feat(gateway): add reversible chat pii redaction

This commit is contained in:
Kayphoon
2026-05-13 18:25:13 +08:00
parent 3c2497f019
commit 2958041dc7
43 changed files with 7770 additions and 126 deletions

View File

@@ -26,3 +26,4 @@ use super::{
mod decision;
mod image;
mod pii_redaction;

View File

@@ -0,0 +1,340 @@
use super::*;
use aether_crypto::{encrypt_python_fernet_plaintext, DEVELOPMENT_ENCRYPTION_KEY};
use aether_data::repository::auth::{
InMemoryAuthApiKeySnapshotRepository, StoredAuthApiKeySnapshot,
};
use aether_data::repository::candidate_selection::InMemoryMinimalCandidateSelectionReadRepository;
use aether_data::repository::candidates::InMemoryRequestCandidateRepository;
use aether_data::repository::provider_catalog::InMemoryProviderCatalogReadRepository;
use aether_data_contracts::repository::candidate_selection::{
StoredMinimalCandidateSelectionRow, StoredProviderModelMapping,
};
use aether_data_contracts::repository::candidates::{
RequestCandidateReadRepository, RequestCandidateStatus,
};
use aether_data_contracts::repository::provider_catalog::{
StoredProviderCatalogEndpoint, StoredProviderCatalogKey, StoredProviderCatalogProvider,
};
use sha2::{Digest, Sha256};
#[derive(Debug, Clone)]
struct SeenProviderStreamRequest {
body: serde_json::Value,
authorization: String,
accept_encoding: String,
accept: String,
}
fn hash_api_key(value: &str) -> String {
let mut hasher = Sha256::new();
hasher.update(value.as_bytes());
format!("{:x}", hasher.finalize())
}
fn auth_snapshot() -> StoredAuthApiKeySnapshot {
StoredAuthApiKeySnapshot::new(
"user-ai-execute-stream-pii-redaction".to_string(),
"alice".to_string(),
Some("alice@example.com".to_string()),
"user".to_string(),
"local".to_string(),
true,
false,
Some(serde_json::json!(["openai"])),
Some(serde_json::json!(["openai:chat"])),
Some(serde_json::json!(["gpt-5"])),
"api-key-ai-execute-stream-pii-redaction".to_string(),
Some("default".to_string()),
true,
false,
false,
Some(60),
Some(5),
Some(4_102_444_800),
Some(serde_json::json!(["openai"])),
Some(serde_json::json!(["openai:chat"])),
Some(serde_json::json!(["gpt-5"])),
)
.expect("auth snapshot should build")
}
fn candidate_row() -> StoredMinimalCandidateSelectionRow {
StoredMinimalCandidateSelectionRow {
provider_id: "provider-ai-execute-stream-pii-redaction".to_string(),
provider_name: "openai".to_string(),
provider_type: "custom".to_string(),
provider_priority: 10,
provider_is_active: true,
endpoint_id: "endpoint-ai-execute-stream-pii-redaction".to_string(),
endpoint_api_format: "openai:chat".to_string(),
endpoint_api_family: Some("openai".to_string()),
endpoint_kind: Some("chat".to_string()),
endpoint_is_active: true,
key_id: "key-ai-execute-stream-pii-redaction".to_string(),
key_name: "prod".to_string(),
key_auth_type: "api_key".to_string(),
key_is_active: true,
key_api_formats: Some(vec!["openai:chat".to_string()]),
key_allowed_models: None,
key_capabilities: None,
key_internal_priority: 5,
key_global_priority_by_format: Some(serde_json::json!({"openai:chat": 1})),
model_id: "model-ai-execute-stream-pii-redaction".to_string(),
global_model_id: "global-model-ai-execute-stream-pii-redaction".to_string(),
global_model_name: "gpt-5".to_string(),
global_model_mappings: None,
global_model_supports_streaming: Some(true),
model_provider_model_name: "gpt-5-upstream".to_string(),
model_provider_model_mappings: Some(vec![StoredProviderModelMapping {
name: "gpt-5-upstream".to_string(),
priority: 1,
api_formats: Some(vec!["openai:chat".to_string()]),
endpoint_ids: Some(vec!["endpoint-ai-execute-stream-pii-redaction".to_string()]),
}]),
model_supports_streaming: Some(true),
model_is_active: true,
model_is_available: true,
}
}
fn provider() -> StoredProviderCatalogProvider {
StoredProviderCatalogProvider::new(
"provider-ai-execute-stream-pii-redaction".to_string(),
"openai".to_string(),
Some("https://example.com".to_string()),
"custom".to_string(),
)
.expect("provider should build")
.with_transport_fields(
true,
false,
true,
None,
Some(2),
None,
Some(20.0),
None,
Some(serde_json::json!({"chat_pii_redaction": {"enabled": true}})),
)
}
fn endpoint(base_url: String) -> StoredProviderCatalogEndpoint {
StoredProviderCatalogEndpoint::new(
"endpoint-ai-execute-stream-pii-redaction".to_string(),
"provider-ai-execute-stream-pii-redaction".to_string(),
"openai:chat".to_string(),
Some("openai".to_string()),
Some("chat".to_string()),
true,
)
.expect("endpoint should build")
.with_transport_fields(base_url, None, None, Some(2), None, None, None, None)
.expect("endpoint transport should build")
}
fn key() -> StoredProviderCatalogKey {
StoredProviderCatalogKey::new(
"key-ai-execute-stream-pii-redaction".to_string(),
"provider-ai-execute-stream-pii-redaction".to_string(),
"prod".to_string(),
"api_key".to_string(),
None,
true,
)
.expect("key should build")
.with_transport_fields(
Some(serde_json::json!(["openai:chat"])),
encrypt_python_fernet_plaintext(DEVELOPMENT_ENCRYPTION_KEY, "sk-upstream-stream-pii")
.expect("api key should encrypt"),
None,
None,
Some(serde_json::json!({"openai:chat": 1})),
None,
None,
None,
None,
)
.expect("key transport should build")
}
fn collect_email_sentinel(text: &str) -> String {
let start = text
.find("<AETHER:EMAIL:")
.expect("email sentinel should exist");
let end = text[start..]
.find('>')
.map(|index| start + index + 1)
.expect("sentinel should close");
text[start..end].to_string()
}
#[tokio::test]
async fn ai_execute_stream_pii_redaction_round_trip() {
let seen_provider_request = Arc::new(Mutex::new(None::<SeenProviderStreamRequest>));
