Rust Server Projects

A project whose Rust model runs natively as tasks on a local server instead of as WebAssembly in the browser. Choose it over the WASM setup when a run takes minutes, needs threads or native dependencies WASM cannot use, or should leave results on disk to keep and compare. The page starts a task, streams its progress and log, plots the outputs, and lists earlier runs to reload. The API and the HTTP protocol are on the Server page.

Prerequisites: Node.js, pnpm, and the Rust toolchain as for Rust projects (no wasm-pack), plus the cfasim CLI: cfasim serve is the server, and it builds and runs your tasks binary.

Scaffold with cfasim init

cfasim init              # pick "Rust server", or pass --template rust-server
cd my-project
pnpm install
pnpm dev

pnpm dev starts Vite and, through the cfasimServer plugin, cfasim serve, which runs cargo build --release and keeps the tasks binary running as a worker; the dev server proxies /api to it. pnpm server runs the server alone (cfasim serve --debug skips --release for faster rebuilds while iterating).

my-project/
├── Cargo.toml              # cfasim-model with the "tasks" feature
├── src/
│   ├── lib.rs              # the model: pure functions, unit-tested
│   └── main.rs             # the tasks binary the page can start
├── output/                 # one folder per run (gitignored)
├── vite.config.ts          # cfasimServer() starts the server with vite
└── interactive/src/App.vue # useServerModel()

The task

The scaffolded src/main.rs registers one task, simulate, with an inputs struct (steps, rate) and a Default for it, and calls compute from src/lib.rs. Writing tasks in Rust walks through the same shape and lists what Context offers.

A task takes the context first and an inputs struct deserialized from the JSON the page sends; the struct's Default describes the parameters (names and defaults) to the page, so give every field a default. Return Outputs, or call ctx.emit(outputs)? for results along the way: emitted tables accumulate, so each call adds what is new and the page shows everything emitted so far. Anything else a run produces (CSVs, figures, a report) goes in ctx.run_dir(), and the page can list and download it. println! inside a task lands in the run log. Inputs must be a struct (or () for none): a misspelled input name is rejected by the server before the run starts.

The model itself stays in src/lib.rs as plain functions, so cargo test never needs the server.

The page

The page is the same as for a Python server project: useServerModel() from cfasim-ui/server starts tasks, follows the current run, and lists saved runs. The scaffolded App.vue is the complete version.

Existing projects and deployment

For a crate that already exists: add cfasim-model = { version = "…", default-features = false, features = ["tasks"] } and serde, add a binary target (src/main.rs, or a [[bin]] named in [package.metadata.cfasim] tasks = "<name>" when the package has several) that registers its tasks and calls Tasks::serve(), then add the UI as in the WASM guide with cfasimServer() in place of cfasimWasm.

Editing anything under src/ or Cargo.toml rebuilds and restarts the worker after the current run; a build error keeps the previous binary running and is printed in the terminal.

To deploy, build the site and serve it with cfasim serve --static dist; see Running the server.