Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Bugfix/usage report #3075

Merged
merged 6 commits into from
Nov 23, 2024
Merged
Show file tree
Hide file tree
Changes from 5 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 8 additions & 8 deletions backend/ee/danswer/db/query_history.py
Original file line number Diff line number Diff line change
Expand Up @@ -20,30 +20,30 @@ def fetch_chat_sessions_eagerly_by_time(
end: datetime.datetime,
db_session: Session,
limit: int | None = 500,
initial_id: int | None = None,
initial_time: datetime.datetime | None = None,
) -> list[ChatSession]:
id_order = desc(ChatSession.id) # type: ignore
time_order = desc(ChatSession.time_created) # type: ignore
message_order = asc(ChatMessage.id) # type: ignore
time_order = desc(ChatSession.time_created)
message_order = asc(ChatMessage.id)

filters: list[ColumnElement | BinaryExpression] = [
ChatSession.time_created.between(start, end)
]

if initial_id:
filters.append(ChatSession.id < initial_id)
if initial_time:
filters.append(ChatSession.time_created > initial_time)

subquery = (
db_session.query(ChatSession.id, ChatSession.time_created)
.filter(*filters)
.order_by(id_order, time_order)
.order_by(ChatSession.id, time_order)
.distinct(ChatSession.id)
.limit(limit)
.subquery()
)

query = (
db_session.query(ChatSession)
.join(subquery, ChatSession.id == subquery.c.id) # type: ignore
.join(subquery, ChatSession.id == subquery.c.id)
.outerjoin(ChatMessage, ChatSession.id == ChatMessage.chat_session_id)
.options(
joinedload(ChatSession.user),
Expand Down
34 changes: 24 additions & 10 deletions backend/ee/danswer/db/usage_export.py
Original file line number Diff line number Diff line change
Expand Up @@ -19,11 +19,15 @@
def get_empty_chat_messages_entries__paginated(
db_session: Session,
period: tuple[datetime, datetime],
limit: int | None = 1,
initial_id: int | None = None,
) -> list[ChatMessageSkeleton]:
limit: int | None = 500,
initial_time: datetime | None = None,
) -> tuple[int, list[ChatMessageSkeleton]]:
chat_sessions = fetch_chat_sessions_eagerly_by_time(
period[0], period[1], db_session, limit=limit, initial_id=initial_id
start=period[0],
end=period[1],
db_session=db_session,
limit=limit,
initial_time=initial_time,
)

message_skeletons: list[ChatMessageSkeleton] = []
Expand All @@ -36,7 +40,7 @@ def get_empty_chat_messages_entries__paginated(
flow_type = FlowType.CHAT

for message in chat_session.messages:
# only count user messages
# Only count user messages
if message.message_type != MessageType.USER:
continue

Expand All @@ -49,24 +53,34 @@ def get_empty_chat_messages_entries__paginated(
time_sent=message.time_sent,
)
)
if len(chat_sessions) == 0:
return None, []

return message_skeletons
return chat_sessions[0].time_created, message_skeletons


def get_all_empty_chat_message_entries(
db_session: Session,
period: tuple[datetime, datetime],
) -> Generator[list[ChatMessageSkeleton], None, None]:
initial_id = None
initial_time = period[0]
ind = 0
while True:
message_skeletons = get_empty_chat_messages_entries__paginated(
db_session, period, initial_id=initial_id
ind += 1

time_created, message_skeletons = get_empty_chat_messages_entries__paginated(
db_session,
period,
initial_time=initial_time,
)

if not message_skeletons:
return

yield message_skeletons
initial_id = message_skeletons[-1].chat_session_id

# Update initial_time for the next iteration
initial_time = time_created


def get_all_usage_reports(db_session: Session) -> list[UsageReportMetadata]:
Expand Down
Loading