let seen_provider_request_clone = Arc::clone(&seen_provider_request);
let provider_app = Router::new().route(
"/v1/chat/completions",
any(move |request: Request| {
let seen_provider_request_inner = Arc::clone(&seen_provider_request_clone);
async move {
let (parts, body) = request.into_parts();
let raw_body = to_bytes(body, usize::MAX).await.expect("body should read");
let payload: serde_json::Value =
serde_json::from_slice(&raw_body).expect("provider payload should parse");
let payload_text = serde_json::to_string(&payload).expect("payload should serialize");
let sentinel = collect_email_sentinel(&payload_text);
*seen_provider_request_inner
.lock()
.expect("mutex should lock") = Some(SeenProviderStreamRequest {
body: payload,
authorization: parts
.headers
.get(http::header::AUTHORIZATION)
.and_then(|value| value.to_str().ok())
.unwrap_or_default()
.to_string(),
accept_encoding: parts
.headers
.get(http::header::ACCEPT_ENCODING)
.and_then(|value| value.to_str().ok())
.unwrap_or_default()
.to_string(),
accept: parts
.headers
.get(http::header::ACCEPT)
.and_then(|value| value.to_str().ok())
.unwrap_or_default()
.to_string(),
});
let split_at = sentinel.len() / 2;
let (first_half, second_half) = sentinel.split_at(split_at);
let first_chunk = format!(
"data: {{\"id\":\"chatcmpl-stream-pii\",\"object\":\"chat.completion.chunk\",\"model\":\"gpt-5-upstream\",\"choices\":[{{\"index\":0,\"delta\":{{\"role\":\"assistant\",\"content\":\"stream {first_half}"
);
let second_chunk = format!(
"{second_half} restored\"}},\"finish_reason\":null}}]}}\n\n"
);
let stream = futures_util::stream::iter([
Ok::<_, Infallible>(Bytes::from(first_chunk)),
Ok::<_, Infallible>(Bytes::from(second_chunk)),
Ok::<_, Infallible>(Bytes::from_static(b"data: [DONE]\n\n")),
]);
let mut response = Response::builder()
.status(StatusCode::OK)
.body(Body::from_stream(stream))
.expect("response should build");
response.headers_mut().insert(
http::header::CONTENT_TYPE,
HeaderValue::from_static("text/event-stream"),
);
response
}
}),
);
let (provider_url, provider_handle) = start_server(provider_app).await;
let auth_repository = Arc::new(InMemoryAuthApiKeySnapshotRepository::seed(vec![(
Some(hash_api_key("sk-client-ai-execute-stream-pii-redaction")),
auth_snapshot(),
)]));
let candidate_selection_repository =
Arc::new(InMemoryMinimalCandidateSelectionReadRepository::seed(vec![
candidate_row(),
]));
let request_candidate_repository = Arc::new(InMemoryRequestCandidateRepository::default());
let provider_catalog_repository = Arc::new(InMemoryProviderCatalogReadRepository::seed(
vec![provider()],
vec![endpoint(provider_url)],
vec![key()],
));
let data_state = crate::data::GatewayDataState::with_auth_candidate_selection_provider_catalog_and_request_candidate_repository_for_tests(
auth_repository,
candidate_selection_repository,
provider_catalog_repository,
Arc::clone(&request_candidate_repository),
DEVELOPMENT_ENCRYPTION_KEY,
)
.with_system_config_values_for_tests(vec![
("module.chat_pii_redaction.enabled".to_string(), json!(true)),
(
"module.chat_pii_redaction.provider_scope".to_string(),
json!("selected_providers"),
),
(
"module.chat_pii_redaction.entities".to_string(),
json!(["email"]),
),
(
"module.chat_pii_redaction.cache_ttl_seconds".to_string(),
json!(300),
),
(
"module.chat_pii_redaction.inject_model_instruction".to_string(),
json!(true),
),
]);
let gateway_state = AppState::new()
.expect("gateway state should build")
.with_data_state_for_tests(data_state);
let gateway = build_router_with_state(gateway_state);
let (gateway_url, gateway_handle) = start_server(gateway).await;
let response = reqwest::Client::new()
.post(format!("{gateway_url}/v1/chat/completions"))
.header(http::header::CONTENT_TYPE, "application/json")
.header(
http::header::AUTHORIZATION,
"Bearer sk-client-ai-execute-stream-pii-redaction",
)
.header(http::header::ACCEPT_ENCODING, "gzip")
.header(TRACE_ID_HEADER, "trace-ai-execute-stream-pii-redaction")
.body(
json!({
"model": "gpt-5",
"messages": [{"role": "user", "content": "Email stream.user@example.com"}],
"stream": true
})
.to_string(),
)
.send()
.await
.expect("request should succeed");
let status = response.status();
let execution_path = response
.headers()
.get(EXECUTION_PATH_HEADER)
.and_then(|value| value.to_str().ok())
.map(ToOwned::to_owned);
let response_text = response.text().await.expect("body should read");
assert_eq!(status, StatusCode::OK, "{response_text}");
assert_eq!(
execution_path.as_deref(),
Some(EXECUTION_PATH_EXECUTION_RUNTIME_STREAM)
);
assert!(response_text.contains("stream stream.user@example.com restored"));
assert!(response_text.contains("data: [DONE]"));
assert!(!response_text.contains("<AETHER:EMAIL:"));
let seen = seen_provider_request
.lock()
.expect("mutex should lock")
.clone()
.expect("provider stream request should be captured");
assert_eq!(seen.authorization, "Bearer sk-upstream-stream-pii");
assert_eq!(seen.accept, "text/event-stream");
assert_eq!(seen.accept_encoding, "identity");
let provider_body_text = serde_json::to_string(&seen.body).expect("body should serialize");
assert!(!provider_body_text.contains("stream.user@example.com"));
assert!(provider_body_text.contains("<AETHER:EMAIL:"));
let stored_candidates = request_candidate_repository
.list_by_request_id("trace-ai-execute-stream-pii-redaction")
.await
.expect("request candidate trace should read");
assert_eq!(stored_candidates.len(), 1);
assert_eq!(stored_candidates[0].status, RequestCandidateStatus::Success);
gateway_handle.abort();
provider_handle.abort();
}

