模拟餐饮订单流水

from faker import Faker
import random
import pandas as pd

# 使用Faker模拟订单编号
f = Faker(locale='zh_CN')


def get_order_count() -> int:
    """Generate the count of orders in one day randomly

    Returns:
        the count of strore orders in one day
    """
    # 如果要控制订单数据，可以在这里调整，15为下限，100为上限
    return random.randint(1, 5)


def get_amount() -> float:
    """Generate the amount of order randomly

    Returns:
        the amount of order
    """
    return round(random.random() * 200 + 200, 2)


def get_orders() -> list:
    order_count = get_order_count()
    data = [get_amount() for i in range(order_count)]
    return data


if __name__ == '__main__':
    # 获取店铺数据
    store_data = pd.read_excel('./store_data.xlsx')
    store_data['key'] = 1
    # 生成时间数据
    date = pd.DataFrame(
        pd.date_range(start='20220101', end='20221231', freq='D')
        .strftime('%Y-%m-%d')
        .values,
        columns=['time'],
        dtype='object'
    )
    date['key'] = 1
    # 合并数据，生成大约30w条数据
    orders = pd.merge(left=store_data, right=date,
                      how='outer', on=['key']).drop('key', axis=1)
    # 模拟订单数据
    orders['amount'] = orders.apply(
        func=lambda x: get_orders(),
        axis=1
    )
    # 扁平化订单数据
    orders = orders.explode('amount', ignore_index=True)
    # 设置订单id号
    orders['order_id'] = orders.reset_index()['index'].apply(
        lambda index: f'HDL-{index:0=7d}')
    # 筛选需要的字段
    orders = orders.loc[:, ['order_id', 'time', 'amount', '店铺编号']]
    # 按时间对结果进行排序
    orders.sort_values(by=['order_id'], inplace=True)
    # 对字段进行重命名
    orders.rename(columns={
        'order_id': '订单号',
        'time': '订单时间',
        'amount': '销售额'
    }, inplace=True)
    orders.to_excel('./orders.xlsx', index=False)
    # orders.to_csv('./orders.csv', sep=',', encoding='utf-8', index=False)