energy-data-trans/utils/df_utils/util.py

# -*- coding: utf-8 -*-
# @Time    : 2024/6/21
# @Author  : 魏志亮
import datetime

import pandas as pd


def estimate_time_interval(df, time_str):
    """
    :return: 查询时间间隔
    """
    begin = datetime.datetime.now()
    df1 = pd.DataFrame(df[time_str])
    df1['chazhi'] = df1[time_str].shift(-1) - df1[time_str]
    result = df1.sample(int(df1.shape[0] / 100))['chazhi'].value_counts().idxmax().seconds
    del df1
    return result


def get_time_space_count(start_time: datetime.datetime, end_time: datetime.datetime, time_space=1):
    """
    获取俩个时间之间的个数
    :return: 查询时间间隔
    """
    delta = end_time - start_time
    total_seconds = delta.days * 24 * 60 * 60 + delta.seconds

    return int(total_seconds / time_space) + 1


def calculate_time_difference(now: datetime.datetime, date: datetime.datetime):
    # 计算时间差
    delta = now - date

    # 分解timedelta对象
    years, days = divmod(delta.days, 365)
    months, days = divmod(days, 30)  # 这里假设一个月有30天
    hours, remainder = divmod(delta.seconds, 3600)
    minutes, seconds = divmod(remainder, 60)

    return years, months, days, hours, minutes, seconds


if __name__ == '__main__':
    df = pd.read_csv(r"D:\data\清理数据\密马风电场\test_11_test\minute\WOG00469.csv")
    df['time_stamp'] = pd.to_datetime(df['time_stamp'])
    space = estimate_time_interval(df, 'time_stamp')
    min = df['time_stamp'].min()
    max = df['time_stamp'].max()
    result = get_time_space_count(min, max, space)
    print(df.shape)
    print(space, min, max, result)

    year, monht, day, _, _, _ = calculate_time_difference(datetime.datetime.now(), max)
    print(f"{year}年{monht}月{day}天")