View File

@@ -10,6 +10,315 @@ use super::{
TRACE_ID_HEADER,
};
#[tokio::test]
async fn proxy_pii_redaction_local_openai_chat_runtime_masks_headers_and_restores_sync_response() {
#[derive(Debug, Clone)]
struct SeenProviderRequest {
body: serde_json::Value,
authorization: String,
accept_encoding: String,
}
fn hash_api_key(value: &str) -> String {
let mut hasher = Sha256::new();
hasher.update(value.as_bytes());
format!("{:x}", hasher.finalize())
}
fn auth_snapshot() -> StoredAuthApiKeySnapshot {
StoredAuthApiKeySnapshot::new(
"user-redaction-1".to_string(),
"alice".to_string(),
Some("alice@example.com".to_string()),
"user".to_string(),
"local".to_string(),
true,
false,
Some(serde_json::json!(["openai"])),
Some(serde_json::json!(["openai:chat"])),
Some(serde_json::json!(["gpt-5"])),
"api-key-redaction-1".to_string(),
Some("default".to_string()),
true,
false,
false,
Some(60),
Some(5),
Some(4_102_444_800),
Some(serde_json::json!(["openai"])),
Some(serde_json::json!(["openai:chat"])),
Some(serde_json::json!(["gpt-5"])),
)
.expect("auth snapshot should build")
}
fn candidate_row() -> StoredMinimalCandidateSelectionRow {
StoredMinimalCandidateSelectionRow {
provider_id: "provider-redaction-1".to_string(),
provider_name: "openai".to_string(),
provider_type: "custom".to_string(),
provider_priority: 10,
provider_is_active: true,
endpoint_id: "endpoint-redaction-1".to_string(),
endpoint_api_format: "openai:chat".to_string(),
endpoint_api_family: Some("openai".to_string()),
endpoint_kind: Some("chat".to_string()),
endpoint_is_active: true,
key_id: "key-redaction-1".to_string(),
key_name: "prod".to_string(),
key_auth_type: "api_key".to_string(),
key_is_active: true,
key_api_formats: Some(vec!["openai:chat".to_string()]),
key_allowed_models: None,
key_capabilities: None,
key_internal_priority: 5,
key_global_priority_by_format: Some(serde_json::json!({"openai:chat": 1})),
model_id: "model-redaction-1".to_string(),
global_model_id: "global-model-redaction-1".to_string(),
global_model_name: "gpt-5".to_string(),
global_model_mappings: None,
global_model_supports_streaming: Some(true),
model_provider_model_name: "gpt-5-upstream".to_string(),
model_provider_model_mappings: Some(vec![StoredProviderModelMapping {
name: "gpt-5-upstream".to_string(),
priority: 1,
api_formats: Some(vec!["openai:chat".to_string()]),
endpoint_ids: Some(vec!["endpoint-redaction-1".to_string()]),
}]),
model_supports_streaming: Some(true),
model_is_active: true,
model_is_available: true,
}
}
fn provider() -> StoredProviderCatalogProvider {
StoredProviderCatalogProvider::new(
"provider-redaction-1".to_string(),
"openai".to_string(),
Some("https://example.com".to_string()),
"custom".to_string(),
)
.expect("provider should build")
.with_transport_fields(
true,
false,
true,
None,
Some(2),
None,
Some(20.0),
None,
Some(serde_json::json!({"chat_pii_redaction": {"enabled": true}})),
)
}
fn endpoint(base_url: String) -> StoredProviderCatalogEndpoint {
StoredProviderCatalogEndpoint::new(
"endpoint-redaction-1".to_string(),
"provider-redaction-1".to_string(),
"openai:chat".to_string(),
Some("openai".to_string()),
Some("chat".to_string()),
true,
)
.expect("endpoint should build")
.with_transport_fields(base_url, None, None, Some(2), None, None, None, None)
.expect("endpoint transport should build")
}
fn key() -> StoredProviderCatalogKey {
StoredProviderCatalogKey::new(
"key-redaction-1".to_string(),
"provider-redaction-1".to_string(),
"prod".to_string(),
"api_key".to_string(),
None,
true,
)
.expect("key should build")
.with_transport_fields(
Some(serde_json::json!(["openai:chat"])),
encrypt_python_fernet_plaintext(DEVELOPMENT_ENCRYPTION_KEY, "sk-upstream-redaction")
.expect("api key should encrypt"),
None,
None,
Some(serde_json::json!({"openai:chat": 1})),
None,
None,
None,
None,
)
.expect("key transport should build")
}
fn sentinel_from_body(body: &serde_json::Value) -> String {
let messages = body["messages"].as_array().expect("messages should exist");
let user_content = messages
.iter()
.find(|message| message["role"] == "user")
.and_then(|message| message["content"].as_str())
.expect("user content should be text");
let start = user_content
.find("<AETHER:EMAIL:")
.expect("redacted content should include email sentinel");
let end = user_content[start..]
.find('>')
.map(|index| start + index + 1)
.expect("sentinel should close");
user_content[start..end].to_string()
}
let seen_provider_request = Arc::new(Mutex::new(None::<SeenProviderRequest>));
let seen_provider_request_clone = Arc::clone(&seen_provider_request);
let provider_app = Router::new().route(
"/v1/chat/completions",
any(move |request: Request| {
let seen_provider_request_inner = Arc::clone(&seen_provider_request_clone);
async move {
let (parts, body) = request.into_parts();
let raw_body = to_bytes(body, usize::MAX).await.expect("body should read");
let payload: serde_json::Value =
serde_json::from_slice(&raw_body).expect("provider payload should parse");
let sentinel = sentinel_from_body(&payload);
*seen_provider_request_inner
.lock()
.expect("mutex should lock") = Some(SeenProviderRequest {
body: payload,
authorization: parts
.headers
.get(http::header::AUTHORIZATION)
.and_then(|value| value.to_str().ok())
.unwrap_or_default()
.to_string(),
accept_encoding: parts
.headers
.get(http::header::ACCEPT_ENCODING)
.and_then(|value| value.to_str().ok())
.unwrap_or_default()
.to_string(),
});
Json(json!({
"id": "chatcmpl-redaction-1",
"object": "chat.completion",
"model": "gpt-5-upstream",
"choices": [{
"index": 0,
"message": {"role": "assistant", "content": format!("restored {sentinel}")},
"finish_reason": "stop"
}],
"usage": {"prompt_tokens": 2, "completion_tokens": 3, "total_tokens": 5}
}))
}
}),
);
let (provider_url, provider_handle) = start_server(provider_app).await;
let auth_repository = Arc::new(InMemoryAuthApiKeySnapshotRepository::seed(vec![(
Some(hash_api_key("sk-client-redaction")),
auth_snapshot(),
)]));
let candidate_selection_repository =
Arc::new(InMemoryMinimalCandidateSelectionReadRepository::seed(vec![
candidate_row(),
]));
let request_candidate_repository = Arc::new(InMemoryRequestCandidateRepository::default());
let provider_catalog_repository = Arc::new(InMemoryProviderCatalogReadRepository::seed(
vec![provider()],
vec![endpoint(provider_url)],
vec![key()],
));
let data_state = crate::data::GatewayDataState::with_auth_candidate_selection_provider_catalog_and_request_candidate_repository_for_tests(
auth_repository,
candidate_selection_repository,
provider_catalog_repository,
Arc::clone(&request_candidate_repository),
DEVELOPMENT_ENCRYPTION_KEY,
)
.with_system_config_values_for_tests(vec![
("module.chat_pii_redaction.enabled".to_string(), json!(true)),
(
"module.chat_pii_redaction.provider_scope".to_string(),
json!("selected_providers"),
),
(
"module.chat_pii_redaction.entities".to_string(),
json!(["email"]),
),
(
"module.chat_pii_redaction.cache_ttl_seconds".to_string(),
json!(300),
),
(
"module.chat_pii_redaction.inject_model_instruction".to_string(),
json!(true),
),
]);
let gateway_state = AppState::new()
.expect("gateway state should build")
.with_data_state_for_tests(data_state);
let gateway = build_router_with_state(gateway_state);
let (gateway_url, gateway_handle) = start_server(gateway).await;
let response = reqwest::Client::new()
.post(format!("{gateway_url}/v1/chat/completions"))
.header(http::header::CONTENT_TYPE, "application/json")
.header(http::header::AUTHORIZATION, "Bearer sk-client-redaction")
.header(http::header::ACCEPT_ENCODING, "gzip")
.header(TRACE_ID_HEADER, "trace-proxy-pii-redaction-sync")
.body(
r#"{"model":"gpt-5","messages":[{"role":"user","content":"Email alice@example.com"}]}"#,
)
.send()
.await
.expect("request should succeed");
let status = response.status();
let execution_path = response
.headers()
.get(EXECUTION_PATH_HEADER)
.and_then(|value| value.to_str().ok())
.map(ToOwned::to_owned);
let response_text = response.text().await.expect("body should read");
assert_eq!(status, StatusCode::OK, "{response_text}");
assert_eq!(
execution_path.as_deref(),
Some(EXECUTION_PATH_EXECUTION_RUNTIME_SYNC)
);
let response_json: serde_json::Value =
serde_json::from_str(&response_text).expect("body should parse");
assert_eq!(
response_json["choices"][0]["message"]["content"],
"restored alice@example.com"
);
let seen = seen_provider_request
.lock()
.expect("mutex should lock")
.clone()
.expect("provider request should be captured");
assert_eq!(seen.authorization, "Bearer sk-upstream-redaction");
assert_eq!(seen.accept_encoding, "identity");
let provider_body_text = serde_json::to_string(&seen.body).expect("body should serialize");
assert!(!provider_body_text.contains("alice@example.com"));
assert!(provider_body_text.contains("<AETHER:EMAIL:"));
assert_eq!(seen.body["messages"][0]["role"], "assistant");
let notice = seen.body["messages"][0]["content"]
.as_str()
.expect("notice should be text");
assert!(notice.contains("not a user request"));
assert!(notice.contains("do not answer"));
assert_eq!(seen.body["messages"][1]["role"], "user");
let stored_candidates = request_candidate_repository
.list_by_request_id("trace-proxy-pii-redaction-sync")
.await
.expect("request candidate trace should read");
assert_eq!(stored_candidates.len(), 1);
assert_eq!(stored_candidates[0].status, RequestCandidateStatus::Success);
gateway_handle.abort();
provider_handle.abort();
}
#[tokio::test]
async fn gateway_executes_openai_chat_sync_via_local_decision_gate_without_execution_runtime_override(
) {

