feat(gateway): harden provider request execution

Preserve exact request payloads and model client surface and API operation explicitly.

Add Anthropic compatibility profiles, bounded stream commitment, and scoped OAuth retry behavior across provider transports.
This commit is contained in:
elky
2026-07-27 09:36:31 +08:00
parent 79b70f7b5c
commit 531cf11025
152 changed files with 13984 additions and 2075 deletions
@@ -990,136 +990,6 @@ async fn gateway_handles_public_gemini_models_without_hitting_fallback_probe() {
fallback_probe_handle.abort();
}
#[tokio::test]
async fn gateway_handles_claude_count_tokens_without_hitting_fallback_probe() {
let fallback_probe_hits = Arc::new(Mutex::new(0usize));
let fallback_probe_hits_clone = Arc::clone(&fallback_probe_hits);
let fallback_probe = Router::new().route(
"/{*path}",
any(move |_request: Request| {
let fallback_probe_hits_inner = Arc::clone(&fallback_probe_hits_clone);
async move {
*fallback_probe_hits_inner.lock().expect("mutex should lock") += 1;
(StatusCode::OK, Json(json!({"proxied": true}))).into_response()
}
}),
);
let auth_repository = Arc::new(InMemoryAuthApiKeySnapshotRepository::seed(vec![(
Some(hash_api_key("sk-claude-count")),
unrestricted_models_snapshot("key-claude-count", "user-claude-count"),
)]));
let (_unused_fallback_probe_url, fallback_probe_handle) = start_server(fallback_probe).await;
let gateway = build_router_with_state(
AppState::new()
.expect("gateway should build")
.with_auth_api_key_data_reader_for_tests(auth_repository),
);
let (gateway_url, gateway_handle) = start_server(gateway).await;
let response = reqwest::Client::new()
.post(format!("{gateway_url}/v1/messages/count_tokens"))
.header("x-api-key", "sk-claude-count")
.header("anthropic-version", "2023-06-01")
.body(
serde_json::to_vec(&json!({
"model": "claude-sonnet-4-5",
"system": [{"type": "text", "text": "abcdefghijklmnop"}],
"messages": [
{
"role": "user",
"content": "abcdefghijkl"
},
{
"role": "assistant",
"content": [
{"type": "text", "text": "abcdefgh"},
{"type": "tool_use", "name": "ignored", "input": {"city": "SF"}}
]
}
]
}))
.expect("request body should encode"),
)
.send()
.await
.expect("request should succeed");
assert_eq!(response.status(), StatusCode::OK);
assert_eq!(
response
.headers()
.get(EXECUTION_PATH_HEADER)
.and_then(|value| value.to_str().ok()),
Some(EXECUTION_PATH_LOCAL_AI_PUBLIC)
);
let payload: serde_json::Value = response.json().await.expect("json body should parse");
assert_eq!(payload["input_tokens"], 17);
assert_eq!(*fallback_probe_hits.lock().expect("mutex should lock"), 0);
gateway_handle.abort();
fallback_probe_handle.abort();
}
#[tokio::test]
async fn gateway_rejects_invalid_claude_count_tokens_payload_without_hitting_fallback_probe() {
let fallback_probe_hits = Arc::new(Mutex::new(0usize));
let fallback_probe_hits_clone = Arc::clone(&fallback_probe_hits);
let fallback_probe = Router::new().route(
"/{*path}",
any(move |_request: Request| {
let fallback_probe_hits_inner = Arc::clone(&fallback_probe_hits_clone);
async move {
*fallback_probe_hits_inner.lock().expect("mutex should lock") += 1;
(StatusCode::OK, Json(json!({"proxied": true}))).into_response()
}
}),
);
let auth_repository = Arc::new(InMemoryAuthApiKeySnapshotRepository::seed(vec![(
Some(hash_api_key("sk-claude-count-invalid")),
unrestricted_models_snapshot("key-claude-count-invalid", "user-claude-count-invalid"),
)]));
let (_unused_fallback_probe_url, fallback_probe_handle) = start_server(fallback_probe).await;
let gateway = build_router_with_state(
AppState::new()
.expect("gateway should build")
.with_auth_api_key_data_reader_for_tests(auth_repository),
);
let (gateway_url, gateway_handle) = start_server(gateway).await;
let response = reqwest::Client::new()
.post(format!("{gateway_url}/v1/messages/count_tokens"))
.header("x-api-key", "sk-claude-count-invalid")
.body(
serde_json::to_vec(&json!({
"model": "claude-sonnet-4-5",
"messages": [{"role": "system", "content": "bad"}]
}))
.expect("request body should encode"),
)
.send()
.await
.expect("request should succeed");
assert_eq!(response.status(), StatusCode::BAD_REQUEST);
assert_eq!(
response
.headers()
.get(EXECUTION_PATH_HEADER)
.and_then(|value| value.to_str().ok()),
Some(EXECUTION_PATH_LOCAL_AI_PUBLIC)
);
let payload: serde_json::Value = response.json().await.expect("json body should parse");
assert_eq!(payload["detail"], "Invalid token count payload");
assert_eq!(*fallback_probe_hits.lock().expect("mutex should lock"), 0);
gateway_handle.abort();
fallback_probe_handle.abort();
}
#[tokio::test]
async fn gateway_handles_antigravity_v1internal_control_plane_without_proxying() {
let fallback_probe_hits = Arc::new(Mutex::new(0usize));