mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-30 19:09:28 +00:00
6b81590c55
Systematic prune per AGENTS.md test policy, one pass over every major test tree (gateway, hermes_cli, tools, agent, run_agent, plugins, cli, cron, tui_gateway, honcho/openviking, root-level): - DELETE: source-reading tests (read_text/getsource on prod files), change-detector tests (exact catalog counts, model-name snapshots, config version literals), mock-echo tests (assert a mock returns what it was told), assertion-free/trivial tests, near-duplicate parametrizations (boundaries + one representative kept), async/sync twin duplicates, cosmetic within-file variations. - KEEP (mandatory): security/redaction/approval guards, message-role alternation invariants, prompt-caching/deterministic-call-id invariants, issue-number regression tests (deduped), E2E tests. - 6 test files deleted outright (script-style/no-assert or fully redundant); conftest.py, fakes/, fixtures/ untouched. - tests/acp/conftest.py added: autouse fixture stubs the live models.dev/GitHub/Copilot/Anthropic inventory fetches that ACP server tests performed on every session create — test_server.py 147s → 3.4s, and the tests are now genuinely hermetic. - Sleep-based slowness shrunk where safe (codex_ttfb_watchdog, compression_concurrent_fork, etc.); no wall-clock assertion tightened. Verification: full hermetic suite via scripts/run_tests.sh — 2439 files, 31,130 tests passed, 0 failed, 0 flaky retries, 315s wall (baseline: 583s wall, 13,564s subprocess CPU).
99 lines
2.9 KiB
Python
99 lines
2.9 KiB
Python
"""Tests for Copilot live /models context-window resolution."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import time
|
|
from unittest.mock import patch
|
|
|
|
import pytest
|
|
|
|
from hermes_cli.models import get_copilot_model_context
|
|
|
|
|
|
# Sample catalog items mimicking the Copilot /models API response
|
|
_SAMPLE_CATALOG = [
|
|
{
|
|
"id": "claude-opus-4.6-1m",
|
|
"capabilities": {
|
|
"type": "chat",
|
|
"limits": {"max_prompt_tokens": 1000000, "max_output_tokens": 64000},
|
|
},
|
|
},
|
|
{
|
|
"id": "gpt-4.1",
|
|
"capabilities": {
|
|
"type": "chat",
|
|
"limits": {"max_prompt_tokens": 128000, "max_output_tokens": 32768},
|
|
},
|
|
},
|
|
{
|
|
"id": "claude-sonnet-4",
|
|
"capabilities": {
|
|
"type": "chat",
|
|
"limits": {"max_prompt_tokens": 200000, "max_output_tokens": 64000},
|
|
},
|
|
},
|
|
{
|
|
"id": "model-without-limits",
|
|
"capabilities": {"type": "chat"},
|
|
},
|
|
{
|
|
"id": "model-zero-limit",
|
|
"capabilities": {
|
|
"type": "chat",
|
|
"limits": {"max_prompt_tokens": 0},
|
|
},
|
|
},
|
|
]
|
|
|
|
|
|
@pytest.fixture(autouse=True)
|
|
def _clear_cache():
|
|
"""Reset module-level cache before each test."""
|
|
import hermes_cli.models as mod
|
|
|
|
mod._copilot_context_cache = {}
|
|
mod._copilot_context_cache_time = 0.0
|
|
yield
|
|
mod._copilot_context_cache = {}
|
|
mod._copilot_context_cache_time = 0.0
|
|
|
|
|
|
class TestGetCopilotModelContext:
|
|
"""Tests for get_copilot_model_context()."""
|
|
|
|
@patch("hermes_cli.models.fetch_github_model_catalog", return_value=_SAMPLE_CATALOG)
|
|
def test_returns_max_prompt_tokens(self, mock_fetch):
|
|
assert get_copilot_model_context("claude-opus-4.6-1m") == 1_000_000
|
|
assert get_copilot_model_context("gpt-4.1") == 128_000
|
|
|
|
|
|
@patch("hermes_cli.models.fetch_github_model_catalog", return_value=_SAMPLE_CATALOG)
|
|
def test_cache_expires(self, mock_fetch):
|
|
import hermes_cli.models as mod
|
|
|
|
get_copilot_model_context("gpt-4.1")
|
|
assert mock_fetch.call_count == 1
|
|
|
|
# Expire the cache
|
|
mod._copilot_context_cache_time = time.time() - 7200
|
|
get_copilot_model_context("gpt-4.1")
|
|
assert mock_fetch.call_count == 2
|
|
|
|
|
|
@patch("hermes_cli.models.fetch_github_model_catalog", return_value=[])
|
|
def test_returns_none_for_empty_catalog(self, mock_fetch):
|
|
assert get_copilot_model_context("gpt-4.1") is None
|
|
|
|
|
|
class TestModelMetadataCopilotIntegration:
|
|
"""Test that get_model_context_length() uses Copilot live API for copilot provider."""
|
|
|
|
@patch("hermes_cli.models.fetch_github_model_catalog", return_value=_SAMPLE_CATALOG)
|
|
def test_copilot_provider_uses_live_api(self, mock_fetch):
|
|
from agent.model_metadata import get_model_context_length
|
|
|
|
ctx = get_model_context_length("claude-opus-4.6-1m", provider="copilot")
|
|
assert ctx == 1_000_000
|
|
|
|
|