chore(config): 添加环境变量示例及.gitignore更新
- 新增 .env.example,包含 Tushare API、钉钉机器人和PostgreSQL数据库配置模板 - 更新.gitignore,忽略本地配置文件如 .env.local 和 config_local.py - 添加对报表文件命名规则的支持,保留示例文件不忽略 - 删除废弃的 chart.py 及相关图表模块代码 - 新增 config/settings.py,实现从环境变量读取配置的统一接口 - 设置数据目录及缓存目录,确保目录存在,提高配置管理规范性
This commit is contained in:
190
core/common/utils.py
Normal file
190
core/common/utils.py
Normal file
@@ -0,0 +1,190 @@
|
||||
"""
|
||||
通用工具函数
|
||||
"""
|
||||
|
||||
import pandas as pd
|
||||
import numpy as np
|
||||
from datetime import datetime, timedelta
|
||||
from typing import Optional
|
||||
|
||||
|
||||
def format_date(date_str: str, output_format: str = "%Y-%m-%d") -> str:
|
||||
"""
|
||||
统一日期格式
|
||||
|
||||
Args:
|
||||
date_str: 输入日期字符串(支持 YYYY-MM-DD 或 YYYYMMDD)
|
||||
output_format: 输出格式
|
||||
|
||||
Returns:
|
||||
str: 格式化后的日期字符串
|
||||
"""
|
||||
# 尝试解析多种格式
|
||||
for fmt in ["%Y-%m-%d", "%Y%m%d", "%Y/%m/%d"]:
|
||||
try:
|
||||
dt = datetime.strptime(date_str, fmt)
|
||||
return dt.strftime(output_format)
|
||||
except ValueError:
|
||||
continue
|
||||
raise ValueError(f"无法解析日期格式: {date_str}")
|
||||
|
||||
|
||||
def get_date_range(
|
||||
start_date: Optional[str] = None,
|
||||
end_date: Optional[str] = None,
|
||||
lookback_days: int = 365,
|
||||
) -> tuple[str, str]:
|
||||
"""
|
||||
获取日期范围
|
||||
|
||||
Args:
|
||||
start_date: 开始日期,None则根据lookback_days计算
|
||||
end_date: 结束日期,None则使用今天
|
||||
lookback_days: 回溯天数
|
||||
|
||||
Returns:
|
||||
tuple: (start_date, end_date) 格式为 YYYY-MM-DD
|
||||
"""
|
||||
if end_date is None:
|
||||
end = datetime.now()
|
||||
else:
|
||||
end = datetime.strptime(format_date(end_date), "%Y-%m-%d")
|
||||
|
||||
if start_date is None:
|
||||
start = end - timedelta(days=lookback_days)
|
||||
else:
|
||||
start = datetime.strptime(format_date(start_date), "%Y-%m-%d")
|
||||
|
||||
return start.strftime("%Y-%m-%d"), end.strftime("%Y-%m-%d")
|
||||
|
||||
|
||||
def calculate_cagr(
|
||||
nav_series: pd.Series,
|
||||
method: str = "natural_days",
|
||||
) -> float:
|
||||
"""
|
||||
计算年化收益率(CAGR)
|
||||
|
||||
Args:
|
||||
nav_series: 净值序列(index=日期)
|
||||
method: 'natural_days' 或 'trading_days'
|
||||
|
||||
Returns:
|
||||
float: CAGR值
|
||||
"""
|
||||
total_return = nav_series.iloc[-1] / nav_series.iloc[0]
|
||||
|
||||
if method == "natural_days":
|
||||
days = (nav_series.index[-1] - nav_series.index[0]).days
|
||||
years = days / 365.0
|
||||
elif method == "trading_days":
|
||||
years = len(nav_series) / 252.0
|
||||
else:
|
||||
raise ValueError(f"不支持的CAGR计算方式: {method}")
|
||||
|
||||
if years <= 0:
|
||||
return 0.0
|
||||
|
||||
return total_return ** (1 / years) - 1
|
||||
|
||||
|
||||
def calculate_max_drawdown(nav_series: pd.Series) -> tuple[float, datetime, datetime]:
|
||||
"""
|
||||
计算最大回撤
|
||||
|
||||
Returns:
|
||||
tuple: (最大回撤比例, 回撤起始日, 回撤结束日)
|
||||
"""
|
||||
cummax = nav_series.cummax()
|
||||
drawdown = (nav_series - cummax) / cummax
|
||||
|
||||
max_dd = drawdown.min()
|
||||
end_idx = drawdown.idxmin()
|
||||
start_idx = nav_series[:end_idx].idxmax()
|
||||
|
||||
return max_dd, start_idx, end_idx
|
||||
|
||||
|
||||
def calculate_sharpe(
|
||||
returns: pd.Series,
|
||||
rf: float = 0.0,
|
||||
periods: int = 252,
|
||||
) -> float:
|
||||
"""
|
||||
计算年化夏普比率
|
||||
|
||||
Args:
|
||||
returns: 日收益率序列
|
||||
rf: 无风险利率(年化)
|
||||
periods: 年化系数
|
||||
|
||||
Returns:
|
||||
float: 夏普比率
|
||||
"""
|
||||
excess_returns = returns - rf / periods
|
||||
if excess_returns.std() == 0:
|
||||
return 0.0
|
||||
return excess_returns.mean() / excess_returns.std() * np.sqrt(periods)
|
||||
|
||||
|
||||
def resample_data(
|
||||
df: pd.DataFrame,
|
||||
timeframe: str,
|
||||
time_col: str = "time",
|
||||
) -> pd.DataFrame:
|
||||
"""
|
||||
对数据进行重采样
|
||||
|
||||
Args:
|
||||
df: 原始数据
|
||||
timeframe: 目标周期 ('1D', '1W', '1M', '1Y')
|
||||
time_col: 时间列名
|
||||
|
||||
Returns:
|
||||
DataFrame: 重采样后的数据
|
||||
"""
|
||||
timeframe_map = {
|
||||
"1D": "D",
|
||||
"1W": "W",
|
||||
"1M": "M",
|
||||
"3M": "3M",
|
||||
"1Y": "Y",
|
||||
}
|
||||
|
||||
if timeframe not in timeframe_map:
|
||||
return df
|
||||
|
||||
df = df.copy()
|
||||
if time_col in df.columns:
|
||||
df[time_col] = pd.to_datetime(df[time_col])
|
||||
df.set_index(time_col, inplace=True)
|
||||
|
||||
rule = timeframe_map[timeframe]
|
||||
|
||||
resampled = (
|
||||
df.resample(rule)
|
||||
.agg(
|
||||
{
|
||||
"open": "first",
|
||||
"high": "max",
|
||||
"low": "min",
|
||||
"close": "last",
|
||||
"volume": "sum",
|
||||
}
|
||||
)
|
||||
.dropna()
|
||||
)
|
||||
|
||||
return resampled.reset_index()
|
||||
|
||||
|
||||
def safe_divide(a: float, b: float, default: float = 0.0) -> float:
|
||||
"""安全除法,避免除以0"""
|
||||
return a / b if b != 0 else default
|
||||
|
||||
|
||||
def truncate_string(s: str, max_length: int = 50, suffix: str = "...") -> str:
|
||||
"""截断字符串"""
|
||||
if len(s) <= max_length:
|
||||
return s
|
||||
return s[: max_length - len(suffix)] + suffix
|
||||
Reference in New Issue
Block a user