Use type safe Mullvad protobuf client over generated dito

Re-write some code in the test framework to prefer the type safe wrapper
around the Mullvad app gRPC client instead of its auto-generated dito.

`ManagementServiceClient` is automatically generated from the protobuf
definitions found in `management_interface.proto`, and contains some
very crude types. The `MullvadProxyClient` is a type-safe wrapper around
`ManagementServiceClient` which performs conversions & validation of the
data types from the gRPC server (the daemon) to their respective
mappings in the `talpid-*` and `mullvad-*` crates. These types are more
ergonomic to work with, and since we already have the conversions in
place we should prefer those.

test/clippy.toml

@@ -0,0 +1,3 @@
+disallowed-types = [
+    { path = "mullvad_management_interface::ManagementServiceClient", reason = "use `mullvad_management_interface::MullvadProxyClient` instead" },
+]

test/test-manager/README.md

@@ -13,7 +13,7 @@ with the `#[test_function]` attribute
 #[test_function]
 pub async fn test(
     rpc: ServiceClient,
-    mut mullvad_client: mullvad_management_interface::ManagementServiceClient,
+    mut mullvad_client: mullvad_management_interface::MullvadProxyClient,
 ) -> Result<(), Error> {
     Ok(())
 }

test/test-manager/src/mullvad_daemon.rs

@@ -1,7 +1,8 @@
+#![allow(clippy::disallowed_types)]
 use std::{io, time::Duration};
 
 use futures::{channel::mpsc, future::BoxFuture, pin_mut, FutureExt, SinkExt, StreamExt};
-use mullvad_management_interface::ManagementServiceClient;
+use mullvad_management_interface::{ManagementServiceClient, MullvadProxyClient};
 use test_rpc::{
     mullvad_daemon::MullvadClientVersion,
     transport::{ConnectionHandle, GrpcForwarder},
@@ -61,7 +62,7 @@ impl RpcClientProvider {
         }
     }
 
-    pub async fn new_client(&self) -> ManagementServiceClient {
+    pub async fn new_client(&self) -> MullvadProxyClient {
         // FIXME: Ugly workaround to ensure that we don't receive stuff from a
         // previous RPC session.
         tokio::time::sleep(std::time::Duration::from_millis(500)).await;
@@ -72,7 +73,7 @@ impl RpcClientProvider {
             .await
             .unwrap();
 
-        ManagementServiceClient::new(channel)
+        MullvadProxyClient::from_rpc_client(ManagementServiceClient::new(channel))
     }
 }
 

test/test-manager/src/run_tests.rs

@@ -8,7 +8,7 @@ use crate::{
 };
 use anyhow::{Context, Result};
 use futures::FutureExt;
-use mullvad_management_interface::ManagementServiceClient;
+use mullvad_management_interface::MullvadProxyClient;
 use std::future::Future;
 use std::panic;
 use std::time::Duration;
@@ -84,7 +84,7 @@ pub async fn run(
             .as_type(test.mullvad_client_version)
             .await;
 
-        if let Some(client) = mclient.downcast_mut::<ManagementServiceClient>() {
+        if let Some(client) = mclient.downcast_mut::<MullvadProxyClient>() {
             crate::tests::init_default_settings(client).await;
         }
 

test/test-manager/src/tests/account.rs

@@ -1,8 +1,8 @@
 use super::config::TEST_CONFIG;
 use super::{helpers, ui, Error, TestContext};
 use mullvad_api::DevicesProxy;
-use mullvad_management_interface::{types, Code, ManagementServiceClient};
-use mullvad_types::device::Device;
+use mullvad_management_interface::{self, client::DaemonEvent, MullvadProxyClient};
+use mullvad_types::device::{Device, DeviceState};
 use mullvad_types::states::TunnelState;
 use std::net::ToSocketAddrs;
 use std::time::Duration;
@@ -17,7 +17,7 @@ const THROTTLE_RETRY_DELAY: Duration = Duration::from_secs(120);
 pub async fn test_login(
     _: TestContext,
     _rpc: ServiceClient,
-    mut mullvad_client: ManagementServiceClient,
+    mut mullvad_client: MullvadProxyClient,
 ) -> Result<(), Error> {
     //
     // Instruct daemon to log in
@@ -33,7 +33,7 @@ pub async fn test_login(
         .expect("login failed");
 
