每天更新python中的数据帧并应用可能的更改

Question

我有一个连接到数据库的数据帧。数据帧每天都在更新，我想写一些代码来比较今天的数据帧和昨天的数据帧。如果今天的数据帧中有任何更改，那么将其用于其余代码，否则只需停止运行代码并退出。我有下面的代码，但我不知道将昨天的数据帧保存在某个地方以将其和今天的数据帧进行比较的最佳方法是什么，或者还有其他有效的方法吗？非常感谢。

import pandas as pd
import numpy as np
pd.options.mode.chained_assignment = None
import os

today_df = pd.DataFrame(columns=['id', 'date', 'count'])
yesterday_df = pd.DataFrame(columns=['id', 'date', 'count'])

def is_dataFrames_equal(today_df, yesterday_df):
if (len(yesterday_df) == len(today_df)):
# len(yesterday_df) == len(today_df) and there is no changes
if (today_df['id'] == yesterday_df['id']).all() and (today_df['date'] == yesterday_df['date']).all() and (today_df['count'] == yesterday_df['count']).all():
return True
# len(yesterday_df) == len(today_df) and there is changes  
else:
return False
# len(yesterday_df) != len(today_df)
else:
return False
if is_dataFrames_equal(today_df, yesterday_df):
print('no new data found.')
exit()
# today_df - yesterday_df
diff_df = today_df[~today_df.isin(yesterday_df).all(1)]

df = today_df.copy(deep=True)

Answer 1

您可以使用pickle:保存变量

import pickle

def save_obj(obj, name):
os.mkdir(f'obj') if not os.path.exists('obj') else None
with open(f'obj/{name}.pkl', 'wb') as f:
pickle.dump(obj, f, pickle.HIGHEST_PROTOCOL)

def load_obj(name):
try:
with open(f'obj/{name}.pkl', 'rb') as f:
return pickle.load(f)
except FileNotFoundError:
return None

today_df = pd.DataFrame(columns=['id', 'date', 'count'])
today_file_name = datetime.datetime.now().strftime("%Y%m%d")
save_obj(today_df, today_file_name)
yesterday_df = pd.DataFrame(columns=['id', 'date', 'count'])
yesterday_file_name = (datetime.datetime.now() - datetime.timedelta(days=1)).strftime("%Y%m%d")
load_obj(yesterday_file_name)