yang/python
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

修正逻辑

Browse Source
This commit is contained in:
mianyang
2026-04-27 21:50:21 +08:00
parent f843197e27
commit 352be50107
3 changed files with 43 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files
limit_up.py
+39 -31
View File
@@ -1,16 +1,11 @@
import pandas as pd
import numpy as np
from sqlalchemy import create_engine, text
# --- 数据库配置 ---
# 请根据实际情况修改数据库连接信息
# DB_CONN_STR = 'mysql+pymysql://root:password@127.0.0.1:3306/stock_data'
def init_db(engine):
"""
初始化数据库表结构
1. 增加 symbol 列，位于 trade_date 之后
2. 建立 UNIQUE KEY 防止物理重复写入
"""
create_table_sql = """
CREATE TABLE IF NOT EXISTS `stock_limit_data` (
@@ -34,6 +29,7 @@ def analyze_and_save_limits(start_date, end_date):
"""
计算并在数据库中落地涨停、炸板详情
"""
# 数据库连接配置
engine = create_engine(
"mysql+pymysql://yangfan@8.159.129.156:10836/level?charset=utf8mb4",
connect_args={
@@ -42,40 +38,61 @@ def analyze_and_save_limits(start_date, end_date):
pool_pre_ping=True
)
# 0. 初始化检查表结构
init_db(engine)
# 1. 获取基础数据和行情数据
# 1. 获取行情数据
query = f"""
SELECT
d.trade_date, d.symbol, d.ts_code, b.name,
d.high, d.close, d.pre_close, b.market
FROM stock_daily d
JOIN stock_basic b ON d.ts_code = b.ts_code
WHERE d.trade_date > '{start_date}'
WHERE d.trade_date >= '{start_date}'
AND d.trade_date <= '{end_date}'
AND d.pre_close IS NOT NULL
ORDER BY d.trade_date ASC
"""
try:
df = pd.read_sql(query, engine)
if df.empty:
print(f"⚠️ 在指定区间 ({start_date} -> {end_date}] 内未获取到数据。")
print(f"⚠️ 区间 [{start_date} , {end_date}] 内未获取到数据。")
return
# 2. 涨停价计算规则
# 强制类型转换
df['pre_close'] = pd.to_numeric(df['pre_close'], errors='coerce')
df['high'] = pd.to_numeric(df['high'], errors='coerce')
df['close'] = pd.to_numeric(df['close'], errors='coerce')
# 2. 涨停价计算规则 (已取消 ST 判定)
def calc_limit(row):
m, p = str(row['market']), float(row['pre_close'])
# 主板10%，创业板/科创板20%，北交所30%
ratio = 1.2 if ('创业板' in m or '科创板' in m) else (1.3 if '北交所' in m else 1.1)
m = str(row['market'])
p = row['pre_close']
if pd.isna(p) or p <= 0: return 0.0
# 仅保留板块判定
if '创业板' in m or '科创板' in m:
ratio = 1.2
elif '北交所' in m:
ratio = 1.3
else:
ratio = 1.1 # 普通主板（含ST股现在也按10%计算）
# 采用标准四舍五入
return round(p * ratio, 2)
df['limit_price'] = df.apply(calc_limit, axis=1)
# 3. 筛选涨停(1)和炸板(2)
df['is_limit'] = (df['high'] == df['limit_price']) & (df['close'] == df['limit_price'])
df['is_broken'] = (df['high'] == df['limit_price']) & (df['close'] < df['limit_price'])
# 3. 筛选逻辑：利用 abs 差值判断，解决浮点数精度问题
# 只要最高价触及理论涨停价（误差0.01以内）即视为触板
is_at_limit = (df['high'] - df['limit_price']).abs() < 0.011
is_close_at_limit = (df['close'] - df['limit_price']).abs() < 0.011
df['is_limit'] = is_at_limit & is_close_at_limit
df['is_broken'] = is_at_limit & (~is_close_at_limit)
# 分离结果
limit_df = df[df['is_limit']].copy()
limit_df['status'] = 1
@@ -85,27 +102,23 @@ def analyze_and_save_limits(start_date, end_date):
save_df = pd.concat([limit_df, broken_df])
if not save_df.empty:
# 4. 字段清洗与对齐
final_df = save_df[[
'trade_date', 'symbol', 'ts_code', 'name',
'status', 'limit_price', 'close', 'market'
]]
# 5. 【核心防重逻辑】：开启事务，先删后插
# 4. 事务入库：先删后插
with engine.begin() as conn:
# 获取待写入的所有交易日期
target_dates = final_df['trade_date'].astype(str).unique()
date_list = ", ".join([f"'{d}'" for d in target_dates])
# 第一层逻辑防重：删除数据库中已存在的相同日期数据
delete_query = f"DELETE FROM stock_limit_data WHERE trade_date IN ({date_list})"
conn.execute(text(delete_query))
# 第二层物理防重：通过 to_sql 写入，若触发 Unique Key 冲突会抛出异常回滚
final_df.to_sql('stock_limit_data', conn, if_exists='append', index=False)
print(f"✅ 处理成功！涉及交易日: {len(target_dates)} 天")
print(f"📊 总计入库: {len(final_df)} 条记录（已自动去重并覆盖历史数据）。")
print(f"📊 记录详情: 总计 {len(final_df)} 条 (涨停: {len(limit_df)} / 炸板: {len(broken_df)})")
else:
print("ℹ️ 该时段内全市场无涨停或炸板记录。")
@@ -113,14 +126,9 @@ def analyze_and_save_limits(start_date, end_date):
print(f"❌ 运行失败: {e}")
# ============================================================
# 用户配置区
# ============================================================
if __name__ == "__main__":
# 开始日期：计算从该日期的下一个交易日开始
START_DATE = '2026-04-23'
# 结束日期
END_DATE = '2026-04-24'
# 配置日期区间
START_DATE = '2024-09-23'
END_DATE = '2026-04-27'
analyze_and_save_limits(START_DATE, END_DATE)