Skip to content

Git

Git is a software for tracking changes in any set of files, usually used for coordinating work among programmers collaboratively developing source code during software development. Its goals include speed, data integrity, and support for distributed, non-linear workflows (thousands of parallel branches running on different systems).

Learning git

Git is a tough nut to crack, no matter how experience you are you'll frequently get surprised. Sadly it's one of the main tools to develop your code, so you must master it as soon as possible.

Depending on how you like to learn I've found these options:

Pull Request Process

This part of the doc is shamefully edited from the source. It was for the k8s project but they are good practices that work for all the projects. It explains the process and best practices for submitting a PR It should serve as a reference for all contributors, and be useful especially to new and infrequent submitters.

Before You Submit a PR

This guide is for contributors who already have a PR to submit. If you're looking for information on setting up your developer environment and creating code to contribute to the project, search the development guide.

Make sure your PR adheres to the projects best practices. These include following project conventions, making small PRs, and commenting thoroughly.

Run Local Verifications

You can run the tests in local before you submit your PR to predict the pass or fail of continuous integration.

Why is my PR not getting reviewed?

A few factors affect how long your PR might wait for review.

If it's the last few weeks of a milestone, we need to reduce churn and stabilize.

Or, it could be related to best practices. One common issue is that the PR is too big to review. Let's say you've touched 39 files and have 8657 insertions. When your would-be reviewers pull up the diffs, they run away - this PR is going to take 4 hours to review and they don't have 4 hours right now. They'll get to it later, just as soon as they have more free time (ha!).

There is a detailed rundown of best practices, including how to avoid too-lengthy PRs, in the next section.

But, if you've already followed the best practices and you still aren't getting any PR love, here are some things you can do to move the process along:

  • Make sure that your PR has an assigned reviewer (assignee in GitHub). If not, reply to the PR comment stream asking for a reviewer to be assigned.

  • Ping the assignee (@username) on the PR comment stream, and ask for an estimate of when they can get to the review.

  • Ping the assignee by email (many of us have publicly available email addresses).

  • If you're a member of the organization ping the team (via @team-name) that works in the area you're submitting code.

  • If you have fixed all the issues from a review, and you haven't heard back, you should ping the assignee on the comment stream with a "please take another look" (PTAL) or similar comment indicating that you are ready for another review.

Read on to learn more about how to get faster reviews by following best practices.

Best Practices for Faster Reviews

You've just had a brilliant idea on how to make a project better. Let's call that idea Feature-X. Feature-X is not even that complicated. You have a pretty good idea of how to implement it. You jump in and implement it, fixing a bunch of stuff along the way. You send your PR - this is awesome! And it sits. And sits. A week goes by and nobody reviews it. Finally, someone offers a few comments, which you fix up and wait for more review. And you wait. Another week or two go by. This is horrible.

Let's talk about best practices so your PR gets reviewed quickly.

Familiarize yourself with project conventions

  • Search for the Development guide
  • Search for the Coding conventions
  • Search for the API conventions

Is the feature wanted? Make a Design Doc or Sketch PR

Are you sure Feature-X is something the project team wants or will accept? Is it implemented to fit with other changes in flight? Are you willing to bet a few days or weeks of work on it?

It's better to get confirmation beforehand. There are two ways to do this:

  • Make a proposal doc (in docs/proposals; for example the QoS proposal), or reach out to the affected special interest group (SIG). Some projects have that
  • Coordinate your effort with SIG Docs ahead of time
  • Make a sketch PR (e.g., just the API or Go interface). Write or code up just enough to express the idea and the design and why you made those choices

Or, do all of the above.

Be clear about what type of feedback you are asking for when you submit a proposal doc or sketch PR.

Now, if we ask you to change the design, you won't have to re-write it all.

Smaller Is Better: Small Commits, Small PRs

Small commits and small PRs get reviewed faster and are more likely to be correct than big ones.

Attention is a scarce resource. If your PR takes 60 minutes to review, the reviewer's eye for detail is not as keen in the last 30 minutes as it was in the first. It might not get reviewed at all if it requires a large continuous block of time from the reviewer.

Breaking up commits

Break up your PR into multiple commits, at logical break points.

