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.
102 lines
3.2 KiB
Python
102 lines
3.2 KiB
Python
"""Regression tests for _rewrite_compound_background.
|
|
|
|
Context: bash parses ``A && B &`` as ``(A && B) &`` — it forks a subshell
|
|
for the compound and backgrounds the subshell. Inside the subshell, B
|
|
runs foreground, so the subshell waits for B. When B never exits on its
|
|
own (HTTP servers, ``yes > /dev/null``, etc.), the subshell is stuck in
|
|
``wait4`` forever and leaks as an orphan process. Pre-fix, we saw this
|
|
pattern leak processes across the fleet (vela, sal, combiagent).
|
|
|
|
The rewriter fixes this by wrapping the tail in a brace group —
|
|
``A && { B & }`` — so B runs as a simple backgrounded command inside
|
|
the current shell. No subshell fork, no wait.
|
|
"""
|
|
|
|
|
|
from tools.terminal_tool import _rewrite_compound_background as rewrite
|
|
|
|
|
|
class TestRewrites:
|
|
"""Commands that trigger the subshell-wait bug MUST be rewritten."""
|
|
|
|
def test_simple_and_background(self):
|
|
assert rewrite("A && B &") == "A && { B & }"
|
|
|
|
def test_or_background(self):
|
|
assert rewrite("A || B &") == "A || { B & }"
|
|
|
|
|
|
def test_multiple_rewrites_in_one_script(self):
|
|
cmd = "A && B &\nfalse || C &"
|
|
assert rewrite(cmd) == "A && { B & }\nfalse || { C & }"
|
|
|
|
|
|
class TestPreserved:
|
|
"""Commands that DON'T have the bug MUST pass through unchanged."""
|
|
|
|
def test_simple_background(self):
|
|
# No compound — just background a single command. Works fine as-is.
|
|
assert rewrite("sleep 5 &") == "sleep 5 &"
|
|
|
|
def test_plain_server_background(self):
|
|
assert rewrite("python3 -m http.server 0 &") == "python3 -m http.server 0 &"
|
|
|
|
|
|
def test_whitespace_only(self):
|
|
assert rewrite(" \n\t") == " \n\t"
|
|
|
|
|
|
class TestRedirectsNotConfused:
|
|
"""``&>``, ``2>&1``, ``>&2`` must not be mistaken for background ``&``."""
|
|
|
|
def test_amp_gt_redirect_alone(self):
|
|
assert rewrite("echo hi &>/dev/null") == "echo hi &>/dev/null"
|
|
|
|
|
|
def test_gt_amp_inside_compound(self):
|
|
cmd = "A && B 2>&1 &"
|
|
assert rewrite(cmd) == "A && { B 2>&1 & }"
|
|
|
|
|
|
class TestQuotingAndParens:
|
|
"""Shell metacharacters inside quotes/parens must not be parsed as operators."""
|
|
|
|
def test_and_and_inside_single_quotes(self):
|
|
cmd = "echo 'A && B &'"
|
|
assert rewrite(cmd) == "echo 'A && B &'"
|
|
|
|
|
|
def test_backslash_escaped_ampersand(self):
|
|
# Escaped & is not a background operator.
|
|
cmd = r"echo A \&\& B"
|
|
assert rewrite(cmd) == cmd
|
|
|
|
def test_comment_line_not_rewritten(self):
|
|
cmd = "# A && B &\nC"
|
|
assert rewrite(cmd) == "# A && B &\nC"
|
|
|
|
|
|
class TestIdempotence:
|
|
"""Running the rewriter twice should be a no-op on its own output."""
|
|
|
|
def test_already_rewritten(self):
|
|
once = rewrite("A && B &")
|
|
twice = rewrite(once)
|
|
assert once == twice
|
|
assert twice == "A && { B & }"
|
|
|
|
def test_multiline_idempotent(self):
|
|
once = rewrite("cd /tmp && server &\nsleep 1")
|
|
assert rewrite(once) == once
|
|
|
|
|
|
class TestEdgeCases:
|
|
def test_only_chain_op_no_second_command(self):
|
|
# Malformed input: bash would error, we shouldn't crash or rewrite.
|
|
cmd = "A && &"
|
|
# Don't assert a specific output; just don't raise.
|
|
rewrite(cmd)
|
|
|
|
|
|
def test_tabs_between_tokens(self):
|
|
assert rewrite("A\t&&\tB\t&") == "A\t&&\t{ B\t& }"
|