[Refactor][OutputProcessor 2/8]: OmniRequestOutput should inherit RequestOutput and no nested wrap-up (#5146)

Signed-off-by: Boao Shi <aoibosh@connect.hku.hk>
Signed-off-by: bowieshi <aoibosh@connect.hku.hk>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
Signed-off-by: R2-Y <ruiruyang2@gmail.com>
Signed-off-by: amy-why-3459 <wuhaiyan17@huawei.com>
Signed-off-by: Shi Boao Bowie <aoibosh@connect.hku.hk>
Co-authored-by: 汪志鹏 <wangzhipeng628@gmail.com>
Co-authored-by: ruirui(rein) yang <73573651+R2-Y@users.noreply.github.com>
Co-authored-by: amy-why-3459 <wuhaiyan17@huawei.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
This commit is contained in:
authored and GitHub committed 2026-08-12 22:58:23 +08:00
1 parent 38a2500bdf
commit fcf1085395
89 files changed
+944 -639

No files matched your search

@@ -171,11 +171,11 @@ def _build_text_output(text: str = "This is a test response.") -> OmniRequestOut
metrics=None,
lora_request=None,
)
return OmniRequestOutput(
return OmniRequestOutput.from_stage_output(
request_output,
request_id="test_req_text",
finished=True,
final_output_type="text",
request_output=request_output,
)
@@ -200,11 +200,11 @@ def _build_audio_chat_output(num_samples: int = 24000) -> OmniRequestOutput:
metrics=None,
lora_request=None,
)
return OmniRequestOutput(
return OmniRequestOutput.from_stage_output(
request_output,
request_id="test_req_audio_chat",
finished=True,
final_output_type="audio",
request_output=request_output,
)
@@ -240,11 +240,11 @@ def _build_diffusion_image_output_for_chat_endpoint() -> OmniRequestOutput:
images=[_build_image_output(color="blue")],
finished=True,
)
return OmniRequestOutput(
return OmniRequestOutput.from_stage_output(
request_output,
request_id="test_req_img_chat",
finished=True,
images=[_build_image_output(color="blue")],
final_output_type="image",
request_output=request_output,
)
@@ -515,8 +515,19 @@ def mock_async_omni(
@pytest.fixture
def api_server(unused_tcp_port_factory, server_case: ServerCase, mock_async_omni):
def api_server(unused_tcp_port_factory, server_case: ServerCase, mock_async_omni, tmp_path):
"""Set up a API server in background process from command line with parametrized model name and mocked AsyncOmni."""
# Override the STORAGE_MANAGER path to a writable temp directory before
# forking the server subprocess. The default /tmp/storage may be owned
# by a different user (e.g. from a previous CI run) and cause
# PermissionError when the server tries to save generated video files.
from vllm_omni.entrypoints.openai.storage import STORAGE_MANAGER
old_storage_path = STORAGE_MANAGER.storage_path
new_storage_path = tmp_path / "omni_storage"
new_storage_path.mkdir(exist_ok=True)
STORAGE_MANAGER.storage_path = str(new_storage_path)
parser = TrackingArgumentParser()
subparsers = parser.add_subparsers(dest="command")
cmd = OmniServeCommand()
@@ -555,6 +566,9 @@ def api_server(unused_tcp_port_factory, server_case: ServerCase, mock_async_omni
yield f"http://127.0.0.1:{port}/v1"
# Restore the original storage path after the test.
STORAGE_MANAGER.storage_path = old_storage_path
if server_process.is_alive():
server_process.terminate()
server_process.join(timeout=10)
@@ -294,7 +294,7 @@ class vLLMOmniHttpServerLocal:
"global_steps": self.global_steps,
}
req_output = final_res.request_output
req_output = final_res
if req_output is not None and hasattr(req_output, "finish_reason"):
finish_reason = req_output.finish_reason or "stop"
else: