mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-31 19:16:29 +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).
121 lines
3.9 KiB
Python
121 lines
3.9 KiB
Python
"""Tests for _detect_tool_failure + _trim_error + get_cute_tool_message
|
|
inline failure suffix rendering.
|
|
|
|
Covers the user-visible promise: when a tool fails, the CLI shows a short,
|
|
specific reason in square brackets at the end of the completion line —
|
|
not a generic "[error]".
|
|
"""
|
|
|
|
import json
|
|
|
|
from agent.display import (
|
|
_detect_tool_failure,
|
|
_trim_error,
|
|
_ERROR_SUFFIX_MAX_LEN,
|
|
get_cute_tool_message,
|
|
)
|
|
|
|
|
|
class TestTrimError:
|
|
"""The helper that shrinks an error message for inline display."""
|
|
|
|
def test_short_message_unchanged(self):
|
|
assert _trim_error("nope") == "nope"
|
|
|
|
|
|
def test_long_message_truncated_to_cap(self):
|
|
msg = "x" * 200
|
|
trimmed = _trim_error(msg)
|
|
assert len(trimmed) <= _ERROR_SUFFIX_MAX_LEN
|
|
assert trimmed.endswith("...")
|
|
|
|
def test_file_not_found_path_collapsed_to_filename(self):
|
|
long_path = "File not found: /home/teknium/.hermes/hermes-agent/very/deep/path/foo.py"
|
|
assert _trim_error(long_path) == "File not found: foo.py"
|
|
|
|
|
|
|
|
|
|
class TestDetectToolFailureTerminal:
|
|
"""terminal: non-zero exit_code is the canonical failure signal."""
|
|
|
|
def test_success_returns_no_suffix(self):
|
|
result = json.dumps({"output": "ok\n", "exit_code": 0})
|
|
assert _detect_tool_failure("terminal", result) == (False, "")
|
|
|
|
|
|
def test_nonzero_exit_with_error_shows_message(self):
|
|
result = json.dumps({
|
|
"output": "",
|
|
"exit_code": 127,
|
|
"error": "ls: cannot access 'foo': No such file or directory",
|
|
})
|
|
is_failure, suffix = _detect_tool_failure("terminal", result)
|
|
assert is_failure is True
|
|
assert "cannot access" in suffix
|
|
# Trimmed to the cap, in brackets
|
|
assert suffix.startswith(" [")
|
|
assert suffix.endswith("]")
|
|
|
|
|
|
|
|
|
|
class TestDetectToolFailureMemory:
|
|
"""memory: 'full' is distinct from real errors."""
|
|
|
|
def test_memory_full_returns_full_suffix(self):
|
|
result = json.dumps({"success": False, "error": "would exceed the limit"})
|
|
assert _detect_tool_failure("memory", result) == (True, " [full]")
|
|
|
|
def test_memory_other_error_returns_specific_message(self):
|
|
# An error that's NOT a "full" overflow falls through to the
|
|
# structured-error path and surfaces the actual message.
|
|
result = json.dumps({"success": False, "error": "invalid action: zap"})
|
|
is_failure, suffix = _detect_tool_failure("memory", result)
|
|
assert is_failure is True
|
|
assert "invalid action" in suffix
|
|
|
|
|
|
class TestDetectToolFailureStructured:
|
|
"""Generic path: any tool that returns {"error": ...} JSON."""
|
|
|
|
def test_read_file_error_surfaced(self):
|
|
result = json.dumps({
|
|
"path": "/nope/missing.py",
|
|
"success": False,
|
|
"error": "File not found: /nope/missing.py",
|
|
})
|
|
is_failure, suffix = _detect_tool_failure("read_file", result)
|
|
assert is_failure is True
|
|
# _trim_error reduces the path to the basename.
|
|
assert suffix == " [File not found: missing.py]"
|
|
|
|
|
|
|
|
def test_successful_result_not_flagged(self):
|
|
result = json.dumps({"success": True, "data": "hello"})
|
|
assert _detect_tool_failure("web_search", result) == (False, "")
|
|
|
|
|
|
|
|
class TestGetCuteToolMessageFailureSuffix:
|
|
"""End-to-end: failure suffix is appended by get_cute_tool_message."""
|
|
|
|
|
|
|
|
|
|
def test_memory_full_suffix(self):
|
|
fail = json.dumps({"success": False, "error": "would exceed the limit"})
|
|
line = get_cute_tool_message(
|
|
"memory",
|
|
{"action": "add", "target": "memory", "content": "x"},
|
|
0.05,
|
|
result=fail,
|
|
)
|
|
assert "[full]" in line
|
|
|
|
def test_success_has_no_suffix(self):
|
|
ok = json.dumps({"success": True, "data": "hi"})
|
|
line = get_cute_tool_message("web_search", {"query": "hi"}, 0.2, result=ok)
|
|
assert "[" not in line.split("0.2s", 1)[1]
|
|
|