, t/comp and t/opbasic
628 Since we don't know if "require" works, or even subroutines, use ad
630 hoc tests for these three. Step carefully to avoid using the
631 feature being tested. Tests in t/opbasic, for instance, have been
632 placed there rather than in t/op because they test functionality
633 which t/test.pl presumes has already been demonstrated to work.
634 · t/cmd, t/run, t/io and t/op
636 Now that basic require() and subroutines are tested, you can use
638 the t/test.pl library.
639 You can also use certain libraries like Config conditionally, but
641 be sure to skip the test gracefully if it's not there.
642 · Everything else
644 Now that the core of Perl is tested, Test::More can and should be
646 used. You can also use the full suite of core modules in the
647 tests.
648 When you say "make test", Perl uses the t/TEST program to run the test
650 suite (except under Win32 where it uses t/harness instead). All tests
651 are run from the t/ directory, not the directory which contains the
652 test. This causes some problems with the tests in lib/, so here's some
653 opportunity for some patching.
654 You must be triply conscious of cross-platform concerns. This usually
656 boils down to using File::Spec, avoiding things like "fork()" and
657 "system()" unless absolutely necessary, and not assuming that a given
658 character has a particular ordinal value (code point) or that its UTF-8
659 representation is composed of particular bytes.
660 There are several functions available to specify characters and code
662 points portably in tests. The always-preloaded functions
663 "utf8::unicode_to_native()" and its inverse "utf8::native_to_unicode()"
664 take code points and translate appropriately. The file
665 t/charset_tools.pl has several functions that can be useful. It has
666 versions of the previous two functions that take strings as inputs --
667 not single numeric code points: "uni_to_native()" and
668 "native_to_uni()". If you must look at the individual bytes comprising
669 a UTF-8 encoded string, "byte_utf8a_to_utf8n()" takes as input a string
670 of those bytes encoded for an ASCII platform, and returns the
671 equivalent string in the native platform. For example,
672 "byte_utf8a_to_utf8n("\xC2\xA0")" returns the byte sequence on the
673 current platform that form the UTF-8 for "U+00A0", since "\xC2\xA0" are
674 the UTF-8 bytes on an ASCII platform for that code point. This
675 function returns "\xC2\xA0" on an ASCII platform, and "\x80\x41" on an
676 EBCDIC 1047 one.
677 But easiest is, if the character is specifiable as a literal, like "A"
679 or "%", to use that; if not so specificable, you can use use "\N{}" ,
680 if the side effects aren't troublesome. Simply specify all your
681 characters in hex, using "\N{U+ZZ}" instead of "\xZZ". "\N{}" is the
682 Unicode name, and so it always gives you the Unicode character.
683 "\N{U+41}" is the character whose Unicode code point is 0x41, hence is
684 'A' on all platforms. The side effects are:
685 · These select Unicode rules. That means that in double-quotish
687 strings, the string is always converted to UTF-8 to force a Unicode
688 interpretation (you can "utf8::downgrade()" afterwards to convert
689 back to non-UTF8, if possible). In regular expression patterns,
690 the conversion isn't done, but if the character set modifier would
691 otherwise be "/d", it is changed to "/u".
692 · If you use the form "\N{character name}", the charnames module gets
694 automatically loaded. This may not be suitable for the test level
695 you are doing.
696 If you are testing locales (see perllocale), there are helper functions
698 in t/loc_tools.pl to enable you to see what locales there are on the
699 current platform.
700 Special "make test" targets
702 There are various special make targets that can be used to test Perl
703 slightly differently than the standard "test" target. Not all them are
704 expected to give a 100% success rate. Many of them have several
705 aliases, and many of them are not available on certain operating
706 systems.
707 · test_porting
709 This runs some basic sanity tests on the source tree and helps
711 catch basic errors before you submit a patch.
712 · minitest
714 Run miniperl on t/base, t/comp, t/cmd, t/run, t/io, t/op, t/uni and
716 t/mro tests.
