Version Control

An experiment to write a version control system from scratch in Rust.

CLI Usage

Usage: revtool <COMMAND>

Commands:
  init      initialize a brand new revision
  diff      check the difference between this branch and another
  changes   shows the files and directories which have been changed since the latest snap
  snap      take a new snapshot
  checkout  switch to branch
  branch    print out current branch
  help      Print this message or the help of the given subcommand(s)

Options:
  -h, --help  Print help

How Does It Work?

Under the hood, it is using a content addressed binary object store in .rev/store using the BLAKE3 hash function to identify binary objects using a type called ObjectId. Building on top of that, we have a Directory data type which represents a directory tree structure where each file's name is mapped to it's ObjectId. We store this type directly, encoded as prettified JSON, in the object store as a commitment to a particular configuration or version, then we have a data type called SnapShot which links these together into a directed, acyclic graph with each vertex having a message attached:

pub struct SnapShot {
  message: String,
  directory: ObjectId,
  previous: ObjectId,
}

We then store this type directly in the object store as well, again as prettified JSON. In the .rev/branches directory, we keep a file for each branch with the ObjectId of a particular encoded SnapShot. In .rev/branch, we keep the name of the current branch we're using.

Finally, when we construct a Directory from the current directory, often we don't care about many files, so we have an ignore list in .rev/ignores which configures which paths we ignore.

Contributing

There are a number of issues on the GitHub repository, please feel free to take any and do them.