View File

@@ -42,3 +42,4 @@ use sha2::{Digest, Sha256};
mod failover;
mod local_decision;
mod pii_redaction;

View File

@@ -0,0 +1,836 @@
use super::*;
use std::sync::atomic::{AtomicUsize, Ordering};
#[derive(Debug, Clone)]
struct SeenProviderRequest {
body: serde_json::Value,
authorization: String,
accept_encoding: String,
}
fn hash_api_key(value: &str) -> String {
let mut hasher = Sha256::new();
hasher.update(value.as_bytes());
format!("{:x}", hasher.finalize())
}
fn auth_snapshot(api_key_id: &str, user_id: &str) -> StoredAuthApiKeySnapshot {
StoredAuthApiKeySnapshot::new(
user_id.to_string(),
"alice".to_string(),
Some("alice@example.com".to_string()),
"user".to_string(),
"local".to_string(),
true,
false,
Some(serde_json::json!(["openai"])),
Some(serde_json::json!(["openai:chat"])),
Some(serde_json::json!(["gpt-5"])),
api_key_id.to_string(),
Some("default".to_string()),
true,
false,
false,
Some(60),
Some(5),
Some(4_102_444_800),
Some(serde_json::json!(["openai"])),
Some(serde_json::json!(["openai:chat"])),
Some(serde_json::json!(["gpt-5"])),
)
.expect("auth snapshot should build")
}
fn candidate_row(test_id: &str) -> StoredMinimalCandidateSelectionRow {
StoredMinimalCandidateSelectionRow {
provider_id: format!("provider-{test_id}"),
provider_name: "openai".to_string(),
provider_type: "custom".to_string(),
provider_priority: 10,
provider_is_active: true,
endpoint_id: format!("endpoint-{test_id}"),
endpoint_api_format: "openai:chat".to_string(),
endpoint_api_family: Some("openai".to_string()),
endpoint_kind: Some("chat".to_string()),
endpoint_is_active: true,
key_id: format!("key-{test_id}"),
key_name: "prod".to_string(),
key_auth_type: "api_key".to_string(),
key_is_active: true,
key_api_formats: Some(vec!["openai:chat".to_string()]),
key_allowed_models: None,
key_capabilities: None,
key_internal_priority: 5,
key_global_priority_by_format: Some(serde_json::json!({"openai:chat": 1})),
model_id: format!("model-{test_id}"),
global_model_id: format!("global-model-{test_id}"),
global_model_name: "gpt-5".to_string(),
global_model_mappings: None,
global_model_supports_streaming: Some(true),
model_provider_model_name: "gpt-5-upstream".to_string(),
model_provider_model_mappings: Some(vec![StoredProviderModelMapping {
name: "gpt-5-upstream".to_string(),
priority: 1,
api_formats: Some(vec!["openai:chat".to_string()]),
endpoint_ids: Some(vec![format!("endpoint-{test_id}")]),
}]),
model_supports_streaming: Some(true),
model_is_active: true,
model_is_available: true,
}
}
fn provider(test_id: &str, redaction_enabled: bool) -> StoredProviderCatalogProvider {
StoredProviderCatalogProvider::new(
format!("provider-{test_id}"),
"openai".to_string(),
Some("https://example.com".to_string()),
"custom".to_string(),
)
.expect("provider should build")
.with_transport_fields(
true,
false,
true,
None,
Some(2),
None,
Some(20.0),
None,
Some(serde_json::json!({"chat_pii_redaction": {"enabled": redaction_enabled}})),
)
}
fn endpoint(test_id: &str, base_url: String) -> StoredProviderCatalogEndpoint {
StoredProviderCatalogEndpoint::new(
format!("endpoint-{test_id}"),
format!("provider-{test_id}"),
"openai:chat".to_string(),
Some("openai".to_string()),
Some("chat".to_string()),
true,
)
.expect("endpoint should build")
.with_transport_fields(base_url, None, None, Some(2), None, None, None, None)
.expect("endpoint transport should build")
}
fn key(test_id: &str) -> StoredProviderCatalogKey {
StoredProviderCatalogKey::new(
format!("key-{test_id}"),
format!("provider-{test_id}"),
"prod".to_string(),
"api_key".to_string(),
None,
true,
)
.expect("key should build")
.with_transport_fields(
Some(serde_json::json!(["openai:chat"])),
encrypt_python_fernet_plaintext(DEVELOPMENT_ENCRYPTION_KEY, "sk-upstream-pii-redaction")
.expect("api key should encrypt"),
None,
None,
Some(serde_json::json!({"openai:chat": 1})),
None,
None,
None,
None,
)
.expect("key transport should build")
}
fn redaction_config(module_enabled: bool) -> Vec<(String, serde_json::Value)> {
redaction_config_with_entities(
module_enabled,
json!(["email", "access_token", "secret_key"]),
)
}
fn redaction_config_with_entities(
module_enabled: bool,
entities: serde_json::Value,
) -> Vec<(String, serde_json::Value)> {
vec![
(
"module.chat_pii_redaction.enabled".to_string(),
json!(module_enabled),
),
(
"module.chat_pii_redaction.provider_scope".to_string(),
json!("selected_providers"),
),
("module.chat_pii_redaction.entities".to_string(), entities),
(
"module.chat_pii_redaction.cache_ttl_seconds".to_string(),
json!(300),
),
(
"module.chat_pii_redaction.inject_model_instruction".to_string(),
json!(true),
),
]
}
fn collect_sentinels(text: &str, kind: &str) -> Vec<String> {
let prefix = format!("<AETHER:{kind}:");
let mut sentinels = Vec::new();
let mut offset = 0;
while let Some(relative_start) = text[offset..].find(&prefix) {
let start = offset + relative_start;
let Some(relative_end) = text[start..].find('>') else {
break;
};
let end = start + relative_end + 1;
sentinels.push(text[start..end].to_string());
offset = end;
}
sentinels
}
async fn run_sync_redaction_case(
test_id: &str,
module_enabled: bool,
provider_enabled: bool,
provider_response: &'static str,
request_body: serde_json::Value,
) -> (serde_json::Value, SeenProviderRequest) {
run_sync_redaction_case_with_system_config(
test_id,
provider_enabled,
provider_response,
request_body,
redaction_config(module_enabled),
)
.await
}
async fn run_sync_redaction_case_with_system_config(
test_id: &str,
provider_enabled: bool,
provider_response: &'static str,
request_body: serde_json::Value,
system_config: Vec<(String, serde_json::Value)>,
) -> (serde_json::Value, SeenProviderRequest) {
let seen_provider_request = Arc::new(Mutex::new(None::<SeenProviderRequest>));
let seen_provider_request_clone = Arc::clone(&seen_provider_request);
let provider_app = Router::new().route(
"/v1/chat/completions",
any(move |request: Request| {
let seen_provider_request_inner = Arc::clone(&seen_provider_request_clone);
async move {
let (parts, body) = request.into_parts();
let raw_body = to_bytes(body, usize::MAX).await.expect("body should read");
let payload: serde_json::Value =
serde_json::from_slice(&raw_body).expect("provider payload should parse");
let payload_text =
serde_json::to_string(&payload).expect("payload should serialize");
let email_sentinels = collect_sentinels(&payload_text, "EMAIL");
let access_token_sentinels = collect_sentinels(&payload_text, "ACCESS_TOKEN");
let secret_key_sentinels = collect_sentinels(&payload_text, "SECRET_KEY");
*seen_provider_request_inner
.lock()
.expect("mutex should lock") = Some(SeenProviderRequest {
body: payload,
authorization: parts
.headers
.get(http::header::AUTHORIZATION)
.and_then(|value| value.to_str().ok())
.unwrap_or_default()
.to_string(),
accept_encoding: parts
.headers
.get(http::header::ACCEPT_ENCODING)
.and_then(|value| value.to_str().ok())
.unwrap_or_default()
.to_string(),
});
let content = match provider_response {
"known" => format!(
"restored {} {} {}",
email_sentinels
.first()
.expect("user email sentinel should exist"),
access_token_sentinels
.first()
.expect("access token sentinel should exist"),
secret_key_sentinels
.first()
.expect("secret key sentinel should exist")
),
"unknown" => "unknown <AETHER:EMAIL:TSRQPONMLKJIHGFEDCBA>".to_string(),
"pass_through" => "pass alice@example.com".to_string(),
_ => unreachable!("provider response mode should be known"),
};
Json(json!({
"id": format!("chatcmpl-{provider_response}"),
"object": "chat.completion",
"model": "gpt-5-upstream",
"choices": [{
"index": 0,
"message": {"role": "assistant", "content": content},
"finish_reason": "stop"
}],
"usage": {"prompt_tokens": 2, "completion_tokens": 3, "total_tokens": 5}
}))
}
}),
);
let (provider_url, provider_handle) = start_server(provider_app).await;
let auth_repository = Arc::new(InMemoryAuthApiKeySnapshotRepository::seed(vec![(
Some(hash_api_key(&format!("sk-client-{test_id}"))),
auth_snapshot(&format!("api-key-{test_id}"), &format!("user-{test_id}")),
)]));
let candidate_selection_repository =
Arc::new(InMemoryMinimalCandidateSelectionReadRepository::seed(vec![
candidate_row(test_id),
]));
let request_candidate_repository = Arc::new(InMemoryRequestCandidateRepository::default());
let provider_catalog_repository = Arc::new(InMemoryProviderCatalogReadRepository::seed(
vec![provider(test_id, provider_enabled)],
vec![endpoint(test_id, provider_url)],
vec![key(test_id)],
));
let data_state = crate::data::GatewayDataState::with_auth_candidate_selection_provider_catalog_and_request_candidate_repository_for_tests(
auth_repository,
candidate_selection_repository,
provider_catalog_repository,
Arc::clone(&request_candidate_repository),
DEVELOPMENT_ENCRYPTION_KEY,
)
.with_system_config_values_for_tests(system_config);
let gateway_state = AppState::new()
.expect("gateway state should build")
.with_data_state_for_tests(data_state);
let gateway = build_router_with_state(gateway_state);
