http-client: add safe route-aware request pool

This commit is contained in:
Michael Bolin
2026-07-09 15:50:15 -07:00
parent 839e7d9e0e
commit 4a0be0b111
5 changed files with 619 additions and 6 deletions

View File

@@ -1,4 +1,4 @@
use http::Error as HttpError;
use http::Error as HttpRequestBuildError;
use http::HeaderMap;
use http::HeaderName;
use http::HeaderValue;
@@ -6,13 +6,15 @@ use opentelemetry::global;
use opentelemetry::propagation::Injector;
use reqwest::IntoUrl;
use reqwest::Method;
use reqwest::Response;
use serde::Serialize;
use std::fmt::Display;
use std::time::Duration;
use tracing::Span;
use tracing_opentelemetry::OpenTelemetrySpanExt;
pub type HttpError = reqwest::Error;
pub type HttpResponse = reqwest::Response;
#[derive(Clone, Debug)]
pub struct HttpClient {
inner: reqwest::Client,
@@ -64,6 +66,43 @@ impl HttpClient {
self.request_logging,
)
}
pub(crate) async fn execute(
&self,
mut request: reqwest::Request,
) -> Result<reqwest::Response, reqwest::Error> {
request.headers_mut().extend(trace_headers());
let method = request.method().clone();
let url = request.url().to_string();
match self.inner.execute(request).await {
Ok(response) => {
if self.request_logging == RequestLogging::Enabled {
tracing::debug!(
method = %method,
url = %url,
status = %response.status(),
headers = ?response.headers(),
version = ?response.version(),
"Request completed"
);
}
Ok(response)
}
Err(error) => {
if self.request_logging == RequestLogging::Enabled {
tracing::debug!(
method = %method,
url = %url,
status = error.status().map(|status| status.as_u16()),
error = %error,
"Request failed"
);
}
Err(error)
}
}
}
}
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
@@ -112,9 +151,9 @@ impl RequestBuilder {
pub fn header<K, V>(self, key: K, value: V) -> Self
where
HeaderName: TryFrom<K>,
<HeaderName as TryFrom<K>>::Error: Into<HttpError>,
<HeaderName as TryFrom<K>>::Error: Into<HttpRequestBuildError>,
HeaderValue: TryFrom<V>,
<HeaderValue as TryFrom<V>>::Error: Into<HttpError>,
<HeaderValue as TryFrom<V>>::Error: Into<HttpRequestBuildError>,
{
self.map(|builder| builder.header(key, value))
}
@@ -144,7 +183,7 @@ impl RequestBuilder {
self.map(|builder| builder.body(body))
}
pub async fn send(self) -> Result<Response, reqwest::Error> {
pub async fn send(self) -> Result<HttpResponse, HttpError> {
let headers = trace_headers();
match self.builder.headers(headers).send().await {
@@ -192,7 +231,7 @@ impl<'a> Injector for HeaderMapInjector<'a> {
}
}
fn trace_headers() -> HeaderMap {
pub(crate) fn trace_headers() -> HeaderMap {
let mut headers = HeaderMap::new();
global::get_text_map_propagator(|prop| {
prop.inject_context(

View File

@@ -5,6 +5,7 @@ mod default_client;
mod error;
mod outbound_proxy;
mod request;
mod route_aware_client_pool;
mod transport;
pub use crate::chatgpt_cloudflare_cookies::with_chatgpt_cloudflare_cookie_store;
@@ -21,6 +22,8 @@ pub use crate::custom_ca::build_reqwest_client_with_custom_ca;
pub use crate::custom_ca::build_rustls_client_config_with_custom_ca;
pub use crate::custom_ca::maybe_build_rustls_client_config_with_custom_ca;
pub use crate::default_client::HttpClient;
pub use crate::default_client::HttpError;
pub use crate::default_client::HttpResponse;
pub use crate::default_client::RequestBuilder;
pub use crate::error::StreamError;
pub use crate::error::TransportError;
@@ -36,6 +39,10 @@ pub use crate::request::Request;
pub use crate::request::RequestBody;
pub use crate::request::RequestCompression;
pub use crate::request::Response;
pub use crate::route_aware_client_pool::RouteAwareClientPool;
pub use crate::route_aware_client_pool::RouteAwareClientPoolError;
pub use crate::route_aware_client_pool::RouteAwareRequestBuilder;
pub use crate::route_aware_client_pool::RouteAwareRequestError;
pub use crate::transport::ByteStream;
pub use crate::transport::HttpTransport;
pub use crate::transport::ReqwestTransport;