717 · test.valgrind check.valgrind
719 (Only in Linux) Run all the tests using the memory leak + naughty
721 memory access tool "valgrind". The log files will be named
722 testname.valgrind.
723 · test_harness
725 Run the test suite with the t/harness controlling program, instead
727 of t/TEST. t/harness is more sophisticated, and uses the
728 Test::Harness module, thus using this test target supposes that
729 perl mostly works. The main advantage for our purposes is that it
730 prints a detailed summary of failed tests at the end. Also, unlike
731 t/TEST, it doesn't redirect stderr to stdout.
732 Note that under Win32 t/harness is always used instead of t/TEST,
734 so there is no special "test_harness" target.
735 Under Win32's "test" target you may use the TEST_SWITCHES and
737 TEST_FILES environment variables to control the behaviour of
738 t/harness. This means you can say
739 nmake test TEST_FILES="op/*.t"
741 nmake test TEST_SWITCHES="-torture" TEST_FILES="op/*.t"
742 · test-notty test_notty
744 Sets PERL_SKIP_TTY_TEST to true before running normal test.
746 Parallel tests
748 The core distribution can now run its regression tests in parallel on
749 Unix-like platforms. Instead of running "make test", set "TEST_JOBS"
750 in your environment to the number of tests to run in parallel, and run
751 "make test_harness". On a Bourne-like shell, this can be done as
752 TEST_JOBS=3 make test_harness # Run 3 tests in parallel
754 An environment variable is used, rather than parallel make itself,
756 because TAP::Harness needs to be able to schedule individual non-
757 conflicting test scripts itself, and there is no standard interface to
758 "make" utilities to interact with their job schedulers.
759 Note that currently some test scripts may fail when run in parallel
761 (most notably dist/IO/t/io_dir.t). If necessary, run just the failing
762 scripts again sequentially and see if the failures go away.
763 Running tests by hand
765 You can run part of the test suite by hand by using one of the
766 following commands from the t/ directory:
767 ./perl -I../lib TEST list-of-.t-files
769 or
771 ./perl -I../lib harness list-of-.t-files
773 (If you don't specify test scripts, the whole test suite will be run.)
775 Using t/harness for testing
777 If you use "harness" for testing, you have several command line options
778 available to you. The arguments are as follows, and are in the order
779 that they must appear if used together.
780 harness -v -torture -re=pattern LIST OF FILES TO TEST
782 harness -v -torture -re LIST OF PATTERNS TO MATCH
783 If "LIST OF FILES TO TEST" is omitted, the file list is obtained from
785 the manifest. The file list may include shell wildcards which will be
786 expanded out.
787 · -v
789 Run the tests under verbose mode so you can see what tests were
791 run, and debug output.
792 · -torture
794 Run the torture tests as well as the normal set.
796 · -re=PATTERN
798 Filter the file list so that all the test files run match PATTERN.
800 Note that this form is distinct from the -re LIST OF PATTERNS form
801 below in that it allows the file list to be provided as well.
802 · -re LIST OF PATTERNS
804 Filter the file list so that all the test files run match
806 /(LIST|OF|PATTERNS)/. Note that with this form the patterns are
807 joined by '|' and you cannot supply a list of files, instead the
808 test files are obtained from the MANIFEST.
809 You can run an individual test by a command similar to
811 ./perl -I../lib path/to/foo.t
813 except that the harnesses set up some environment variables that may
815 affect the execution of the test:
816 · PERL_CORE=1
818 indicates that we're running this test as part of the perl core
820 test suite. This is useful for modules that have a dual life on
821 CPAN.
822 · PERL_DESTRUCT_LEVEL=2
824 is set to 2 if it isn't set already (see "PERL_DESTRUCT_LEVEL" in
826 perlhacktips).
827 · PERL
829 (used only by t/TEST) if set, overrides the path to the perl
831 executable that should be used to run the tests (the default being
832 ./perl).
833 · PERL_SKIP_TTY_TEST
835 if set, tells to skip the tests that need a terminal. It's
837 actually set automatically by the Makefile, but can also be forced
838 artificially by running 'make test_notty'.
