PERLGIT(1perl)
π perlgit β the Perl Programmers Reference Guide, but make it a group chat π
okay bestie, sit down, hydrate, because this is perlgit in the Perl Programmers Reference Guide and it's literally the whole lore on doing Perl dev with git. π
β¨ NAME β¨
perlgit β Detailed information about git and the Perl repository π¦
πΈ DESCRIPTION πΈ
So here's the deal: this doc gives you the deets on using git to develop Perl. If you're just here for a quick lil patch moment, go read perlhack first, ok? Like, this one is for the regular contributors β the people who are in the repo all the time, including the ones with write access to the git repository. π
π CLONING THE REPOSITORY π
All of Perl's source code lives centrally in a Git repository over at github.com. no cap, all of it, one place. π
You can make a read-only clone of the repository by running:
% git clone git@github.com:Perl/perl5.git perl
And if that one won't work for you because of, like, firewall reasons π, you can also clone via http:
% git clone https://github.com/Perl/perl5.git perl
either way you get the goods. π«Ά
π WORKING WITH THE REPOSITORY π
Once you've changed into the repository directory, you can inspect it. After a clone the repository will contain a single local branch, which will also be the current branch β the asterisk is how you know which one is the main character. β
% git branch
* blead
Using the -a switch to branch will also show the remote tracking branches in the repository:
% git branch -a
* blead
origin/HEAD
origin/blead
...
The branches that begin with origin correspond to the git remote that you cloned from (which is named origin). Each branch on the remote will be exactly tracked by these branches. Now listen: you should NEVER do work on these remote tracking branches. π Never ever. You only ever do work in a local branch. Local branches can be configured to automerge (on pull) from a designated remote tracking branch. This is the case with the default branch blead, which will be configured to merge from the remote tracking branch origin/blead. iykyk.
You can see recent commits:
% git log
And pull new changes from the repository, and update your local repository (must be clean first):
% git pull
Assuming we are on the branch blead immediately after a pull, this command would be more or less equivalent to:
% git fetch
% git merge origin/blead
In fact if you want to update your local repository without touching your working directory you do:
% git fetch
And if you want to update your remote-tracking branches for all defined remotes simultaneously you can do:
% git remote update
Neither of these last two commands will update your working directory, however both will update the remote-tracking branches in your repository. π
To make a local branch of a remote branch:
% git checkout -b maint-5.10 origin/maint-5.10
To switch back to blead:
% git checkout blead
π Finding out your status π
The most common git command you will use will probably be:
% git status
This command will produce as output a description of the current state of the repository, including modified files and unignored untracked files, and in addition it will show things like what files have been staged for the next commit, and usually some useful information about how to change things. For instance the following:
% git status
On branch blead
Your branch is ahead of 'origin/blead' by 1 commit.
Changes to be committed:
(use "git reset HEAD <file>..." to unstage)
modified: pod/perlgit.pod
Changes not staged for commit:
(use "git add <file>..." to update what will be committed)
(use "git checkout -- <file>..." to discard changes in working
directory)
modified: pod/perlgit.pod
Untracked files:
(use "git add <file>..." to include in what will be committed)
deliberate.untracked
This shows that there were changes to this document staged for commit, and that there were further changes in the working directory not yet staged. It also shows that there was an untracked file in the working directory, and as you can see shows how to change all of this. It also shows that there is one commit on the working branch blead which has not been pushed to the origin remote yet.
