Crate ya_runtime_wasi

This crate allows you to embed Yagna WASI runtime inside your application.

Quick start

The usage is pretty straightforward. In your Cargo.toml, put ya-runtime-wasi as your dependency

# Cargo.toml
[dependencies]
ya-runtime-wasi = "0.2"

You can now embed the runtime in your app like so

use std::path::Path;
use ya_runtime_wasi::*;

// In this example, we assume that `package.zip` contains a WASI binary
// called `hello.wasm` which also our entrypoint into the package, and
// maps input/output to `/workdir`
let workspace = Path::new("workspace");
let entrypoint = "hello";
let package = Path::new("package.zip");

// Deploy package
deploy(&workspace, &package).unwrap();

// Start the runtime
start(&workspace).unwrap();

// Execute the binary
run(
    &workspace,
    &entrypoint,
    vec![
        "/workdir/input".into(),
        "/workdir/output".into(),
    ],
).unwrap();

Obtaining handle to the deployed image

A handle to the deployed image, i.e., the path to the image as well as a list of mapped container volumes, can be obtained using DeployFile::load function invoked after ya_runtime_wasi::deploy and ya_runtime_wasi::start were run.

Examples

A good example of using ya-runtime-wasi embedding API can be found in the gfaas crate.

Structs

ContainerVolume	Represents name and path to the mapped volume in the container.
DeployFile	Represents deployed Yagna Wasm image with set up volumes inside the container.

Functions

deploy	Deploys the Wasm image into the workspace.
run	Instantiates and executes the deployed image using Wasmtime runtime.
start	Validates the deployed image.