mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-15 14:22:43 +00:00
* test: deflake CI and dev-machine flaky tests in bulk Fixes ten distinct flake sources found by mining recent CI failures and running the full suite on a dev machine with real user state: CI-observed races: - tests/conftest.py live-system guard: allow signal 0 (pure liveness probe) through _guarded_kill/_guarded_killpg. psutil.pid_exists() probes a just-killed grandchild reparented to init; the subtree check fails for it and the guard RuntimeError'd test_entire_tree_is_sigkilled_not_just_parent intermittently on unrelated PRs. Hermeticity flakes (fail on dev machines with real state, pass on CI): - agent/coding_context.py: _marker_root() now skips the shared temp root (tempfile.gettempdir()) like it skips $HOME — a stray /tmp/package.json flipped every tmp_path test into the coding posture (9 failures in test_coding_context.py). - test_agent_guardrails.py: pin MAX_CONCURRENT_CHILDREN=3 via autouse monkeypatch instead of freezing the user's real config value at import time (import-time vs call-time config mismatch). - test_web_tools_config.py: TestCheckWebApiKey now neutralizes the ddgs package probe and registry providers — the optional ddgs package in a dev venv lit up the fallback backend. - test_credential_pool.py: block claude_code/hermes-oauth credential autodiscovery in the two pool-merge tests that assert exact id lists (a real ~/.claude/.credentials.json seeded an extra entry). - test_modal_sandbox_fixes.py: clear _permanent_approved / _session_approved — the user's real command_allowlist silently approved the guard-escalation commands under test. - test_setup_irc.py: stub prompt_checklist to select only the IRC row; the non-TTY cancel fallback re-ran the real configured platforms' interactive setup_fn, which hit input() under captured stdin. - test_doctor.py: TestGitHubTokenCheck now patches the module-level HERMES_HOME constant (the file's established pattern) instead of only setenv — doctor was running PRAGMA integrity_check against the real multi-GB state.db and blowing the 300s per-file budget. Latent atexit-duplication (same _enter_buffered_busy class as #34217): - test_undo_command.py: drop importlib.reload(tui_gateway.server) in fixture teardown; reload re-registers the module's atexit hooks. - test_session_platform_resolution.py: drop per-test reload of tui_gateway.server; every resolver reads env at call time. * test: sentinel model value in ignore-user-config fallback assertion With HERMES_IGNORE_USER_CONFIG=1, load_cli_config() falls back to the repo-root cli-config.yaml (untracked, gitignored). On a dev machine that file can legitimately set the same popular model the test hardcoded (anthropic/claude-sonnet-4.6), flipping the != assertion locally while CI (no cli-config.yaml) stayed green. Use an impossible sentinel model name instead.
151 lines
6.4 KiB
Python
151 lines
6.4 KiB
Python
"""Platform/source tagging for the desktop chat surface.
|
|
|
|
The desktop app's chat panel uses ``hermes serve`` (the ``tui_gateway``
|
|
backend), so every chat session historically got ``platform="tui"`` stamped
|
|
on it — even though the user is in a graphical chat surface, not a
|
|
terminal. That mis-tag is why the agent suggested TUI-only slash commands
|
|
(like ``/reload-mcp``) to desktop chat users.
|
|
|
|
These tests pin the env-var matrix that resolves the session platform at
|
|
``tui_gateway`` session-creation time:
|
|
|
|
HERMES_DESKTOP=1, HERMES_DESKTOP_TERMINAL unset -> platform="desktop"
|
|
HERMES_DESKTOP=1, HERMES_DESKTOP_TERMINAL=1 -> platform="tui" (embedded pane)
|
|
neither set -> platform="tui" (standalone)
|
|
|
|
The resolver helper is import-safe (no heavy module side effects) so it
|
|
can be unit-tested without spinning up the full gateway.