NOTE: This output is also what you see as a template if you do not provide a message to git commit. π
π§‘ Patch workflow π§‘
First, please read perlhack for details on hacking the Perl core. That document covers many details on how to create a good patch. π
If you already have a Perl repository, you should ensure that you're on the blead branch, and your repository is up to date:
% git checkout blead
% git pull
It's preferable to patch against the latest blead version, since this is where new development occurs for all changes other than critical bug fixes. Critical bug fix patches should be made against the relevant maint branches, or should be submitted with a note indicating all the branches where the fix should be applied. π
Now that we have everything up to date, we need to create a temporary new branch</italic> for these changes and switch into it:
% git checkout -b orange
which is the short form of:
% git branch orange
% git checkout orange
Creating a topic branch makes it easier for the maintainers to rebase or merge back into the master blead for a more linear history. If you don't work on a topic branch the maintainer has to manually cherry pick your changes onto blead before they can be applied. π
That'll get you scolded on perl5-porters, so don't do that. Be Awesome. π
Then make your changes. For example, if Leon Brocard changes his name to Orange Brocard, we should change his name in the AUTHORS file:
% perl -pi -e 's{Leon Brocard}{Orange Brocard}' AUTHORS
You can see what files are changed:
% git status
On branch orange
Changes to be committed:
(use "git reset HEAD <file>..." to unstage)
modified: AUTHORS
And you can see the changes:
% git diff
diff --git a/AUTHORS b/AUTHORS
index 293dd70..722c93e 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -541,7 +541,7 @@ Lars Hecking <lhecking@nmrc.ucc.ie>
Laszlo Molnar <laszlo.molnar@eth.ericsson.se>
Leif Huhn <leif@hale.dkstat.com>
Len Johnson <lenjay@ibm.net>
-Leon Brocard <acme@astray.com>
+Orange Brocard <acme@astray.com>
Les Peters <lpeters@aol.net>
Lesley Binks <lesley.binks@gmail.com>
Lincoln D. Stein <lstein@cshl.org>
Now commit your change locally:
% git commit -a -m 'Rename Leon Brocard to Orange Brocard'
Created commit 6196c1d: Rename Leon Brocard to Orange Brocard
1 files changed, 1 insertions(+), 1 deletions(-)
The -a option is used to include all files that git tracks that you have changed. If at this time, you only want to commit some of the files you have worked on, you can omit the -a and use the command git add FILE ... before doing the commit. git add --interactive allows you to even just commit portions of files instead of all the changes in them. π€―
The -m option is used to specify the commit message. If you omit it, git will open a text editor for you to compose the message interactively. This is useful when the changes are more complex than the sample given here, and, depending on the editor, to know that the first line of the commit message doesn't exceed the 50 character legal maximum. See "Commit message" in perlhack for more information about what makes a good commit message. βοΈ
Once you've finished writing your commit message and exited your editor, git will write your change to disk and tell you something like this:
Created commit daf8e63: explain git status and stuff about remotes
1 files changed, 83 insertions(+), 3 deletions(-)
If you re-run git status, you should see something like this:
% git status
On branch orange
Untracked files:
(use "git add <file>..." to include in what will be committed)
deliberate.untracked
nothing added to commit but untracked files present (use "git add" to
track)
When in doubt, before you do anything else, check your status and read it carefully β many questions are answered directly by the git status output. π§ β¨
You can examine your last commit with:
% git show HEAD
and if you are not happy with either the description or the patch itself you can fix it up by editing the files once more and then issue:
% git commit -a --amend
Now, create a fork on GitHub to push your branch to, and add it as a remote if you haven't already, as described in the GitHub documentation at <https://help.github.com/en/articles/working-with-forks>:
% git remote add fork git@github.com:MyUser/perl5.git
And push the branch to your fork:
% git push -u fork orange
You should now submit a Pull Request (PR) on GitHub from the new branch to blead. For more information, see the GitHub documentation at <https://help.github.com/en/articles/creating-a-pull-request-from-a-fork>. π«Ά
You can also send patch files to perl5-porters@perl.org <mailto:perl5-porters@perl.org> directly if the patch is not ready to be applied, but intended for discussion. π¬
To create a patch file for all your local changes:
% git format-patch -M blead..
0001-Rename-Leon-Brocard-to-Orange-Brocard.patch
Or for a lot of changes, e.g. from a topic branch:
% git format-patch --stdout -M blead.. > topic-branch-changes.patch
If you want to delete your temporary branch, you may do so with:
% git checkout blead
% git branch -d orange
error: The branch 'orange' is not an ancestor of your current HEAD.
