perlhack - How to hack on Perl
This document explains how Perl development works. It includes details about the Perl 5 Porters email list, the Perl repository, the Perlbug bug tracker, patch guidelines, and commentary on Perl development philosophy.
If you just want to submit a single small patch like a pod fix, a test for a bug, comment fixes, etc., it's easy! Here's how:
The perl source is in a git repository. You can clone the repository with the following command:
% git clone git://perl5.git.perl.org/perl.git perl
In case the advice in this guide has been updated recently, read the latest version directly from the perl source:
% perldoc pod/perlhack.pod
Hack, hack, hack.
You can run all the tests with the following commands:
% ./Configure -des -Dusedevel % make test
Keep hacking until the tests pass.
Committing your work will save the change on your local system:
% git commit -a -m 'Commit message goes here'
Make sure the commit message describes your change in a single sentence. For example, "Fixed spelling errors in perlhack.pod".
The next step is to submit your patch to the Perl core ticket system via email.
If your changes are in a single git commit, run the following commands to write the file as a MIME attachment and send it with a meaningful subject:
% git format-patch -1 --attach % ./perl -Ilib utils/perlbug -s "[PATCH] $( git log -1 --oneline HEAD)" -f 0001-*.patch
The perlbug program will ask you a few questions about your email address and the patch you're submitting. Once you've answered them it will submit your patch via email.
If your changes are in multiple commits, generate a patch file containing them all, and attach that:
% git format-patch origin/blead --attach --stdout > patches % ./perl -Ilib utils/perlbug -f patches
When prompted, pick a subject that summarizes your changes overall and has "[PATCH]" at the beginning.
The porters appreciate the time you spent helping to make Perl better. Thank you!
The next time you wish to make a patch, you need to start from the latest perl in a pristine state. Check you don't have any local changes or added files in your perl check-out which you wish to keep, then run these commands:
% git pull % git reset --hard origin/blead % git clean -dxf
If you want to report a bug in Perl, you must use the perlbug command line tool. This tool will ensure that your bug report includes all the relevant system and configuration information.
To browse existing Perl bugs and patches, you can use the web interface at http://rt.perl.org/.
Please check the archive of the perl5-porters list (see below) and/or the bug tracking system before submitting a bug report. Often, you'll find that the bug has been reported already.
You can log in to the bug tracking system and comment on existing bug reports. If you have additional information regarding an existing bug, please add it. This will help the porters fix the bug.
The perl5-porters (p5p) mailing list is where the Perl standard distribution is maintained and developed. The people who maintain Perl are also referred to as the "Perl 5 Porters", "p5p" or just the "porters".
A searchable archive of the list is available at http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/. There is also another archive at http://firstname.lastname@example.org/.
The perl5-changes mailing list receives a copy of each patch that gets submitted to the maintenance and development branches of the perl repository. See http://lists.perl.org/list/perl5-changes.html for subscription and archive information.
Many porters are also active on the irc://irc.perl.org/#p5p channel. Feel free to join the channel and ask questions about hacking on the Perl core.
All of Perl's source code is kept centrally in a Git repository at perl5.git.perl.org. The repository contains many Perl revisions from Perl 1 onwards and all the revisions from Perforce, the previous version control system.
For much more detail on using git with the Perl repository, please see perlgit.
You will need a copy of Git for your computer. You can fetch a copy of the repository using the git protocol:
% git clone git://perl5.git.perl.org/perl.git perl
This clones the repository and makes a local copy in the perl directory.
If you cannot use the git protocol for firewall reasons, you can also clone via http, though this is much slower:
% git clone http://perl5.git.perl.org/perl.git perl
You may access the repository over the web. This allows you to browse the tree, see recent commits, subscribe to RSS feeds for the changes, search for particular commits and more. You may access it at http://perl5.git.perl.org/perl.git. A mirror of the repository is found at http://github.com/mirrors/perl.
You can also choose to use rsync to get a copy of the current source tree for the bleadperl branch and all maintenance branches:
% rsync -avz rsync://perl5.git.perl.org/perl-current . % rsync -avz rsync://perl5.git.perl.org/perl-5.12.x . % rsync -avz rsync://perl5.git.perl.org/perl-5.10.x . % rsync -avz rsync://perl5.git.perl.org/perl-5.8.x . % rsync -avz rsync://perl5.git.perl.org/perl-5.6.x . % rsync -avz rsync://perl5.git.perl.org/perl-5.005xx .
--delete option to remove leftover files.)
