remove dead code (#251)

src/litserve/loops.py

@@ -15,7 +15,6 @@
 import inspect
 import logging
 import multiprocessing as mp
-import os
 import pickle
 import sys
 import time
@@ -44,12 +43,6 @@
 
 logger = logging.getLogger(__name__)
 
-# if defined, it will require clients to auth with X-API-Key in the header
-LIT_SERVER_API_KEY = os.environ.get("LIT_SERVER_API_KEY")
-
-# timeout when we need to poll or wait indefinitely for a result in a loop.
-LONG_TIMEOUT = 100
-
 # FastAPI writes form files to disk over 1MB by default, which prevents serialization by multiprocessing
 MultiPartParser.max_file_size = sys.maxsize
 
@@ -61,17 +54,6 @@ def _inject_context(context: Union[List[dict], dict], func, *args, **kwargs):
     return func(*args, **kwargs)
 
 
-def get_batch_from_uid(uids, lit_api, request_buffer):
-    batches = []
-    for uid in uids:
-        try:
-            x_enc, pipe_s = request_buffer.pop(uid)
-        except KeyError:
-            continue
-        batches.append((x_enc, pipe_s))
-    return batches
-
-
 def collate_requests(
     lit_api: LitAPI, request_queue: Queue, max_batch_size: int, batch_timeout: float
 ) -> Tuple[List, List]:

src/litserve/server.py

@@ -30,7 +30,7 @@
 from typing import Callable, Dict, Optional, Sequence, Tuple, Union
 
 import uvicorn
-from fastapi import BackgroundTasks, Depends, FastAPI, HTTPException, Request, Response
+from fastapi import Depends, FastAPI, HTTPException, Request, Response
 from fastapi.responses import StreamingResponse
 from fastapi.security import APIKeyHeader
 from starlette.formparsers import MultiPartParser
@@ -58,9 +58,6 @@
 # if defined, it will require clients to auth with X-API-Key in the header
 LIT_SERVER_API_KEY = os.environ.get("LIT_SERVER_API_KEY")
 
-# timeout when we need to poll or wait indefinitely for a result in a loop.
-LONG_TIMEOUT = 100
-
 # FastAPI writes form files to disk over 1MB by default, which prevents serialization by multiprocessing
 MultiPartParser.max_file_size = sys.maxsize
 
@@ -306,7 +303,7 @@ async def health(request: Request) -> Response:
 
             return Response(content="not ready", status_code=503)
 
-        async def predict(request: self.request_type, background_tasks: BackgroundTasks) -> self.response_type:
+        async def predict(request: self.request_type) -> self.response_type:
             response_queue_id = self.app.response_queue_id
             uid = uuid.uuid4()
             event = asyncio.Event()
@@ -331,7 +328,7 @@ async def predict(request: self.request_type, background_tasks: BackgroundTasks)
                 load_and_raise(response)
             return response
 
-        async def stream_predict(request: self.request_type, background_tasks: BackgroundTasks) -> self.response_type:
+        async def stream_predict(request: self.request_type) -> self.response_type:
             response_queue_id = self.app.response_queue_id
             uid = uuid.uuid4()
             event = asyncio.Event()