If you are sure you want to delete it, run 'git branch -D orange'.
% git branch -D orange
Deleted branch orange.
π¦ A note on derived files π¦
Be aware that many files in the distribution are derivative β avoid patching them, because git won't see the changes to them, and the build process will overwrite them. π Patch the originals instead. Most utilities (like perldoc) are in this category, i.e. patch utils/perldoc.PL rather than utils/perldoc. Similarly, don't create patches for files under $src_root/ext from their copies found in $install_root/lib. If you are unsure about the proper location of a file that may have gotten copied while building the source distribution, consult the MANIFEST. π
π§Ό Cleaning a working directory π§Ό
The command git clean can with varying arguments be used as a replacement for make clean.
To reset your working directory to a pristine condition you can do:
% git clean -dxf
However, be aware this will delete ALL untracked content. π You can use:
% git clean -Xf
the way that this next bit is SO satisfying, i cannot π
To nuke all your ignored untracked files β like build and test byproducts, the little guys that multiply in the night β but leave anything you hand-made alone, that's the move above. Your actual files? Untouched. Queen behaviour.
Only wanna cancel some uncommitted edits? No stress bestie, hit up git checkout with a list of the files you want reverted, or git checkout -f if you're feeling feral and want them all gone.
And if it's one or several commits you wanna cancel? git reset is your girl. β¨
π¦ Bisecting π¦
okay so this is genuinely iconic. git ships with a built-in way to figure out which commit is to blame for sneaking a bug into your life. git bisect runs a binary search through history to hunt down the first failing commit. It's fast, it's powerful, it's flexible β but it needs some setup, and if you want it automated you need an auxiliary shell script. Nothing in life is free, babe.
Now, the core gives you a wrapper program, Porting/bisect.pl, which tries to make this as simple as humanly possible β like, bisecting becomes as easy as running a Perl one-liner. Say you wanna know when this became an error:
perl -e 'my $x := 2'
you literally just run:
.../Porting/bisect.pl -e 'my $x := 2;'
With Porting/bisect.pl and one single command (and no other files, no cap), it's easy to find out:
- Which commit caused this example code to break?
- Which commit caused this example code to start working?
- Which commit added the first file to match this regex?
- Which commit removed the last file to match this regex?
β¦and usually you don't even need to know which perl versions to use as your start and end revisions, because Porting/bisect.pl automatically searches to find the earliest stable version where your test case passes. It just knows. Main character energy. Run Porting/bisect.pl --help for the full docs, including how to set the Configure and build-time options.
If you need more flexibility than Porting/bisect.pl is handing out, you're gonna have to run git bisect yourself, bestie. The most useful flavour is git bisect run for automating the building and testing of perl revisions. For that you need a shell script for git to call so it can test a particular revision. There's an example script at Porting/bisect-example.sh β and please copy it outside the repository, because the bisect process resets the state to a clean checkout as it runs. It will not hesitate. It will not feel bad. Copy it elsewhere first. The instructions below assume you copied it as ~/run and edited it to do your thing.
First, you enter bisect mode with:
% git bisect start
For example, if the bug is present on HEAD but wasn't in 5.10.0, git learns that when you enter:
% git bisect bad
% git bisect good perl-5.10.0
Bisecting: 853 revisions left to test after this
That checks out the median commit between HEAD and perl-5.10.0. Then you run the whole bisecting process with:
% git bisect run ~/run
When the first bad commit finally gets isolated, git bisect tells you like the good friend it is:
ca4cfd28534303b82a216cfe83a1c80cbc3b9dc5 is first bad commit
commit ca4cfd28534303b82a216cfe83a1c80cbc3b9dc5
Author: Dave Mitchell <davem@fdisolutions.com>
Date: Sat Feb 9 14:56:23 2008 +0000
[perl #49472] Attributes + Unknown Error
...
bisect run success
You can snoop on the bisecting process with git bisect log and git bisect visualize. And git bisect reset gets you out of bisect mode. Free at last.