To get a full list of the available sync points:
% rsync perl5.git.perl.org::
If you have a commit bit, please see perlgit for more details on using git.
If you're planning to do more extensive work than a single small fix, we encourage you to read the documentation below. This will help you focus your work and make your patches easier to incorporate into the Perl source.
If you have a small patch to submit, please submit it via perlbug. You can also send email directly to email@example.com. Please note that messages sent to perlbug may be held in a moderation queue, so you won't receive a response immediately.
You'll know your submission has been processed when you receive an email from our ticket tracking system. This email will give you a ticket number. Once your patch has made it to the ticket tracking system, it will also be sent to the firstname.lastname@example.org list.
Patches are reviewed and discussed on the p5p list. Simple, uncontroversial patches will usually be applied without any discussion. When the patch is applied, the ticket will be updated and you will receive email. In addition, an email will be sent to the p5p list.
In other cases, the patch will need more work or discussion. That will happen on the p5p list.
You are encouraged to participate in the discussion and advocate for your patch. Sometimes your patch may get lost in the shuffle. It's appropriate to send a reminder email to p5p if no action has been taken in a month. Please remember that the Perl 5 developers are all volunteers, and be polite.
Changes are always applied directly to the main development branch, called "blead". Some patches may be backported to a maintenance branch. If you think your patch is appropriate for the maintenance branch (see "MAINTENANCE BRANCHES" in perlpolicy), please explain why when you submit it.
If you are submitting a code patch there are several things that you can do to help the Perl 5 Porters accept your patch.
If you used git to check out the Perl source, then using
git format-patch will produce a patch in a style suitable for Perl. The
format-patch command produces one patch file for each commit you made. If you prefer to send a single patch for all commits, you can use
% git checkout blead % git pull % git diff blead my-branch-name
This produces a patch based on the difference between blead and your current branch. It's important to make sure that blead is up to date before producing the diff, that's why we call
git pull first.
We strongly recommend that you use git if possible. It will make your life easier, and ours as well.
However, if you're not using git, you can still produce a suitable patch. You'll need a pristine copy of the Perl source to diff against. The porters prefer unified diffs. Using GNU
diff, you can produce a diff like this:
% diff -Npurd perl.pristine perl.mine
Make sure that you
make realclean in your copy of Perl to remove any build artifacts, or you may get a confusing result.
As you craft each patch you intend to submit to the Perl core, it's important to write a good commit message. This is especially important if your submission will consist of a series of commits.
The first line of the commit message should be a short description without a period. It should be no longer than the subject line of an email, 50 characters being a good rule of thumb.
A lot of Git tools (Gitweb, GitHub, git log --pretty=oneline, ...) will only display the first line (cut off at 50 characters) when presenting commit summaries.
The commit message should include a description of the problem that the patch corrects or new functionality that the patch adds.
As a general rule of thumb, your commit message should help a programmer who knows the Perl core quickly understand what you were trying to do, how you were trying to do it, and why the change matters to Perl.
Your commit message should describe why the change you are making is important. When someone looks at your change in six months or six years, your intent should be clear.
If you're deprecating a feature with the intent of later simplifying another bit of code, say so. If you're fixing a performance problem or adding a new feature to support some other bit of the core, mention that.
Your commit message should describe what part of the Perl core you're changing and what you expect your patch to do.
While it's not necessary for documentation changes, new tests or trivial patches, it's often worth explaining how your change works. Even if it's clear to you today, it may not be clear to a porter next month or next year.
A commit message isn't intended to take the place of comments in your code. Commit messages should describe the change you made, while code comments should describe the current state of the code.
If you've just implemented a new feature, complete with doc, tests and well-commented code, a brief commit message will often suffice. If, however, you've just changed a single character deep in the parser or lexer, you might need to write a small novel to ensure that future readers understand what you did and why you did it.
Be sure to adequately comment your code. While commenting every line is unnecessary, anything that takes advantage of side effects of operators, that creates changes that will be felt outside of the function being patched, or that others may find confusing should be documented. If you are going to err, it is better to err on the side of adding too many comments than too few.
The best comments explain why the code does what it does, not what it does.
In general, please follow the particular style of the code you are patching.
In particular, follow these general guidelines for patching Perl sources:
-Wswitch-defaultwhich warns whenever a switch statement does not have a "default" case. The use of these extra flags is to catch potential problems in legal C code, and is often used by Perl aggregators, such as Linux distributors.
