Simon
ef09d5b419
- DailyBarRepository.get_range_many / stream_range_many_columns 增加 adjust 参数 （默认 'none'）→ SQL 层过滤口径，消除 stock_daily 混 source/adjust 污染因子的风险 - ResearchSpec / SelectionQuery 增加 price_adjustment（none|qfq），随 config_snapshot 落库可溯源；ResearchService._load_daily 与 SelectionService 装配按口径取数 - tests/test_price_adjustment.py：repo 读取按 adjust 过滤（none/qfq 各自命中）、 spec 默认与字段记录；全量 pytest 通过
117 lines
4.3 KiB
Python
117 lines
4.3 KiB
Python
"""选股用例入口（ARCHITECTURE_v2 §14 Selection Engine · 业务层）。
|
||
|
||
- 输入：SelectionQuery（universe + method + factors/conditions + top_n/pct + as_of）
|
||
- 装配：股票池（universe 过滤）→ 行情长表（含预热窗口）→ Selection Engine
|
||
（method=score 因子评分 / method=condition 结构化条件）
|
||
- 输出：SelectionResult（可解释：factor_values / filter_status / selection_reason）
|
||
- 未来函数红线：行情只取 <= as_of；财务条件只取 announce_date <= as_of 的已公告值（v2 §9）
|
||
|
||
MVP 为同步执行（单日全市场因子/条件计算量轻）；如需异步可复用 Job 链路。
|
||
"""
|
||
|
||
from __future__ import annotations
|
||
|
||
from datetime import date, timedelta
|
||
|
||
import pandas as pd
|
||
|
||
from app.domain.entities.market import FinancialIndicator
|
||
from app.domain.entities.selection import SelectionQuery, SelectionResult
|
||
from app.domain.repositories.market import (
|
||
DailyBarRepository,
|
||
FinancialRepository,
|
||
StockRepository,
|
||
)
|
||
from app.quant.selection import (
|
||
condition_needed_columns,
|
||
factor_columns,
|
||
run_condition_selection,
|
||
run_score_selection,
|
||
)
|
||
from app.quant.service import filter_stocks, load_daily_df
|
||
|
||
_FUNDAMENTAL_PREFIX = "fundamental."
|
||
|
||
|
||
class SelectionService:
|
||
"""选股用例入口：select(query) → SelectionResult（当前或历史 as_of）。"""
|
||
|
||
def __init__(
|
||
self,
|
||
stock_repo: StockRepository,
|
||
daily_repo: DailyBarRepository,
|
||
financial_repo: FinancialRepository | None = None,
|
||
) -> None:
|
||
self._stock_repo = stock_repo
|
||
self._daily_repo = daily_repo
|
||
self._financial_repo = financial_repo
|
||
|
||
def select(self, query: SelectionQuery) -> SelectionResult:
|
||
as_of = query.as_of or date.today()
|
||
stocks = filter_stocks(self._stock_repo.list(), query.universe, as_of=as_of)
|
||
if not stocks:
|
||
return self._run(query, pd.DataFrame(), stocks, as_of, financial={})
|
||
symbols = [s.symbol for s in stocks]
|
||
if query.method == "score":
|
||
columns = sorted(factor_columns(query))
|
||
else:
|
||
columns = sorted(condition_needed_columns(query))
|
||
daily = load_daily_df(
|
||
self._daily_repo,
|
||
symbols,
|
||
as_of - timedelta(days=query.warmup_days),
|
||
as_of,
|
||
columns,
|
||
adjust=query.price_adjustment,
|
||
)
|
||
financial: dict[str, FinancialIndicator] = {}
|
||
if query.method == "condition" and self._uses_fundamental(query):
|
||
financial = self._load_financial(symbols, as_of)
|
||
return self._run(query, daily, stocks, as_of, financial)
|
||
|
||
# ---- 内部 ----
|
||
|
||
def _run(
|
||
self,
|
||
query: SelectionQuery,
|
||
daily: pd.DataFrame,
|
||
stocks: list,
|
||
as_of: date,
|
||
financial: dict[str, FinancialIndicator],
|
||
) -> SelectionResult:
|
||
if query.method == "score":
|
||
return run_score_selection(daily, query, as_of)
|
||
return run_condition_selection(daily, stocks, query, as_of, financial)
|
||
|
||
@staticmethod
|
||
def _uses_fundamental(query: SelectionQuery) -> bool:
|
||
for c in query.conditions:
|
||
if c.field.startswith(_FUNDAMENTAL_PREFIX) or (
|
||
c.ref is not None and c.ref.startswith(_FUNDAMENTAL_PREFIX)
|
||
):
|
||
return True
|
||
return False
|
||
|
||
def _load_financial(
|
||
self, symbols: list[str], as_of: date
|
||
) -> dict[str, FinancialIndicator]:
|
||
"""按 announce_date <= as_of 批量取财务，每 symbol 保留最新一版。"""
|
||
if self._financial_repo is None:
|
||
raise ValueError("condition 引用了 fundamental.* 字段，但未注入 FinancialRepository")
|
||
getter = getattr(self._financial_repo, "list_announced_many", None)
|
||
if getter is not None:
|
||
rows = list(getter(symbols, as_of))
|
||
else: # 回退逐只
|
||
rows = []
|
||
for sym in symbols:
|
||
rows.extend(self._financial_repo.list_announced(sym, as_of))
|
||
by_symbol: dict[str, FinancialIndicator] = {}
|
||
for row in rows:
|
||
cur = by_symbol.get(row.symbol)
|
||
if cur is None or (row.announce_date, row.report_date) > (
|
||
cur.announce_date,
|
||
cur.report_date,
|
||
):
|
||
by_symbol[row.symbol] = row
|
||
return by_symbol
|