Please note: the first "good" state must be an ancestor of the first "bad" state. If you're searching for the commit that solved some bug, you have to negate your test case (i.e. exit with 1 if OK and 0 if not) and still mark the lower bound as "good" and the upper as "bad". The "first bad commit" then has to be understood as the "first commit where the bug is solved". π₯
git help bisect has so much more on how you can tweak your binary searches.
After bisection you might wanna configure, build and test perl at the commits the bisection process pointed at. Sometimes β especially with older perls β make may fail during this process. In that case you might be able to patch the source code at the older commit point. To do that, please follow the suggestions in "Building perl at older commits" in perlhack. π«Ά
π Topic branches and rewriting history π
Individual committers should make topic branches under yourname/some_descriptive_name:
% branch="$yourname/$some_descriptive_name"
% git checkout -b $branch
... do local edits, commits etc ...
% git push origin -u $branch
If you're stuck with an ancient version of git (prior to 1.7), then git push won't have the -u switch, and you have to replace that last step with this sequence instead:
% git push origin $branch:refs/heads/$branch
% git config branch.$branch.remote origin
% git config branch.$branch.merge refs/heads/$branch
If you wanna change someone else's topic branch, check with its creator first. Like, communicate. It's not that hard bestie. π
Sometimes you'll find that the original author edited the branch's history. There are lots of good reasons for this. Sometimes an author is just rebasing the branch onto a newer source point. Sometimes an author found an error in an early commit that they wanted to fix before merging the branch into blead.
Right now the master repository is configured to forbid non-fast-forward merges. Which means the branches inside it can't be rebased and pushed as a single step. Nope. Blocked. Not happening.
The only way you will ever be allowed to rebase or modify the history of a pushed branch is to delete it and push it back as a new branch under the same name. Please think carefully before doing this. It may be better to sequentially rename your branches so it's easier for other people working with you to cherry-pick their local changes onto the new version. (XXX: needs explanation).
If you wanna rebase a personal topic branch, you'll have to delete your existing topic branch and push a new version of it. After you've rebased your branch, you can do it with this formula (peep the explanation about refspec's in the git push documentation for the deets):
# first rebase
% git checkout $user/$topic
% git fetch
% git rebase origin/blead
# then "delete-and-push"
% git push origin :$user/$topic
% git push origin $user/$topic
NOTE: it is forbidden at the repository level to delete any of the "primary" branches. That's any branch matching m!^(blead|maint|perl)!. Any attempt to do so makes git produce an error like this:
% git push origin :blead
*** It is forbidden to delete blead/maint branches in this repository
error: hooks/update exited with error code 1
error: hook declined to update refs/heads/blead
To ssh://perl5.git.perl.org/perl
! [remote rejected] blead (hook declined)
error: failed to push some refs to 'ssh://perl5.git.perl.org/perl'
As a matter of policy we do not edit the history of the blead and maint-* branches. If a typo (or something worse π) sneaks into a commit on blead or maint-*, we fix it in another commit, simple as that. The only type of updates allowed on these branches are "fast-forwards", where all history is preserved. History stays intact, periodt.
Annotated tags in the canonical perl.git repository will never be deleted or modified. Think long and hard about whether you wanna push a local tag to perl.git before you do it. (Pushing simple tags is not allowed.)
πΈ Grafts πΈ
The perl history contains one mistake that wasn't caught during conversion: a merge was recorded between blead and maint-5.10 where no merge actually happened. Due to the nature of git, this is now impossible to fix in the public repository. We're stuck with it, bestie, it's baked in.
