Spaces:
Running on Zero
Running on Zero
File size: 3,271 Bytes
7f9dfed | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 | from __future__ import annotations
import unittest
import requests
from models.llama_cpp_service import LlamaCppConfig, LlamaCppService, local_file_status
from models.model_catalog import load_model_catalog
from models.service_factory import create_text_service
class LlamaCppServiceTest(unittest.TestCase):
def test_factory_creates_llama_cpp_service(self) -> None:
catalog = load_model_catalog("config/models.yaml")
service = create_text_service(catalog["minicpm5_1b"], "llama.cpp")
self.assertIsInstance(service, LlamaCppService)
def test_status_reports_missing_executable(self) -> None:
status = LlamaCppService.status(which_func=lambda _name: None)
self.assertFalse(status.available)
self.assertIn("not found", status.detail)
def test_launch_command_includes_model_and_mmproj(self) -> None:
catalog = load_model_catalog("config/models.yaml")
service = LlamaCppService(
catalog["minicpm_v46"],
LlamaCppConfig(
server_path="C:/llama/llama-server.exe",
model_path="model.gguf",
mmproj_path="mmproj.gguf",
),
)
self.assertEqual(
service.launch_command(),
[
"C:/llama/llama-server.exe",
"-m",
"model.gguf",
"--mmproj",
"mmproj.gguf",
],
)
def test_launch_command_is_empty_without_model_path(self) -> None:
catalog = load_model_catalog("config/models.yaml")
service = LlamaCppService(catalog["minicpm5_1b"])
self.assertEqual(service.launch_command(), [])
def test_chat_returns_clear_unavailable_message(self) -> None:
catalog = load_model_catalog("config/models.yaml")
response = LlamaCppService(catalog["minicpm5_1b"]).chat("", "Hello")
self.assertIn("[llama.cpp unavailable]", response)
def test_extracts_openai_compatible_response(self) -> None:
response = LlamaCppService._extract_response(
{"choices": [{"message": {"content": "hello"}}]}
)
self.assertEqual(response, "hello")
def test_status_reports_unreachable_server(self) -> None:
def raise_request_error(*_args, **_kwargs):
raise requests.ConnectionError("offline")
status = LlamaCppService.status(
which_func=lambda _name: "llama-server",
get_func=raise_request_error,
)
self.assertFalse(status.available)
self.assertIn("not reachable", status.detail)
def test_status_uses_configured_server_path(self) -> None:
def ok_response(*_args, **_kwargs):
response = requests.Response()
response.status_code = 200
return response
status = LlamaCppService.status(
which_func=lambda _name: None,
get_func=ok_response,
server_path=__file__,
)
self.assertTrue(status.available)
def test_local_file_status(self) -> None:
self.assertEqual(local_file_status(""), "not configured")
self.assertEqual(local_file_status("missing-file.gguf"), "missing")
if __name__ == "__main__":
unittest.main()
|