     // Wait for the relay list to be updated
-    helpers::ensure_updated_relay_list(&mut mullvad_client).await;
+    helpers::ensure_updated_relay_list(&mut mullvad_client).await?;
 
     Ok(())
 }
@@ -44,12 +44,12 @@ pub async fn test_login(
 pub async fn test_logout(
     _: TestContext,
     _rpc: ServiceClient,
-    mut mullvad_client: ManagementServiceClient,
+    mut mullvad_client: MullvadProxyClient,
 ) -> Result<(), Error> {
     log::info!("Removing device");
 
     mullvad_client
-        .logout_account(())
+        .logout_account()
         .await
         .expect("logout failed");
 
@@ -61,7 +61,7 @@ pub async fn test_logout(
 pub async fn test_too_many_devices(
     _: TestContext,
     rpc: ServiceClient,
-    mut mullvad_client: ManagementServiceClient,
+    mut mullvad_client: MullvadProxyClient,
 ) -> Result<(), Error> {
     log::info!("Using up all devices");
 
@@ -97,7 +97,10 @@ pub async fn test_too_many_devices(
     log::info!("Log in with too many devices");
     let login_result = login_with_retries(&mut mullvad_client).await;
 
-    assert!(matches!(login_result, Err(status) if status.code() == Code::ResourceExhausted));
+    assert!(matches!(
+        login_result,
+        Err(mullvad_management_interface::Error::TooManyDevices)
+    ));
 
     // Run UI test
     let ui_result = ui::run_test_env(
@@ -128,22 +131,20 @@ pub async fn test_too_many_devices(
 pub async fn test_revoked_device(
     _: TestContext,
     rpc: ServiceClient,
-    mut mullvad_client: ManagementServiceClient,
+    mut mullvad_client: MullvadProxyClient,
 ) -> Result<(), Error> {
     log::info!("Logging in/generating device");
     login_with_retries(&mut mullvad_client)
         .await
         .expect("login failed");
 
     let device_id = mullvad_client
-        .get_device(())
+        .get_device()
         .await
         .expect("failed to get device data")
-        .into_inner()
-        .device
+        .into_device()
         .unwrap()
         .device
-        .unwrap()
         .id;
 
     helpers::connect_and_wait(&mut mullvad_client).await?;
@@ -165,16 +166,16 @@ pub async fn test_revoked_device(
     // Begin listening to tunnel state changes first, so that we catch changes due to
     // `update_device`.
     let events = mullvad_client
-        .events_listen(())
+        .events_listen()
         .await
-        .expect("failed to begin listening for state changes")
-        .into_inner();
+        .expect("failed to begin listening for state changes");
     let next_state =
         helpers::find_next_tunnel_state(events, |state| matches!(state, TunnelState::Error(..),));
 
     log::debug!("Update device state");
 
-    let _update_status = mullvad_client.update_device(()).await;
+    // Update the device status, which performs a device validation.
+    let _ = mullvad_client.update_device().await;
 
     // Ensure that the tunnel state transitions to "error". Fail if it transitions to some other
     // state.
@@ -186,12 +187,11 @@ pub async fn test_revoked_device(
 
     // Verify that the device state is `Revoked`.
     let device_state = mullvad_client
-        .get_device(())
+        .get_device()
         .await
         .expect("failed to get device data");
-    assert_eq!(
-        device_state.into_inner().state,
-        i32::from(types::device_state::State::Revoked),
+    assert!(
+        matches!(device_state, DeviceState::Revoked),
         "expected device to be revoked"
     );
 
