fix: decompress body before forward and support zstd (#3817)

* fix(proxy): decompress Codex request body before forward, support zstd

Codex Desktop sends zstd-compressed request bodies when authenticated
against the Codex backend, which broke local proxy routing because the
handlers parsed the raw bytes with serde_json directly.

Reworked on top of current main so it preserves the response_processor
behavior that landed after this PR was first opened:

- Extract content-encoding helpers into a shared proxy::content_encoding
  module. decompress_body keeps returning Option<Vec<u8>> so unknown
  encodings stay pass-through with their content-encoding header intact,
  and keeps the deflate zlib-then-raw fallback (RFC 9110).
- Add zstd/zst support (zstd 0.13) and disable reqwest's auto zstd
  decompression via .no_zstd() for parity with gzip/br/deflate.
- Decompress the request body before JSON parsing in the three Codex
  handlers (chat_completions / responses / responses_compact) and strip
  the stale content-encoding / content-length / transfer-encoding headers
  so the forwarder regenerates them.
- Support stacked codings (e.g. "gzip, zstd") by decoding in reverse
  order and merge repeated Content-Encoding headers via get_all.

Fixes #3764
Fixes #3696

Co-authored-by: chenx-dust <16610294+chenx-dust@users.noreply.github.com>

* fix(proxy): decompress upstream error bodies before reading them

The forwarder error branch consumes non-2xx responses via String::from_utf8
directly, bypassing read_decoded_body. reqwest has no auto-decompression
feature enabled, so a compressed error body (gzip/br/deflate/zstd) arrives
as raw bytes, fails from_utf8, and gets dropped, hiding upstream rate-limit
and auth details from the client.

Decode the error body with the shared proxy::content_encoding helper,
mirroring the success path. Falls back to the raw bytes when the encoding is
unsupported or decoding fails.

Co-authored-by: chenx-dust <16610294+chenx-dust@users.noreply.github.com>

---------

Co-authored-by: Jason <farion1231@gmail.com>
Co-authored-by: chenx-dust <16610294+chenx-dust@users.noreply.github.com>
This commit is contained in:
Chenx Dust
2026-06-26 08:50:59 +08:00
committed by GitHub
parent 524b9d9825
commit 1a0e8c7a44
8 changed files with 306 additions and 92 deletions
+2 -87
View File
@@ -3,6 +3,7 @@
//! 统一处理流式和非流式 API 响应
use super::{
content_encoding::{decompress_body, get_content_encoding},
forwarder::ActiveConnectionGuard,
handler_config::{StreamUsageEventFilter, UsageParserConfig},
handler_context::{RequestContext, StreamingTimeoutConfig},
@@ -19,7 +20,6 @@ use bytes::Bytes;
use futures::stream::{Stream, StreamExt};
use serde_json::Value;
use std::{
io::Read,
sync::{
atomic::{AtomicBool, Ordering},
Arc,
@@ -29,60 +29,9 @@ use std::{
use tokio::sync::Mutex;
// ============================================================================
// 响应解压
// 响应头处理
// ============================================================================
/// 根据 content-encoding 解压响应体字节
///
/// reqwest 自动解压已禁用(为了透传 accept-encoding),需要手动解压。
/// 返回 `Ok(None)` 表示编码不受支持、原样透传——此时调用方必须保留
/// content-encoding 头,否则下游(诊断/客户端)会把压缩字节误当明文。
fn decompress_body(content_encoding: &str, body: &[u8]) -> Result<Option<Vec<u8>>, std::io::Error> {
match content_encoding {
"gzip" | "x-gzip" => {
let mut decoder = flate2::read::GzDecoder::new(body);
let mut decompressed = Vec::new();
decoder.read_to_end(&mut decompressed)?;
Ok(Some(decompressed))
}
"deflate" => {
// RFC 9110: deflate 指 zlib 包裹格式;但部分上游发 raw deflate 流。
// 先按规范尝试 zlib,失败再回退 raw —— 否则合规上游必然解压失败,
// 原始压缩字节会被 fail-open 透传给 JSON 解析(#2234 形态 C 之一)。
let mut decompressed = Vec::new();
let mut zlib = flate2::read::ZlibDecoder::new(body);
match zlib.read_to_end(&mut decompressed) {
Ok(_) => Ok(Some(decompressed)),
Err(zlib_err) => {
log::debug!("deflate 按 zlib 解压失败({zlib_err}),回退 raw deflate");
let mut decompressed = Vec::new();
let mut raw = flate2::read::DeflateDecoder::new(body);
raw.read_to_end(&mut decompressed)?;
Ok(Some(decompressed))
}
}
}
"br" => {
let mut decompressed = Vec::new();
brotli::BrotliDecompress(&mut std::io::Cursor::new(body), &mut decompressed)?;
Ok(Some(decompressed))
}
_ => {
log::warn!("未知的 content-encoding: {content_encoding},跳过解压");
Ok(None)
}
}
}
/// 从响应头提取 content-encoding(忽略 identity 和 chunked
fn get_content_encoding(headers: &HeaderMap) -> Option<String> {
headers
.get("content-encoding")
.and_then(|v| v.to_str().ok())
.map(|s| s.trim().to_lowercase())
.filter(|s| !s.is_empty() && s != "identity")
}
/// RFC 2616 / RFC 7230 中定义的不应被代理继续转发的响应头。
const HOP_BY_HOP_RESPONSE_HEADERS: &[&str] = &[
"connection",
@@ -878,40 +827,6 @@ mod tests {
use std::sync::Arc;
use tokio::sync::RwLock;
#[test]
fn decompress_body_deflate_handles_zlib_wrapped_per_rfc9110() {
// RFC 9110 规范的 deflate = zlib 包裹格式(合规上游发的就是这个)
let payload = br#"{"ok":true}"#;
let mut encoder =
flate2::write::ZlibEncoder::new(Vec::new(), flate2::Compression::default());
std::io::Write::write_all(&mut encoder, payload).unwrap();
let compressed = encoder.finish().unwrap();
let decompressed = decompress_body("deflate", &compressed).unwrap().unwrap();
assert_eq!(decompressed, payload);
}
#[test]
fn decompress_body_deflate_falls_back_to_raw_stream() {
// 部分上游违规发 raw deflate 流,保持兼容
let payload = br#"{"ok":true}"#;
let mut encoder =
flate2::write::DeflateEncoder::new(Vec::new(), flate2::Compression::default());
std::io::Write::write_all(&mut encoder, payload).unwrap();
let compressed = encoder.finish().unwrap();
let decompressed = decompress_body("deflate", &compressed).unwrap().unwrap();
assert_eq!(decompressed, payload);
}
#[test]
fn decompress_body_unknown_encoding_returns_none_to_keep_headers() {
// 未知编码必须返回 None(而非伪装成"已解码"),否则 content-encoding
// 头被剥掉,下游诊断会把压缩字节误报成明文
let result = decompress_body("zstd", b"\x28\xb5\x2f\xfd").unwrap();
assert!(result.is_none());
}
#[test]
fn test_strip_sse_field_accepts_optional_space() {
assert_eq!(