If your patch changes code (rather than just changing documentation), you should also include one or more test cases which illustrate the bug you're fixing or validate the new functionality you're adding. In general, you should update an existing test file rather than create a new one.
Your test suite additions should generally follow these guidelines (courtesy of Gurusamy Sarathy <email@example.com>):
Test all optional arguments.
Test return values in various contexts (boolean, scalar, list, lvalue).
Use both global and lexical variables.
Don't forget the exceptional, pathological cases.
This works just like patching anything else, with one extra consideration.
Modules in the cpan/ directory of the source tree are maintained outside of the Perl core. When the author updates the module, the updates are simply copied into the core. See that module's documentation or its listing on http://search.cpan.org/ for more information on reporting bugs and submitting patches.
In most cases, patches to modules in cpan/ should be sent upstream and should not be applied to the Perl core individually. If a patch to a file in cpan/ absolutely cannot wait for the fix to be made upstream, released to CPAN and copied to blead, you must add (or update) a
CUSTOMIZED entry in the "Porting/Maintainers.pl" file to flag that a local modification has been made. See "Porting/Maintainers.pl" for more details.
In contrast, modules in the dist/ directory are maintained in the core.
For changes significant enough to warrant a pod/perldelta.pod entry, the porters will greatly appreciate it if you submit a delta entry along with your actual change. Significant changes include, but are not limited to:
Please make sure you add the perldelta entry to the right section within pod/perldelta.pod. More information on how to write good perldelta entries is available in the
Style section of Porting/how_to_write_a_perldelta.pod.
New features and extensions to the language can be contentious. There is no specific set of criteria which determine what features get added, but here are some questions to consider when developing a patch:
Our goals include, but are not limited to:
All the talk in the world is useless without an implementation. In almost every case, the person or people who argue for a new feature will be expected to be the ones who implement it. Porters capable of coding new features have their own agendas, and are not available to implement your (possibly good) idea.
It's a cardinal sin to break existing Perl programs. New warnings can be contentious--some say that a program that emits warnings is not broken, while others say it is. Adding keywords has the potential to break programs, changing the meaning of existing token sequences or functions might break programs.
Perl 5 has extension mechanisms, modules and XS, specifically to avoid the need to keep changing the Perl interpreter. You can write modules that export functions, you can give those functions prototypes so they can be called like built-in functions, you can even write XS code to mess with the runtime data structures of the Perl interpreter if you want to implement really complicated things.
Whenever possible, new features should be prototyped in a CPAN module before they will be considered for the core.
Is this something that only the submitter wants added to the language, or is it broadly useful? Sometimes, instead of adding a feature with a tight focus, the porters might decide to wait until someone implements the more generalized feature.
Radical rewrites of large chunks of the Perl interpreter have the potential to introduce new bugs.
The smaller and more localized the change, the better. Similarly, a series of small patches is greatly preferred over a single large patch.
A patch is likely to be rejected if it closes off future avenues of development. For instance, a patch that placed a true and final interpretation on prototypes is likely to be rejected because there are still options for the future of prototypes that haven't been addressed.
Good patches (tight code, complete, correct) stand more chance of going in. Sloppy or incorrect patches might be placed on the back burner until the pumpking has time to fix, or might be discarded altogether without further notice.
The worst patches make use of system-specific features. It's highly unlikely that non-portable additions to the Perl language will be accepted.
Patches which change behaviour (fixing bugs or introducing new features) must include regression tests to verify that everything works as expected.
Without tests provided by the original author, how can anyone else changing perl in the future be sure that they haven't unwittingly broken the behaviour the patch implements? And without tests, how can the patch's author be confident that his/her hard work put into the patch won't be accidentally thrown away by someone in the future?
Patches without documentation are probably ill-thought out or incomplete. No features can be added or changed without documentation, so submitting a patch for the appropriate pod docs as well as the source code is important.
Larry said "Although the Perl Slogan is There's More Than One Way to Do It, I hesitate to make 10 ways to do something". This is a tricky heuristic to navigate, though--one man's essential addition is another man's pointless cruft.
Work for the pumpking, work for Perl programmers, work for module authors, ... Perl is supposed to be easy.
Working code is always preferred to pie-in-the-sky ideas. A patch to add a feature stands a much higher chance of making it to the language than does a random feature request, no matter how fervently argued the request might be. This ties into "Will it be useful?", as the fact that someone took the time to make the patch demonstrates a strong desire for the feature.
The core uses the same testing style as the rest of Perl, a simple "ok/not ok" run through Test::Harness, but there are a few special considerations.
