A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 06:00 UTC
[Misc] Add __str__ for RequestStatus (#19780)
Signed-off-by: Linkun Chen <github@lkchen.net>
lkchen committed
Jun 17, 2025 at 20:03 UTC
d4629dc43f92c189e90a3e2f2f8a52648aed4d9d
2 files changed
+18
tests/v1/test_request.py
new
+15
@@ -0,0 +1,15 @@
1
+# SPDX-License-Identifier: Apache-2.0
2
+from vllm.v1.request import RequestStatus
3
+
4
+
5
+def test_request_status_fmt_str():
6
+ """Test that the string representation of RequestStatus is correct."""
7
+ assert f"{RequestStatus.WAITING}" == "WAITING"
8
+ assert f"{RequestStatus.WAITING_FOR_FSM}" == "WAITING_FOR_FSM"
9
+ assert f"{RequestStatus.WAITING_FOR_REMOTE_KVS}" == "WAITING_FOR_REMOTE_KVS"
10
+ assert f"{RequestStatus.RUNNING}" == "RUNNING"
11
+ assert f"{RequestStatus.PREEMPTED}" == "PREEMPTED"
12
+ assert f"{RequestStatus.FINISHED_STOPPED}" == "FINISHED_STOPPED"
13
+ assert f"{RequestStatus.FINISHED_LENGTH_CAPPED}" == "FINISHED_LENGTH_CAPPED"
14
+ assert f"{RequestStatus.FINISHED_ABORTED}" == "FINISHED_ABORTED"
15
+ assert f"{RequestStatus.FINISHED_IGNORED}" == "FINISHED_IGNORED"
vllm/v1/request.py
+3
@@ -171,6 +171,9 @@ class RequestStatus(enum.IntEnum):
171
FINISHED_ABORTED = enum.auto()
172
FINISHED_IGNORED = enum.auto()
173
174
+ def __str__(self):
175
+ return self.name
176
+
177
@staticmethod
178
def is_finished(status: "RequestStatus") -> bool:
179
return status > RequestStatus.PREEMPTED