Making a series of discrete commits is a powerful way to express the evolution of an idea or the different ideas that make up a single feature. Strive to group logically distinct ideas into separate commits.

For example, if you found that Feature-X needed some prefactoring to fit in, make a commit that JUST does that prefactoring. Then make a new commit for Feature-X.

Strike a balance with the number of commits. A PR with 25 commits is still very cumbersome to review, so use judgment.

Breaking up PRs

Or, going back to our prefactoring example, you could also fork a new branch, do the prefactoring there and send a PR for that. If you can extract whole ideas from your PR and send those as PRs of their own, you can avoid the painful problem of continually rebasing.

Multiple small PRs are often better than multiple commits. Don't worry about flooding us with PRs. We'd rather have 100 small, obvious PRs than 10 unreviewable monoliths.

We want every PR to be useful on its own, so use your best judgment on what should be a PR vs. a commit.

As a rule of thumb, if your PR is directly related to Feature-X and nothing else, it should probably be part of the Feature-X PR. If you can explain why you are doing seemingly no-op work ("it makes the Feature-X change easier, I promise") we'll probably be OK with it. If you can imagine someone finding value independently of Feature-X, try it as a PR. (Do not link pull requests by # in a commit description, because GitHub creates lots of spam. Instead, reference other PRs via the PR your commit is in.)

Open a Different PR for Fixes and Generic Features

Put changes that are unrelated to your feature into a different PR.

Often, as you are implementing Feature-X, you will find bad comments, poorly named functions, bad structure, weak type-safety, etc.

You absolutely should fix those things (or at least file issues, please) - but not in the same PR as your feature. Otherwise, your diff will have way too many changes, and your reviewer won't see the forest for the trees.

Look for opportunities to pull out generic features.

For example, if you find yourself touching a lot of modules, think about the dependencies you are introducing between packages. Can some of what you're doing be made more generic and moved up and out of the Feature-X package? Do you need to use a function or type from an otherwise unrelated package? If so, promote! We have places for hosting more generic code.

Likewise, if Feature-X is similar in form to Feature-W which was checked in last month, and you're duplicating some tricky stuff from Feature-W, consider prefactoring the core logic out and using it in both Feature-W and Feature-X. (Do that in its own commit or PR, please.)

Comments Matter

In your code, if someone might not understand why you did something (or you won't remember why later), comment it. Many code-review comments are about this exact issue.

If you think there's something pretty obvious that we could follow up on, add a TODO.

Test

Nothing is more frustrating than starting a review, only to find that the tests are inadequate or absent. Very few PRs can touch code and NOT touch tests.

If you don't know how to test Feature-X, please ask! We'll be happy to help you design things for easy testing or to suggest appropriate test cases.

Squashing and Commit Titles

Your reviewer has finally sent you feedback on Feature-X.

Make the fixups, and don't squash yet. Put them in a new commit, and re-push. That way your reviewer can look at the new commit on its own, which is much faster than starting over.

We might still ask you to clean up your commits at the very end for the sake of a more readable history, but don't do this until asked: typically at the point where the PR would otherwise be tagged LGTM.

Each commit should have a good title line (<70 characters) and include an additional description paragraph describing in more detail the change intended.

General squashing guidelines:

  • Sausage => squash

Do squash when there are several commits to fix bugs in the original commit(s), address reviewer feedback, etc. Really we only want to see the end state and commit message for the whole PR.

  • Layers => don't squash

Don't squash when there are independent changes layered to achieve a single goal. For instance, writing a code munger could be one commit, applying it could be another, and adding a precommit check could be a third. One could argue they should be separate PRs, but there's really no way to test/review the munger without seeing it applied, and there needs to be a precommit check to ensure the munged output doesn't immediately get out of date.

A commit, as much as possible, should be a single logical change.

KISS, YAGNI, MVP, etc.

Sometimes we need to remind each other of core tenets of software design - Keep It Simple, You Aren't Gonna Need It, Minimum Viable Product, and so on. Adding a feature "because we might need it later" is antithetical to software that ships. Add the things you need NOW and (ideally) leave room for things you might need later - but don't implement them now.

