hermes-agent/tests/run_agent/test_compression_boundary_hook.py
Teknium 6b81590c55
test: prune low-value tests suite-wide (wave 1) — 46,820 → 28,106 test functions
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).
2026-07-29 13:10:23 -07:00

325 lines
13 KiB
Python

"""Test: the context engine is notified of a compression-boundary rollover.
When _compress_context rotates session_id (compression split), the active
context engine receives on_session_start(new_sid, boundary_reason="compression",
old_session_id=<old>). This lets plugin engines (e.g. hermes-lcm) preserve
DAG lineage across the split instead of treating it as a fresh /new.
See hermes-lcm#68: after Hermes compresses and mints a new physical session,
LCM was losing continuity (compression_count: 1, store_messages: 0,
dag_nodes: 0). With boundary_reason="compression" plugins can distinguish
this from a real user-initiated /new.
"""
import os
import tempfile
from pathlib import Path
from unittest.mock import MagicMock, patch
import pytest
from agent.conversation_compression import (
finalize_context_engine_compression_notification,
)
class TestCompressionBoundaryHook:
def _make_agent(self, session_db):
with patch.dict(os.environ, {"OPENROUTER_API_KEY": "test-key"}):
from run_agent import AIAgent
agent = AIAgent(
api_key="test-key",
base_url="https://openrouter.ai/api/v1",
model="test/model",
quiet_mode=True,
session_db=session_db,
session_id="original-session",
skip_context_files=True,
skip_memory=True,
)
# ROTATION fallback — pin in_place=False regardless of default (#38763).
agent.compression_in_place = False
return agent
def test_on_session_start_called_with_compression_boundary(self):
from hermes_state import SessionDB
with tempfile.TemporaryDirectory() as tmpdir:
db = SessionDB(db_path=Path(tmpdir) / "test.db")
agent = self._make_agent(db)
# Stub the context compressor: we only need to observe the hook.
compressor = MagicMock()
compressor.compress.return_value = [
{"role": "user", "content": "[CONTEXT COMPACTION] summary"},
{"role": "user", "content": "tail question"},
]
compressor.compression_count = 1
compressor.last_prompt_tokens = 0
compressor.last_completion_tokens = 0
# Avoid the summary-error warning path
compressor._last_summary_error = None
# MagicMock auto-creates truthy attrs; explicitly clear the abort
# flag so the post-compress abort branch in
# conversation_compression.py does not short-circuit before the
# session-id rotation we are asserting on.
compressor._last_compress_aborted = False
agent.context_compressor = compressor
original_sid = agent.session_id
messages = [
{"role": "user", "content": f"m{i}"} for i in range(10)
]
agent._compress_context(messages, "sys", approx_tokens=10_000)
# Session_id rotated
assert agent.session_id != original_sid, \
"compression should rotate session_id when session_db is set"
# Hook fired with boundary_reason="compression" and old_session_id
calls = [
c for c in compressor.on_session_start.call_args_list
]
assert calls, "on_session_start was never called on the context engine"
# Find the compression boundary call (there may be others from init)
comp_calls = [
c for c in calls
if c.kwargs.get("boundary_reason") == "compression"
]
assert comp_calls, (
f"Expected an on_session_start call with "
f"boundary_reason='compression', got {calls!r}"
)
call = comp_calls[-1]
# Positional new session_id
assert call.args and call.args[0] == agent.session_id, \
f"Expected new session_id as first positional arg, got {call!r}"
assert call.kwargs.get("old_session_id") == original_sid, \
f"Expected old_session_id={original_sid!r}, got {call.kwargs!r}"
assert len(comp_calls) == 1
def test_automatic_notification_follows_core_persistence(self):
from hermes_state import SessionDB
events = []
with tempfile.TemporaryDirectory() as tmpdir:
db = SessionDB(db_path=Path(tmpdir) / "test.db")
agent = self._make_agent(db)
compressor = MagicMock()
compressor.compress.return_value = [
{"role": "user", "content": "summary"}
]
compressor.compression_count = 1
compressor.last_prompt_tokens = 0
compressor.last_completion_tokens = 0
compressor._last_summary_error = None
compressor._last_compress_aborted = False
compressor.on_session_start.side_effect = (
lambda *_args, **kwargs: events.append(
kwargs.get("boundary_reason")
)
)
agent.context_compressor = compressor
original_publish = db.publish_compression_child
def _record_publish(*args, **kwargs):
result = original_publish(*args, **kwargs)
events.append("persist")
return result
with patch.object(
db, "publish_compression_child", side_effect=_record_publish
):
agent._compress_context(
[{"role": "user", "content": "request"}],
"sys",
approx_tokens=100,
)
assert events == ["persist", "compression"]
def test_failure_before_persistence_does_not_notify(self):
from hermes_state import SessionDB
with tempfile.TemporaryDirectory() as tmpdir:
db = SessionDB(db_path=Path(tmpdir) / "test.db")
agent = self._make_agent(db)
compressor = MagicMock()
compressor.compress.side_effect = RuntimeError("synthetic compression failure")
agent.context_compressor = compressor
with pytest.raises(RuntimeError, match="synthetic compression failure"):
agent._compress_context(
[{"role": "user", "content": "request"}],
"sys",
approx_tokens=100,
)
compressor.on_session_start.assert_not_called()
def test_no_progress_does_not_notify(self):
from hermes_state import SessionDB
with tempfile.TemporaryDirectory() as tmpdir:
db = SessionDB(db_path=Path(tmpdir) / "test.db")
agent = self._make_agent(db)
compressor = MagicMock()
compressor.compress.side_effect = lambda messages, **_kwargs: messages
compressor._last_compress_aborted = False
agent.context_compressor = compressor
messages = [{"role": "user", "content": "request"}]
returned, _ = agent._compress_context(
messages,
"sys",
approx_tokens=100,
)
assert returned is messages
compressor.on_session_start.assert_not_called()
def test_no_hook_when_no_session_db(self):
"""Without session_db, session_id does not rotate and the hook is not fired."""