View File

@@ -253,6 +253,16 @@ impl HttpClientFactory {
self.outbound_proxy_policy,
)
}
pub(crate) fn build_reqwest_client_for_resolved_route(
&self,
builder: reqwest::ClientBuilder,
route_class: ClientRouteClass,
route: &OutboundProxyRoute,
) -> Result<reqwest::Client, BuildRouteAwareHttpClientError> {
let builder = configure_builder_for_resolved_route(builder, route_class, route)?;
build_reqwest_client_with_custom_ca(builder).map_err(Into::into)
}
}
fn resolve_proxy_route(

View File

@@ -0,0 +1,399 @@
use std::collections::HashMap;
use std::fmt;
use std::future::Future;
use std::io;
use std::sync::Arc;
use std::sync::Mutex;
use std::time::Duration;
use http::HeaderMap;
use http::HeaderName;
use http::HeaderValue;
use http::Method;
use http::StatusCode;
use http::header::AUTHORIZATION;
use http::header::CONTENT_TYPE;
use reqwest::IntoUrl;
use serde::Serialize;
use crate::BuildRouteAwareHttpClientError;
use crate::ClientRouteClass;
use crate::HttpClient;
use crate::HttpClientFactory;
use crate::OutboundProxyPolicy;
use crate::OutboundProxyRoute;
use crate::with_chatgpt_cloudflare_cookie_store;
const MAX_CACHED_ROUTES: usize = 16;
/// Reuses transport clients by resolved route while selecting a route for every request URL.
///
/// Request creation stays on the pool so the URL used for PAC or system-proxy resolution cannot
/// differ from the URL that is sent.
#[derive(Clone)]
pub struct RouteAwareClientPool {
http_client_factory: HttpClientFactory,
route_class: ClientRouteClass,
builder_factory: Arc<dyn Fn() -> reqwest::ClientBuilder + Send + Sync>,
request_logging: PoolRequestLogging,
clients: Arc<Mutex<HashMap<OutboundProxyRoute, HttpClient>>>,
}
impl fmt::Debug for RouteAwareClientPool {
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
formatter
.debug_struct("RouteAwareClientPool")
.field("http_client_factory", &self.http_client_factory)
.field("route_class", &self.route_class)
.field("request_logging", &self.request_logging)
.finish_non_exhaustive()
}
}
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
enum PoolRequestLogging {
Enabled,
Disabled,
}
/// Error returned when selecting a route or constructing its pooled HTTP client.
#[derive(Debug, thiserror::Error)]
pub enum RouteAwareClientPoolError {
#[error("failed to resolve the outbound proxy route: {0}")]
Resolve(#[source] io::Error),
#[error(transparent)]
Build(#[from] BuildRouteAwareHttpClientError),
}
/// Error returned while building, routing, or sending a route-aware request.
#[derive(Debug, thiserror::Error)]
pub enum RouteAwareRequestError {
#[error(transparent)]
Request(#[from] reqwest::Error),
#[error(transparent)]
Route(#[from] RouteAwareClientPoolError),
#[error("failed to build route-aware request: {0}")]
Build(String),
}
impl RouteAwareRequestError {
pub fn status(&self) -> Option<StatusCode> {
match self {
Self::Request(error) => error.status(),
Self::Route(_) | Self::Build(_) => None,
}
}
pub fn is_timeout(&self) -> bool {
matches!(self, Self::Request(error) if error.is_timeout())
}
pub fn is_connect(&self) -> bool {
matches!(self, Self::Request(error) if error.is_connect())
}
}
#[must_use = "requests are not sent unless `send` is awaited"]
pub struct RouteAwareRequestBuilder {
pool: RouteAwareClientPool,
request: Result<reqwest::Request, RouteAwareRequestError>,
}
impl fmt::Debug for RouteAwareRequestBuilder {
fn fmt(&self, formatter: &mut fmt::Formatter<'_>) -> fmt::Result {
formatter
.debug_struct("RouteAwareRequestBuilder")