839 Other environment variables that may influence tests
841 · PERL_TEST_Net_Ping
843 Setting this variable runs all the Net::Ping modules tests,
845 otherwise some tests that interact with the outside world are
846 skipped. See perl58delta.
847 · PERL_TEST_NOVREXX
849 Setting this variable skips the vrexx.t tests for OS2::REXX.
851 · PERL_TEST_NUMCONVERTS
853 This sets a variable in op/numconvert.t.
855 · PERL_TEST_MEMORY
857 Setting this variable includes the tests in t/bigmem/. This should
859 be set to the number of gigabytes of memory available for testing,
860 eg. "PERL_TEST_MEMORY=4" indicates that tests that require 4GiB of
861 available memory can be run safely.
862 See also the documentation for the Test and Test::Harness modules, for
864 more environment variables that affect testing.
865 Performance testing
867 The file t/perf/benchmarks contains snippets of perl code which are
868 intended to be benchmarked across a range of perls by the
869 Porting/bench.pl tool. If you fix or enhance a performance issue, you
870 may want to add a representative code sample to the file, then run
871 bench.pl against the previous and current perls to see what difference
872 it has made, and whether anything else has slowed down as a
873 consequence.
874 The file t/perf/opcount.t is designed to test whether a particular code
876 snippet has been compiled into an optree containing specified numbers
877 of particular op types. This is good for testing whether optimisations
878 which alter ops, such as converting an "aelem" op into an "aelemfast"
879 op, are really doing that.
880 The files t/perf/speed.t and t/re/speed.t are designed to test things
882 that run thousands of times slower if a particular optimisation is
883 broken (for example, the utf8 length cache on long utf8 strings). Add
884 a test that will take a fraction of a second normally, and minutes
885 otherwise, causing the test file to time out on failure.
886
888 To hack on the Perl guts, you'll need to read the following things:
889 · perlsource
891 An overview of the Perl source tree. This will help you find the
893 files you're looking for.
894 · perlinterp
896 An overview of the Perl interpreter source code and some details on
898 how Perl does what it does.
899 · perlhacktut
901 This document walks through the creation of a small patch to Perl's
903 C code. If you're just getting started with Perl core hacking,
904 this will help you understand how it works.
905 · perlhacktips
907 More details on hacking the Perl core. This document focuses on
909 lower level details such as how to write tests, compilation issues,
910 portability, debugging, etc.
911 If you plan on doing serious C hacking, make sure to read this.
913 · perlguts
915 This is of paramount importance, since it's the documentation of
917 what goes where in the Perl source. Read it over a couple of times
918 and it might start to make sense - don't worry if it doesn't yet,
919 because the best way to study it is to read it in conjunction with
920 poking at Perl source, and we'll do that later on.
921 Gisle Aas's "illustrated perlguts", also known as illguts, has very
923 helpful pictures:
924 <http://search.cpan.org/dist/illguts/>
926 · perlxstut and perlxs
928 A working knowledge of XSUB programming is incredibly useful for
930 core hacking; XSUBs use techniques drawn from the PP code, the
931 portion of the guts that actually executes a Perl program. It's a
932 lot gentler to learn those techniques from simple examples and
933 explanation than from the core itself.
934 · perlapi
936 The documentation for the Perl API explains what some of the
938 internal functions do, as well as the many macros used in the
939 source.
940 · Porting/pumpkin.pod
942 This is a collection of words of wisdom for a Perl porter; some of
944 it is only useful to the pumpkin holder, but most of it applies to
945 anyone wanting to go about Perl development.
946
948 The CPAN testers ( <http://testers.cpan.org/> ) are a group of
949 volunteers who test CPAN modules on a variety of platforms.
950 Perl Smokers ( <http://www.nntp.perl.org/group/perl.daily-build/> and
952 <http://www.nntp.perl.org/group/perl.daily-build.reports/> )
953 automatically test Perl source releases on platforms with various
954 configurations.
