mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-30 19:09:28 +00:00
39975613b1
Second, deeper pass over tools/gateway/hermes_cli plus first pass over the trees wave 1 missed (acp, acp_adapter, skills, computer_use, docker, dashboard, conformance, monitoring, secret_sources, hermes_state, providers). Same rubric as wave 1 (AGENTS.md test policy); security, alternation/caching invariants, issue-number regressions, and E2E kept. Real test-quality fixes found and rooted out along the way: - tests/tools/test_command_guards.py made real auxiliary-LLM HTTPS calls (DEFAULT_CONFIG smart-approval leaked in) — pinned approval mode=manual via autouse fixture: 17.4s → 0.4s. - test_model_switch_custom_providers.py / test_user_providers_model_switch.py silently probed live provider catalogs (~2s/test) — stubbed cached_provider_model_ids/provider_model_ids/fetch_api_models. - test_telegram_noise_filter.py: 15-platform copy-paste matrix over shared gateway.run logic → 3 representative platforms (55s → 3.9s). - test_gateway_shutdown.py: stop()'s 5s interrupt-deadline loop spun on MagicMock agents — interrupt.side_effect now clears _running_agents (22s → 1.0s). - test_gateway_inactivity_timeout.py poll-harness timings shrunk 3-5x (24s → 1.1s); test_mcp_stability.py backoff/SIGTERM-grace sleeps patched (15.4s → 2.5s); test_async_delegation.py negative-drain wait 5s → 0.5s. - test_telegram_init_deadline.py: loop-block margin restored to 1.0s with rationale comment — the watchdog-dump assertion needs the loop blocked well past deadline+grace under parallel load (flaked once in the 40-worker verification run at a 0.2s margin). Verification: full hermetic suite via scripts/run_tests.sh — 2,438 files, 21,718 tests passed, 0 failed, 293.9s wall. Suite totals vs original baseline: 46,820 → 19,757 test functions (−57.8%), wall 583.5s → 293.9s (−50%), subprocess CPU 13,564s → 11,623s.
108 lines
3.7 KiB
Python
108 lines
3.7 KiB
Python
"""Tests for the shared interactive-prompt formatting cores in BasePlatformAdapter.
|
||
|
||
Covers ``_format_exec_approval`` (template-attr driven exec-approval text),
|
||
``_format_choice_page`` (picker pagination core), ``_truncate_preview``, and
|
||
byte-parity of the rewired adapters (telegram/feishu/matrix) against their
|
||
historical inline formatting.
|
||
"""
|
||
|
||
import html as _html
|
||
|
||
from gateway.platforms.base import BasePlatformAdapter
|
||
|
||
|
||
def _bare(cls):
|
||
"""Bare instance without running __init__ (documented test pattern)."""
|
||
return object.__new__(cls)
|
||
|
||
|
||
class _DefaultAdapter(BasePlatformAdapter):
|
||
"""Concrete subclass using only base-class template attrs."""
|
||
|
||
async def connect(self): # pragma: no cover - not used
|
||
pass
|
||
|
||
async def disconnect(self): # pragma: no cover - not used
|
||
pass
|
||
|
||
async def get_chat_info(self, chat_id): # pragma: no cover - not used
|
||
return {}
|
||
|
||
async def send(self, *a, **k): # pragma: no cover - not used
|
||
raise NotImplementedError
|
||
|
||
|
||
class TestTruncatePreview:
|
||
def test_short_text_unchanged(self):
|
||
assert BasePlatformAdapter._truncate_preview("abc", 10) == "abc"
|
||
|
||
def test_exact_budget_unchanged(self):
|
||
assert BasePlatformAdapter._truncate_preview("x" * 10, 10) == "x" * 10
|
||
|
||
|
||
class TestFormatExecApproval:
|
||
def test_default_template(self):
|
||
ad = _bare(_DefaultAdapter)
|
||
text = ad._format_exec_approval("rm -rf /", "scary")
|
||
assert text == (
|
||
"⚠️ Command Approval Required\n\n"
|
||
"```\nrm -rf /\n```\n"
|
||
"Reason: scary"
|
||
)
|
||
|
||
|
||
def test_escape_hook_applied_to_command_and_reason(self):
|
||
class Escaping(_DefaultAdapter):
|
||
def _ea_escape(self, text: str) -> str:
|
||
return _html.escape(text)
|
||
|
||
ad = _bare(Escaping)
|
||
text = ad._format_exec_approval("echo <hi>", "a & b")
|
||
assert "echo <hi>" in text
|
||
assert "a & b" in text
|
||
|
||
|
||
class TestFormatChoicePage:
|
||
def test_single_page_no_page_info(self):
|
||
opts, meta = BasePlatformAdapter._format_choice_page([1, 2, 3], 0, 10)
|
||
assert opts == [1, 2, 3]
|
||
assert meta["page_info"] == ""
|
||
assert meta["total_pages"] == 1
|
||
assert meta["page"] == 0
|
||
|
||
|
||
def test_page_clamped_high(self):
|
||
opts, meta = BasePlatformAdapter._format_choice_page(list(range(25)), 99, 10)
|
||
assert meta["page"] == 2
|
||
assert opts == list(range(20, 25))
|
||
assert meta["page_info"] == " (21–25 of 25)"
|
||
|
||
|
||
class TestAdapterParity:
|
||
"""Rewired adapters produce byte-identical text vs their historical inline code."""
|
||
|
||
|
||
def test_telegram_pagination_parity(self):
|
||
"""_format_choice_page matches the old _build_*_keyboard arithmetic."""
|
||
|
||
def old(options, page, page_size):
|
||
total = len(options)
|
||
total_pages = max(1, (total + page_size - 1) // page_size)
|
||
page = max(0, min(page, total_pages - 1))
|
||
start = page * page_size
|
||
end = min(start + page_size, total)
|
||
page_info = f" ({start + 1}–{end} of {total})" if total_pages > 1 else ""
|
||
return options[start:end], page, total_pages, page_info
|
||
|
||
for n in (0, 1, 8, 9, 10, 25):
|
||
options = list(range(n))
|
||
for page in (-3, 0, 1, 2, 99):
|
||
for per in (8, 10):
|
||
o_opts, o_page, o_tp, o_info = old(options, page, per)
|
||
n_opts, meta = BasePlatformAdapter._format_choice_page(
|
||
options, page, per
|
||
)
|
||
assert n_opts == o_opts
|
||
assert meta["page"] == o_page
|
||
assert meta["total_pages"] == o_tp
|
||
assert meta["page_info"] == o_info
|