[AZ-180] Refactor detection event handling and improve SSE support

- Updated the detection image endpoint to require a channel ID for event streaming.
- Introduced a new endpoint for streaming detection events, allowing clients to receive real-time updates.
- Enhanced the internal buffering mechanism for detection events to manage multiple channels.
- Refactored the inference module to support the new event handling structure.

Made-with: Cursor

e2e/tests/test_async_sse.py

@@ -4,6 +4,7 @@ import time
 import uuid
 
 import pytest
+import sseclient
 
 
 def _ai_config_video() -> dict:
@@ -19,36 +20,43 @@ def test_ft_p08_immediate_async_response(
 ):
     media_id = f"async-{uuid.uuid4().hex}"
     body = _ai_config_image()
-    headers = {"Authorization": f"Bearer {jwt_token}"}
+    channel_id = str(uuid.uuid4())
+    headers = {"Authorization": f"Bearer {jwt_token}", "X-Channel-Id": channel_id}
     t0 = time.monotonic()
     r = http_client.post(f"/detect/{media_id}", json=body, headers=headers)
     elapsed = time.monotonic() - t0
     assert elapsed < 2.0
-    assert r.status_code == 200
-    assert r.json() == {"status": "started", "mediaId": media_id}
+    assert r.status_code == 202
 
 
 @pytest.mark.timeout(10)
 def test_ft_p09_sse_event_delivery(
-    warm_engine, http_client, jwt_token, sse_client_factory
+    warm_engine, http_client, jwt_token
 ):
     media_id = f"sse-{uuid.uuid4().hex}"
+    channel_id = str(uuid.uuid4())
     body = _ai_config_video()
-    headers = {"Authorization": f"Bearer {jwt_token}"}
+    auth_header = {"Authorization": f"Bearer {jwt_token}"}
+    post_headers = {**auth_header, "X-Channel-Id": channel_id}
     collected: list[dict] = []
     thread_exc: list[BaseException] = []
     first_event = threading.Event()
+    connected = threading.Event()
 
     def _listen():
         try:
-            with sse_client_factory(media_id) as sse:
-                time.sleep(0.3)
-                for event in sse.events():
+            with http_client.get(
+                f"/detect/events/{channel_id}",
+                stream=True,
+                timeout=600,
+                headers=auth_header,
+            ) as resp:
+                resp.raise_for_status()
+                connected.set()
+                for event in sseclient.SSEClient(resp).events():
                     if not event.data or not str(event.data).strip():
                         continue
                     data = json.loads(event.data)
-                    if data.get("mediaId") != media_id:
-                        continue
                     collected.append(data)
                     first_event.set()
                     if len(collected) >= 5:
@@ -56,13 +64,14 @@ def test_ft_p09_sse_event_delivery(
         except BaseException as e:
             thread_exc.append(e)
         finally:
+            connected.set()
             first_event.set()
 
     th = threading.Thread(target=_listen, daemon=True)
     th.start()
-    time.sleep(0.5)
-    r = http_client.post(f"/detect/{media_id}", json=body, headers=headers)
-    assert r.status_code == 200
+    connected.wait(timeout=5)
+    r = http_client.post(f"/detect/{media_id}", json=body, headers=post_headers)
+    assert r.status_code == 202
     first_event.wait(timeout=5)
     th.join(timeout=5)
     assert not thread_exc, thread_exc
@@ -74,8 +83,9 @@ def test_ft_n04_duplicate_media_id_409(
 ):
     media_id = "dup-test"
     body = _ai_config_image()
-    headers = {"Authorization": f"Bearer {jwt_token}"}
-    r1 = http_client.post(f"/detect/{media_id}", json=body, headers=headers)
-    assert r1.status_code == 200
-    r2 = http_client.post(f"/detect/{media_id}", json=body, headers=headers)
+    headers1 = {"Authorization": f"Bearer {jwt_token}", "X-Channel-Id": str(uuid.uuid4())}
+    headers2 = {"Authorization": f"Bearer {jwt_token}", "X-Channel-Id": str(uuid.uuid4())}
+    r1 = http_client.post(f"/detect/{media_id}", json=body, headers=headers1)
+    assert r1.status_code == 202
+    r2 = http_client.post(f"/detect/{media_id}", json=body, headers=headers2)
     assert r2.status_code == 409