955 Both efforts welcome volunteers. In order to get involved in smoke
957 testing of the perl itself visit
958 <http://search.cpan.org/dist/Test-Smoke/>. In order to start smoke
959 testing CPAN modules visit
960 <http://search.cpan.org/dist/CPANPLUS-YACSmoke/> or
961 <http://search.cpan.org/dist/minismokebox/> or
962 <http://search.cpan.org/dist/CPAN-Reporter/>.
963
965 If you've read all the documentation in the document and the ones
966 listed above, you're more than ready to hack on Perl.
967 Here's some more recommendations
969 · Subscribe to perl5-porters, follow the patches and try and
971 understand them; don't be afraid to ask if there's a portion you're
972 not clear on - who knows, you may unearth a bug in the patch...
973 · Do read the README associated with your operating system, e.g.
975 README.aix on the IBM AIX OS. Don't hesitate to supply patches to
976 that README if you find anything missing or changed over a new OS
977 release.
978 · Find an area of Perl that seems interesting to you, and see if you
980 can work out how it works. Scan through the source, and step over
981 it in the debugger. Play, poke, investigate, fiddle! You'll
982 probably get to understand not just your chosen area but a much
983 wider range of perl's activity as well, and probably sooner than
984 you'd think.
985 "The Road goes ever on and on, down from the door where it began."
987 If you can do these things, you've started on the long road to Perl
988 porting. Thanks for wanting to help make Perl better - and happy
989 hacking!
990 Metaphoric Quotations
992 If you recognized the quote about the Road above, you're in luck.
993 Most software projects begin each file with a literal description of
995 each file's purpose. Perl instead begins each with a literary allusion
996 to that file's purpose.
997 Like chapters in many books, all top-level Perl source files (along
999 with a few others here and there) begin with an epigrammatic
1000 inscription that alludes, indirectly and metaphorically, to the
1001 material you're about to read.
1002 Quotations are taken from writings of J.R.R. Tolkien pertaining to his
1004 Legendarium, almost always from The Lord of the Rings. Chapters and
1005 page numbers are given using the following editions:
1006 · The Hobbit, by J.R.R. Tolkien. The hardcover, 70th-anniversary
1008 edition of 2007 was used, published in the UK by Harper Collins
1009 Publishers and in the US by the Houghton Mifflin Company.
1010 · The Lord of the Rings, by J.R.R. Tolkien. The hardcover,
1012 50th-anniversary edition of 2004 was used, published in the UK by
1013 Harper Collins Publishers and in the US by the Houghton Mifflin
1014 Company.
1015 · The Lays of Beleriand, by J.R.R. Tolkien and published posthumously
1017 by his son and literary executor, C.J.R. Tolkien, being the 3rd of
1018 the 12 volumes in Christopher's mammoth History of Middle Earth.
1019 Page numbers derive from the hardcover edition, first published in
1020 1983 by George Allen & Unwin; no page numbers changed for the
1021 special 3-volume omnibus edition of 2002 or the various trade-paper
1022 editions, all again now by Harper Collins or Houghton Mifflin.
1023 Other JRRT books fair game for quotes would thus include The Adventures
1025 of Tom Bombadil, The Silmarillion, Unfinished Tales, and The Tale of
1026 the Children of Hurin, all but the first posthumously assembled by
1027 CJRT. But The Lord of the Rings itself is perfectly fine and probably
1028 best to quote from, provided you can find a suitable quote there.
1029 So if you were to supply a new, complete, top-level source file to add
1031 to Perl, you should conform to this peculiar practice by yourself
1032 selecting an appropriate quotation from Tolkien, retaining the original
1033 spelling and punctuation and using the same format the rest of the
1034 quotes are in. Indirect and oblique is just fine; remember, it's a
1035 metaphor, so being meta is, after all, what it's for.
1036
1038 This document was originally written by Nathan Torkington, and is
1039 maintained by the perl5-porters mailing list.
1040perl v5.26.3 2018-03-23 PERLHACK(1)