[Serve] Use serve_instance for test_http_headers to reduce the test t…

…ime (ray-project#36988)

rewrite test to use shared instance instead of ray shutdown.
In my local, time is reduced from 40s to 23s.

python/ray/serve/tests/test_http_headers.py

@@ -8,7 +8,7 @@
 from ray.serve._private.constants import RAY_SERVE_REQUEST_ID_HEADER
 
 
-def test_request_id_header_by_default(ray_shutdown):
+def test_request_id_header_by_default(serve_instance):
     """Test that a request_id is generated by default and returned as a header."""
 
     @serve.deployment
@@ -24,50 +24,52 @@ def __call__(self):
     assert resp.text == resp.headers[RAY_SERVE_REQUEST_ID_HEADER]
 
 
-@pytest.mark.parametrize("deploy_type", ["basic", "fastapi", "starlette_resp"])
-def test_user_provided_request_id_header(ray_shutdown, deploy_type):
-    """Test that a user-provided request_id is propagated to the
-    replica and returned as a header."""
-
-    if deploy_type == "fastapi":
-        app = FastAPI()
+class TestUserProvidedRequestIDHeader:
+    def verify_result(self):
+        resp = requests.get(
+            "http://localhost:8000", headers={RAY_SERVE_REQUEST_ID_HEADER: "123-234"}
+        )
+        assert resp.status_code == 200
+        assert resp.json() == 1
+        assert RAY_SERVE_REQUEST_ID_HEADER in resp.headers
+        assert resp.headers[RAY_SERVE_REQUEST_ID_HEADER] == "123-234"
 
+    def test_basic(self, serve_instance):
         @serve.deployment
-        @serve.ingress(app)
         class Model:
-            @app.get("/")
-            def say_hi(self) -> int:
+            def __call__(self) -> int:
                 request_id = ray.serve.context._serve_request_context.get().request_id
                 assert request_id == "123-234"
                 return 1
 
-    elif deploy_type == "basic":
+        serve.run(Model.bind())
+        self.verify_result()
+
+    def test_fastapi(self, serve_instance):
+        app = FastAPI()
 
         @serve.deployment
+        @serve.ingress(app)
         class Model:
-            def __call__(self) -> int:
+            @app.get("/")
+            def say_hi(self) -> int:
                 request_id = ray.serve.context._serve_request_context.get().request_id
                 assert request_id == "123-234"
                 return 1
 
-    else:
+        serve.run(Model.bind())
+        self.verify_result()
 
+    def test_starlette_resp(self, serve_instance):
         @serve.deployment
         class Model:
             def __call__(self) -> int:
                 request_id = ray.serve.context._serve_request_context.get().request_id
                 assert request_id == "123-234"
                 return starlette.responses.Response("1", media_type="application/json")
 
-    serve.run(Model.bind())
-
-    resp = requests.get(
-        "http://localhost:8000", headers={RAY_SERVE_REQUEST_ID_HEADER: "123-234"}
-    )
-    assert resp.status_code == 200
-    assert resp.json() == 1
-    assert RAY_SERVE_REQUEST_ID_HEADER in resp.headers
-    assert resp.headers[RAY_SERVE_REQUEST_ID_HEADER] == "123-234"
+        serve.run(Model.bind())
+        self.verify_result()
 
 
 if __name__ == "__main__":