from run_agent import AIAgent
with patch.dict(os.environ, {"OPENROUTER_API_KEY": "test-key"}):
agent = AIAgent(
api_key="test-key",
base_url="https://openrouter.ai/api/v1",
model="test/model",
quiet_mode=True,
session_db=None,
session_id="original-session",
skip_context_files=True,
skip_memory=True,
)
compressor = MagicMock()
compressor.compress.return_value = [{"role": "user", "content": "x"}]
compressor.compression_count = 1
compressor.last_prompt_tokens = 0
compressor.last_completion_tokens = 0
compressor._last_summary_error = None
agent.context_compressor = compressor
original_sid = agent.session_id
agent._compress_context([{"role": "user", "content": "m"}], "sys", approx_tokens=100)
# No DB => no rotation => no compression-boundary hook
assert agent.session_id == original_sid
comp_calls = [
c for c in compressor.on_session_start.call_args_list
if c.kwargs.get("boundary_reason") == "compression"
]
assert not comp_calls, (
f"No compression hook should fire without session_db rotation, "
f"got {comp_calls!r}"
)
def test_hook_failure_does_not_break_compression(self):
"""If the context engine raises from on_session_start, compression still completes."""
from hermes_state import SessionDB
with tempfile.TemporaryDirectory() as tmpdir:
db = SessionDB(db_path=Path(tmpdir) / "test.db")
agent = self._make_agent(db)
compressor = MagicMock()
compressor.compress.return_value = [{"role": "user", "content": "summary"}]
compressor.compression_count = 1
compressor.last_prompt_tokens = 0
compressor.last_completion_tokens = 0
compressor._last_summary_error = None
compressor._last_compress_aborted = False
# Raise only on the compression-boundary call, not on earlier calls.
def _raise_on_compression(*args, **kwargs):
if kwargs.get("boundary_reason") == "compression":
raise RuntimeError("plugin exploded")
return None
compressor.on_session_start.side_effect = _raise_on_compression
agent.context_compressor = compressor
original_sid = agent.session_id
# Must not raise
compressed, _prompt = agent._compress_context(
[{"role": "user", "content": "m"}], "sys", approx_tokens=100
)
assert compressed
assert agent.session_id != original_sid
class TestSessionCompressEvent:
"""The session:compress event_callback fires after a compression split."""
def _make_agent(self, session_db, event_callback=None):
with patch.dict(os.environ, {"OPENROUTER_API_KEY": "test-key"}):
from run_agent import AIAgent
agent = AIAgent(
api_key="test-key",
base_url="https://openrouter.ai/api/v1",
model="test/model",
quiet_mode=True,
session_db=session_db,
session_id="original-session",
skip_context_files=True,
skip_memory=True,
event_callback=event_callback,
)
# ROTATION fallback — pin in_place=False regardless of default (#38763).
agent.compression_in_place = False
return agent
def _stub_compressor(self):
compressor = MagicMock()
compressor.compress.return_value = [
{"role": "user", "content": "[CONTEXT COMPACTION] summary"},
{"role": "user", "content": "tail"},
]
compressor.compression_count = 1
compressor.last_prompt_tokens = 0
compressor.last_completion_tokens = 0
compressor._last_summary_error = None
compressor._last_compress_aborted = False
return compressor
def test_event_emitted_on_compression(self):
from hermes_state import SessionDB
events = []
with tempfile.TemporaryDirectory() as tmpdir:
db = SessionDB(db_path=Path(tmpdir) / "test.db")
agent = self._make_agent(
db, event_callback=lambda et, ctx: events.append((et, ctx))
)
original_sid = agent.session_id
agent.context_compressor = self._stub_compressor()
agent._compress_context(
[{"role": "user", "content": f"m{i}"} for i in range(10)],
"sys",
approx_tokens=10_000,
)
compress_events = [e for e in events if e[0] == "session:compress"]
assert compress_events, f"session:compress not emitted, got {events!r}"
_, ctx = compress_events[-1]
assert ctx["session_id"] == agent.session_id
assert ctx["old_session_id"] == original_sid
assert ctx["compression_count"] == 1
def test_no_callback_is_safe(self):
"""Compression must work when no event_callback is wired."""
from hermes_state import SessionDB
with tempfile.TemporaryDirectory() as tmpdir:
db = SessionDB(db_path=Path(tmpdir) / "test.db")
agent = self._make_agent(db, event_callback=None)
agent.context_compressor = self._stub_compressor()
compressed, _ = agent._compress_context(
[{"role": "user", "content": "m"}], "sys", approx_tokens=100
)
assert compressed