It's OK to Push Back

Sometimes reviewers make mistakes. It's OK to push back on changes your reviewer requested. If you have a good reason for doing something a certain way, you are absolutely allowed to debate the merits of a requested change. Both the reviewer and reviewee should strive to discuss these issues in a polite and respectful manner.

You might be overruled, but you might also prevail. We're pretty reasonable people. Mostly.

Another phenomenon of open-source projects (where anyone can comment on any issue) is the dog-pile - your PR gets so many comments from so many people it becomes hard to follow. In this situation, you can ask the primary reviewer (assignee) whether they want you to fork a new PR to clear out all the comments. You don't HAVE to fix every issue raised by every person who feels like commenting, but you should answer reasonable comments with an explanation.

Common Sense and Courtesy

No document can take the place of common sense and good taste. Use your best judgment, while you put a bit of thought into how your work can be made easier to review. If you do these things your PRs will get merged with less friction.

Split long PR into smaller ones

  • Start a new branch from where you want to merge.

  • Start an interactive rebase on HEAD:

git rebase -i HEAD
  • Get the commits you want: Now comes the clever part, we are going to pick out all the commits we care about from 112-new-feature-branch using the following command:
git log --oneline --reverse HEAD..112-new-feature-branch -- app/models/ spec/models

Woah thats quite the line! Let’s dissect it first:

  • git log shows a log of what you have done in your project.
  • --online formats the output from a few lines (including author and time of commit), to just “[sha-hash-of-commit] [description-of-commit]”
  • --reverse reverses the log output chronologically (so oldest commit first, newest last).
  • 112-new-feature-branch..HEAD shows the difference in commits from your current branch (HEAD) and the branch you are interested in 112-new-feature-branch.
  • -- app/models/ spec/models Only show commits that changed files in app/models/ or spec/models So that we confine the changes to our model and its tests.

Now if you are using vim (or vi or neovim) you can put the results of this command directly into your rebase-todo (which was opened when starting the rebase) using the :r command like so:

:r !git log --oneline --reverse HEAD..112-new-feature-branch -- app/models/
  • Review the commits you want: Now you have a chance to go though your todo once again. First you should remove the noop from above, since you actually do something now. Second you should check the diffs of the sha-hashes.

Note: If you are using vim, you might already have the fugitive plug-in. If you haven’t changed the standard configuration, you can just move your cursor over the sha-hashes and press K (note that its capitalized) to see the diff of that commit.

If you don’t have fugitive or don’t use vim, you can check the diff using git show SHA-HASH (for example git show c4f74d0), which shows the commits data.

Now you can prepend and even rearrange the commits (Be careful rearranging or leaving out commits, you might have to fix conflicts later).

  • Execute the rebase: Now you can save and exit the editor and git will try to execute the rebase. If you have conflicts you can fix them just like you do with merges and then continue using git rebase --continue.

If you feel like something is going terribly wrong (for example you have a bunch of conflicts in just a few commits), you can abort the rebase using git rebase --abort and it will be like nothing ever happened.

Git workflow

There are many ways of using git, one of the most popular is git flow, please read this article to understand it before going on.

Unless you are part of a big team that delivers software that needs to maintain many versions, it's not worth using git flow as it's too complex and cumbersome. Instead I suggest a variation of the Github workflow.

To carry out a reliable continuous delivery we must work to comply with the following list of best practices:

  • Everything must be in the git server: source code, tests, pipelines, scripts, templates and documentation.
  • There is only a main branch (main) whose key is that everything is in this branch must be always stable and deployable into production at any time.
  • New branches are created from main in order to develop new features that should be merged into main branch in short development cycles.
  • It is highly recommended to do small commits to have more control over what is being done and to avoid discarding many lines of code if a rollback has to be done.
  • A commit message policy should be set so that they are clear and conform the same pattern, for example semantic versioning.
  • main is blocked to reject direct pushes as well as to protect it of catastrophic deletion. Only pre-validated merge requests are accepted.
  • When a feature is ready, we will open a merge request to merge changes into main branch.
  • Use webhooks to automate the execution of tests and validation tasks in the CI server before/after adding changes in main.
  • It is not needed to discard a merge request if any of the validation tasks failed. We check the code and when the changes are pushed, the CI server will relaunch the validation tasks.
  • If all validation tasks pass, we will assign the merge request to two team developers to review the feature code.
  • After both reviewers validate the code, the merge request can be accepted and the feature branch may be deleted.
  • A clear versioning policy must be adopted for all generated artifacts.
  • Each artifact must be generated once and be promoted to the different environments in different stages.