.field("pool", &self.pool)
.field(
"url",
&self.request.as_ref().ok().map(reqwest::Request::url),
)
.finish_non_exhaustive()
}
}
impl RouteAwareRequestBuilder {
fn new<U>(pool: RouteAwareClientPool, method: Method, url: U) -> Self
where
U: IntoUrl,
{
let request = url
.into_url()
.map(|url| reqwest::Request::new(method, url))
.map_err(RouteAwareRequestError::Request);
Self { pool, request }
}
pub fn headers(mut self, headers: HeaderMap) -> Self {
if let Ok(request) = &mut self.request {
request.headers_mut().extend(headers);
}
self
}
pub fn header<K, V>(mut self, key: K, value: V) -> Self
where
HeaderName: TryFrom<K>,
<HeaderName as TryFrom<K>>::Error: Into<http::Error>,
HeaderValue: TryFrom<V>,
<HeaderValue as TryFrom<V>>::Error: Into<http::Error>,
{
if let Ok(request) = &mut self.request {
let header = HeaderName::try_from(key)
.map_err(Into::into)
.and_then(|key| {
HeaderValue::try_from(value)
.map(|value| (key, value))
.map_err(Into::into)
});
match header {
Ok((key, value)) => {
request.headers_mut().append(key, value);
}
Err(error) => {
self.request = Err(RouteAwareRequestError::Build(error.to_string()));
}
}
}
self
}
pub fn bearer_auth<T>(mut self, token: T) -> Self
where
T: fmt::Display,
{
let value = HeaderValue::from_str(&format!("Bearer {token}"));
match (&mut self.request, value) {
(Ok(request), Ok(mut value)) => {
value.set_sensitive(true);
request.headers_mut().append(AUTHORIZATION, value);
}
(Ok(_), Err(error)) => {
self.request = Err(RouteAwareRequestError::Build(error.to_string()));
}
(Err(_), _) => {}
}
self
}
pub fn timeout(mut self, timeout: Duration) -> Self {
if let Ok(request) = &mut self.request {
*request.timeout_mut() = Some(timeout);
}
self
}
pub fn json<T>(mut self, value: &T) -> Self
where
T: ?Sized + Serialize,
{
if let Ok(request) = &mut self.request {
match serde_json::to_vec(value) {
Ok(body) => {
if !request.headers().contains_key(CONTENT_TYPE) {
request
.headers_mut()
.insert(CONTENT_TYPE, HeaderValue::from_static("application/json"));
}
*request.body_mut() = Some(body.into());
}
Err(error) => {
self.request = Err(RouteAwareRequestError::Build(error.to_string()));
}
}
}
self
}
pub fn body<B>(mut self, body: B) -> Self
where
B: Into<reqwest::Body>,
{
if let Ok(request) = &mut self.request {
*request.body_mut() = Some(body.into());
}
self
}
pub async fn send(self) -> Result<reqwest::Response, RouteAwareRequestError> {
self.pool.send(self.request?).await
}
}
impl RouteAwareClientPool {
pub fn outbound_proxy_policy(&self) -> OutboundProxyPolicy {
self.http_client_factory.outbound_proxy_policy()
}
/// Creates a pool with the shared default HTTP transport settings.
pub fn new(http_client_factory: HttpClientFactory, route_class: ClientRouteClass) -> Self {
Self::with_builder_factory(
http_client_factory,
route_class,
reqwest::Client::builder,
PoolRequestLogging::Enabled,
)
}
/// Creates a pool with the shared defaults but without URL or response-header diagnostics.
pub fn new_without_request_logging(
http_client_factory: HttpClientFactory,
route_class: ClientRouteClass,
) -> Self {
Self::with_builder_factory(
http_client_factory,
route_class,
reqwest::Client::builder,
PoolRequestLogging::Disabled,
)
}
/// Creates a pool that retains the Cloudflare cookies required by ChatGPT endpoints.
pub fn with_chatgpt_cloudflare_cookies(
http_client_factory: HttpClientFactory,
route_class: ClientRouteClass,
) -> Self {
Self::with_builder_factory(
http_client_factory,
route_class,
