You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
125 lines
4.1 KiB
Python
125 lines
4.1 KiB
Python
"""Feature orchestrator: combines market, event, and financial features."""
|
|
from __future__ import annotations
|
|
|
|
import datetime as dt
|
|
|
|
from sqlalchemy import select
|
|
from sqlalchemy.ext.asyncio import AsyncSession
|
|
|
|
from libs.common.logging import get_logger
|
|
from libs.db.models import Event, EventParse, FeatureSnapshot
|
|
from libs.features.event_features import compute_event_features
|
|
from libs.features.financial_features import compute_financial_features
|
|
from libs.features.market_features import compute_market_features
|
|
from libs.oracle_client.financial import FinancialService
|
|
from libs.oracle_client.price import PriceService
|
|
|
|
logger = get_logger(__name__)
|
|
|
|
SNAPSHOT_VERSION = "1.0.0"
|
|
|
|
|
|
async def build_features_for_event(
|
|
session: AsyncSession,
|
|
event: Event,
|
|
price_service: PriceService,
|
|
financial_service: FinancialService | None = None,
|
|
) -> tuple[FeatureSnapshot, FeatureSnapshot] | None:
|
|
"""Build market_v1 and event_v1 feature snapshots for an event.
|
|
|
|
Returns (market_snapshot, event_snapshot) or None on failure.
|
|
"""
|
|
# Get ticker from event (via symbol)
|
|
if not event.symbol_id:
|
|
logger.warning("event_no_symbol", event_id=event.event_id)
|
|
return None
|
|
|
|
# Extract ticker from symbol_id (format: SYM::{ticker}::{venue})
|
|
parts = event.symbol_id.split("::")
|
|
ticker = parts[1] if len(parts) >= 2 else None
|
|
if not ticker:
|
|
logger.warning("event_bad_symbol_id", event_id=event.event_id, symbol_id=event.symbol_id)
|
|
return None
|
|
|
|
event_date_str = event.event_date.isoformat()
|
|
|
|
# Fetch price bars from Stock Oracle
|
|
try:
|
|
# 30 trading days before event
|
|
start_date = (event.event_date - dt.timedelta(days=45)).isoformat()
|
|
end_date = (event.event_date + dt.timedelta(days=5)).isoformat()
|
|
price_response = await price_service.get_daily_bars(
|
|
ticker, start=start_date, end=end_date
|
|
)
|
|
bars = price_response.bars
|
|
except Exception as exc:
|
|
logger.error("price_fetch_failed", event_id=event.event_id, error=str(exc))
|
|
return None
|
|
|
|
# Compute market features
|
|
mf = compute_market_features(bars, event_date_str)
|
|
|
|
# Get latest valid event parse
|
|
result = await session.execute(
|
|
select(EventParse)
|
|
.where(EventParse.event_id == event.event_id)
|
|
.where(EventParse.validation_status == "valid")
|
|
.order_by(EventParse.event_parse_id.desc())
|
|
.limit(1)
|
|
)
|
|
parse = result.scalar_one_or_none()
|
|
|
|
if parse is None:
|
|
logger.warning("no_valid_parse", event_id=event.event_id)
|
|
return None
|
|
|
|
ef = compute_event_features(parse.output_json)
|
|
|
|
market_snapshot = FeatureSnapshot(
|
|
event_id=event.event_id,
|
|
snapshot_name="market_v1",
|
|
snapshot_version=SNAPSHOT_VERSION,
|
|
feature_json=mf,
|
|
)
|
|
event_snapshot = FeatureSnapshot(
|
|
event_id=event.event_id,
|
|
snapshot_name="event_v1",
|
|
snapshot_version=SNAPSHOT_VERSION,
|
|
feature_json=ef,
|
|
)
|
|
|
|
session.add(market_snapshot)
|
|
session.add(event_snapshot)
|
|
await session.flush()
|
|
|
|
# Optional: financial features (non-fatal if unavailable)
|
|
if financial_service is not None:
|
|
try:
|
|
fin_response = await financial_service.get_financial_data(ticker)
|
|
ff = compute_financial_features(fin_response)
|
|
if ff:
|
|
financial_snapshot = FeatureSnapshot(
|
|
event_id=event.event_id,
|
|
snapshot_name="financial_v1",
|
|
snapshot_version=SNAPSHOT_VERSION,
|
|
feature_json=ff,
|
|
)
|
|
session.add(financial_snapshot)
|
|
await session.flush()
|
|
logger.info("financial_features_built", event_id=event.event_id, ticker=ticker)
|
|
except Exception as exc:
|
|
logger.warning(
|
|
"financial_features_skipped",
|
|
event_id=event.event_id,
|
|
ticker=ticker,
|
|
error=str(exc),
|
|
)
|
|
|
|
logger.info(
|
|
"features_built",
|
|
event_id=event.event_id,
|
|
ticker=ticker,
|
|
market_features=list(mf.keys()),
|
|
)
|
|
return market_snapshot, event_snapshot
|