mirrors/hermes-agent
1
0
Fork 0
mirror of https://github.com/NousResearch/hermes-agent.git synced 2026-05-30 06:41:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 15

feat(state.db): persist platform_message_id; restore yuanbao exact-id recall

Browse source 
PR #29211 dropped JSONL gateway transcripts and noted that the platform's
own `message_id` field (used by Yuanbao's recall guard to redact a
message by exact platform id) was no longer preserved — falling back to
content-match.  That fallback works for the common case but redacts the
wrong row when two messages share text (or fails to match when content
is post-processed).

Restore exact-id matching by giving state.db a column for it:

- New `platform_message_id TEXT` column on the messages table
  (SCHEMA_VERSION bump 11 → 12; column added via declarative reconciler
  on existing DBs, no version-gated migration block needed)
- Partial index `idx_messages_platform_msg_id` on
  (session_id, platform_message_id) to keep recall's point-lookup cheap
  even on large sessions
- `append_message()` and `replace_messages()` accept the new value:
  the gateway-facing `append_to_transcript` in `gateway/session.py`
  forwards either `message["platform_message_id"]` or the legacy
  `message["message_id"]` key (yuanbao's existing convention)
- `get_messages_as_conversation()` surfaces the column back on the
  message dict as `message_id` so platform code reads the same shape
  it used to read from JSONL
- Yuanbao `_patch_transcript`: restore branch A1 (exact id match)
  ahead of A2 (content match) ahead of B (system-note).  Both branches
  log which one fired so operators can tell from gateway.log whether
  recall hit the canonical path or had to fall back.

Tests:
- New low-level round-trip tests in `test_hermes_state.py` for both
  `append_message` and `replace_messages` paths
- The PR's `test_yuanbao_recall_db_only.py` was rewritten to assert
  the new contract: branch A1 (id match) works against DB-only
  transcripts, and branch A2 (content match) still recovers rows that
  were observed without a platform id (e.g. agent-processed @bot
  messages where run.py doesn't carry msg_id through)
This commit is contained in:
Teknium 2026-05-20 12:55:01 -07:00
parent 0cc1a1d2d9
commit 31a0100104
5 changed files with 185 additions and 38 deletions
Download patch file Download diff file Expand all files Collapse all files

89
tests/gateway/platforms/test_yuanbao_recall_db_only.py
View file

@ -1,31 +1,88 @@
"""Yuanbao recall: branch A (content-match) works against DB-only transcripts."""
"""Yuanbao recall: branch A1 (exact id) and A2 (content-match) against DB-only transcripts.
state.db persists the platform-side ``message_id`` via the
``platform_message_id`` column (added in the salvage of PR #29211) and
``load_transcript`` surfaces it back on each message dict as ``message_id``
— so the recall guard's exact-id match path stays canonical even with the
JSONL file gone. When a row has no platform id (e.g. agent-processed
@bot messages whose adapter didn't carry a msg_id, or pre-column legacy
rows), recall falls through to content-match.
"""
from gateway.session import SessionStore
from gateway.config import GatewayConfig
def test_recall_content_match_finds_target_in_db_transcript(tmp_path, monkeypatch):
"""state.db doesn't preserve message_id, so recall uses content-match.
Pin DEFAULT_DB_PATH to tmp_path so SessionDB() can't write to the real
~/.hermes/state.db. (Module-level constant snapshot, see test_load_transcript_db_only.)
"""
def _pin_db(monkeypatch, tmp_path):
"""Force SessionDB() to write into tmp_path instead of the real ~/.hermes."""
import hermes_state
monkeypatch.setattr(hermes_state, "DEFAULT_DB_PATH", tmp_path / "state.db")
def test_recall_branch_a1_exact_id_match_round_trips_through_db(tmp_path, monkeypatch):
"""A user message persisted with ``message_id`` must round-trip through
state.db so recall can find and redact it by exact id (branch A1)."""
_pin_db(monkeypatch, tmp_path)
config = GatewayConfig()
store = SessionStore(sessions_dir=tmp_path, config=config)
sid = "test-yuanbao-recall"
sid = "test-yuanbao-recall-a1"
store._db.create_session(session_id=sid, source="yuanbao:group:G")
store.append_to_transcript(sid, {"role": "user", "content": "sensitive content", "timestamp": 1.0})
store.append_to_transcript(sid, {"role": "assistant", "content": "ack", "timestamp": 2.0})
store.append_to_transcript(sid, {
"role": "user",
"content": "sensitive content",
"timestamp": 1.0,
"message_id": "platform-msg-abc",
})
store.append_to_transcript(sid, {
"role": "assistant",
"content": "ack",
"timestamp": 2.0,
})
# DB-only history carries no platform message_id (PR #29211 dropped that path).
history = store.load_transcript(sid)
assert all("message_id" not in msg for msg in history)
# The user row must carry its platform id back so the recall guard can
# match by exact id; the assistant row had no platform id so it should
# not gain one spuriously.
user_msg = next(m for m in history if m["role"] == "user")
assistant_msg = next(m for m in history if m["role"] == "assistant")
assert user_msg.get("message_id") == "platform-msg-abc"
assert "message_id" not in assistant_msg
# Branch A: content match finds the target row that recall would redact.
target = next((m for m in history
if m.get("role") == "user" and m.get("content") == "sensitive content"), None)
# Branch A1: locate the row by exact platform id — no content heuristics.
target = next(
(m for m in history if m.get("message_id") == "platform-msg-abc"),
None,
)
assert target is not None
assert target["content"] == "sensitive content"
def test_recall_branch_a2_content_match_when_no_platform_id(tmp_path, monkeypatch):
"""Rows that lack a platform_message_id (e.g. agent-processed @bot
messages) still match by content as a fallback."""
_pin_db(monkeypatch, tmp_path)
config = GatewayConfig()
store = SessionStore(sessions_dir=tmp_path, config=config)
sid = "test-yuanbao-recall-a2"
store._db.create_session(session_id=sid, source="yuanbao:group:G")
# No message_id on the dict — simulates an agent-processed message
# that did not carry the platform msg_id through.
store.append_to_transcript(sid, {
"role": "user",
"content": "sensitive content",
"timestamp": 1.0,
})
history = store.load_transcript(sid)
assert all("message_id" not in m for m in history)
# Branch A2: content match recovers the target.
target = next(
(m for m in history
if m.get("role") == "user" and m.get("content") == "sensitive content"),
None,
)
assert target is not None
# Caller would then redact: target["content"] = REDACTED; store.rewrite_transcript(sid, history)

45
tests/test_hermes_state.py
View file

@ -316,6 +316,42 @@ class TestMessageStorage:
assert conv[0] == {"role": "user", "content": "Hello"}
assert conv[1] == {"role": "assistant", "content": "Hi!"}
def test_platform_message_id_round_trips(self, db):
"""Platform-side message ids (yuanbao msg_id, telegram update_id, …)
survive append → get_messages_as_conversation under the
``message_id`` key so platform recall flows can match by exact id."""
db.create_session(session_id="s_pmi", source="yuanbao")
db.append_message(
"s_pmi",
role="user",
content="hi",
platform_message_id="abc-123",
)
db.append_message("s_pmi", role="assistant", content="hello")
conv = db.get_messages_as_conversation("s_pmi")
user_msg = next(m for m in conv if m["role"] == "user")
assistant_msg = next(m for m in conv if m["role"] == "assistant")
assert user_msg.get("message_id") == "abc-123"
# Assistant row had no platform id — must not gain one spuriously.
assert "message_id" not in assistant_msg
def test_replace_messages_preserves_platform_message_id(self, db):
"""``rewrite_transcript`` (which goes through replace_messages) must
keep the platform_message_id round-trip working for /retry, /undo,
/compress and yuanbao's recall rewrite path."""
db.create_session(session_id="s_rep", source="yuanbao")
db.replace_messages(
"s_rep",
[
{"role": "user", "content": "x", "message_id": "ext-1"},
{"role": "assistant", "content": "y"},
],
)
conv = db.get_messages_as_conversation("s_rep")
assert next(m for m in conv if m["role"] == "user").get("message_id") == "ext-1"
assert "message_id" not in next(m for m in conv if m["role"] == "assistant")
def test_get_messages_as_conversation_includes_ancestor_chain(self, db):
db.create_session("root", "tui")
db.append_message("root", role="user", content="first prompt")
@ -1462,9 +1498,10 @@ class TestSchemaInit:
assert "schema_version" in tables
def test_schema_version(self, db):
from hermes_state import SCHEMA_VERSION
cursor = db._conn.execute("SELECT version FROM schema_version")
version = cursor.fetchone()[0]
assert version == 11
assert version == SCHEMA_VERSION
def test_title_column_exists(self, db):
"""Verify the title column was created in the sessions table."""
@ -1760,8 +1797,9 @@ class TestSchemaInit:
migrated_db = SessionDB(db_path=db_path)
# Verify migration
from hermes_state import SCHEMA_VERSION
cursor = migrated_db._conn.execute("SELECT version FROM schema_version")
assert cursor.fetchone()[0] == 11
assert cursor.fetchone()[0] == SCHEMA_VERSION
# Verify title column exists and is NULL for existing sessions
session = migrated_db.get_session("existing")
@ -2952,11 +2990,12 @@ class TestFTS5ToolCallMigration:
assert len(session_db.search_messages("LEGACYARG")) == 1, \
"v11 migration must backfill tool_calls JSON into FTS"
# schema_version bumped
from hermes_state import SCHEMA_VERSION
row = session_db._conn.execute(
"SELECT version FROM schema_version LIMIT 1"
).fetchone()
version = row["version"] if hasattr(row, "keys") else row[0]
assert version == 11
assert version == SCHEMA_VERSION
finally:
session_db.close()