cirry/stock
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

看股功能更新

Browse Source
This commit is contained in:
cirry
2026-08-15 15:36:11 +08:00
parent c1c43d2ff7
commit 9cce670b74
26 changed files with 957 additions and 427 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
backend/app/data/aggregation.py
View File

@@ -4,6 +4,7 @@
MVP 在应用层用 pandas resample 即可，逻辑等价、便于切换。
OHLCV 聚合规则：开=周期内首根开、高=最高、低=最低、收=末根收、量=求和。
成交额/换手率为名义量：求和（全缺则保持 None，不伪造 0）。
"""
from __future__ import annotations
@@ -22,6 +23,13 @@ def bars_per_year(timeframe: str) -> int:
return _BARS_PER_YEAR.get(timeframe, 252)
def _sum_or_none(s: pd.Series):
"""求和；全为 NaN 返回 None（部分缺失则忽略缺失项求和）。"""
if s.isna().all():
return None
return float(s.sum())
def resample_bars(bars: list[Bar], timeframe: str) -> list[Bar]:
"""把日线 bars 聚合为目标周期；日线或未知周期原样返回。"""
if not bars or timeframe in ("1d", "d", "day", "", None):
@@ -31,14 +39,18 @@ def resample_bars(bars: list[Bar], timeframe: str) -> list[Bar]:
return bars
df = pd.DataFrame(
[{"ts": b.ts, "open": b.open, "high": b.high, "low": b.low, "close": b.close, "volume": b.volume}
[{"ts": b.ts, "open": b.open, "high": b.high, "low": b.low, "close": b.close,
"volume": b.volume,
"amount": b.amount if b.amount is not None else float("nan"),
"turnover": b.turnover if b.turnover is not None else float("nan")}
for b in bars]
).set_index("ts").sort_index()
agg = (
df.resample(rule)
.agg({"open": "first", "high": "max", "low": "min", "close": "last", "volume": "sum"})
.dropna()
.agg({"open": "first", "high": "max", "low": "min", "close": "last",
"volume": "sum", "amount": _sum_or_none, "turnover": _sum_or_none})
.dropna(subset=["open"])
)
return [
@@ -49,6 +61,8 @@ def resample_bars(bars: list[Bar], timeframe: str) -> list[Bar]:
low=float(row["low"]),
close=float(row["close"]),
volume=float(row["volume"]),
amount=row["amount"] if row["amount"] == row["amount"] else None, # NaN -> None
turnover=row["turnover"] if row["turnover"] == row["turnover"] else None,
)
for ts, row in agg.iterrows()
]

2
backend/app/data/akshare_provider.py
View File

@@ -27,12 +27,14 @@ def fetch_daily(code: str, start: str = "20200101", end: str | None = None,
bars: list[Bar] = []
for _, r in df.iterrows():
amt = r.get("成交额")
bars.append(
Bar(
ts=datetime.strptime(str(r["日期"]), "%Y-%m-%d"),
open=float(r["开盘"]), high=float(r["最高"]),
low=float(r["最低"]), close=float(r["收盘"]),
volume=float(r["成交量"]) * 100.0, # AKShare 成交量单位为手 -> 股
amount=float(amt) if amt is not None and amt == amt else None, # AKShare 成交额单位为元
)
)
return bars