When a developer wants to add code to main should proceed as follows:

  • Wait until the pipeline execution ends if it exists. If that process fails, then the developer must help to other team members to fix the issue before requesting a new merge request.
  • Pull the changes from main and resolve the conflicts locally before pushing the code to the new feature branch.
  • Run a local script that compiles and executes the tests before committing changes. This task can be done executing it manually by the developer or using a git precommit.
  • Open a new merge request setting the feature branch as source branch and main as target branch.
  • The CI server is notified of the new merge request and executes the pipeline which compiles the source code, executes the tests, deploys the artifact, etc.
  • If there are errors in the previous step, the developer must fix the code and push it to the git server as soon as possible so that the CI server validate once again the merge request.
  • If no errors, the CI server will mark the merge request as OK and the developer can assign it to two other team members to review the feature code.
  • At this point, the developer can start with other task.

Considerations

The build process and the execution of the tests have to be pretty fast. It should not exceed about 10 minutes.

Unit tests must be guarantee that they are completely unitary; they must be executed without starting the context of the application, they must not access to the DDBB, external systems, file system, etc.

Naming conventions

The best idea is to use Semantic Versioning to define the names of the branches, for example: feat/add-command-line-support or fix/correct-security-issue, and also for the commit messages.

Tag versioning policy

We will also adopt semantic versioning policy on version management.

Versioning control

When a branch is merged into main, the CI server launches a job which generates a new artifact release as follow:

  • The new version number is calculated taken into account the above considerations.
  • Generates a new artifact named as appname-major.minor.patch.build
  • Upload the previous artifact to the artifact repository manager.
  • Create a git tag on the repository with the same version identifier, major.minor.patch.build
  • Automatically deploy the artifact on the desired environment (dev, pre, etc)

Hotfixing

Hotfix should be developed and fixed using one of the next cases, which has been defined by preference order:

Case 1

In this case, we have pushed new code to "main" branch since the last deploy on production and we want to deploy the new code with the fix code.

We have to follow the next steps:

  • Create a branch "Hotfix" from commit/tag of the last deploy
  • Fix the bug in "hotfix" branch
  • Merge the new branch to "main"
  • Deploy main branch

Case 2

In this case, we have pushed new code to "main" branch since the last deploy on production but we don't want to deploy the new code with the fix code.

We have to follow the next steps:

  • Create a branch "Hotfix" from commit/tag of the last deploy
  • Fix the bug in "hotfix" branch
  • Deploy main branch
  • Merge the new branch to "main.

Case 3

In this case, we have pushed new code to "main" branch since the last deploy on production but we don't want to deploy the new code with the fix code.

We have to follow the next steps:

  • Create a branch "Hotfix" from commit/tag of the last deploy
  • Fix the bug in "hotfix" branch
  • Deploy main branch
  • Merge the new branch to "main.

Git housekeeping

The best option is to:

git fetch --prune
git-sweep cleanup

To remove the local branches you can:

cd myrepo
git remote add local $(pwd)
git-sweep cleanup --origin=local

The tool is no longer maintained but there is still no good alternative. I've found some but are either not popular and/or not maintained:

Submodules

Shamefully edited from the docs

It often happens that while working on one project, you need to use another project from within it. Perhaps it’s a library that a third party developed or that you’re developing separately and using in multiple parent projects. A common issue arises in these scenarios: you want to be able to treat the two projects as separate yet still be able to use one from within the other.

Here’s an example. Suppose you’re developing a website and creating Atom feeds. Instead of writing your own Atom-generating code, you decide to use a library. You’re likely to have to either include this code from a shared library like a CPAN install or Ruby gem, or copy the source code into your own project tree. The issue with including the library is that it’s difficult to customize the library in any way and often more difficult to deploy it, because you need to make sure every client has that library available. The issue with copying the code into your own project is that any custom changes you make are difficult to merge when upstream changes become available.

