claude_dev
1ed7b72aca
根因: daily_dir mixed-adjust(hfq bulk+akshare raw tail)致3-30 -94%假跌。 方案(Linus三问简化单raw, 除权留分期项#3): - datareader read_parquet_daily/15min 加 dir_key 参数 - data_source iter_bars/fetch_day: adjust=raw→raw_dir(缺配置报错防混源), qfq→daily_dir - engine PaperEngine 默认 adjust=raw - config 加 raw_dir; scripts/raw_redownload.py 新浪源adjust='' 直连+单线程限速 - 验证: 浦发606行close 6.5/14.6 mean10.08 0跳变, 撮合成交价9.71-10.25真实 - 测试9/9+trader全量108/108通过
81 lines
3.0 KiB
Python
81 lines
3.0 KiB
Python
"""模拟盘行情统一接口（raw 真实价 / qfq 前复权，spec §3.3 / §5）。
|
||
|
||
raw：撮合/涨跌停/成交价用真实价（adjustflag=3 / akshare adjust=""），
|
||
从 cfg.data_paths['raw_dir'] 读 parquet（raw 重下脚本生成，单一 adjust 不混源）。
|
||
qfq：策略信号用前复权（无除权缺口），从 cfg.data_paths['daily_dir'] 读。
|
||
|
||
_read_fn 内 lazy import datareader，避免模块级依赖 vnpy 链（tzlocal 等），
|
||
本机无 vnpy 完整依赖时仍可 import + 单测（mock _read_fn）。
|
||
"""
|
||
import logging
|
||
from collections.abc import Iterator
|
||
|
||
logger = logging.getLogger(__name__)
|
||
|
||
|
||
def _read_fn(interval: str):
|
||
"""按 interval 返回读取函数（lazy import）。"""
|
||
if interval == "d":
|
||
from sanguo_data.datareader import read_parquet_daily
|
||
return read_parquet_daily
|
||
if interval == "15m":
|
||
from sanguo_data.datareader import read_parquet_15min
|
||
return read_parquet_15min
|
||
raise ValueError(f"不支持的 interval: {interval}（首版仅 d / 15m）")
|
||
|
||
|
||
def _resolve_dir_key(adjust: str, interval: str) -> str:
|
||
"""adjust → cfg.data_paths 的目录 key。
|
||
|
||
raw 仅支持日线（raw 15min 待分红除权分期项）；qfq/默认按 interval 选日线/15min。
|
||
"""
|
||
if adjust == "raw":
|
||
if interval != "d":
|
||
raise ValueError(
|
||
f"raw 模式暂仅支持日线（interval='d'），got '{interval}'"
|
||
"（raw 15min 待分期项）"
|
||
)
|
||
return "raw_dir"
|
||
return "daily_dir" if interval == "d" else "minute_15_dir"
|
||
|
||
|
||
def _check_raw_cfg(adjust: str, cfg) -> None:
|
||
"""raw 模式需 raw_dir 配置，缺失明确报错（不静默 fallback 到 qfq，避免混源）。"""
|
||
if adjust == "raw" and (not cfg or "raw_dir" not in getattr(cfg, "data_paths", {})):
|
||
raise ValueError(
|
||
"raw 模式需 cfg.data_paths['raw_dir']（未配置；"
|
||
"先用 scripts/data_platform/raw_redownload.py 生成 raw parquet）"
|
||
)
|
||
|
||
|
||
def iter_bars(
|
||
symbols: list[str],
|
||
start: str,
|
||
end: str,
|
||
interval: str,
|
||
adjust: str = "qfq",
|
||
cfg=None,
|
||
) -> Iterator[tuple]:
|
||
"""按日期 cross-section yield (date, {symbol: BarData})。"""
|
||
dir_key = _resolve_dir_key(adjust, interval)
|
||
_check_raw_cfg(adjust, cfg)
|
||
read_fn = _read_fn(interval)
|
||
by_date: dict = {}
|
||
for sym in symbols:
|
||
for bar in read_fn(sym, start, end, cfg, dir_key):
|
||
dt = bar.datetime
|
||
key = dt.date() if hasattr(dt, "date") else dt
|
||
by_date.setdefault(key, {})[sym] = bar
|
||
for date in sorted(by_date.keys()):
|
||
yield date, by_date[date]
|
||
|
||
|
||
def fetch_day(symbol: str, date: str, interval: str,
|
||
adjust: str = "qfq", cfg=None):
|
||
"""实走模式拉当日 bar（C-S3 用）。"""
|
||
dir_key = _resolve_dir_key(adjust, interval)
|
||
_check_raw_cfg(adjust, cfg)
|
||
read_fn = _read_fn(interval)
|
||
bars = read_fn(symbol, date, date, cfg, dir_key)
|
||
return bars[-1] if bars else None
|