|| with_chatgpt_cloudflare_cookie_store(reqwest::Client::builder()),
PoolRequestLogging::Enabled,
)
}
/// Creates a ChatGPT Cloudflare-cookie pool without URL or response-header diagnostics.
pub fn with_chatgpt_cloudflare_cookies_without_request_logging(
http_client_factory: HttpClientFactory,
route_class: ClientRouteClass,
) -> Self {
Self::with_builder_factory(
http_client_factory,
route_class,
|| with_chatgpt_cloudflare_cookie_store(reqwest::Client::builder()),
PoolRequestLogging::Disabled,
)
}
pub fn get<U>(&self, url: U) -> RouteAwareRequestBuilder
where
U: IntoUrl,
{
self.request(Method::GET, url)
}
pub fn post<U>(&self, url: U) -> RouteAwareRequestBuilder
where
U: IntoUrl,
{
self.request(Method::POST, url)
}
pub fn put<U>(&self, url: U) -> RouteAwareRequestBuilder
where
U: IntoUrl,
{
self.request(Method::PUT, url)
}
pub fn delete<U>(&self, url: U) -> RouteAwareRequestBuilder
where
U: IntoUrl,
{
self.request(Method::DELETE, url)
}
pub fn request<U>(&self, method: Method, url: U) -> RouteAwareRequestBuilder
where
U: IntoUrl,
{
RouteAwareRequestBuilder::new(self.clone(), method, url)
}
fn with_builder_factory(
http_client_factory: HttpClientFactory,
route_class: ClientRouteClass,
builder_factory: impl Fn() -> reqwest::ClientBuilder + Send + Sync + 'static,
request_logging: PoolRequestLogging,
) -> Self {
Self {
http_client_factory,
route_class,
builder_factory: Arc::new(builder_factory),
request_logging,
clients: Arc::new(Mutex::new(HashMap::new())),
}
}
async fn send(
&self,
request: reqwest::Request,
) -> Result<reqwest::Response, RouteAwareRequestError> {
let client = self.client_for_url(request.url().as_str()).await?;
Ok(client.execute(request).await?)
}
async fn client_for_url(
&self,
request_url: &str,
) -> Result<HttpClient, RouteAwareClientPoolError> {
let http_client_factory = self.http_client_factory.clone();
self.client_for_url_with_resolver(request_url, move |request_url| async move {
http_client_factory
.resolve_proxy_route_async(request_url)
.await
})
.await
}
async fn client_for_url_with_resolver<F, Fut>(
&self,
request_url: &str,
resolve_route: F,
) -> Result<HttpClient, RouteAwareClientPoolError>
where
F: FnOnce(String) -> Fut,
Fut: Future<Output = io::Result<OutboundProxyRoute>>,
{
let route = resolve_route(request_url.to_string())
.await
.map_err(RouteAwareClientPoolError::Resolve)?;
let clients = match self.clients.lock() {
Ok(clients) => clients,
Err(error) => panic!("route-aware client cache lock should not be poisoned: {error}"),
};
if let Some(client) = clients.get(&route) {
return Ok(client.clone());
}
drop(clients);
let client = self
.http_client_factory
.build_reqwest_client_for_resolved_route(
(self.builder_factory)().redirect(reqwest::redirect::Policy::none()),
self.route_class,
&route,
)?;
let client = match self.request_logging {
PoolRequestLogging::Enabled => HttpClient::new(client),
PoolRequestLogging::Disabled => HttpClient::new_without_request_logging(client),
};
let mut clients = match self.clients.lock() {
Ok(clients) => clients,
Err(error) => panic!("route-aware client cache lock should not be poisoned: {error}"),
};
if let Some(existing_client) = clients.get(&route) {
return Ok(existing_client.clone());
}
if clients.len() >= MAX_CACHED_ROUTES
&& let Some(route_to_evict) = clients.keys().next().cloned()
{
clients.remove(&route_to_evict);
}
clients.insert(route, client.clone());
Ok(client)
}
}
#[cfg(test)]
#[path = "route_aware_client_pool_tests.rs"]
mod tests;

