A Git-compatible VCS that is both simple and powerful
Find a file
Martin von Zweigbergk e48ace56d1 conflicts: replace missing files by empty in materialized conflict
When we materialize modify/delete conflicts, we currently don't
include any context lines. That's because modify/delete conflicts have
only two sides, so there's no common base to compare to. Hunks that
are unchanged on the "modify" side are therefore not considered
conflicting, and since they they don't contribute new changes, they're
simply skipped (here:
3dfedf5814/lib/src/files.rs (L228-L230)).

It seems more useful to instead pretend that the missing side is an
empty file. That way we'll get a conflict in the entire file.

We can still decide later to make e.g. `jj resolve` prompt the user on
modify/delete conflicts just like `hg resolve` does (or maybe it
actually happens earlier there, I don't remember).

Closes #1244.
2023-02-17 22:19:04 -08:00
.github release: build Linux (musl) release using vendored OpenSSL 2023-02-16 15:52:00 -08:00
benches cleanup: let new Clippy move variables into format strings 2022-12-14 21:30:58 -08:00
demos demos: minor changes to operation log demo 2023-02-15 00:25:56 -08:00
docs docs/config.md: add example for using pipes 2023-02-17 08:35:35 -08:00
examples backend: let backend decide length of change id 2023-02-07 22:31:34 -08:00
lib conflicts: replace missing files by empty in materialized conflict 2023-02-17 22:19:04 -08:00
src templater: extract function that dispatches method call by property type 2023-02-18 01:16:43 +09:00
testing Add a config edit command to open jj config in editor 2023-01-12 01:10:07 -06:00
tests conflicts: replace missing files by empty in materialized conflict 2023-02-17 22:19:04 -08:00
.editorconfig editorconfig: disable trim_trailing_whitespace due to multi-line bugs 2022-03-07 20:23:55 -08:00
.envrc nix: re-enable the .envrc file 2023-01-28 07:10:12 +01:00
.gitattributes build: mark Cargo.lock as binary for merging 2022-04-25 22:58:15 -07:00
.gitignore Configure nix development shell environment 2023-01-19 06:45:50 +01:00
AUTHORS copyright: change from "Google LLC" to "The Jujutsu Authors" 2022-11-28 06:05:45 -10:00
Cargo.lock release: release version 0.7.0 2023-02-16 12:50:52 -08:00
Cargo.toml release: release version 0.7.0 2023-02-16 12:50:52 -08:00
CHANGELOG.md conflicts: replace missing files by empty in materialized conflict 2023-02-17 22:19:04 -08:00
deny.toml cargo: replace dependency on Sapling's dag by renderdag 2023-01-22 00:03:48 -08:00
flake.lock nix: replace manual system iteration by flake-utils 2023-01-28 07:10:12 +01:00
flake.nix nix: use default list of systems 2023-01-28 07:10:12 +01:00
LICENSE Boilerplate for new Google open source project 2020-12-11 23:37:59 -08:00
README.md readme: add --locked to cargo install step 2023-01-27 18:42:09 -08:00
rustfmt.toml rustfmt: set edition to 2021 2022-06-09 14:41:11 -07:00

Jujutsu VCS


Disclaimer

This is not a Google product. It is an experimental version-control system (VCS). It was written by me, Martin von Zweigbergk (martinvonz@google.com). It is my personal hobby project and my 20% project at Google. It does not indicate any commitment or direction from Google. However, my presentation from Git Merge 2022 does include some information about Google's plans. See the slides or the recording.

Introduction

Jujutsu is a Git-compatible DVCS. It combines features from Git (data model, speed), Mercurial (anonymous branching, simple CLI free from "the index", revsets, powerful history-rewriting), and Pijul/Darcs (first-class conflicts), with features not found in most of them (working-copy-as-a-commit, undo functionality, automatic rebase, safe replication via rsync, Dropbox, or distributed file system).

The command-line tool is called jj for now because it's easy to type and easy to replace (rare in English). The project is called "Jujutsu" because it matches "jj".

If you have any questions, please join us on Discord Discord.

Features

Compatible with Git

Jujutsu has two backends. One of them is a Git backend (the other is a native one 1). This lets you use Jujutsu as an alternative interface to Git. The commits you create will look like regular Git commits. You can always switch back to Git. The Git support uses the libgit2 C library.

The working copy is automatically committed

Most Jujutsu commands automatically commit the working copy. This leads to a simpler and more powerful interface, since all commands work the same way on the working copy or any other commit. It also means that you can always check out a different commit without first explicitly committing the working copy changes (you can even check out a different commit while resolving merge conflicts).

Operations update the repo first, then possibly the working copy

The working copy is only updated at the end of an operation, after all other changes have already been recorded. This means that you can run any command (such as jj rebase) even if the working copy is dirty.

Entire repo is under version control

All operations you perform in the repo are recorded, along with a snapshot of the repo state after the operation. This means that you can easily revert to an earlier repo state, or to simply undo a particular operation (which does not necessarily have to be the most recent operation).

Conflicts can be recorded in commits

If an operation results in conflicts, information about those conflicts will be recorded in the commit(s). The operation will succeed. You can then resolve the conflicts later. One consequence of this design is that there's no need to continue interrupted operations. Instead, you get a single workflow for resolving conflicts, regardless of which command caused them. This design also lets Jujutsu rebase merge commits correctly (unlike both Git and Mercurial).

Basic conflict resolution:

Juggling conflicts:

Automatic rebase

Whenever you modify a commit, any descendants of the old commit will be rebased onto the new commit. Thanks to the conflict design described above, that can be done even if there are conflicts. Branches pointing to rebased commits will be updated. So will the working copy if it points to a rebased commit.

Comprehensive support for rewriting history

Besides the usual rebase command, there's jj describe for editing the description (commit message) of an arbitrary commit. There's also jj diffedit, which lets you edit the changes in a commit without checking it out. To split a commit into two, use jj split. You can even move part of the changes in a commit to any other commit using jj move.

Status

The tool is quite feature-complete, but some important features like (the equivalent of) git blame are not yet supported. There are also several performance bugs. It's also likely that workflows and setups different from what the core developers use are not well supported.

I (Martin von Zweigbergk) have almost exclusively used jj to develop the project itself since early January 2021. I haven't had to re-clone from source (I don't think I've even had to restore from backup).

There will be changes to workflows and backward-incompatible changes to the on-disk formats before version 1.0.0. Even the binary's name may change (i.e. away from jj). For any format changes, we'll try to implement transparent upgrades (as we've done with recent changes), or provide upgrade commands or scripts if requested.

Installation

See below for how to build from source. There are also pre-built binaries for Windows, Mac, or Linux (musl).

If you're installing from source, you need to use Rust version 1.61 or higher, or you will get a cryptic message like this:

error: failed to select a version for the requirement `libgit2-sys = "=0.14.0"``
candidate versions found which didn't match: 0.13.2+1.4.2, 0.13.1+1.4.2, 0.13.0+1.4.1, ...

Linux

On most distributions, you'll need to build from source using cargo directly.

Build using cargo

First make sure that you have the libssl-dev, openssl, and pkg-config packages installed by running something like this:

sudo apt-get install libssl-dev openssl pkg-config

Now run:

cargo install --git https://github.com/martinvonz/jj.git --locked --bin jj jujutsu

Nix OS

If you're on Nix OS you can use the flake for this repository. For example, if you want to run jj loaded from the flake, use:

nix run 'github:martinvonz/jj'

You can also add this flake url to your system input flakes. Or you can install the flake to your user profile:

nix profile install 'github:martinvonz/jj'

Homebrew

If you use linuxbrew, you can run:

brew install jj

Mac

Homebrew

If you use Homebrew, you can run:

brew install jj

MacPorts

You can also install jj via MacPorts (as the jujutsu port):

sudo port install jujutsu

(port page)

From Source

You may need to run some or all of these:

xcode-select --install
brew install openssl
brew install pkg-config
export PKG_CONFIG_PATH="$(brew --prefix)/opt/openssl@3/lib/pkgconfig"

Now run:

cargo install --git https://github.com/martinvonz/jj.git --locked --bin jj jujutsu

Windows

Run:

cargo install --git https://github.com/martinvonz/jj.git --locked --bin jj jujutsu --features vendored-openssl

Initial configuration

You may want to configure your name and email so commits are made in your name. Create a file at ~/.jjconfig.toml and make it look something like this:

$ cat ~/.jjconfig.toml
[user]
name = "Martin von Zweigbergk"
email = "martinvonz@google.com"

Command-line completion

To set up command-line completion, source the output of jj debug completion --bash/--zsh/--fish. Exactly how to source it depends on your shell.

Bash

source <(jj debug completion)  # --bash is the default

Zsh

autoload -U compinit
compinit
source <(jj debug completion --zsh | sed '$d')  # remove the last line
compdef _jj jj

Fish

jj debug completion --fish | source

Xonsh

source-bash $(jj debug completion)

Getting started

The best way to get started is probably to go through the tutorial. Also see the Git comparison, which includes a table of jj vs. git commands.

There are several tools trying to solve similar problems as Jujutsu. See related work for details.


  1. At this time, there's practically no reason to use the native backend (the only minor reason might be #27). The backend exists mainly to make sure that it's possible to eventually add functionality that cannot easily be added to the Git backend. ↩︎