let (gateway_url, gateway_handle) = start_server(gateway).await;
let response = reqwest::Client::new()
.post(format!("{gateway_url}/v1/chat/completions"))
.header(http::header::CONTENT_TYPE, "application/json")
.header(
http::header::AUTHORIZATION,
format!("Bearer sk-client-{test_id}"),
)
.header(http::header::ACCEPT_ENCODING, "gzip")
.header(TRACE_ID_HEADER, format!("trace-{test_id}"))
.body(request_body.to_string())
.send()
.await
.expect("request should succeed");
let status = response.status();
let execution_path = response
.headers()
.get(EXECUTION_PATH_HEADER)
.and_then(|value| value.to_str().ok())
.map(ToOwned::to_owned);
let response_text = response.text().await.expect("body should read");
assert_eq!(status, StatusCode::OK, "{response_text}");
assert_eq!(
execution_path.as_deref(),
Some(EXECUTION_PATH_EXECUTION_RUNTIME_SYNC)
);
let response_json: serde_json::Value =
serde_json::from_str(&response_text).expect("response body should parse");
let stored_candidates = request_candidate_repository
.list_by_request_id(&format!("trace-{test_id}"))
.await
.expect("request candidate trace should read");
assert_eq!(stored_candidates.len(), 1);
assert_eq!(stored_candidates[0].status, RequestCandidateStatus::Success);
let seen = seen_provider_request
.lock()
.expect("mutex should lock")
.clone()
.expect("provider request should be captured");
gateway_handle.abort();
provider_handle.abort();
(response_json, seen)
}
fn rich_pii_request() -> serde_json::Value {
json!({
"model": "gpt-5",
"messages": [
{"role": "system", "content": "Be concise."},
{
"role": "user",
"content": [
{"type": "text", "text": "Contact alice@example.com now"},
{"type": "input_audio", "input_audio": {"data": "AAAA", "format": "wav"}}
]
},
{
"role": "assistant",
"content": "Preparing lookup.",
"tool_calls": [{
"id": "call_1",
"type": "function",
"function": {
"name": "lookup_contact",
"arguments": "{\"email\":\"bob@example.net\"}"
}
}]
},
{
"role": "tool",
"tool_call_id": "call_1",
"content": "Tool returned access_token=accessValueABCDEF1234567890abcdef secret_key=secretValueABCDEF1234567890abcdef"
}
],
"tools": [{
"type": "function",
"function": {
"name": "lookup_contact",
"parameters": {
"type": "object",
"properties": {"email": {"type": "string"}}
}
}
}]
})
}
#[tokio::test]
async fn ai_execute_sync_pii_redaction_round_trip() {
let (response_json, seen) = run_sync_redaction_case(
"ai-execute-sync-pii-redaction-round-trip",
true,
true,
"known",
rich_pii_request(),
)
.await;
assert_eq!(seen.authorization, "Bearer sk-upstream-pii-redaction");
assert_eq!(seen.accept_encoding, "identity");
let provider_body_text = serde_json::to_string(&seen.body).expect("body should serialize");
for original in [
"alice@example.com",
"bob@example.net",
"access_token=accessValueABCDEF1234567890abcdef",
"secret_key\\\":\\\"secretValueABCDEF1234567890abcdef",
] {
assert!(
!provider_body_text.contains(original),
"leaked {original} in {provider_body_text}"
);
}
assert!(provider_body_text.contains("<AETHER:EMAIL:"));
assert!(provider_body_text.contains("<AETHER:ACCESS_TOKEN:"));
assert!(provider_body_text.contains("<AETHER:SECRET_KEY:"));
assert_eq!(seen.body["messages"][0]["role"], "system");
assert_eq!(seen.body["messages"][1]["role"], "assistant");
let notice = seen.body["messages"][1]["content"]
.as_str()
.expect("notice should be text");
assert!(notice.contains("not a user request"));
assert_eq!(seen.body["messages"][2]["role"], "user");
assert_eq!(seen.body["messages"][3]["role"], "assistant");
assert_eq!(seen.body["messages"][4]["role"], "tool");
let response_content = response_json["choices"][0]["message"]["content"]
.as_str()
.expect("assistant content should be text");
assert!(response_content.contains("alice@example.com"));
assert!(response_content.contains("access_token=accessValueABCDEF1234567890abcdef"));
assert!(response_content.contains("secretValueABCDEF1234567890abcdef"));
assert!(!response_content.contains("<AETHER:"));
}
#[tokio::test]
async fn ai_execute_pii_redaction_disabled_module_passes_original_chat_through() {
let (response_json, seen) = run_sync_redaction_case(
"ai-execute-pii-redaction-disabled-module",
false,
true,
"pass_through",
rich_pii_request(),
)
.await;
let provider_body_text = serde_json::to_string(&seen.body).expect("body should serialize");
assert!(provider_body_text.contains("alice@example.com"));
assert!(provider_body_text.contains("bob@example.net"));
assert!(provider_body_text.contains("access_token=accessValueABCDEF1234567890abcdef"));
assert!(provider_body_text.contains("secretValueABCDEF1234567890abcdef"));
assert!(!provider_body_text.contains("<AETHER:"));
assert_eq!(
response_json["choices"][0]["message"]["content"],
"pass alice@example.com"
);
}
#[tokio::test]
async fn ai_execute_pii_redaction_disabled_provider_passes_original_chat_through() {
let (response_json, seen) = run_sync_redaction_case(
"ai-execute-pii-redaction-disabled-provider",
true,
false,
"pass_through",
rich_pii_request(),
)
.await;
let provider_body_text = serde_json::to_string(&seen.body).expect("body should serialize");
assert!(provider_body_text.contains("alice@example.com"));
assert!(provider_body_text.contains("bob@example.net"));
assert!(provider_body_text.contains("access_token=accessValueABCDEF1234567890abcdef"));
assert!(provider_body_text.contains("secretValueABCDEF1234567890abcdef"));
assert!(!provider_body_text.contains("<AETHER:"));
assert_eq!(
response_json["choices"][0]["message"]["content"],
"pass alice@example.com"
);
}
#[tokio::test]
async fn ai_execute_pii_redaction_empty_entities_passes_original_chat_through() {
let (response_json, seen) = run_sync_redaction_case_with_system_config(
"ai-execute-pii-redaction-empty-entities",
true,
"pass_through",
rich_pii_request(),
redaction_config_with_entities(true, json!([])),
)
.await;
let provider_body_text = serde_json::to_string(&seen.body).expect("body should serialize");
assert!(provider_body_text.contains("alice@example.com"));
assert!(provider_body_text.contains("bob@example.net"));
assert!(provider_body_text.contains("access_token=accessValueABCDEF1234567890abcdef"));
assert!(provider_body_text.contains("secretValueABCDEF1234567890abcdef"));
assert!(!provider_body_text.contains("<AETHER:"));
assert_eq!(
response_json["choices"][0]["message"]["content"],
"pass alice@example.com"
);
}
#[tokio::test]
async fn ai_execute_pii_redaction_unknown_sentinel_like_output_is_not_restored() {
let (response_json, seen) = run_sync_redaction_case(
"ai-execute-pii-redaction-unknown-sentinel",
true,
true,
"unknown",
rich_pii_request(),
)
.await;
let provider_body_text = serde_json::to_string(&seen.body).expect("body should serialize");
assert!(!provider_body_text.contains("alice@example.com"));
assert!(provider_body_text.contains("<AETHER:EMAIL:"));
assert_eq!(
response_json["choices"][0]["message"]["content"],
"unknown <AETHER:EMAIL:TSRQPONMLKJIHGFEDCBA>"
);
}
#[tokio::test]
async fn ai_execute_pii_redaction_restores_executed_candidate_session_after_later_candidate_planning(
) {
let seen_provider_request = Arc::new(Mutex::new(None::<SeenProviderRequest>));
let seen_provider_request_clone = Arc::clone(&seen_provider_request);
let provider_app = Router::new().route(
"/v1/chat/completions",
any(move |request: Request| {
let seen_provider_request_inner = Arc::clone(&seen_provider_request_clone);
async move {
let (parts, body) = request.into_parts();
let raw_body = to_bytes(body, usize::MAX).await.expect("body should read");
let payload: serde_json::Value =
serde_json::from_slice(&raw_body).expect("provider payload should parse");
let payload_text = serde_json::to_string(&payload).expect("payload should serialize");
let email_sentinel = collect_sentinels(&payload_text, "EMAIL")
.into_iter()
.next()
.expect("email sentinel should exist");
*seen_provider_request_inner.lock().expect("mutex should lock") = Some(
SeenProviderRequest {
body: payload,
authorization: parts
.headers
.get(http::header::AUTHORIZATION)
.and_then(|value| value.to_str().ok())
.unwrap_or_default()
.to_string(),
accept_encoding: parts
.headers
.get(http::header::ACCEPT_ENCODING)
.and_then(|value| value.to_str().ok())
.unwrap_or_default()
.to_string(),
},
);
Json(json!({
"id": "chatcmpl-redaction-candidate-session",
"object": "chat.completion",
"model": "gpt-5-upstream",
"choices": [{
"index": 0,
"message": {"role": "assistant", "content": format!("restored {email_sentinel}")},
"finish_reason": "stop"
}],
"usage": {"prompt_tokens": 2, "completion_tokens": 3, "total_tokens": 5}
}))
}
}),
);
let (provider_url, provider_handle) = start_server(provider_app).await;
let auth_repository = Arc::new(InMemoryAuthApiKeySnapshotRepository::seed(vec![(
Some(hash_api_key("sk-client-redaction-candidate-session")),
auth_snapshot(
"api-key-redaction-candidate-session",
"user-redaction-candidate-session",
),
)]));
let mut later_candidate = candidate_row("redaction-candidate-session");
later_candidate.provider_id = "provider-redaction-candidate-session-later".to_string();