Git addresses this issue using submodules. Submodules allow you to keep a Git repository as a subdirectory of another Git repository. This lets you clone another repository into your project and keep your commits separate.

It often happens that while working on one project, you need to use another project from within it. Perhaps it’s a library that a third party developed or that you’re developing separately and using in multiple parent projects. A common issue arises in these scenarios: you want to be able to treat the two projects as separate yet still be able to use one from within the other.

Here’s an example. Suppose you’re developing a website and creating Atom feeds. Instead of writing your own Atom-generating code, you decide to use a library. You’re likely to have to either include this code from a shared library like a CPAN install or Ruby gem, or copy the source code into your own project tree. The issue with including the library is that it’s difficult to customize the library in any way and often more difficult to deploy it, because you need to make sure every client has that library available. The issue with copying the code into your own project is that any custom changes you make are difficult to merge when upstream changes become available.

Git addresses this issue using submodules. Submodules allow you to keep a Git repository as a subdirectory of another Git repository. This lets you clone another repository into your project and keep your commits separate.

Submodule tips

Update all git submodules

If it's the first time you check-out a repo you need to use --init first:

git submodule update --init --recursive

To update to latest tips of remote branches use:

git submodule update --recursive --remote

Submodule Foreach

There is a foreach submodule command to run some arbitrary command in each submodule. This can be really helpful if you have a number of submodules in the same project.

For example, let’s say we want to start a new feature or do a bugfix and we have work going on in several submodules. We can easily stash all the work in all our submodules.

git submodule foreach 'git stash'

Then we can create a new branch and switch to it in all our submodules.

git submodule foreach 'git checkout -b featureA'

You get the idea. One really useful thing you can do is produce a nice unified diff of what is changed in your main project and all your subprojects as well.

git diff; git submodule foreach 'git diff'

Useful Aliases

You may want to set up some aliases for some of these commands as they can be quite long and you can’t set configuration options for most of them to make them defaults. We covered setting up Git aliases in Git Aliases, but here is an example of what you may want to set up if you plan on working with submodules in Git a lot.

git config alias.sdiff '!'"git diff && git submodule foreach 'git diff'"
git config alias.spush 'push --recurse-submodules=on-demand'
git config alias.supdate 'submodule update --remote --merge'

This way you can simply run git supdate when you want to update your submodules, or git spush to push with submodule dependency checking.

Encrypt sensitive information

Use git-crypt.

Use different git configs

Include in your ~/.gitconfig

[includeIf "gitdir:~/company_A/"]
  path = ~/.config/git/company_A.config

Every repository you create under that directory it will append the other configuration

Renaming from master to main

There's been a movement to migrate from master to main, the reason behind it is that the initial branch name, master, is offensive to some people and we empathize with those hurt by the use of that term.

Existing versions of Git are capable of working with any branch name; there's nothing special about master except that it has historically been the name used for the first branch when creating a new repository from scratch (with the git init command). Thus many projects use it to represent the primary line of development. We support and encourage projects to switch to branch names that are meaningful and inclusive.

To configure git to use main by default run:

git config --global init.defaultBranch main

It only works on since git version 2.28.0, so you're stuck with manually changing it if you have an earlier version.

Change's Controversy

The change is not free of controversy, for example in the PDM project some people are not sure that it's needed for many reasons. Let's see each of them:

As we're not part of the deciding organisms of the collectives doing the changes, all we can use are their statements and discussions to guess what are the reasons behind their support of the change. Despite that some of them do use the argument that other communities do support the change to emphasize the need of the change, all of them mention that the main reason is that the term is offensive to some people.

  • I don't see an issue using the term master: If you relate to this statement it can be because you're not part of the communities that suffer the oppression tied to the term, and that makes you blind to the issue. It's a lesson I learned on my own skin throughout the years. There are thousand of situations, gestures, double meaning words and sentences that went unnoticed by me until I started discussing it with the people that are suffering them (women, racialized people, LGTBQI+, ...). Throughout my experience I've seen that the more privileged you are, the blinder you become. You can read more on privileged blindness here, here or here (I've skimmed through the articles, and are the first articles I've found, there are probably better references).

