teknium1
d10bb2ab6f
`scripts/run_tests.sh tests/<dir>/` is how a change gets its regression
coverage run, so a test filed under the wrong directory is a test nobody
runs when that code changes. Two kinds of drift had accumulated.
Parallel directories for one source package, folded into the mirror:
tests/acp -> tests/acp_adapter (its __init__/conftest move with it)
tests/cli -> tests/hermes_cli (prompt_toolkit fixture merged into
hermes_cli/conftest.py)
tests/run_agent -> tests/agent (backoff fixture becomes
agent/conftest.py)
tests/relay -> tests/gateway/relay
tests/state -> tests/hermes_state
246 loose files at tests/ root, routed by the package they import/patch:
hermes_cli, hermes_state, agent, gateway, tools, plugins, tui_gateway, cron.
Installer and desktop-update script tests go to tests/scripts/{install,
desktop_update}/. 43 tests of root-level modules (batch_runner, utils,
hermes_constants, packaging) stay at the root.
Filenames drop their issue numbers (95 files: test_89315_x.py -> test_x.py);
the number stays in the module docstring where it has context.
Collisions: test_cli_skin_integration.py existed in both tests/ and tests/cli
with different subsets — merged into one (10 tests, all kept);
run_agent/test_pre_compress_memory_context.py -> agent/..._handoff.py;
tests/test_account_usage.py -> agent/test_account_usage_fetch.py;
tests/test_web_server.py -> hermes_cli/test_web_server_ws_ping.py.
Deleted: test_minisweagent_path.py (empty since PR #2804),
test_model_picker_scroll.py (tested a private copy of the logic, imported
nothing), test_process_loop_event_loop_warning.py (asserted asyncio behaviour,
imported nothing from Hermes).
Repo-root path arithmetic (Path(__file__).parents[N], dirname chains) is
bumped for the 202 files that changed depth and verified by evaluating every
such expression against the new location. classify_changes' desktop-updater
lane prefix, tests-os.yml's ignore glob and every in-tree path comment follow
the moves. tests/test_tests_tree_layout.py keeps the tree from drifting back.
85 lines
4.1 KiB
Python
85 lines
4.1 KiB
Python
"""Tests for MiniMax model validation via static catalog (issues #12611, #12460, #12399, #12547).
|
|
|
|
MiniMax and MiniMax-CN providers don't expose /v1/models, so validate_requested_model()
|
|
must validate against the static catalog instead of probing the live API.
|
|
"""
|
|
|
|
from unittest.mock import patch
|
|
|
|
import pytest
|
|
|
|
from hermes_cli.models_validate import validate_requested_model
|
|
|
|
|
|
class TestMiniMaxModelValidation:
|
|
"""Test that validate_requested_model handles MiniMax providers correctly."""
|
|
|
|
@pytest.fixture(autouse=True)
|
|
def _isolate_minimax(self):
|
|
"""Ensure MiniMax catalog is used even if a live /v1/models endpoint exists."""
|
|
# Simulate fetch_api_models returning None (i.e., /v1/models is unreachable),
|
|
# proving that the catalog path is taken.
|
|
probe_payload = {
|
|
"models": None,
|
|
"probed_url": "https://api.minimax.io/v1/models",
|
|
"resolved_base_url": "https://api.minimax.io/v1",
|
|
"suggested_base_url": None,
|
|
"used_fallback": False,
|
|
}
|
|
with patch("hermes_cli.models.fetch_api_models", return_value=None), \
|
|
patch("hermes_cli.models.probe_api_models", return_value=probe_payload):
|
|
yield
|
|
|
|
# -------------------------------------------------------------------------
|
|
# Test 1: A known MiniMax model is accepted with recognized=True
|
|
# -------------------------------------------------------------------------
|
|
def test_valid_minimax_model_accepted(self):
|
|
result = validate_requested_model("MiniMax-M2.7", "minimax")
|
|
assert result["accepted"] is True
|
|
assert result["persist"] is True
|
|
assert result["recognized"] is True
|
|
assert result["message"] is None
|
|
|
|
# -------------------------------------------------------------------------
|
|
# Test 1b: Case-insensitive lookup matches catalog entries
|
|
# -------------------------------------------------------------------------
|
|
|
|
def test_valid_minimax_model_uppercase(self):
|
|
result = validate_requested_model("MINIMAX-M2.7", "minimax")
|
|
assert result["accepted"] is True
|
|
assert result["recognized"] is True
|
|
|
|
# -------------------------------------------------------------------------
|
|
# Test 2: A near-match model on minimax-cn triggers a suggestion (not auto-correct)
|
|
# -------------------------------------------------------------------------
|
|
def test_near_match_minimax_cn_suggests_similar(self):
|
|
# "MiniMax-M2.7-highspeed" is somewhat similar to "MiniMax-M2.7" (ratio ~0.71)
|
|
# but below the 0.9 auto-correct cutoff. It should be accepted with a
|
|
# recognized=False and a similar-models suggestion (ratio > 0.5).
|
|
result = validate_requested_model("MiniMax-M2.7-highspeed", "minimax-cn")
|
|
assert result["accepted"] is True
|
|
assert result["persist"] is True
|
|
assert result["recognized"] is False
|
|
# Should NOT auto-correct (ratio 0.71 < 0.9)
|
|
assert "corrected_model" not in result
|
|
# But should suggest similar models (ratio 0.71 > 0.5)
|
|
assert "MiniMax-M2.7" in result["message"]
|
|
|
|
# -------------------------------------------------------------------------
|
|
# Test 3: A completely unknown model is accepted (not rejected) with a warning
|
|
# -------------------------------------------------------------------------
|
|
|
|
# -------------------------------------------------------------------------
|
|
# Test 4: Verify catalog path is used (probe_api_models returns None)
|
|
# -------------------------------------------------------------------------
|
|
def test_minimax_uses_catalog_not_api_probe(self):
|
|
"""Ensure that when fetch_api_models returns None, the catalog is still checked."""
|
|
# The _isolate_minimax fixture already patches fetch_api_models to return None.
|
|
# If we reach the catalog path, MiniMax-M2.5 should be found and recognized.
|
|
result = validate_requested_model("MiniMax-M2.5", "minimax")
|
|
assert result["accepted"] is True
|
|
assert result["recognized"] is True
|
|
assert result["message"] is None
|
|
|
|
|