[Rust Frontend] Split engine core DTOs into separate modules (#47265)
Signed-off-by: Bugen Zhao <i@bugenzhao.com>
This commit is contained in:
@@ -5,7 +5,7 @@ use clap::Parser;
|
||||
use futures::StreamExt as _;
|
||||
use tokio::time::timeout;
|
||||
use tracing_subscriber::EnvFilter;
|
||||
use vllm_engine_core_client::protocol::EngineCoreSamplingParams;
|
||||
use vllm_engine_core_client::protocol::sampling::EngineCoreSamplingParams;
|
||||
use vllm_engine_core_client::{EngineCoreClient, EngineCoreClientConfig, TransportMode};
|
||||
use vllm_llm::{FinishReason, GenerateOutputStream, GenerateRequest, Llm};
|
||||
|
||||
|
||||
@@ -8,7 +8,7 @@ use futures::stream::FusedStream;
|
||||
use futures::{Stream, StreamExt as _, pin_mut};
|
||||
use serde::{Deserialize, Serialize};
|
||||
use vllm_engine_core_client::protocol::logprobs::Logprobs;
|
||||
use vllm_engine_core_client::protocol::{EngineCoreFinishReason, StopReason};
|
||||
use vllm_engine_core_client::protocol::output::{EngineCoreFinishReason, StopReason};
|
||||
use vllm_engine_core_client::{AbortCause, EngineCoreOutputStream};
|
||||
|
||||
use crate::error::Result;
|
||||
|
||||
@@ -4,9 +4,8 @@ use std::time::{SystemTime, UNIX_EPOCH};
|
||||
use uuid::Uuid;
|
||||
use vllm_engine_core_client::protocol::lora::LoraRequest;
|
||||
use vllm_engine_core_client::protocol::multimodal::MmFeatures;
|
||||
use vllm_engine_core_client::protocol::{
|
||||
EngineCoreRequest, EngineCoreSamplingParams, ReasoningParserKwargs,
|
||||
};
|
||||
use vllm_engine_core_client::protocol::request::{EngineCoreRequest, ReasoningParserKwargs};
|
||||
use vllm_engine_core_client::protocol::sampling::EngineCoreSamplingParams;
|
||||
|
||||
use crate::error::{Error, Result};
|
||||
|
||||
@@ -134,7 +133,8 @@ fn current_unix_timestamp_secs() -> f64 {
|
||||
mod tests {
|
||||
use std::collections::BTreeMap;
|
||||
|
||||
use vllm_engine_core_client::protocol::{EngineCoreSamplingParams, ReasoningParserKwargs};
|
||||
use vllm_engine_core_client::protocol::request::ReasoningParserKwargs;
|
||||
use vllm_engine_core_client::protocol::sampling::EngineCoreSamplingParams;
|
||||
|
||||
use super::GenerateRequest;
|
||||
use crate::error::Error;
|
||||
|
||||
@@ -1,7 +1,9 @@
|
||||
use std::time::{SystemTime, UNIX_EPOCH};
|
||||
|
||||
use vllm_engine_core_client::protocol::output::{
|
||||
EngineCoreEvent, EngineCoreEventType, EngineCoreOutput,
|
||||
};
|
||||
use vllm_engine_core_client::protocol::stats::PrefillStats;
|
||||
use vllm_engine_core_client::protocol::{EngineCoreEvent, EngineCoreEventType, EngineCoreOutput};
|
||||
use vllm_metrics::{
|
||||
EngineLabels, FinishedReasonLabels, METRICS, PromptTokenSourceLabels, RequestMetrics,
|
||||
};
|
||||
@@ -328,8 +330,8 @@ pub(crate) fn current_unix_timestamp_secs() -> f64 {
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use vllm_engine_core_client::protocol::output::{EngineCoreEvent, EngineCoreEventType};
|
||||
use vllm_engine_core_client::protocol::stats::PrefillStats;
|
||||
use vllm_engine_core_client::protocol::{EngineCoreEvent, EngineCoreEventType};
|
||||
|
||||
use super::{RequestMetricsTracker, diff_or_zero};
|
||||
|
||||
@@ -341,7 +343,7 @@ mod tests {
|
||||
2,
|
||||
10.0,
|
||||
100.2,
|
||||
&vllm_engine_core_client::protocol::EngineCoreOutput {
|
||||
&vllm_engine_core_client::protocol::output::EngineCoreOutput {
|
||||
request_id: "req-1".to_string(),
|
||||
new_token_ids: vec![1],
|
||||
finish_reason: None,
|
||||
@@ -369,7 +371,7 @@ mod tests {
|
||||
2,
|
||||
11.5,
|
||||
100.4,
|
||||
&vllm_engine_core_client::protocol::EngineCoreOutput {
|
||||
&vllm_engine_core_client::protocol::output::EngineCoreOutput {
|
||||
request_id: "req-1".to_string(),
|
||||
new_token_ids: vec![2, 3],
|
||||
finish_reason: None,
|
||||
|
||||
@@ -9,11 +9,13 @@ use uuid::Uuid;
|
||||
use vllm_engine_core_client::protocol::logprobs::{
|
||||
Logprobs, MaybeWireLogprobs, PositionLogprobs, TokenLogprob,
|
||||
};
|
||||
use vllm_engine_core_client::protocol::stats::PrefillStats;
|
||||
use vllm_engine_core_client::protocol::{
|
||||
use vllm_engine_core_client::protocol::output::{
|
||||
EngineCoreEvent, EngineCoreEventType, EngineCoreFinishReason, EngineCoreOutput,
|
||||
EngineCoreOutputs, EngineCoreRequest, EngineCoreSamplingParams,
|
||||
EngineCoreOutputs,
|
||||
};
|
||||
use vllm_engine_core_client::protocol::request::EngineCoreRequest;
|
||||
use vllm_engine_core_client::protocol::sampling::EngineCoreSamplingParams;
|
||||
use vllm_engine_core_client::protocol::stats::PrefillStats;
|
||||
use vllm_engine_core_client::test_utils::{IpcNamespace, spawn_mock_engine_task};
|
||||
use vllm_engine_core_client::{EngineCoreClient, EngineCoreClientConfig};
|
||||
use vllm_llm::{
|
||||
|
||||
Reference in New Issue
Block a user