mirror of
https://github.com/fawney19/Aether.git
synced 2026-10-06 17:37:47 +08:00
- 新增 cache_creation_ephemeral_5m/1h_input_tokens 字段,区分不同 TTL 的缓存写入 token - 引入 effective_input_tokens(扣除 cache read 后的有效输入 token),暴露给 usage 接口 - billing 规则生成器支持 5m/1h ephemeral cache 独立定价与分级计费 - usage_mapper 增加 Claude/Anthropic 格式映射,修复 OpenAI responses 格式字段兼容性 - 迁移逻辑增强:支持 checksum 容错、applied/pending 数量日志、逐步执行信息输出 - executor 抽离 LocalExecutionRequestOutcome 类型,统一 sync/stream 路径返回语义 - provider-transport auth 层新增 complete passthrough headers 构建逻辑 - 前端 usage 类型全面补充 effective_input_tokens、cache_creation_tokens、total_input_context 字段
69 lines
2.2 KiB
Rust
69 lines
2.2 KiB
Rust
use axum::body::{Body, Bytes};
|
|
use axum::http::{HeaderName, HeaderValue, Response};
|
|
|
|
use crate::constants::CONTROL_EXECUTED_HEADER;
|
|
use crate::control::GatewayControlDecision;
|
|
use crate::executor::{
|
|
maybe_execute_stream_request, maybe_execute_sync_request, LocalExecutionRequestOutcome,
|
|
};
|
|
use crate::{AppState, GatewayError};
|
|
|
|
use super::resolve_execution_runtime_auth_context;
|
|
|
|
pub(crate) fn allows_control_execute_emergency(decision: &GatewayControlDecision) -> bool {
|
|
decision.is_execution_runtime_candidate()
|
|
}
|
|
|
|
pub(crate) async fn maybe_execute_via_control(
|
|
state: &AppState,
|
|
parts: &http::request::Parts,
|
|
body_bytes: Bytes,
|
|
trace_id: &str,
|
|
decision: Option<&GatewayControlDecision>,
|
|
require_stream: bool,
|
|
) -> Result<LocalExecutionRequestOutcome, GatewayError> {
|
|
let Some(decision) = decision else {
|
|
return Ok(LocalExecutionRequestOutcome::NoPath);
|
|
};
|
|
|
|
let mut local_decision = decision.clone();
|
|
if let Some(auth_context) = resolve_execution_runtime_auth_context(
|
|
state,
|
|
&local_decision,
|
|
&parts.headers,
|
|
&parts.uri,
|
|
trace_id,
|
|
)
|
|
.await?
|
|
{
|
|
local_decision.auth_context = Some(auth_context);
|
|
local_decision.local_auth_rejection = None;
|
|
}
|
|
|
|
let response = if require_stream {
|
|
maybe_execute_stream_request(state, parts, &body_bytes, trace_id, Some(&local_decision))
|
|
.await?
|
|
} else {
|
|
maybe_execute_sync_request(state, parts, &body_bytes, trace_id, Some(&local_decision))
|
|
.await?
|
|
};
|
|
|
|
Ok(match response {
|
|
LocalExecutionRequestOutcome::Responded(response) => {
|
|
LocalExecutionRequestOutcome::Responded(mark_control_executed(response))
|
|
}
|
|
LocalExecutionRequestOutcome::Exhausted(outcome) => {
|
|
LocalExecutionRequestOutcome::Exhausted(outcome)
|
|
}
|
|
LocalExecutionRequestOutcome::NoPath => LocalExecutionRequestOutcome::NoPath,
|
|
})
|
|
}
|
|
|
|
fn mark_control_executed(mut response: Response<Body>) -> Response<Body> {
|
|
response.headers_mut().insert(
|
|
HeaderName::from_static(CONTROL_EXECUTED_HEADER),
|
|
HeaderValue::from_static("true"),
|
|
);
|
|
response
|
|
}
|