feat: demonstrate RetryBackoffLayer

README.md

@@ -41,6 +41,7 @@ This repository contains the following examples:
   - [x] [Wallet management filler](./examples/fillers/examples/wallet_filler.rs)
 - [x] Layers
   - [x] [Request / response logging layer](./examples/layers/examples/logging_layer.rs)
+  - [x] [Retry-backoff layer](./examples/layers/examples/retry_layer.rs)
 - [x] Node Bindings
   - [x] [Deploy contract on local Anvil instance](./examples/node-bindings/examples/anvil_deploy_contract.rs)
   - [x] [Fork instance on Anvil](./examples/node-bindings/examples/anvil_fork_instance.rs)

examples/layers/examples/retry_layer.rs

@@ -0,0 +1,38 @@
+//! This example demonstrates how to use the [`RetryBackoffLayer`] in the provider.
+
+use alloy::{
+    node_bindings::Anvil,
+    providers::{Provider, ProviderBuilder},
+    rpc::client::RpcClient,
+    transports::layers::RetryBackoffLayer,
+};
+
+#[tokio::main]
+async fn main() -> eyre::Result<()> {
+    let anvil = Anvil::new().spawn();
+
+    // The maximum number of retries for rate limit errors
+    let max_retry = 10;
+
+    //  The initial backoff in milliseconds
+    let backoff = 1000;
+
+    //  The number of compute units per second for this provider
+    let cups = 100;
+
+    // Instantiate the RetryBackoffLayer with the configuration
+    let retry_layer = RetryBackoffLayer::new(max_retry, backoff, cups);
+
+    // Add the layer to the transport client.
+    // The layer will retry all requests that return a rate limit error (eg. 429) until max_retries
+    // have been reached.
+    let client = RpcClient::builder().layer(retry_layer).http(anvil.endpoint_url());
+
+    let provider = ProviderBuilder::new().on_client(client);
+
+    let latest_block = provider.get_block_number().await?;
+
+    assert_eq!(latest_block, 0);
+
+    Ok(())
+}