Update dataframe from csv file when updated

☁️ Streamlit Community Cloud
,
1

Hello !

I would like to share data between different user on my streamlit cloud app.
Let’s consider this example :

import streamlit as st
import pandas as pd
import os

## User input
number_1 = st.number_input("Number 1 :",value=0)
number_2 = st.number_input("Number 2 :",value=0)

## Button to update values
button_state=st.button('Update !')

if button_state:
	# Storing data as dataframe
	df=pd.DataFrame()
	df.loc['Number 1','Value']=number_1
	df.loc['Number 2','Value']=number_2

	# Showing df
	st.dataframe(df.loc)

	# Saving as csv
	df.to_csv('my_df.csv')

else:
	if os.path.exists('my_df.csv'):
		df=pd.read_csv('my_df.csv')
		st.dataframe(df.loc)

How can I tell streamlit to listen for changes to ‘my_df.csv’ and update the display, so my users can share their values with others ?

Thanks in advance !

2

Do these users have access to your app?

3

I have an app where on first load I check the time the file was last updated using pathlib.Path.stat().st_mtime and then store that in session_state as last_update_time.

I then on rerun use the same function and compare the output to last_update_time, and if different I refresh the dataframe from the updated file.

I’m sure it’s far from the most efficient way of doing this but it works.

4

Yes they have access, i add them with the sharing option :slight_smile:

5

Good idea indeed, thanks for sharing :slight_smile:
In my case i’m not sure it helps because if I rerun my app, it will reload the csv file, and as it is a light file, it doesn’t consume a lot of time. I just want it to be reloaded without reloading the whole code when changes are noticed.