There are three ways to write a test in the core: Test::More, t/test.pl and ad hoc
print $test ? "ok 42\n" : "not ok 42\n". The decision of which to use depends on what part of the test suite you're working on. This is a measure to prevent a high-level failure (such as Config.pm breaking) from causing basic functionality tests to fail.
The t/test.pl library provides some of the features of Test::More, but avoids loading most modules and uses as few core features as possible.
If you write your own test, use the Test Anything Protocol.
Since we don't know if require works, or even subroutines, use ad hoc tests for these three. Step carefully to avoid using the feature being tested. Tests in t/opbasic, for instance, have been placed there rather than in t/op because they test functionality which t/test.pl presumes has already been demonstrated to work.
Now that basic require() and subroutines are tested, you can use the t/test.pl library.
You can also use certain libraries like Config conditionally, but be sure to skip the test gracefully if it's not there.
Now that the core of Perl is tested, Test::More can and should be used. You can also use the full suite of core modules in the tests.
When you say "make test", Perl uses the t/TEST program to run the test suite (except under Win32 where it uses t/harness instead). All tests are run from the t/ directory, not the directory which contains the test. This causes some problems with the tests in lib/, so here's some opportunity for some patching.
You must be triply conscious of cross-platform concerns. This usually boils down to using File::Spec and avoiding things like
system() unless absolutely necessary.
There are various special make targets that can be used to test Perl slightly differently than the standard "test" target. Not all them are expected to give a 100% success rate. Many of them have several aliases, and many of them are not available on certain operating systems.
This runs some basic sanity tests on the source tree and helps catch basic errors before you submit a patch.
Run miniperl on t/base, t/comp, t/cmd, t/run, t/io, t/op, t/uni and t/mro tests.
(Only in Linux) Run all the tests using the memory leak + naughty memory access tool "valgrind". The log files will be named testname.valgrind.
Run the test suite with the t/harness controlling program, instead of t/TEST. t/harness is more sophisticated, and uses the Test::Harness module, thus using this test target supposes that perl mostly works. The main advantage for our purposes is that it prints a detailed summary of failed tests at the end. Also, unlike t/TEST, it doesn't redirect stderr to stdout.
Note that under Win32 t/harness is always used instead of t/TEST, so there is no special "test_harness" target.
Under Win32's "test" target you may use the TEST_SWITCHES and TEST_FILES environment variables to control the behaviour of t/harness. This means you can say
nmake test TEST_FILES="op/*.t" nmake test TEST_SWITCHES="-torture" TEST_FILES="op/*.t"
Sets PERL_SKIP_TTY_TEST to true before running normal test.
The core distribution can now run its regression tests in parallel on Unix-like platforms. Instead of running
make test, set
TEST_JOBS in your environment to the number of tests to run in parallel, and run
make test_harness. On a Bourne-like shell, this can be done as
TEST_JOBS=3 make test_harness # Run 3 tests in parallel
An environment variable is used, rather than parallel make itself, because TAP::Harness needs to be able to schedule individual non-conflicting test scripts itself, and there is no standard interface to
make utilities to interact with their job schedulers.
Note that currently some test scripts may fail when run in parallel (most notably ext/IO/t/io_dir.t). If necessary, run just the failing scripts again sequentially and see if the failures go away.
You can run part of the test suite by hand by using one of the following commands from the t/ directory:
./perl -I../lib TEST list-of-.t-files
./perl -I../lib harness list-of-.t-files
(If you don't specify test scripts, the whole test suite will be run.)
If you use
harness for testing, you have several command line options available to you. The arguments are as follows, and are in the order that they must appear if used together.
harness -v -torture -re=pattern LIST OF FILES TO TEST harness -v -torture -re LIST OF PATTERNS TO MATCH
LIST OF FILES TO TEST is omitted, the file list is obtained from the manifest. The file list may include shell wildcards which will be expanded out.
Run the tests under verbose mode so you can see what tests were run, and debug output.
Run the torture tests as well as the normal set.
Filter the file list so that all the test files run match PATTERN. Note that this form is distinct from the -re LIST OF PATTERNS form below in that it allows the file list to be provided as well.
Filter the file list so that all the test files run match /(LIST|OF|PATTERNS)/. Note that with this form the patterns are joined by '|' and you cannot supply a list of files, instead the test files are obtained from the MANIFEST.