You can remove this mis-merge locally though β just add the following line to your .git/info/grafts file:
296f12bbbbaa06de9be9d09d3dcf8f4528898a49 434946e0cb7a32589ed92d18008aaa1d88515930
It's particularly important to have this graft line if any bisecting gets done in the area of the merge in question. Don't skip it. π
π WRITE ACCESS TO THE GIT REPOSITORY π
So you got write access. Congrats, queen, you ate. π
Now you need to modify the URL for the origin remote to enable pushing. Edit .git/config with the git-config(1) command:
% git config remote.origin.url git@github.com:Perl/perl5.git
You can also set up your user name and e-mail address. Most people do this once, globally, in their ~/.gitconfig, by doing something like:
% git config --global user.name "Γvar ArnfjΓΆrΓ° Bjarmason"
% git config --global user.email avarab@gmail.com
But if you wanna override that just for perl, run something like this inside perl:
% git config user.email avar@cpan.org
It's also possible to keep origin as a git remote and add a new remote for ssh access:
% git remote add camel git@github.com:Perl/perl5.git
This lets you update your local repository by pulling from origin β which is faster and doesn't require you to authenticate β and push your changes back through the camel remote:
% git fetch camel
% git push camel
The fetch command only updates the camel refs, since the objects themselves should've been fetched already when you pulled from origin. Efficiency. We love her.
π¦ Working with GitHub pull requests π¦
Pull requests typically come from outside the Perl/perl.git repository, so if you wanna test or work with one locally, a vanilla git fetch from the Perl/perl5.git repository won't fetch it. It simply will not. It has no idea it exists.
But GitHub does give you a mechanism to fetch a pull request into a local branch. They live on GitHub remotes under pull/, so you can use git fetch pull/PRID/head:localname to make a local copy. e.g. to fetch pull request 9999 to the local branch local-branch-name, run:
git fetch origin pull/9999/head:local-branch-name
and then:
git checkout local-branch-name
Note: this branch is not rebased on blead, so instead of that checkout above, you might want:
git rebase origin/blead local-branch-name
which rebases local-branch-name on blead, and checks it out. β¨
Alternatively, you can configure the remote to fetch all pull requests as remote-tracking branches. To do that, edit the remote in .git/config β for example, if your GitHub remote is origin, you'd have:
[remote "origin"]
url = git@github.com:/Perl/perl5.git
fetch = +refs/heads/*:refs/remotes/origin/*
Add a line to map the remote pull request branches to remote-tracking branches:
[remote "origin"]
url = git@github.com:/Perl/perl5.git
fetch = +refs/heads/*:refs/remotes/origin/*
fetch = +refs/pull/*/head:refs/remotes/origin/pull/*
and then do a fetch like normal:
git fetch origin
This creates a remote-tracking branch for every pull request, including closed ones. All of them. Every single one. We're collecting them like stickers.
To remove those remote-tracking branches, delete the line you added above and prune:
git fetch -p origin # or git remote prune origin
π©Ή Accepting a patch
So someone slid a patch file into your mentions, made with the section above? Bestie, you gotta try it out first. Never trust, always test. π
Step one: make yourself a lil temporary branch and hop into it:
% git checkout -b experimental
Patch files that got formatted by git format-patch? Those get applied with git am:
% git am 0001-Rename-Leon-Brocard-to-Orange-Brocard.patch
Applying Rename Leon Brocard to Orange Brocard
Heads up tho: some UNIX mail systems get messy with text attachments that contain From . Not the vibe. This fixes them right up:
% perl -pi -e's/^>From /From /' \
0001-Rename-Leon-Brocard-to-Orange-Brocard.patch
And if all you got handed was a raw diff, no stress β you can do this whole two-step situation instead:
% git apply bugfix.diff
% git commit -a -m "Some fixing" \
--author="That Guy <that.guy@internets.com>"
Now let's peek at the change like a nosy queen π:
% git show HEAD
commit b1b3dab48344cff6de4087efca3dbd63548ab5e2
Author: Leon Brocard <acme@astray.com>
Date: Fri Dec 19 17:02:59 2008 +0000
Rename Leon Brocard to Orange Brocard
diff --git a/AUTHORS b/AUTHORS
index 293dd70..722c93e 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -541,7 +541,7 @@ Lars Hecking <lhecking@nmrc.ucc.ie>
Laszlo Molnar <laszlo.molnar@eth.ericsson.se>
Leif Huhn <leif@hale.dkstat.com>
Len Johnson <lenjay@ibm.net>
-Leon Brocard <acme@astray.com>
+Orange Brocard <acme@astray.com>
Les Peters <lpeters@aol.net>
Lesley Binks <lesley.binks@gmail.com>
Lincoln D. Stein <lstein@cshl.org>
If you're a committer to Perl and the patch is giving absolute main character energy, you merge it into blead and push it out to the main repository:
% git checkout blead
% git merge experimental
% git push origin blead
Wanna yeet that temporary branch into the void? Totally valid:
% git checkout blead
% git branch -d experimental
error: The branch 'experimental' is not an ancestor of your current
HEAD. If you are sure you want to delete it, run 'git branch -D
experimental'.