View File

@@ -0,0 +1,158 @@
use std::collections::HashMap;
use std::io;
use std::sync::Arc;
use std::sync::Mutex;
use std::sync::atomic::AtomicUsize;
use std::sync::atomic::Ordering;
use pretty_assertions::assert_eq;
use super::*;
use crate::OutboundProxyPolicy;
#[tokio::test]
async fn forwards_exact_urls_and_reuses_clients_by_resolved_route() {
let builder_count = Arc::new(AtomicUsize::new(0));
let observed_builder_count = Arc::clone(&builder_count);
let pool = RouteAwareClientPool::with_builder_factory(
HttpClientFactory::new(OutboundProxyPolicy::ReqwestDefault),
ClientRouteClass::Api,
move || {
observed_builder_count.fetch_add(1, Ordering::SeqCst);
reqwest::Client::builder()
},
PoolRequestLogging::Enabled,
);
let direct_url = "https://example.com/first?target=direct";
let same_route_url = "https://example.com/second?target=direct%202";
let proxy_url = "https://example.com/third?target=proxy";
let resolver = FakeRouteResolver::new(HashMap::from([
(direct_url.to_string(), OutboundProxyRoute::Direct),
(same_route_url.to_string(), OutboundProxyRoute::Direct),
(
proxy_url.to_string(),
OutboundProxyRoute::Proxy {
url: "http://proxy.example".to_string(),
no_proxy: None,
},
),
]));
resolve_with(&pool, &resolver, direct_url)
.await
.expect("first client should build");
resolve_with(&pool, &resolver, same_route_url)
.await
.expect("second client should reuse the route");
resolve_with(&pool, &resolver, proxy_url)
.await
.expect("proxy client should build separately");
assert_eq!(builder_count.load(Ordering::SeqCst), 2);
assert_eq!(
resolver.observed_urls(),
vec![
direct_url.to_string(),
same_route_url.to_string(),
proxy_url.to_string(),
]
);
}
#[tokio::test]
async fn bounds_cached_routes_and_rebuilds_an_evicted_route() {
let builder_count = Arc::new(AtomicUsize::new(0));
let observed_builder_count = Arc::clone(&builder_count);
let pool = RouteAwareClientPool::with_builder_factory(
HttpClientFactory::new(OutboundProxyPolicy::ReqwestDefault),
ClientRouteClass::Api,
move || {
observed_builder_count.fetch_add(1, Ordering::SeqCst);
reqwest::Client::builder()
},
PoolRequestLogging::Enabled,
);
let routes = (0..=MAX_CACHED_ROUTES)
.map(|index| {
(
format!("https://target-{index}.example"),
OutboundProxyRoute::Proxy {
url: format!("http://proxy-{index}.example"),
no_proxy: None,
},
)
})
.collect::<HashMap<_, _>>();
let resolver = FakeRouteResolver::new(routes.clone());
for request_url in routes.keys() {
resolve_with(&pool, &resolver, request_url)
.await
.expect("client should build");
}
let evicted_route = {
let clients = pool.clients.lock().expect("client cache lock");
assert_eq!(clients.len(), MAX_CACHED_ROUTES);
routes
.iter()
.find(|(_, route)| !clients.contains_key(*route))
.map(|(request_url, _)| request_url.clone())
.expect("one route should have been evicted")
};
resolve_with(&pool, &resolver, &evicted_route)
.await
.expect("evicted client should rebuild");
assert_eq!(builder_count.load(Ordering::SeqCst), MAX_CACHED_ROUTES + 2);
assert_eq!(
pool.clients.lock().expect("client cache lock").len(),
MAX_CACHED_ROUTES
);
}
#[derive(Clone)]
struct FakeRouteResolver {
routes: Arc<HashMap<String, OutboundProxyRoute>>,
observed_urls: Arc<Mutex<Vec<String>>>,
}
impl FakeRouteResolver {
fn new(routes: HashMap<String, OutboundProxyRoute>) -> Self {
Self {
routes: Arc::new(routes),
observed_urls: Arc::new(Mutex::new(Vec::new())),
}
}
async fn resolve(&self, request_url: String) -> io::Result<OutboundProxyRoute> {
self.observed_urls
.lock()
.expect("observed URL lock")
.push(request_url.clone());
self.routes
.get(&request_url)
.cloned()
.ok_or_else(|| io::Error::other(format!("no route for {request_url}")))
}
fn observed_urls(&self) -> Vec<String> {
self.observed_urls
.lock()
.expect("observed URL lock")
.clone()
}
}
async fn resolve_with(
pool: &RouteAwareClientPool,
resolver: &FakeRouteResolver,
request_url: &str,
) -> Result<HttpClient, RouteAwareClientPoolError> {
let resolver = resolver.clone();
pool.client_for_url_with_resolver(request_url, move |request_url| async move {
resolver.resolve(request_url).await
})
.await
}