later_candidate.endpoint_id = "endpoint-redaction-candidate-session-later".to_string();
later_candidate.key_id = "key-redaction-candidate-session-later".to_string();
later_candidate.provider_priority = 20;
later_candidate.key_internal_priority = 6;
later_candidate.model_id = "model-redaction-candidate-session-later".to_string();
later_candidate.global_model_id = "global-model-redaction-candidate-session-later".to_string();
let candidate_selection_repository =
Arc::new(InMemoryMinimalCandidateSelectionReadRepository::seed(vec![
candidate_row("redaction-candidate-session"),
later_candidate,
]));
let request_candidate_repository = Arc::new(InMemoryRequestCandidateRepository::default());
let mut later_provider = provider("redaction-candidate-session", false);
later_provider.id = "provider-redaction-candidate-session-later".to_string();
let mut later_endpoint = endpoint(
"redaction-candidate-session",
"http://127.0.0.1:9".to_string(),
);
later_endpoint.id = "endpoint-redaction-candidate-session-later".to_string();
later_endpoint.provider_id = "provider-redaction-candidate-session-later".to_string();
let mut later_key = key("redaction-candidate-session");
later_key.id = "key-redaction-candidate-session-later".to_string();
later_key.provider_id = "provider-redaction-candidate-session-later".to_string();
let provider_catalog_repository = Arc::new(InMemoryProviderCatalogReadRepository::seed(
vec![
provider("redaction-candidate-session", true),
later_provider,
],
vec![
endpoint("redaction-candidate-session", provider_url),
later_endpoint,
],
vec![key("redaction-candidate-session"), later_key],
));
let data_state = crate::data::GatewayDataState::with_auth_candidate_selection_provider_catalog_and_request_candidate_repository_for_tests(
auth_repository,
candidate_selection_repository,
provider_catalog_repository,
Arc::clone(&request_candidate_repository),
DEVELOPMENT_ENCRYPTION_KEY,
)
.with_system_config_values_for_tests(redaction_config(true));
let gateway_state = AppState::new()
.expect("gateway state should build")
.with_data_state_for_tests(data_state);
let gateway = build_router_with_state(gateway_state);
let (gateway_url, gateway_handle) = start_server(gateway).await;
let response = reqwest::Client::new()
.post(format!("{gateway_url}/v1/chat/completions"))
.header(http::header::CONTENT_TYPE, "application/json")
.header(
http::header::AUTHORIZATION,
"Bearer sk-client-redaction-candidate-session",
)
.header(TRACE_ID_HEADER, "trace-redaction-candidate-session")
.body(
json!({
"model": "gpt-5",
"messages": [{"role": "user", "content": "Contact alice@example.com"}]
})
.to_string(),
)
.send()
.await
.expect("request should succeed");
let status = response.status();
let response_text = response.text().await.expect("body should read");
assert_eq!(status, StatusCode::OK, "{response_text}");
let response_json: serde_json::Value =
serde_json::from_str(&response_text).expect("response body should parse");
assert_eq!(
response_json["choices"][0]["message"]["content"],
"restored alice@example.com"
);
let seen = seen_provider_request
.lock()
.expect("mutex should lock")
.clone()
.expect("provider request should be captured");
let provider_body_text = serde_json::to_string(&seen.body).expect("body should serialize");
assert!(!provider_body_text.contains("alice@example.com"));
assert!(provider_body_text.contains("<AETHER:EMAIL:"));
gateway_handle.abort();
provider_handle.abort();
}
#[tokio::test]
async fn pii_redaction_performance_limits_do_not_forward_unredacted_body_upstream() {
let provider_hits = Arc::new(AtomicUsize::new(0));
let provider_hits_clone = Arc::clone(&provider_hits);
let provider_app = Router::new().route(
"/v1/chat/completions",
any(move |_request: Request| {
let provider_hits_inner = Arc::clone(&provider_hits_clone);
async move {
provider_hits_inner.fetch_add(1, Ordering::SeqCst);
Json(json!({
"id": "unexpected",
"choices": [{"message": {"role": "assistant", "content": "unexpected"}}]
}))
}
}),
);
let (provider_url, provider_handle) = start_server(provider_app).await;
let auth_repository = Arc::new(InMemoryAuthApiKeySnapshotRepository::seed(vec![(
Some(hash_api_key("sk-client-pii-redaction-limit")),
auth_snapshot("api-key-pii-redaction-limit", "user-pii-redaction-limit"),
)]));
let candidate_selection_repository =
Arc::new(InMemoryMinimalCandidateSelectionReadRepository::seed(vec![
candidate_row("pii-redaction-limit"),
]));
let request_candidate_repository = Arc::new(InMemoryRequestCandidateRepository::default());
let provider_catalog_repository = Arc::new(InMemoryProviderCatalogReadRepository::seed(
vec![provider("pii-redaction-limit", true)],
vec![endpoint("pii-redaction-limit", provider_url)],
vec![key("pii-redaction-limit")],
));
let data_state = crate::data::GatewayDataState::with_auth_candidate_selection_provider_catalog_and_request_candidate_repository_for_tests(
auth_repository,
candidate_selection_repository,
provider_catalog_repository,
Arc::clone(&request_candidate_repository),
DEVELOPMENT_ENCRYPTION_KEY,
)
.with_system_config_values_for_tests(redaction_config(true));
let gateway_state = AppState::new()
.expect("gateway state should build")
.with_data_state_for_tests(data_state);
let gateway = build_router_with_state(gateway_state);
let (gateway_url, gateway_handle) = start_server(gateway).await;
let original = format!("alice@example.com {}", "x".repeat(2 * 1024 * 1024));
let response = reqwest::Client::new()
.post(format!("{gateway_url}/v1/chat/completions"))
.header(http::header::CONTENT_TYPE, "application/json")
.header(
http::header::AUTHORIZATION,
"Bearer sk-client-pii-redaction-limit",
)
.header(TRACE_ID_HEADER, "trace-pii-redaction-limit")
.body(
json!({
"model": "gpt-5",
"messages": [{"role": "user", "content": original}]
})
.to_string(),
)
.send()
.await
.expect("request should complete");
let status = response.status();
let response_text = response.text().await.expect("body should read");
assert_eq!(status, StatusCode::PAYLOAD_TOO_LARGE, "{response_text}");
assert!(response_text.contains("scanned text limit exceeded"));
assert!(!response_text.contains("alice@example.com"));
assert_eq!(provider_hits.load(Ordering::SeqCst), 0);
gateway_handle.abort();
provider_handle.abort();
}
#[tokio::test]
async fn ai_execute_pii_redaction_missing_encryption_key_fails_closed_before_provider() {
let execution_runtime_hits = Arc::new(AtomicUsize::new(0));
let execution_runtime_hits_clone = Arc::clone(&execution_runtime_hits);
let execution_runtime = Router::new().route(
"/v1/execute/sync",
any(move |_request: Request| {
let execution_runtime_hits_inner = Arc::clone(&execution_runtime_hits_clone);
async move {
execution_runtime_hits_inner.fetch_add(1, Ordering::SeqCst);
Json(json!({"ok": true}))
}
}),
);
let (execution_runtime_url, execution_runtime_handle) = start_server(execution_runtime).await;
let test_id = "ai-execute-pii-redaction-missing-encryption-key";
let auth_repository = Arc::new(InMemoryAuthApiKeySnapshotRepository::seed(vec![(
Some(hash_api_key(&format!("sk-client-{test_id}"))),
auth_snapshot(&format!("api-key-{test_id}"), &format!("user-{test_id}")),
)]));
let candidate_selection_repository =
Arc::new(InMemoryMinimalCandidateSelectionReadRepository::seed(vec![
candidate_row(test_id),
]));
let request_candidate_repository = Arc::new(InMemoryRequestCandidateRepository::default());
let provider_catalog_repository = Arc::new(InMemoryProviderCatalogReadRepository::seed(
vec![provider(test_id, true)],
vec![endpoint(test_id, "https://example.com".to_string())],
vec![key(test_id)],
));
let gateway_state = build_state_with_execution_runtime_override(execution_runtime_url.clone())
.with_data_state_for_tests(
crate::data::GatewayDataState::with_auth_candidate_selection_provider_catalog_and_request_candidate_repository_for_tests(
auth_repository,
candidate_selection_repository,
provider_catalog_repository,
Arc::clone(&request_candidate_repository),
"",
)
.with_system_config_values_for_tests(redaction_config(true)),
);
let gateway = build_router_with_state(gateway_state);
let (gateway_url, gateway_handle) = start_server(gateway).await;
let response = reqwest::Client::new()
.post(format!("{gateway_url}/v1/chat/completions"))
.header(http::header::CONTENT_TYPE, "application/json")
.header(
http::header::AUTHORIZATION,
format!("Bearer sk-client-{test_id}"),
)
.header(TRACE_ID_HEADER, format!("trace-{test_id}"))
.body(rich_pii_request().to_string())
.send()
.await
.expect("request should succeed");
assert_eq!(response.status(), StatusCode::SERVICE_UNAVAILABLE);
let response_text = response.text().await.expect("body should read");
for original in [
"alice@example.com",
"bob@example.net",
"accessValueABCDEF1234567890abcdef",
"secretValueABCDEF1234567890abcdef",
] {
assert!(!response_text.contains(original));
}
assert!(!response_text.contains("<AETHER:"));
assert_eq!(execution_runtime_hits.load(Ordering::SeqCst), 0);
gateway_handle.abort();
execution_runtime_handle.abort();
}