% git branch -D experimental
Deleted branch experimental.
π Committing to blead
The 'blead' branch is literally gonna become the next production release of Perl. That's not a small thing.
Before you push any local change to blead, it's incredibly important that you do a few things β otherwise other committers will come for you with pitchforks and torches. π₯ Not a metaphor. Okay maybe a little a metaphor.
- Make sure you have a good commit message. See
"Commit message"in perlhack for the deets. - Run the test suite. You might be sitting there thinking "bestie it's one typo fix, how could it possibly break a test file" β and you'd be wrong. Here's a real example of not running the suite causing problems: a patch was submitted that added a couple of tests to an existing
.t. It couldn't possibly affect anything else, so no need to test beyond the single affected.t, right? But the submitter's email address had changed since the last of their submissions, and this caused other tests to fail. Running the test target given in the next item would have caught this problem. π - If you don't run the full test suite, at least run
make test_porting. It runs basic sanity checks. Wanna see which sanity checks? Go look int/porting. - If you make any changes that affect
miniperlor core routines that have different code paths forminiperl, be sure to runmake minitest. This catches problems that even the full test suite will not catch, because it runs a subset of tests underminiperlrather thanperl. Iconic and thorough, we love to see it. β¨
π On merging and rebasing
Simple, one-off commits pushed to the 'blead' branch should be simple commits that apply cleanly. In other words, make sure your work is committed against the current position of blead, so you can push back to the master repository without merging. Clean, linear, no drama. πΈ
Sometimes blead moves while you're building or testing your changes. Rude, but it happens. When it does, your push gets rejected with a message like this:
To ssh://perl5.git.perl.org/perl.git
! [rejected] blead -> blead (non-fast-forward)
error: failed to push some refs to 'ssh://perl5.git.perl.org/perl.git'
To prevent you from losing history, non-fast-forward updates were
rejected Merge the remote changes (e.g. 'git pull') before pushing
again. See the 'Note about fast-forwards' section of 'git push --help'
for details.
When this happens you can just rebase your work against the new position of blead, like this (assuming your remote for the master repository is "p5p"):
% git fetch p5p
% git rebase p5p/blead
You'll watch your commits get re-applied, and then you can push safely. More info about rebasing lives in the documentation for the git-rebase(1) command. π¦
Now for larger sets of commits that only make sense together, or that would benefit from a summary of the set's purpose β those should get a merge commit. Do your work on a topic branch, and regularly rebase that branch against blead so your code doesn't get broken by blead moving. When you're done, please do a final rebase and test. Linear history is something that gets lost with every commit on blead, but a final rebase makes the history linear again, which makes it easier for future maintainers to see what happened. Rebase like so (assuming your work was on the branch "committer/somework"):
% git checkout committer/somework
% git rebase blead
Then you can merge it into master like this:
% git checkout blead
% git merge --no-ff --no-commit committer/somework
% git commit -a
Those switches deserve an explanation, so here's the tea β:
--no-ff indicates that even if all your work can be applied linearly against blead, a merge commit should still be prepared. This ensures all your work will be shown as a side branch, with all its commits merged into the mainstream blead by the merge commit.
--no-commit means the merge commit will be prepared but not committed. The commit actually happens when you run the next command, which brings up your editor so you can describe the commit. Without --no-commit, the commit would be made with nearly no useful message, which would greatly diminish the value of the merge commit as a placeholder for the work's description.
