| """Tests for /models endpoints (Phase 2)."""
|
|
|
| import pytest
|
| from unittest.mock import patch, AsyncMock
|
|
|
|
|
| @pytest.mark.asyncio
|
| async def test_models_list(client, auth_headers):
|
| mock_models = [
|
| {"name": "qwen2.5-coder:7b", "model": "qwen2.5-coder:7b", "size": 4000000000}
|
| ]
|
| with patch("mac.services.llm_service.list_ollama_models", new_callable=AsyncMock, return_value=mock_models):
|
| resp = await client.get("/api/v1/models", headers=auth_headers)
|
| assert resp.status_code == 200
|
| data = resp.json()
|
| assert "models" in data
|
| assert "total" in data
|
|
|
|
|
| @pytest.mark.asyncio
|
| async def test_models_list_unauthenticated(client):
|
| resp = await client.get("/api/v1/models")
|
|
|
| assert resp.status_code == 200
|
|
|
|
|
| @pytest.mark.asyncio
|
| async def test_model_detail(client, auth_headers):
|
| with patch("mac.services.llm_service.list_ollama_models", new_callable=AsyncMock, return_value=[]):
|
| with patch("mac.services.llm_service.get_ollama_model_detail", new_callable=AsyncMock, return_value={"modelfile": "FROM qwen", "parameters": "temp 0.7"}):
|
| resp = await client.get("/api/v1/models/qwen2.5-coder:7b", headers=auth_headers)
|
| assert resp.status_code == 200
|
| data = resp.json()
|
| assert data["id"] == "qwen2.5-coder:7b"
|
|
|
|
|
| @pytest.mark.asyncio
|
| async def test_model_detail_not_found(client, auth_headers):
|
| resp = await client.get("/api/v1/models/nonexistent-model", headers=auth_headers)
|
| assert resp.status_code == 404
|
|
|
|
|
| @pytest.mark.asyncio
|
| async def test_model_load_requires_admin(client, auth_headers):
|
| resp = await client.post("/api/v1/models/qwen2.5-coder:7b/load", headers=auth_headers)
|
| assert resp.status_code == 403
|
|
|
|
|
| @pytest.mark.asyncio
|
| async def test_model_load_as_admin(client, admin_headers):
|
| mock_resp = AsyncMock()
|
| mock_resp.status_code = 200
|
| mock_resp.raise_for_status = lambda: None
|
| mock_resp.json = lambda: {"done": True}
|
|
|
| with patch("mac.services.model_service.httpx.AsyncClient") as MockClient:
|
| mock_client_instance = AsyncMock()
|
| mock_client_instance.post = AsyncMock(return_value=mock_resp)
|
| mock_client_instance.__aenter__ = AsyncMock(return_value=mock_client_instance)
|
| mock_client_instance.__aexit__ = AsyncMock(return_value=None)
|
| MockClient.return_value = mock_client_instance
|
|
|
| resp = await client.post("/api/v1/models/qwen2.5-coder:7b/load", headers=admin_headers)
|
| assert resp.status_code == 200
|
|
|
|
|
| @pytest.mark.asyncio
|
| async def test_model_health(client, auth_headers):
|
| with patch("mac.services.model_service.get_model_health", new_callable=AsyncMock, return_value={
|
| "model_id": "qwen2.5-coder:7b", "status": "ready", "ready": True, "latency_ms": 100, "memory_mb": 4096
|
| }):
|
| resp = await client.get("/api/v1/models/qwen2.5-coder:7b/health", headers=auth_headers)
|
| assert resp.status_code == 200
|
|
|
|
|
| @pytest.mark.asyncio
|
| async def test_model_download_requires_admin(client, auth_headers):
|
| resp = await client.post("/api/v1/models/download", headers=auth_headers,
|
| json={"model_name": "test:latest"})
|
| assert resp.status_code == 403
|
|
|
