Awesome
<!-- markdownlint-disable-next-line --> <div align="center"> <img src="https://github.com/helsing-ai/buffrs/assets/37018485/76c51445-b5a6-4f4e-a39c-7de7e31a0613" onerror="this.style.display='none'" /> <br/>Buffrs
Modern protobuf package management
</div>Quickstart
$ cargo install buffrs
$ buffrs login
$ buffrs init --api
$ buffrs add <dependency>
$ buffrs install
Useful resources:
- The Buffrs Book
- Crate Documentation
- Announcement Post
buffrs help
Synopsis
Modern protobuf package management
Usage: buffrs <COMMAND>
Commands:
init Initializes a buffrs setup
new Creates a new buffrs package in the current directory
lint Check rule violations for this package
add Adds dependencies to a manifest file
remove Removes dependencies from a manifest file
package Exports the current package into a distributable tgz archive
publish Packages and uploads this api to the registry
install Installs dependencies
uninstall Uninstalls dependencies
list Lists all protobuf files managed by Buffrs to stdout
login Logs you in for a registry
logout Logs you out from a registry
lock Lockfile related commands
help Print this message or the help of the given subcommand(s)
Options:
-h, --help Print help
-V, --version Print version
Motivation
Protocol buffers are agreeably a great way to define fully typed, language-independent API schemas with strong backward compatibility guarantees. They offer a neat experience for API consumers through generated bindings. The biggest problem associated with Protocol Buffers is their distribution.
- How do you consume the raw protobuf files of one project reliably in another one?
- How do you prevent transitive dependencies?
- How do you publish to a unified registry with package format across languages?
One obvious way is to generate code bindings in the repository containing the Protocol Buffers and publish the generated bindings, but this is associated with problems such as language lock-in. You need to proactively publish bindings for any possible language your API consumers may use. Also, in strongly typed languages like Rust, it is hard to extend the behavior of generated code in consuming projects due to the orphan rule. Summing up: this approach works somehow but hurts frequently.
This is where Buffrs comes in: Buffrs solves this by defining a strict, package-based distribution mechanism and treats Protocol Buffers as a first-class citizen.
This allows you to publish Buffrs packages to a registry and properly depend on them in other projects.
Roadmap
- Support project manifests and dependency declaration
- Support package distribution via Artifactory
- Support tonic as code generation backend
- Support protoc as code generation backend
- Implement
buffrs-registry
, a self-hostable, S3-based registry. - Supply tooling around Protocol Buffers, such as bindgen, linting, validation and formatting.