I'm not saying that privileged people are not aware of the issues or that they can even raise them. We can do so and more we read, discuss and train ourselves, the better we'll detect them. All I'm saying is that a non privileged person will always detect more because they suffer them daily.

I understand that for you there is no issue using the word master, there wasn't an issue for me either until I saw these projects doing the change, again I was blinded to the issue as I'm not suffering it. That's because change is not meant for us, as we're not triggered by it. The change is targeted to the people that do perceive that master is an offensive term. What we can do is empathize with them and follow this tiny tiny tiny gesture. It's the least we can do.

Think of a term that triggers you, such as heil hitler, imagine that those words were being used to define the main branch of your code, and that everyday you sit in front of your computer you see them. You'll probably be reminded of the historic events, concepts, feelings that are tied to that term each time you see it, and being them quite negative it can slowly mine you. Therefore it's legit that you wouldn't want to be exposed to that negative effects.

  • I don't see who will benefit from this change: Probably the people that belongs to communities that are and have been under constant oppression for a very long time, in this case, specially the racialized ones which have suffered slavery.

Sadly you will probably won't see many the affected people speak in these discussions, first because there are not that many, sadly the IT world is dominated by middle aged, economically comfortable, white, cis, hetero, males. Small changes like this are meant to foster diversity in the community by allowing them being more comfortable. Secondly because when they see these debates they move on as they are so fed up on teaching privileged people of their privileges. They not only have to suffer the oppression, we also put the burden on their shoulders to teach us.

As and ending thought, if you see yourself being specially troubled by the change, having a discomfort feeling and strong reactions. In my experience these signs are characteristic of privileged people that feel that their privileges are being threatened, I've felt them myself countless times. When I feel it, I usually do two things, fight them as strong as I can, or embrace them, analyze them, and go to the root of them. Depending on how much energy I have I go with the easy or the hard one. I'm not saying that it's you're case, but it could be.

Configuration

Set the upstream remote by default

git config --global --add push.default current
git config --global --add push.autoSetupRemote true

Snippets

Remove tags

To delete a tag you can run:

git tag -d {{tag_name}}

To remove them remotely do

git push --delete origin {{ tag_name }}

Revert a commit

git revert commit_id

Get interesting stats of the repo

Number of commits of the last year per user:

git shortlog -sne --since="31 Dec 2020" --before="31 Dec 2021"

You can also use git-fame to extract a more detailed report:

$: git-fame --since 1.year --cost hour --loc ins -w -M -C

| Author          |   hrs |   loc |   coms |   fils |  distribution   |
|:----------------|------:|------:|-------:|-------:|:----------------|
| Lyz             |    10 | 28933 |    112 |    238 | 64.1/33.3/75.8  |
| GitHub Action   |     2 | 16194 |    220 |     73 | 35.9/65.5/23.2  |
| Alexander Gil   |     2 |     9 |      1 |      1 | 0.0/ 0.3/ 0.3   |
| n0rt3y5ur       |     2 |     1 |      1 |      1 | 0.0/ 0.3/ 0.3   |
| Guilherme Danno |     2 |     1 |      1 |      1 | 0.0/ 0.3/ 0.3   |
| lyz-code        |     2 |     0 |      1 |      0 | 0.0/ 0.3/ 0.0   |

You can use pipx install git-fame to install it.

Troubleshooting

fatal: A git directory for X is found locally with remote

If you get this error when you're adding a git submodule follow the next steps:

  • Remove the submodule files
    git rm --cached path_to_submodule (no trailing slash) as well as
    rm -rf path_to_submodule
    
  • Delete the relevant lines from the .gitmodules
    [submodule "path_to_submodule"]
      path = path_to_submodule
      url = https://github.com/path_to_submodule
    
  • Delete the relevant section from .git/config
    [submodule "path_to_submodule"]
      url = https://github.com/path_to_submodule
    
  • Remove the cached submodule
    rm -rf .git/modules/path_to_submodule
    
  • Then you can add the submodule again
    git submodule add https://github.com/path_to_submodule
    

References

Courses

Tools