View File

@@ -782,6 +782,19 @@ async fn gateway_handles_admin_modules_status_locally_with_trusted_admin_princip
assert_eq!(payload["oauth"]["active"], json!(true));
assert_eq!(payload["oauth"]["config_validated"], json!(true));
assert_eq!(payload["management_tokens"]["active"], json!(true));
assert_eq!(payload["chat_pii_redaction"]["enabled"], json!(false));
assert_eq!(
payload["chat_pii_redaction"]["display_name"],
"敏感信息替换保护"
);
assert_eq!(
payload["chat_pii_redaction"]["config_validated"],
json!(true)
);
assert_eq!(
payload["chat_pii_redaction"]["admin_route"],
"/admin/modules/chat-pii-redaction"
);
assert_eq!(
payload["notification_email"]["config_validated"],
json!(true)
@@ -908,6 +921,63 @@ async fn gateway_handles_admin_module_status_detail_locally_with_trusted_admin_p
upstream_handle.abort();
}
#[tokio::test]
async fn gateway_handles_chat_pii_redaction_module_status_detail_locally_with_trusted_admin_principal(
) {
let upstream_hits = Arc::new(Mutex::new(0usize));
let upstream_hits_clone = Arc::clone(&upstream_hits);
let upstream = Router::new().route(
"/api/admin/modules/status/chat_pii_redaction",
any(move |_request: Request| {
let upstream_hits_inner = Arc::clone(&upstream_hits_clone);
async move {
*upstream_hits_inner.lock().expect("mutex should lock") += 1;
(StatusCode::OK, Body::from("unexpected upstream hit"))
}
}),
);
let auth_module_repository = Arc::new(InMemoryAuthModuleReadRepository::default());
let data_state = GatewayDataState::with_auth_module_reader_for_tests(auth_module_repository)
.with_system_config_values_for_tests(vec![(
"module.chat_pii_redaction.enabled".to_string(),
json!(true),
)]);
let (upstream_url, upstream_handle) = start_server(upstream).await;
let gateway = build_router_with_state(
AppState::new()
.expect("gateway should build")
.with_data_state_for_tests(data_state),
);
let (gateway_url, gateway_handle) = start_server(gateway).await;
let response = reqwest::Client::new()
.get(format!(
"{gateway_url}/api/admin/modules/status/chat_pii_redaction"
))
.header(crate::constants::GATEWAY_HEADER, "rust-phase3b")
.header(TRUSTED_ADMIN_USER_ID_HEADER, "admin-user-123")
.header(TRUSTED_ADMIN_USER_ROLE_HEADER, "admin")
.header(TRUSTED_ADMIN_SESSION_ID_HEADER, "session-123")
.send()
.await
.expect("request should succeed");
assert_eq!(response.status(), StatusCode::OK);
let payload: serde_json::Value = response.json().await.expect("json body should parse");
assert_eq!(payload["name"], "chat_pii_redaction");
assert_eq!(payload["display_name"], "敏感信息替换保护");
assert_eq!(payload["enabled"], json!(true));
assert_eq!(payload["active"], json!(true));
assert_eq!(payload["config_validated"], json!(true));
assert_eq!(payload["admin_route"], "/admin/modules/chat-pii-redaction");
assert_eq!(*upstream_hits.lock().expect("mutex should lock"), 0);
gateway_handle.abort();
upstream_handle.abort();
}
#[tokio::test]
async fn gateway_sets_admin_module_enabled_locally_with_trusted_admin_principal() {
let upstream_hits = Arc::new(Mutex::new(0usize));

View File

@@ -131,7 +131,7 @@ async fn gateway_handles_admin_providers_locally_with_trusted_admin_principal()
.expect("gateway should build")
.with_data_state_for_tests(
GatewayDataState::with_provider_catalog_repository_for_tests(
provider_catalog_repository,
provider_catalog_repository.clone(),
),
),
);
@@ -240,6 +240,7 @@ async fn gateway_handles_admin_provider_summary_locally_with_trusted_admin_princ
"claude_code_advanced": {"pool_size": 3},
"pool_advanced": {"enabled": true},
"failover_rules": {"strategy": "ordered"},
"chat_pii_redaction": {"enabled": true},
"provider_ops": {"architecture_id": "anyrouter"}
})),
);
@@ -351,6 +352,7 @@ async fn gateway_handles_admin_provider_summary_locally_with_trusted_admin_princ
);
assert_eq!(payload["ops_configured"], true);
assert_eq!(payload["ops_architecture_id"], "anyrouter");
assert_eq!(payload["chat_pii_redaction"], json!({"enabled": true}));
assert_eq!(payload["created_at"], "2024-03-21T05:46:40Z");
assert_eq!(payload["updated_at"], "2024-03-21T05:48:20Z");
assert_eq!(
@@ -774,6 +776,20 @@ async fn gateway_updates_admin_provider_locally_with_trusted_admin_principal() {
let provider_catalog_repository = Arc::new(InMemoryProviderCatalogReadRepository::seed(
vec![
sample_provider("provider-openai", "openai", 10)
.with_transport_fields(
true,
false,
false,
None,
None,
None,
None,
None,
Some(json!({
"pool_advanced": {},
"failover_rules": {"strategy": "ordered"}
})),
)
.with_timestamps(Some(1_711_000_000), Some(1_711_000_100)),
sample_provider("provider-other", "other", 20),
],
@@ -792,7 +808,7 @@ async fn gateway_updates_admin_provider_locally_with_trusted_admin_principal() {
.expect("gateway should build")
.with_data_state_for_tests(
GatewayDataState::with_provider_catalog_repository_for_tests(
provider_catalog_repository,
provider_catalog_repository.clone(),
),
),
);
@@ -817,10 +833,11 @@ async fn gateway_updates_admin_provider_locally_with_trusted_admin_principal() {
"request_timeout": 55.0,
"stream_first_byte_timeout": 11.0,
"enable_format_conversion": false,
"config": {"provider_ops": {"architecture_id": "cubence"}},
"config": {
"provider_ops": {"architecture_id": "cubence"},
"chat_pii_redaction": {"enabled": true}
},
"claude_code_advanced": {"pool_size": 2},
"pool_advanced": {},
"failover_rules": {"strategy": "ordered"},
"proxy": {"url": "https://proxy.example"}
}))
.send()
@@ -847,8 +864,88 @@ async fn gateway_updates_admin_provider_locally_with_trusted_admin_principal() {
assert_eq!(payload["claude_code_advanced"], json!({"pool_size": 2}));
assert_eq!(payload["pool_advanced"], json!({}));
assert_eq!(payload["failover_rules"], json!({"strategy": "ordered"}));
assert_eq!(payload["chat_pii_redaction"], json!({"enabled": true}));
assert_eq!(payload["ops_configured"], true);
assert_eq!(payload["ops_architecture_id"], "cubence");
let disable_response = reqwest::Client::new()
.patch(format!("{gateway_url}/api/admin/providers/provider-openai"))
.header(crate::constants::GATEWAY_HEADER, "rust-phase3b")
.header(TRUSTED_ADMIN_USER_ID_HEADER, "admin-user-123")
.header(TRUSTED_ADMIN_USER_ROLE_HEADER, "admin")
.header(TRUSTED_ADMIN_SESSION_ID_HEADER, "session-123")
.json(&json!({
"config": {
"chat_pii_redaction": {"enabled": false}
}
}))
.send()
.await
.expect("request should succeed");
let disable_status = disable_response.status();
let disable_body = disable_response.text().await.expect("body should read");
assert_eq!(disable_status, StatusCode::OK, "body={disable_body}");
let disable_payload: serde_json::Value =
serde_json::from_str(&disable_body).expect("json body should parse");
assert_eq!(
disable_payload["chat_pii_redaction"],
json!({"enabled": false})
);
assert_eq!(disable_payload["pool_advanced"], json!({}));
assert_eq!(
disable_payload["failover_rules"],
json!({"strategy": "ordered"})
);
assert_eq!(disable_payload["ops_architecture_id"], "cubence");
let invalid_response = reqwest::Client::new()
.patch(format!("{gateway_url}/api/admin/providers/provider-openai"))
.header(crate::constants::GATEWAY_HEADER, "rust-phase3b")
.header(TRUSTED_ADMIN_USER_ID_HEADER, "admin-user-123")
.header(TRUSTED_ADMIN_USER_ROLE_HEADER, "admin")
.header(TRUSTED_ADMIN_SESSION_ID_HEADER, "session-123")
.json(&json!({
"config": {
"chat_pii_redaction": {"enabled": true, "entities": ["email"]}
}
}))
.send()
.await
.expect("request should succeed");
assert_eq!(invalid_response.status(), StatusCode::BAD_REQUEST);
let providers = provider_catalog_repository
.list_providers(false)
.await
.expect("providers should list");
let updated_provider = providers
.iter()
.find(|provider| provider.id == "provider-openai")
.expect("provider should exist");
assert_eq!(
updated_provider
.config
.as_ref()
.and_then(|value| value.get("chat_pii_redaction"))
.cloned(),
Some(json!({"enabled": false}))
);
assert_eq!(
updated_provider
.config
.as_ref()
.and_then(|value| value.get("pool_advanced"))
.cloned(),
Some(json!({}))
);
assert_eq!(
updated_provider
.config
.as_ref()
.and_then(|value| value.get("failover_rules"))
.cloned(),
Some(json!({"strategy": "ordered"}))
);
assert_eq!(*upstream_hits.lock().expect("mutex should lock"), 0);
gateway_handle.abort();
@@ -901,6 +998,7 @@ async fn gateway_creates_admin_provider_locally_with_trusted_admin_principal() {
"website": "codex.example",
"keep_priority_on_conversion": true,
"max_retries": 7,
"config": {"chat_pii_redaction": {"enabled": true}},
"pool_advanced": {},
"failover_rules": {"strategy": "ordered"},
"proxy": {"url": "https://proxy.example"}
@@ -943,6 +1041,38 @@ async fn gateway_creates_admin_provider_locally_with_trusted_admin_principal() {
.cloned(),
Some(json!({}))
);
assert_eq!(
created
.config
.as_ref()
.and_then(|value| value.get("chat_pii_redaction"))
.cloned(),
Some(json!({"enabled": true}))
);
assert_eq!(
created
.config
.as_ref()
.and_then(|value| value.get("failover_rules"))
.cloned(),
Some(json!({"strategy": "ordered"}))
);
let invalid_response = reqwest::Client::new()
.post(format!("{gateway_url}/api/admin/providers/"))
.header(crate::constants::GATEWAY_HEADER, "rust-phase3b")
.header(TRUSTED_ADMIN_USER_ID_HEADER, "admin-user-123")
.header(TRUSTED_ADMIN_USER_ROLE_HEADER, "admin")
.header(TRUSTED_ADMIN_SESSION_ID_HEADER, "session-123")
.json(&json!({
"name": "invalid-redaction-provider",
"provider_type": "custom",
"config": {"chat_pii_redaction": {"enabled": true, "entities": ["email"]}}
}))
.send()
.await
.expect("request should succeed");
assert_eq!(invalid_response.status(), StatusCode::BAD_REQUEST);
let endpoints = provider_catalog_repository
.list_endpoints_by_provider_ids(std::slice::from_ref(&created.id))