61
backend/app/data/fetcher.py
View File

@@ -1,12 +1,15 @@
"""数据编排：拉取（Tushare 主 -> AKShare 兜底）+ 本地缓存。
真实行情落库到 candles 表（timeframe='1d'），回测统一从库读。
真实行情落库到 candles 表（timeframe='1d'，**不复权底座**），回测统一从库读。
复权（qfq/hfq）在读取时按 adj_factor 表本地换算，见 api._adjust_bars。
"""
from __future__ import annotations
import asyncio
from sqlalchemy import delete, func, select
from sqlalchemy import func, select
from sqlalchemy.dialects.postgresql import insert as pg_insert
from sqlalchemy.ext.asyncio import AsyncSession
from ..config import settings
@@ -40,6 +43,13 @@ async def is_cached(session: AsyncSession, symbol: str) -> bool:
return await count_cached(session, symbol) > 0
async def _last_cached_ts(session: AsyncSession, symbol: str):
res = await session.execute(
select(func.max(Candle.ts)).where(Candle.symbol == symbol, Candle.timeframe == "1d")
)
return res.scalar()
async def sync_symbol(
session: AsyncSession,
code: str,
@@ -48,34 +58,57 @@ async def sync_symbol(
source: str = "auto",
force: bool = False,
) -> dict:
"""拉取并缓存某标的日线。已缓存且非 force 时直接返回缓存计数。"""
if not force and await is_cached(session, code):
"""增量拉取并 upsert 某标的日线（**不复权**底座）。
- 永不删除已有行：按 (symbol, timeframe, ts) 主键 upsert，
不会把 TDX 导入的 30 年历史冲掉；
- 已缓存时从最后一根的次日开始增量拉取（force 仅跳过「有缓存就返回」
的短路，用于缓存落后于最新交易日时的刷新）；
- 拉不到新行时保持原缓存不动。
"""
last_ts = await _last_cached_ts(session, code)
if last_ts is not None and not force and not start:
return {"symbol": code, "bars": await count_cached(session, code), "source": "cache"}
if last_ts is not None and not start:
# 增量：从缓存最后一根当天开始（重叠一天重新拉取，容忍数据源漏行/盘后修订）
start = last_ts.strftime("%Y%m%d")
start = start or DEFAULT_START
adjust = settings.data_adjust
errors: list[str] = []
bars: list[Bar] = []
used = None
for name, fn in _providers(source):
try:
# tushare/akshare 是同步网络 IO，丢到线程池避免阻塞事件循环
bars = await asyncio.to_thread(fn, code, start, end, adjust)
# tushare/akshare 是同步网络 IO，丢到线程池避免阻塞事件循环；
# adjust=None -> 不复权（复权在读取时按 adj_factor 换算）
bars = await asyncio.to_thread(fn, code, start, end, None)
used = name
break
except Exception as e: # noqa: BLE001
errors.append(f"{name}: {e}")
if not bars:
if last_ts is not None:
# 增量失败（如停牌/新股无新行）：保留缓存，不算错误
return {"symbol": code, "bars": await count_cached(session, code), "source": "cache"}
raise RuntimeError("所有数据源均失败 -> " + " | ".join(errors) if errors else "无可用数据源")
# 全量替换该标的日线（避免重复主键）
await session.execute(delete(Candle).where(Candle.symbol == code, Candle.timeframe == "1d"))
for b in bars:
session.add(
Candle(symbol=code, timeframe="1d", ts=b.ts, open=b.open, high=b.high,
low=b.low, close=b.close, volume=b.volume)
)
# upsert：不 delete，避免破坏既有底座（TDX 全量历史）
stmt = pg_insert(Candle).values([
{"symbol": code, "timeframe": "1d", "ts": b.ts, "open": b.open, "high": b.high,
"low": b.low, "close": b.close, "volume": b.volume,
"amount": b.amount, "turnover": b.turnover}
for b in bars
])
stmt = stmt.on_conflict_do_update(
index_elements=["symbol", "timeframe", "ts"],
set_={"open": stmt.excluded.open, "high": stmt.excluded.high, "low": stmt.excluded.low,
"close": stmt.excluded.close, "volume": stmt.excluded.volume,
# 增量源缺失额/换手时保留库里的旧值（如 TDX 已回补的 30 年成交额）
"amount": func.coalesce(stmt.excluded.amount, Candle.amount),
"turnover": func.coalesce(stmt.excluded.turnover, Candle.turnover)},
)
await session.execute(stmt)
await session.commit()
return {"symbol": code, "bars": len(bars), "source": used}

35
backend/app/data/repository.py
View File

@@ -32,3 +32,38 @@ async def get_candles(
stmt = stmt.order_by(Candle.ts.asc()).limit(limit)
result = await session.execute(stmt)
return list(result.scalars().all())
async def get_recent_candles(
session: AsyncSession,
symbol: str,
timeframe: str = "1d",
limit: int = 5000,
) -> list[Candle]:
"""取最近 limit 根 K 线（含最新交易日），按时间升序返回。"""
stmt = (
select(Candle)
.where(Candle.symbol == symbol, Candle.timeframe == timeframe)
.order_by(Candle.ts.desc())
.limit(limit)
)
result = await session.execute(stmt)
return list(reversed(result.scalars().all()))
async def get_candles_before(
session: AsyncSession,
symbol: str,
timeframe: str,
before: datetime,
limit: int = 5000,
) -> list[Candle]:
"""取 before 之前（不含）的最近 limit 根 K 线，按时间升序返回（历史向前翻页用）。"""
stmt = (
select(Candle)
.where(Candle.symbol == symbol, Candle.timeframe == timeframe, Candle.ts < before)
.order_by(Candle.ts.desc())
.limit(limit)
)
result = await session.execute(stmt)
return list(reversed(result.scalars().all()))

2
backend/app/data/tushare_provider.py
View File

@@ -40,12 +40,14 @@ def fetch_daily(code: str, start: str = "20200101", end: str | None = None,
df = df.sort_values("trade_date")
bars: list[Bar] = []
for _, r in df.iterrows():
amt = r.get("amount")
bars.append(
Bar(
ts=_parse(r["trade_date"]),
open=float(r["open"]), high=float(r["high"]),
low=float(r["low"]), close=float(r["close"]),
volume=float(r["vol"]) * 100.0, # Tushare vol 单位为手 -> 股
amount=float(amt) * 1000.0 if amt is not None and amt == amt else None, # 千元 -> 元
)
)
return bars