mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-30 19:09:28 +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).
108 lines
3.3 KiB
Python
108 lines
3.3 KiB
Python
"""Regression tests for live Codex app-server events.
|
|
|
|
The history projector is completion-only. These tests protect the parallel
|
|
display bridge (make_codex_app_server_event_bridge) that makes deltas and
|
|
tool cards visible before resume: it fires both the tool_progress bubbles
|
|
AND the authoritative stable-ID tool_start/tool_complete callbacks the TUI
|
|
tool cards depend on.
|
|
|
|
Grafted from PR #65412 (@HaiderSultanArc) onto the merged bridge.
|
|
"""
|
|
|
|
from types import SimpleNamespace
|
|
|
|
from agent.codex_runtime import (
|
|
_codex_item_completion_payload,
|
|
make_codex_app_server_event_bridge,
|
|
)
|
|
from agent.transports.codex_event_projector import _deterministic_call_id
|
|
|
|
|
|
def _recording_agent():
|
|
calls = {
|
|
"stream": [],
|
|
"reasoning": [],
|
|
"tool_progress": [],
|
|
"tool_start": [],
|
|
"tool_complete": [],
|
|
}
|
|
agent = SimpleNamespace(
|
|
_fire_stream_delta=lambda text: calls["stream"].append(text),
|
|
_fire_reasoning_delta=lambda text: calls["reasoning"].append(text),
|
|
tool_progress_callback=lambda *args, **kwargs: calls["tool_progress"].append((
|
|
args,
|
|
kwargs,
|
|
)),
|
|
tool_start_callback=lambda call_id, name, args: calls["tool_start"].append((
|
|
call_id,
|
|
name,
|
|
args,
|
|
)),
|
|
tool_complete_callback=lambda call_id, name, args, result: calls[
|
|
"tool_complete"
|
|
].append((call_id, name, args, result)),
|
|
_emit_interim_assistant_message=None,
|
|
show_commentary=True,
|
|
)
|
|
return agent, calls
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_stable_ids_match_history_projector():
|
|
"""The bridge's stable call ids mirror CodexEventProjector so a live
|
|
TUI tool card correlates with the projected history entry after
|
|
resume."""
|
|
agent, calls = _recording_agent()
|
|
bridge = make_codex_app_server_event_bridge(agent)
|
|
|
|
mcp = {
|
|
"type": "mcpToolCall",
|
|
"id": "m1",
|
|
"server": "filesystem",
|
|
"tool": "read",
|
|
"arguments": {"path": "a.py"},
|
|
}
|
|
bridge({"method": "item/started", "params": {"item": mcp}})
|
|
call_id, name, args = calls["tool_start"][0]
|
|
assert call_id == _deterministic_call_id("mcp__filesystem__read", "m1")
|
|
assert name == "mcp.filesystem.read"
|
|
assert args == {"path": "a.py"}
|
|
|
|
calls["tool_start"].clear()
|
|
patch = {
|
|
"type": "fileChange",
|
|
"id": "p1",
|
|
"changes": [{"kind": {"type": "add"}, "path": "a.py"}],
|
|
}
|
|
bridge({"method": "item/started", "params": {"item": patch}})
|
|
call_id, name, args = calls["tool_start"][0]
|
|
assert call_id == _deterministic_call_id("apply_patch", "p1")
|
|
assert name == "apply_patch"
|
|
assert args == {"changes": [{"kind": "add", "path": "a.py"}]}
|
|
|
|
|
|
def test_failed_command_result_and_error_flag_are_preserved():
|
|
agent, calls = _recording_agent()
|
|
bridge = make_codex_app_server_event_bridge(agent)
|
|
item = {
|
|
"type": "commandExecution",
|
|
"id": "failed",
|
|
"command": "false",
|
|
"aggregatedOutput": "boom",
|
|
"exitCode": 2,
|
|
}
|
|
|
|
bridge({"method": "item/completed", "params": {"item": item}})
|
|
|
|
result, is_error = _codex_item_completion_payload(item)
|
|
assert result == "[exit 2]\nboom"
|
|
assert is_error is True
|
|
assert calls["tool_progress"][0][1]["is_error"] is True
|
|
assert calls["tool_complete"][0][3] == "[exit 2]\nboom"
|
|
|
|
|
|
|
|
|