View File

@@ -1324,6 +1324,260 @@ async fn gateway_handles_admin_system_model_directives_default_as_disabled() {
gateway_handle.abort();
}
#[tokio::test]
async fn gateway_validates_chat_pii_redaction_system_config_locally_with_trusted_admin_principal() {
let upstream_hits = Arc::new(Mutex::new(0usize));
let upstream_hits_clone = Arc::clone(&upstream_hits);
let upstream = Router::new().route(
"/api/admin/system/configs/module.chat_pii_redaction.cache_ttl_seconds",
any(move |_request: Request| {
let upstream_hits_inner = Arc::clone(&upstream_hits_clone);
async move {
*upstream_hits_inner.lock().expect("mutex should lock") += 1;
(StatusCode::OK, Body::from("unexpected upstream hit"))
}
}),
);
let data_state =
GatewayDataState::disabled()
.with_system_config_values_for_tests(Vec::<(String, serde_json::Value)>::new());
let (upstream_url, upstream_handle) = start_server(upstream).await;
let gateway = build_router_with_state(
AppState::new()
.expect("gateway should build")
.with_data_state_for_tests(data_state),
);
let (gateway_url, gateway_handle) = start_server(gateway).await;
let client = reqwest::Client::new();
let get_config = |key: &'static str| {
let client = client.clone();
let gateway_url = gateway_url.clone();
async move {
let response = client
.get(format!("{gateway_url}/api/admin/system/configs/{key}"))
.header(crate::constants::GATEWAY_HEADER, "rust-phase3b")
.header(TRUSTED_ADMIN_USER_ID_HEADER, "admin-user-123")
.header(TRUSTED_ADMIN_USER_ROLE_HEADER, "admin")
.header(TRUSTED_ADMIN_SESSION_ID_HEADER, "session-123")
.send()
.await
.expect("request should succeed");
assert_eq!(response.status(), StatusCode::OK, "key={key}");
response
.json::<serde_json::Value>()
.await
.expect("json body should parse")
}
};
let put_config = |key: &'static str, value: serde_json::Value| {
let client = client.clone();
let gateway_url = gateway_url.clone();
async move {
client
.put(format!("{gateway_url}/api/admin/system/configs/{key}"))
.header(crate::constants::GATEWAY_HEADER, "rust-phase3b")
.header(TRUSTED_ADMIN_USER_ID_HEADER, "admin-user-123")
.header(TRUSTED_ADMIN_USER_ROLE_HEADER, "admin")
.header(TRUSTED_ADMIN_SESSION_ID_HEADER, "session-123")
.json(&json!({ "value": value }))
.send()
.await
.expect("request should succeed")
}
};
assert_eq!(
get_config("module.chat_pii_redaction.enabled").await["value"],
json!(false)
);
assert_eq!(
get_config("module.chat_pii_redaction.provider_scope").await["value"],
json!("selected_providers")
);
assert_eq!(
get_config("module.chat_pii_redaction.inject_model_instruction").await["value"],
json!(true)
);
assert_eq!(
get_config("module.chat_pii_redaction.cache_ttl_seconds").await["value"],
json!(300)
);
assert_eq!(
get_config("module.chat_pii_redaction.entities").await["value"],
json!([
"email",
"cn_phone",
"global_phone",
"cn_id",
"payment_card",
"ipv4",
"ipv6",
"api_key",
"access_token",
"secret_key",
"bearer_token",
"jwt"
])
);
let enabled_response = put_config("module.chat_pii_redaction.enabled", json!(true)).await;
assert_eq!(enabled_response.status(), StatusCode::OK);
let enabled_payload: serde_json::Value = enabled_response
.json()
.await
.expect("json body should parse");
assert_eq!(enabled_payload["value"], json!(true));
let scope_response = put_config(
"module.chat_pii_redaction.provider_scope",
json!("all_providers"),
)
.await;
assert_eq!(scope_response.status(), StatusCode::OK);
let scope_payload: serde_json::Value =
scope_response.json().await.expect("json body should parse");
assert_eq!(scope_payload["value"], json!("all_providers"));
let selected_entities_response = put_config(
"module.chat_pii_redaction.entities",
json!(["email", "jwt", "cn_phone"]),
)
.await;
assert_eq!(selected_entities_response.status(), StatusCode::OK);
let selected_entities_payload: serde_json::Value = selected_entities_response
.json()
.await
.expect("json body should parse");
assert_eq!(
selected_entities_payload["value"],
json!(["email", "cn_phone", "jwt"])
);
let ttl_response = put_config("module.chat_pii_redaction.cache_ttl_seconds", json!(3600)).await;
assert_eq!(ttl_response.status(), StatusCode::OK);
let ttl_payload: serde_json::Value = ttl_response.json().await.expect("json body should parse");
assert_eq!(ttl_payload["value"], json!(3600));
let instruction_response = put_config(
"module.chat_pii_redaction.inject_model_instruction",
json!(false),
)
.await;
assert_eq!(instruction_response.status(), StatusCode::OK);
let instruction_payload: serde_json::Value = instruction_response
.json()
.await
.expect("json body should parse");
assert_eq!(instruction_payload["value"], json!(false));
let invalid_scope_response = put_config(
"module.chat_pii_redaction.provider_scope",
json!("enabled_providers"),
)
.await;
assert_eq!(invalid_scope_response.status(), StatusCode::BAD_REQUEST);
let invalid_entities_response = put_config(
"module.chat_pii_redaction.entities",
json!(["email", "name"]),
)
.await;
assert_eq!(invalid_entities_response.status(), StatusCode::BAD_REQUEST);
let invalid_ttl_response =
put_config("module.chat_pii_redaction.cache_ttl_seconds", json!(600)).await;
assert_eq!(invalid_ttl_response.status(), StatusCode::BAD_REQUEST);
let invalid_instruction_response = put_config(
"module.chat_pii_redaction.inject_model_instruction",
json!("yes"),
)
.await;
assert_eq!(
invalid_instruction_response.status(),
StatusCode::BAD_REQUEST
);
let enabled_default_response =
put_config("module.chat_pii_redaction.enabled", serde_json::Value::Null).await;
assert_eq!(enabled_default_response.status(), StatusCode::OK);
let enabled_default_payload: serde_json::Value = enabled_default_response
.json()
.await
.expect("json body should parse");
assert_eq!(enabled_default_payload["value"], json!(false));
let scope_default_response = put_config(
"module.chat_pii_redaction.provider_scope",
serde_json::Value::Null,
)
.await;
assert_eq!(scope_default_response.status(), StatusCode::OK);
let scope_default_payload: serde_json::Value = scope_default_response
.json()
.await
.expect("json body should parse");
assert_eq!(scope_default_payload["value"], json!("selected_providers"));
let entities_default_response = put_config(
"module.chat_pii_redaction.entities",
serde_json::Value::Null,
)
.await;
assert_eq!(entities_default_response.status(), StatusCode::OK);
let entities_default_payload: serde_json::Value = entities_default_response
.json()
.await
.expect("json body should parse");
assert_eq!(
entities_default_payload["value"],
json!([
"email",
"cn_phone",
"global_phone",
"cn_id",
"payment_card",
"ipv4",
"ipv6",
"api_key",
"access_token",
"secret_key",
"bearer_token",
"jwt"
])
);
let ttl_default_response = put_config(
"module.chat_pii_redaction.cache_ttl_seconds",
serde_json::Value::Null,
)
.await;
assert_eq!(ttl_default_response.status(), StatusCode::OK);
let ttl_default_payload: serde_json::Value = ttl_default_response
.json()
.await
.expect("json body should parse");
assert_eq!(ttl_default_payload["value"], json!(300));
let instruction_default_response = put_config(
"module.chat_pii_redaction.inject_model_instruction",
serde_json::Value::Null,
)
.await;
assert_eq!(instruction_default_response.status(), StatusCode::OK);
let instruction_default_payload: serde_json::Value = instruction_default_response
.json()
.await
.expect("json body should parse");
assert_eq!(instruction_default_payload["value"], json!(true));
assert_eq!(*upstream_hits.lock().expect("mutex should lock"), 0);
gateway_handle.abort();
upstream_handle.abort();
}
#[tokio::test]
async fn gateway_handles_admin_system_provider_priority_mode_locally_with_bearer_admin_session() {
let upstream_hits = Arc::new(Mutex::new(0usize));