mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-31 19:16:29 +00:00
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).
67 lines
2.3 KiB
Python
67 lines
2.3 KiB
Python
"""Regression tests confirming the setup model/provider/reasoning pickers route
|
|
through the shared curses radiolist (ESC + arrow-key handling that works across
|
|
terminals, incl. Ghostty) instead of simple_term_menu.
|
|
|
|
Guards against silently regressing back to simple_term_menu, whose ESC/arrow
|
|
handling was unreliable in `hermes setup` (the provider->model sub-menu).
|
|
"""
|
|
from unittest.mock import patch
|
|
|
|
|
|
def test_prompt_model_selection_uses_curses_radiolist():
|
|
from hermes_cli.auth import _prompt_model_selection
|
|
from hermes_cli.curses_ui import radio_item_plain
|
|
|
|
seen = {}
|
|
|
|
def _fake(
|
|
title,
|
|
items,
|
|
*,
|
|
selected=0,
|
|
cancel_returns=None,
|
|
description=None,
|
|
searchable=False,
|
|
search_labels=None,
|
|
):
|
|
seen["title"] = title
|
|
seen["items"] = items
|
|
seen["search_labels"] = search_labels
|
|
return 1 # pick second model
|
|
|
|
with patch("hermes_cli.curses_ui.curses_radiolist", side_effect=_fake), \
|
|
patch("builtins.print"):
|
|
result = _prompt_model_selection(["model-a", "model-b"])
|
|
|
|
assert result == "model-b"
|
|
assert seen["title"] == "Select default model:"
|
|
# Items are the models plus the custom/skip entries. Model rows may be
|
|
# rich (text, style) segments for sale chrome — compare the plain text.
|
|
plain = [radio_item_plain(item) for item in seen["items"]]
|
|
assert plain[:2] == ["model-a", "model-b"]
|
|
assert "Skip (keep current)" in plain
|
|
assert seen["search_labels"] is not None
|
|
assert len(seen["search_labels"]) == len(seen["items"])
|
|
|
|
|
|
def test_prompt_model_selection_esc_cancels():
|
|
from hermes_cli.auth import _prompt_model_selection
|
|
|
|
# curses_radiolist returns the cancel sentinel (-1) on ESC.
|
|
with patch("hermes_cli.curses_ui.curses_radiolist", return_value=-1), \
|
|
patch("builtins.print"):
|
|
result = _prompt_model_selection(["model-a", "model-b"])
|
|
|
|
assert result is None
|
|
|
|
|
|
def test_reasoning_effort_uses_curses_radiolist():
|
|
from hermes_cli.main import _prompt_reasoning_effort_selection
|
|
|
|
with patch("hermes_cli.curses_ui.curses_radiolist", return_value=2), \
|
|
patch("builtins.print"):
|
|
result = _prompt_reasoning_effort_selection(["low", "medium", "high"], current_effort="")
|
|
|
|
assert result == "high"
|
|
|
|
|