Hive plugin for Apollo-Router

Hive is a fully open-source schema registry, analytics, metrics and gateway for GraphQL federation and other GraphQL APIs.

This project includes a Hive integration plugin for Apollo-Router.

At the moment, the following are implemented:

• Fetching Supergraph from Hive CDN
• Sending usage information from a running Apollo Router instance to Hive
• Persisted Operations using Hive's App Deployments

This project is constructed as a Rust project that implements Apollo-Router plugin interface.

This build of this project creates an artifact identical to Apollo-Router releases, with additional features provided by Hive.

Getting Started

Binary/Docker

We provide a custom build of Apollo-Router that acts as a drop-in replacement, and adds Hive integration to Apollo-Router.

Please follow this guide and documentation for integrating Hive with Apollo Router

Library

If you are building a custom Apollo-Router with your own native plugins, you can use the Hive plugin as a dependency from Crates.io:

[dependencies]
hive-apollo-router-plugin = "..."

And then in your codebase, make sure to import and register the Hive plugin:

// import the registry instance and the plugin registration function
use hive_apollo_router_plugin::registry::HiveRegistry;
use hive_apollo_router_plugin::usage::register;

// In your main function, make sure to register the plugin before you create or initialize Apollo-Router
fn main() {
    // Register the Hive usage_reporting plugin
    register();

    // Initialize the Hive Registry instance and start the Apollo Router
    match HiveRegistry::new(None).and(apollo_router::main()) {
        Ok(_) => {}
        Err(e) => {
            eprintln!("{}", e);
            std::process::exit(1);
        }
    }
}

Development

0. Install latest version of Rust
1. To get started with development, it is recommended to ensure Rust-analyzer extension is enabled on your VSCode instance.
2. Validate project status by running cargo check
3. To start the server with the demo config file (./router.yaml), use cargo run -- --config router.yaml. Make sure to set environment variables required for your setup and development process (docs).
4. You can also just run cargo run -- --config router.yaml --log debug --dev --supergraph some.supergraph.graphql for running it with a test supergraph file.