|
|
"""
|
|
|
|
import pytest
|
|
|
|
|
|
def _reload_resolver():
|
|
# Plain import — every resolver under test reads the env at CALL time, so
|
|
# no reload is needed. importlib.reload(tui_gateway.server) would
|
|
# re-register the module's atexit hooks (thread-pool shutdown +
|
|
# _shutdown_sessions) on every test; duplicated hooks race the stderr
|
|
# buffer at interpreter shutdown (Fatal Python error:
|
|
# _enter_buffered_busy) — same flake class as PR #34217. Name kept for
|
|
# the existing call sites.
|
|
import tui_gateway.server as _srv
|
|
return _srv
|
|
|
|
|
|
@pytest.fixture
|
|
def clean_env(monkeypatch):
|
|
monkeypatch.delenv("HERMES_DESKTOP", raising=False)
|
|
monkeypatch.delenv("HERMES_DESKTOP_TERMINAL", raising=False)
|
|
return monkeypatch
|
|
|
|
|
|
class TestResolveSessionPlatform:
|
|
def test_standalone_tui_neither_env_set(self, clean_env):
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_platform() == "tui"
|
|
|
|
def test_desktop_chat_backend_gets_desktop_tag(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_platform() == "desktop"
|
|
|
|
def test_desktop_embedded_terminal_pane_stays_tui(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
clean_env.setenv("HERMES_DESKTOP_TERMINAL", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_platform() == "tui"
|
|
|
|
def test_desktop_terminal_alone_means_standalone_tui(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP_TERMINAL", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_platform() == "tui"
|
|
|
|
@pytest.mark.parametrize("val", ["1", "true", "yes", "on", "TRUE", "Yes", "ON"])
|
|
def test_truthy_variants_recognized(self, clean_env, val):
|
|
clean_env.setenv("HERMES_DESKTOP", val)
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_platform() == "desktop"
|
|
|
|
@pytest.mark.parametrize("val", ["0", "false", "", "no", "off", "False"])
|
|
def test_falsy_variants_fall_back_to_tui(self, clean_env, val):
|
|
clean_env.setenv("HERMES_DESKTOP", val)
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_platform() == "tui"
|
|
|
|
def test_embedded_terminal_overrides_desktop_when_both_set(self, clean_env):
|
|
"""The terminal-pane qualifier must short-circuit the desktop-backend
|
|
marker. An embedded TUI is a TUI, not a desktop chat surface."""
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
clean_env.setenv("HERMES_DESKTOP_TERMINAL", "true")
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_platform() == "tui"
|
|
|
|
|
|
class TestResolveSessionSource:
|
|
def test_explicit_source_param_wins(self, clean_env):
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_source("telegram") == "telegram"
|
|
|
|
def test_explicit_empty_source_falls_back_to_env(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_source("") == "desktop"
|
|
|
|
def test_explicit_none_source_falls_back_to_env(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_source(None) == "desktop"
|
|
|
|
def test_no_env_no_param_defaults_to_tui(self, clean_env):
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_source(None) == "tui"
|
|
|
|
def test_embedded_terminal_default_is_tui(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
clean_env.setenv("HERMES_DESKTOP_TERMINAL", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_source(None) == "tui"
|
|
|
|
def test_explicit_source_param_resists_env_drift(self, clean_env):
|
|
"""A caller that explicitly passes source="cli" must not be silently
|
|
rewritten to "desktop" by env vars — the resolver only fills in the
|
|
default when one is missing."""
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_session_source("cli") == "cli"
|
|
|
|
|
|
class TestResolveAgentPlatform:
|
|
def test_explicit_desktop_source_drives_agent_platform_without_env(self, clean_env):
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_agent_platform("desktop") == "desktop"
|
|
|
|
def test_missing_source_falls_back_to_env_resolved_platform(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_agent_platform(None) == "desktop"
|
|
|
|
def test_explicit_tui_source_keeps_embedded_terminal_as_tui(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._resolve_agent_platform("tui") == "tui"
|
|
|
|
|
|
class TestSessionSourceFallback:
|
|
def test_session_source_uses_existing_session_value(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._session_source({"source": "telegram"}) == "telegram"
|
|
|
|
def test_session_source_defaults_to_desktop_under_desktop_backend(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._session_source({}) == "desktop"
|
|
assert _srv._session_source(None) == "desktop"
|
|
|
|
def test_session_source_defaults_to_tui_for_embedded_terminal(self, clean_env):
|
|
clean_env.setenv("HERMES_DESKTOP", "1")
|
|
clean_env.setenv("HERMES_DESKTOP_TERMINAL", "1")
|
|
_srv = _reload_resolver()
|
|
assert _srv._session_source({}) == "tui"
|
|
assert _srv._session_source(None) == "tui"
|