When you're describing the merge commit, explain the purpose of the branch, and keep in mind this description will probably be used by the eventual release engineer when reviewing the next perldelta document. No pressure, but also. Some pressure. π
If the subsequent push fails, you must be careful about how you rebase. If you use
% git rebase p5p/blead
or
% git pull --rebase
then your carefully created merge commit will be lost. The audacity. π To avoid that, use:
% git fetch p5p
% git rebase --rebase-merges p5p/blead
This will recreate your merge commit. Saved. π«Ά
(Should you be stuck with an older version of git (prior to 2.18), then"git rebase"will not have the"--rebase-merges"switch; instead you have to use the"--preserve-merges"switch.)
π§° Committing to maintenance versions
Maintenance versions should only be altered to add critical bug fixes β see perlpolicy. Critical only. Don't be out here remodelling the whole house.
To commit to a maintenance version of perl, you need to create a local tracking branch:
% git checkout --track -b maint-5.005 origin/maint-5.005
This creates a local branch named "maint-5.005", which tracks the remote branch "origin/maint-5.005". Then you can pull, commit, merge and push as before. Business as usual, bestie. π
You can also cherry-pick commits from blead and another branch using the "git cherry-pick" command. It is recommended to use the -x option to "git cherry-pick" in order to record the SHA1 of the original commit in the new commit message. Receipts matter, periodt.
Before pushing any change to a maint version, make sure you've satisfied the steps in "Committing to blead" above. Same rules, still apply, no skipping. π
π¨ Using a smoke-me branch to test changes
Sometimes a change affects code paths you cannot test on the OSes directly available to you, and it'd be wise to have users on other OSes test the change before you commit it to blead. Can't test what you don't have access to, and that's not a skill issue, that's just hardware.
Fortunately there's a way to get your change smoke-tested on various OSes: push it to a "smoke-me" branch and wait for certain automated smoke-testers to report the results from their OSes. A "smoke-me" branch is identified by the branch name: specifically, as seen on github.com, it must be a local branch whose first name component is precisely "smoke-me". Precisely. Character for character. π
The procedure goes roughly like this (using the example of tonyc's smoke-me branch called win32stat):
First, make a local branch and switch to it:
% git checkout -b win32stat
Make some changes, build perl and test your changes, then commit them to your local branch. Then push your local branch to a remote smoke-me branch:
% git push origin win32stat:smoke-me/tonyc/win32stat
Now you can switch back to blead locally:
% git checkout blead
and keep working on other things while you wait a day or two, keeping an eye on the results reported for your "smoke-me" branch at <https://perl.develop-help.com/?b=smoke-me/tonyc/win32state>. π
If all is well then update your blead branch:
% git pull
then check out your smoke-me branch once more and rebase it on blead:
% git rebase blead win32stat
Now switch back to blead and merge your smoke-me branch into it:
% git checkout blead
% git merge win32stat
As described earlier, if there are many changes on your smoke-me branch then you should prepare a merge commit in which to give an overview of those changes, by using the following command instead of the last command above:
% git merge win32stat --no-ff --no-commit
You should now build perl and test your (merged) changes one last time β ideally run the whole test suite, but failing that at least run the t/porting/*.t tests β before pushing your changes as usual:
% git push origin blead
Finally, you should then delete the remote smoke-me branch:
% git push origin :smoke-me/tonyc/win32stat
(which is likely to produce a warning like this, which can be ignored:
remote: fatal: ambiguous argument
'refs/heads/smoke-me/tonyc/win32stat':
unknown revision or path not in the working tree.
remote: Use '--' to separate paths from revisions
) and then delete your local branch:
% git branch -d win32stat
And that's the whole thing, bestie β you are now certified perl git-literate, no cap. π¦β¨ This page was shipped in perl v5.44.0, dated 2026-10-17, as PERLGIT(1perl). Go forth and push responsibly. ππ
π¦ where this page comes from
- arch rolling: perl 5.44.0-1 Β· Artistic-1.0-Perl AND GPL-1.0-or-later
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms