perf(state): batch message hydration in export_all
This commit is contained in:
@@ -258,7 +258,22 @@ class SessionPortabilityMixin:
|
||||
|
||||
def export_all(self, source: str = None) -> List[Dict[str, Any]]:
|
||||
"""Export all sessions (with messages) as dicts, e.g. for JSONL backup."""
|
||||
return [self._with_messages(s) for s in self.search_sessions(source=source, limit=100000)]
|
||||
sessions = self.search_sessions(source=source, limit=100000)
|
||||
messages_by_session = {session["id"]: [] for session in sessions}
|
||||
session_ids = list(messages_by_session)
|
||||
# Stay below SQLite's legacy 999-variable limit while replacing the per-session N+1 reads.
|
||||
for start in range(0, len(session_ids), 900):
|
||||
chunk = session_ids[start:start + 900]
|
||||
rows = self._read_all(
|
||||
f"SELECT * FROM messages WHERE session_id IN ({','.join('?' for _ in chunk)}) "
|
||||
"AND active = 1 ORDER BY session_id, id",
|
||||
chunk,
|
||||
)
|
||||
for row in rows:
|
||||
messages_by_session[row["session_id"]].append(
|
||||
self._row_to_message_dict(row, warn_context="get_messages", summary_flag=True)
|
||||
)
|
||||
return [{**session, "messages": messages_by_session[session["id"]]} for session in sessions]
|
||||
|
||||
def adopt_session_lineage_from(self, donor_db: Any, session_id: str, *, retire_donor: bool = True) -> Dict[str, Any]:
|
||||
"""Adopt *session_id*'s full compression lineage from *donor_db* (stranded-bot-session
|
||||
|
||||
@@ -0,0 +1,41 @@
|
||||
from hermes_state import SessionDB
|
||||
|
||||
|
||||
def test_export_all_batches_message_reads_without_changing_export_rows(tmp_path, monkeypatch):
|
||||
db = SessionDB(tmp_path / "state.db")
|
||||
try:
|
||||
for index, source in enumerate(("cli", "telegram", "cli", "cli")):
|
||||
session_id = f"session-{index}"
|
||||
db.create_session(session_id=session_id, source=source)
|
||||
db.append_messages_batch(
|
||||
session_id,
|
||||
[
|
||||
{"role": "user", "content": f"question {index}"},
|
||||
{
|
||||
"role": "assistant",
|
||||
"content": f"answer {index}",
|
||||
"tool_calls": [{"id": f"call-{index}", "type": "function"}],
|
||||
},
|
||||
],
|
||||
)
|
||||
|
||||
sessions = db.search_sessions(source="cli", limit=100000)
|
||||
expected = [
|
||||
{**session, "messages": db.get_messages(session["id"])}
|
||||
for session in sessions
|
||||
]
|
||||
|
||||
original_read_all = db._read_all
|
||||
read_calls = 0
|
||||
|
||||
def counted_read_all(*args, **kwargs):
|
||||
nonlocal read_calls
|
||||
read_calls += 1
|
||||
return original_read_all(*args, **kwargs)
|
||||
|
||||
monkeypatch.setattr(db, "_read_all", counted_read_all)
|
||||
|
||||
assert db.export_all(source="cli") == expected
|
||||
assert read_calls <= 2
|
||||
finally:
|
||||
db.close()
|
||||
Reference in New Issue
Block a user