test: deflake the JEV classifier select and two router tests that inherited leaked state (#44840)

* test(ui): wait for the classifier model popup before picking its option

Base UI exposes its select option asynchronously, and the helper waits for the option and its positioner to become clickable before selection

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* test: isolate two router tests from work leaked by earlier tests

Collect earlier tests' garbage before warning capture so their unawaited coroutines cannot be attributed to the target's warning assertion

Filter success-event callbacks by the request's litellm_call_id so queued logging work cannot replace the current request's captured messages

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

---------

Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

tests/unit/router_utils/pre_call_checks/test_prompt_caching_deployment_check.py

@@ -1,6 +1,7 @@
 import asyncio
 import copy
 import functools
+import uuid
 from typing import Final, cast
 
 import pytest
@@ -350,10 +351,13 @@ def _affinity_messages(messages: list[AllMessageValues]) -> list[AllMessageValue
 
 
 class _SentMessagesCapture(CustomLogger):
-    def __init__(self):
+    def __init__(self, litellm_call_id: str):
+        self.litellm_call_id = litellm_call_id
         self.messages: list[AllMessageValues] | None = None
 
     async def async_log_success_event(self, kwargs, response_obj, start_time, end_time):
+        if kwargs.get("litellm_call_id") != self.litellm_call_id:
+            return
         standard_logging_object = kwargs.get("standard_logging_object")
         if standard_logging_object is not None:
             self.messages = standard_logging_object["messages"]
@@ -380,7 +384,8 @@ async def test_affinity_key_matches_the_messages_auto_caching_actually_sends(mon
     request was actually sent with, otherwise auto-injected caching gets no affinity at all.
     """
     monkeypatch.setattr(litellm, "enable_anthropic_prompt_caching", True)
-    capture = _SentMessagesCapture()
+    call_id: Final = str(uuid.uuid4())
+    capture = _SentMessagesCapture(call_id)
     monkeypatch.setattr(litellm, "callbacks", [capture])
     messages = _auto_caching_messages()
 
@@ -389,6 +394,7 @@ async def test_affinity_key_matches_the_messages_auto_caching_actually_sends(mon
         messages=copy.deepcopy(messages),
         mock_response="ok",
         api_key="sk-fake",
+        litellm_call_id=call_id,
     )
     sent_messages = await _eventually(lambda: capture.messages)
     assert sent_messages is not None
@@ -904,13 +910,18 @@ async def test_pin_matches_when_the_success_event_truncated_an_image_payload(mon
     replaced by size placeholders, while routing sees the raw request. Hashing the raw bytes on the
     read side would key every image-carrying session past its own pin.
     """
-    capture = _SentMessagesCapture()
+    call_id: Final = str(uuid.uuid4())
+    capture = _SentMessagesCapture(call_id)
     monkeypatch.setattr(litellm, "callbacks", [capture])
     image = {"type": "image_url", "image_url": {"url": ONE_PIXEL_PNG}}
     turn_one = _turn({"role": "user", "content": [image, _marked(LONG_PROMPT)]})
 
     await litellm.acompletion(
-        model=AUTO_CACHING_MODEL, messages=copy.deepcopy(turn_one), mock_response="ok", api_key="sk-fake"
+        model=AUTO_CACHING_MODEL,
+        messages=copy.deepcopy(turn_one),
+        mock_response="ok",
+        api_key="sk-fake",
+        litellm_call_id=call_id,
     )
     logged = await _eventually(lambda: capture.messages)
     assert logged is not None

tests/unit/test_router/test_router.py

@@ -18968,6 +18968,8 @@ async def test_router_embedding_path_rejects_past_max_parallel_requests_without_
             },
         )
 
+    # Reap earlier tests' garbage first so only this test's coroutines are recorded
+    gc.collect()
     with respx.mock() as respx_mock, warnings.catch_warnings(record=True) as caught:
         warnings.simplefilter("always")
         route: Final = respx_mock.post("https://max-parallel-embed.local/v1/embeddings").mock(side_effect=upstream)

ui/litellm-dashboard/src/components/add_model/JevClassifierConfig.integration.test.tsx

@@ -1,7 +1,7 @@
 import React, { useState } from "react";
 import userEvent from "@testing-library/user-event";
 import { afterEach, describe, expect, it, vi } from "vitest";
-import { fireEvent, renderWithProviders, screen } from "../../../tests/test-utils";
+import { chooseSelectOption, fireEvent, renderWithProviders, screen } from "../../../tests/test-utils";
 import useAuthorized from "@/app/(dashboard)/hooks/useAuthorized";
 import ClassificationMethodConfig from "./ClassificationMethodConfig";
 import AutoRouterClassifierTabs from "./AutoRouterClassifierTabs";
@@ -124,8 +124,7 @@ describe("JEV classifier editor", () => {
       fireEvent.click(screen.getByRole("radio", { name: label }));
       if (provider === "bespoke") expect(screen.getByLabelText("Classifier Model")).toHaveTextContent("nimble-latest");
       if (provider !== "jev") {
-        await userEvent.click(screen.getByLabelText("Classifier Model"));
-        await userEvent.click(screen.getByRole("option", { name: model }));
+        await chooseSelectOption(userEvent, screen.getByLabelText("Classifier Model"), model);
       } else {
         fireEvent.change(screen.getByLabelText("Classifier Model"), { target: { value: model } });
       }