@@ -244,28 +244,26 @@ pub async fn new_device_client() -> DevicesProxy {
 
 /// Log in and retry if it fails due to throttling
 pub async fn login_with_retries(
-    mullvad_client: &mut ManagementServiceClient,
-) -> Result<(), mullvad_management_interface::Status> {
+    mullvad_client: &mut MullvadProxyClient,
+) -> Result<(), mullvad_management_interface::Error> {
     loop {
-        let result = mullvad_client
+        match mullvad_client
             .login_account(TEST_CONFIG.account_number.clone())
-            .await;
+            .await
+        {
+            Err(mullvad_management_interface::Error::Rpc(status))
+                if status.message().to_uppercase().contains("THROTTLED") =>
+            {
+                // Work around throttling errors by sleeping
+                log::debug!(
+                    "Login failed due to throttling. Sleeping for {} seconds",
+                    THROTTLE_RETRY_DELAY.as_secs()
+                );
 
-        if let Err(error) = result {
-            if !error.message().contains("THROTTLED") {
-                return Err(error);
+                tokio::time::sleep(THROTTLE_RETRY_DELAY).await;
             }
-
-            // Work around throttling errors by sleeping
-
-            log::debug!(
-                "Login failed due to throttling. Sleeping for {} seconds",
-                THROTTLE_RETRY_DELAY.as_secs()
-            );
-
-            tokio::time::sleep(THROTTLE_RETRY_DELAY).await;
-        } else {
-            break Ok(());
+            Err(err) => break Err(err),
+            Ok(_) => break Ok(()),
         }
     }
 }
@@ -306,18 +304,17 @@ pub async fn retry_if_throttled<
 pub async fn test_automatic_wireguard_rotation(
     ctx: TestContext,
     rpc: ServiceClient,
-    mut mullvad_client: ManagementServiceClient,
+    mut mullvad_client: MullvadProxyClient,
 ) -> Result<(), Error> {
+    use futures::StreamExt;
     // Make note of current WG key
     let old_key = mullvad_client
-        .get_device(())
+        .get_device()
         .await
-        .expect("Could not get device")
-        .into_inner()
-        .device
         .unwrap()
+        .into_device()
+        .expect("Could not get device")
         .device
-        .unwrap()
         .pubkey;
 
     // Stop daemon
@@ -343,29 +340,29 @@ pub async fn test_automatic_wireguard_rotation(
     // Verify rotation has happened after a minute
     const KEY_ROTATION_TIMEOUT: Duration = Duration::from_secs(100);
 
-    let mut event_stream = mullvad_client.events_listen(()).await.unwrap().into_inner();
+    let mut event_stream = mullvad_client.events_listen().await.unwrap();
     let get_pub_key_event = async {
         loop {
-            let message = event_stream.message().await;
-            if let Ok(Some(event)) = message {
-                match event.event.unwrap() {
-                    mullvad_management_interface::types::daemon_event::Event::Device(
-                        device_event,
-                    ) => {
-                        let pubkey = device_event
-                            .new_state
-                            .unwrap()
-                            .device
-                            .unwrap()
-                            .device
-                            .unwrap()
-                            .pubkey;
-                        return Ok(pubkey);
-                    }
-                    _ => continue,
+            // TODO(markus): See if this can be refactored. This is exactly the same as helpers:274.
+            match event_stream.next().await {
+                Some(Ok(DaemonEvent::Device(device_event))) => {
+                    let pubkey = device_event
+                        .new_state
+                        .into_device()
+                        .expect("Could not get device")
+                        .device
+                        .pubkey;
+                    return Ok(pubkey);
+                }
+                Some(Ok(_)) => continue,
+                Some(Err(status)) => {
+                    break Err(Error::Daemon(format!(
+                        "Failed to get next event: {}",
+                        status
+                    )))
                 }
+                None => break Err(Error::Daemon(String::from("Lost daemon event stream"))),
             }
-            return Err(message);
         }
     };