You can run an individual test by a command similar to
./perl -I../lib path/to/foo.t
except that the harnesses set up some environment variables that may affect the execution of the test:
indicates that we're running this test as part of the perl core test suite. This is useful for modules that have a dual life on CPAN.
is set to 2 if it isn't set already (see "PERL_DESTRUCT_LEVEL" in perlhacktips).
(used only by t/TEST) if set, overrides the path to the perl executable that should be used to run the tests (the default being ./perl).
if set, tells to skip the tests that need a terminal. It's actually set automatically by the Makefile, but can also be forced artificially by running 'make test_notty'.
Setting this variable runs all the Net::Ping modules tests, otherwise some tests that interact with the outside world are skipped. See perl58delta.
Setting this variable skips the vrexx.t tests for OS2::REXX.
This sets a variable in op/numconvert.t.
Setting this variable includes the tests in t/bigmem/. This should be set to the number of gigabytes of memory available for testing, eg.
PERL_TEST_MEMORY=4 indicates that tests that require 4GiB of available memory can be run safely.
See also the documentation for the Test and Test::Harness modules, for more environment variables that affect testing.
To hack on the Perl guts, you'll need to read the following things:
An overview of the Perl source tree. This will help you find the files you're looking for.
An overview of the Perl interpreter source code and some details on how Perl does what it does.
This document walks through the creation of a small patch to Perl's C code. If you're just getting started with Perl core hacking, this will help you understand how it works.
More details on hacking the Perl core. This document focuses on lower level details such as how to write tests, compilation issues, portability, debugging, etc.
If you plan on doing serious C hacking, make sure to read this.
This is of paramount importance, since it's the documentation of what goes where in the Perl source. Read it over a couple of times and it might start to make sense - don't worry if it doesn't yet, because the best way to study it is to read it in conjunction with poking at Perl source, and we'll do that later on.
Gisle Aas's "illustrated perlguts", also known as illguts, has very helpful pictures:
A working knowledge of XSUB programming is incredibly useful for core hacking; XSUBs use techniques drawn from the PP code, the portion of the guts that actually executes a Perl program. It's a lot gentler to learn those techniques from simple examples and explanation than from the core itself.
The documentation for the Perl API explains what some of the internal functions do, as well as the many macros used in the source.
This is a collection of words of wisdom for a Perl porter; some of it is only useful to the pumpkin holder, but most of it applies to anyone wanting to go about Perl development.
The CPAN testers ( http://testers.cpan.org/ ) are a group of volunteers who test CPAN modules on a variety of platforms.
Perl Smokers ( http://www.nntp.perl.org/group/perl.daily-build/ and http://www.nntp.perl.org/group/perl.daily-build.reports/ ) automatically test Perl source releases on platforms with various configurations.
Both efforts welcome volunteers. In order to get involved in smoke testing of the perl itself visit http://search.cpan.org/dist/Test-Smoke/. In order to start smoke testing CPAN modules visit http://search.cpan.org/dist/CPANPLUS-YACSmoke/ or http://search.cpan.org/dist/minismokebox/ or http://search.cpan.org/dist/CPAN-Reporter/.
If you've read all the documentation in the document and the ones listed above, you're more than ready to hack on Perl.
Here's some more recommendations
If you can do these things, you've started on the long road to Perl porting. Thanks for wanting to help make Perl better - and happy hacking!
If you recognized the quote about the Road above, you're in luck.
Most software projects begin each file with a literal description of each file's purpose. Perl instead begins each with a literary allusion to that file's purpose.
Like chapters in many books, all top-level Perl source files (along with a few others here and there) begin with an epigrammatic inscription that alludes, indirectly and metaphorically, to the material you're about to read.
Quotations are taken from writings of J.R.R. Tolkien pertaining to his Legendarium, almost always from The Lord of the Rings. Chapters and page numbers are given using the following editions:
Other JRRT books fair game for quotes would thus include The Adventures of Tom Bombadil, The Silmarillion, Unfinished Tales, and The Tale of the Children of Hurin, all but the first posthumously assembled by CJRT. But The Lord of the Rings itself is perfectly fine and probably best to quote from, provided you can find a suitable quote there.
So if you were to supply a new, complete, top-level source file to add to Perl, you should conform to this peculiar practice by yourself selecting an appropriate quotation from Tolkien, retaining the original spelling and punctuation and using the same format the rest of the quotes are in. Indirect and oblique is just fine; remember, it's a metaphor, so being meta is, after all, what it's for.
This document was originally written by Nathan Torkington, and is maintained by the perl5-porters mailing list.