From 13fb55de9a7a5c7b066b2d7166fdc74c2caa96fa Mon Sep 17 00:00:00 2001
From: imbolc <me@imbolc.name>
Date: Sun, 16 Jun 2024 12:43:14 +0600
Subject: [PATCH] Example

---
 examples/auto-vary.rs | 40 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)
 create mode 100644 examples/auto-vary.rs

diff --git a/examples/auto-vary.rs b/examples/auto-vary.rs
new file mode 100644
index 0000000..b7868d5
--- /dev/null
+++ b/examples/auto-vary.rs
@@ -0,0 +1,40 @@
+//! Using `auto-vary` middleware
+//!
+//! Don't forget about the feature while running it:
+//! `cargo run --features auto-vary --example auto-vary`
+use std::time::Duration;
+
+use axum::{response::Html, routing::get, serve, Router};
+use axum_htmx::{AutoVaryLayer, HxRequest};
+use tokio::{net::TcpListener, time::sleep};
+
+#[tokio::main]
+async fn main() {
+    let app = Router::new()
+        .route("/", get(handler))
+        // Add the middleware
+        .layer(AutoVaryLayer);
+
+    let listener = TcpListener::bind("0.0.0.0:3000").await.unwrap();
+    serve(listener, app).await.unwrap();
+}
+
+// Our handler differentiates full-page GET requests from HTMx-based ones by looking at the `hx-request`
+// requestheader.
+//
+// The middleware sees the usage of the `HxRequest` extractor and automatically adds the
+// `Vary: hx-request` response header.
+async fn handler(HxRequest(hx_request): HxRequest) -> Html<&'static str> {
+    if hx_request {
+        // For HTMx-based GET request, it returns a partial page update
+        sleep(Duration::from_secs(3)).await;
+        return Html("HTMx response");
+    }
+    // While for a normal GET request, it returns the whole page
+    Html(
+        r#"
+        <script src="https://unpkg.com/htmx.org@1"></script>
+        <p hx-get="/" hx-trigger="load">Loading ...</p>
+        "#,
+    )
+}