nautilus_network/http/

mod.rs

// -------------------------------------------------------------------------------------------------
//  Copyright (C) 2015-2026 Nautech Systems Pty Ltd. All rights reserved.
//  https://nautechsystems.io
//
//  Licensed under the GNU Lesser General Public License Version 3.0 (the "License");
//  You may not use this file except in compliance with the License.
//  You may obtain a copy of the License at https://www.gnu.org/licenses/lgpl-3.0.en.html
//
//  Unless required by applicable law or agreed to in writing, software
//  distributed under the License is distributed on an "AS IS" BASIS,
//  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
//  See the License for the specific language governing permissions and
//  limitations under the License.
// -------------------------------------------------------------------------------------------------

//! Asynchronous HTTP requests with rate limiting, connection reuse, and bounded responses.
//!
//! # Architecture
//!
//! [`HttpClient`] applies quota policy before delegating requests to [`InnerHttpClient`]. The inner
//! client owns one reusable Hyper client, preserving its connection pool across requests and clones.
//!
//! # Rate limiting and requests
//!
//! Requests can await default and per-key quotas from one or more shared
//! [`RateLimiter`](crate::ratelimiter::RateLimiter) instances. Sharing a limiter across clients
//! enforces one process-wide budget for scopes such as an IP address or account. The client accepts
//! default and per-request headers, repeated query values, raw bodies, and client-level and
//! per-request timeouts.
//!
//! # Proxy routing
//!
//! By default the client honors ambient proxy configuration: with `use_system_proxy` left at its
//! default of `true` and no explicit `proxy_url`, requests are routed through the proxy named by
//! `HTTP_PROXY`, `HTTPS_PROXY`, or `ALL_PROXY` (lowercase variants included), except destinations
//! matched by `NO_PROXY`.
//!
//! This default is a deliberate trust decision on the process environment: an actor who controls
//! it chooses the proxy that observes all plaintext HTTP traffic and the CONNECT tunnels that
//! carry HTTPS.
//!
//! An optional explicit `proxy_url` applies to both HTTP and HTTPS traffic and always takes
//! precedence over ambient lookup. Passing `use_system_proxy(false)` disables ambient lookup, so
//! requests route directly only when `proxy_url` is unset.
//!
//! HTTP status errors remain [`HttpResponse`] values for adapter-specific handling. The transport
//! retries requests canceled before transmission on reused connections, and allows two retries for
//! remote HTTP/2 `GOAWAY(NO_ERROR)` or `REFUSED_STREAM` errors. Other transport failures and HTTP
//! status codes do not trigger retries. Adapters can apply [`crate::retry::RetryManager`] when the
//! operation and venue error are safe to retry.
//!
//! # Connection and response policy
//!
//! Production clients built through [`HttpClient::builder`] enable `TCP_NODELAY`, pooled idle
//! connections, and HTTP/2 keepalive while idle.
//!
//! HTTP/2 connections use fixed flow-control windows of 16 MiB per stream and 32 MiB per
//! connection, larger than Hyper's fixed defaults of 2 MiB and 5 MiB. Setting
//! `NAUTILUS_HTTP2_ADAPTIVE_WINDOW=true` before building a client enables Hyper's adaptive windows
//! instead, which start at 65,535 bytes and grow toward 16 MiB through PING probes and SETTINGS
//! updates. Cloudflare-fronted endpoints can reset large response bodies mid-transfer under
//! adaptive windows. The value `false` keeps fixed windows, and any other value fails the build.
//!
//! Buffered requests advertise gzip unless default or per-request headers set `Accept-Encoding`.
//! Gzip responses are decoded before return; unsupported encodings and corrupt bodies are errors.
//! The request deadline includes gzip decoding. Responses whose method or status forbids content
//! preserve their encoding metadata. Decoded responses omit `Content-Encoding` and `Content-Length`,
//! even when selected through `header_keys`.
//!
//! Buffered responses reject wire bodies larger than 100 MiB, including chunked bodies without a
//! declared length. Decoded bodies have a separate 100 MiB default limit, configurable through
//! `max_decompressed_bytes`. [`HttpClient::get_stream`] returns raw body chunks without decoding,
//! adding `Accept-Encoding`, or applying a total size limit. Request URLs in transport errors and
//! response-stream diagnostics are limited to their origin (scheme, host, and port). Setting `redact_url(true)` on the
//! builder omits the origin too. This policy controls the URL context added by this client;
//! diagnostic messages from lower layers are unchanged. The client logs request metadata only, never URLs.
//!
//! Hyper owns the lifecycle of individual pooled connections, so this client exposes no socket
//! state sink or explicit reconnect operation. Callers observe connection failure through each
//! request result and retain the client to preserve its pool.

pub mod client;
pub mod error;
pub mod types;

#[cfg(not(all(feature = "simulation", madsim)))]
mod connector;
#[cfg(all(feature = "simulation", madsim))]
mod simulation;
#[cfg(not(all(feature = "simulation", madsim)))]
mod transport;

mod stream;

#[cfg(all(test, not(all(feature = "simulation", madsim))))]
mod tests;

// Re-exports
pub use client::{HttpClient, HttpRedirectPolicy, InnerHttpClient};
pub use error::HttpClientError;
pub use http::{
    Method, StatusCode,
    header::{RETRY_AFTER, USER_AGENT},
};
use nautilus_core::consts::NAUTILUS_USER_AGENT;
pub use stream::HttpResponseStream;
pub use types::{HttpMethod, HttpResponse, HttpStatus, parse_retry_after};
pub use url::Url;

/// Returns the standard headers every NautilusTrader HTTP and WebSocket client sends.
///
/// Carries the NautilusTrader [`USER_AGENT`] with the compile-time version.
#[must_use]
pub fn create_standard_nautilus_headers() -> Vec<(String, String)> {
    vec![(USER_AGENT.to_string(), NAUTILUS_USER_AGENT.to_string())]
}