Tauri Invoke HTTP

This is a crate that provides a custom invoke system for Tauri using a localhost server. Each message is delivered through a XMLHttpRequest and the server is responsible for replying to it.

Usage

First, add the dependency to your src-tauri/Cargo.toml file:

[dependencies]
tauri-invoke-http = { git = "https://github.com/tauri-apps/tauri-invoke-http", branch = "dev" }

Then, setup the HTTP invoke system on the main.rs file:

fn main() {
  // initialize the custom invoke system as a HTTP server, allowing the given origins to access it.
  let http = tauri_invoke_http::Invoke::new(if cfg!(feature = "custom-protocol") {
    ["tauri://localhost"]
  } else {
    ["http://localhost:8080"]
  });
  tauri::Builder::default()
    .invoke_system(http.initialization_script(), http.responder())
    .setup(move |app| {
      http.start(app.handle());
      Ok(())
    })
    .run(tauri::generate_context!())
    .expect("error while running tauri application")
}

To invoke a custom command from your own or remote system you can use curl or similar tooling. See examples/vanilla to test this on your system.

An example command to invoke the exit command in the example Tauri app exposing port 18436 (randomly chosen port) could look like:

curl localhost:18436/main -H 'Content-Type: application/json' -d '{ "__tauriModule": "Process", "cmd": "exit", "callback": 1234, "error": 1234, "message": {"cmd": "exit", "exitCode": 1  } }'