New features are incomplete? Unable to achieve the function of st.experimental_data_editor

🎈 Using Streamlit
1

I see you guys showing these effect.

image
image554×744 78 KB

So how to display widgets in st.experimental_data_editor, could you write a demo? I didn’t find a way to do it in any docs.

1 Like
2

Have you checked the docs?

3

I have read this document many times, could you show a simple demo how to put chart widgets in table cell? Just like these?
image
Or a button or something else?

4

We are currently finalizing these more advanced column types. You can already try it out here. This will be released within the next few weeks.

4 Likes
5

It’s cool. Will you add a button(button/upload button/ download button/ clickable link[just like button]) to data_editor? If so, we can make many exciting features!

6

I went into a styling issue when rendering decimals.

import streamlit as st
import pandas as pd
df = pd.DataFrame(
    [
       {"item": "Operating Income", 
        "2022Amount": 518224096.84, 
        "2021Amount": 541102193.80},
       {"item": "Operating Costs", 
        "2022Amount": 138946684.89, 
        "2021Amount": 158904121.43},
       {"item": "Taxes & Surcharges", 
        "2022Amount": 2113607.54, 
        "2021Amount": 2123422.01},
   ]
)

edited_df = st.experimental_data_editor(df.style.format({"2021Amount":"{:.2f}"}))

the 2021Amout of Operating Income shows the styling doesn’t work:
image
It seems the experimental_data_editor widget have a default style for float type, st.DataFrame’s styler got ignored.

2 Likes
7

Good idea :slight_smile: We will definitely explore adding support for buttons in cells.

8

Unfortunately, pandas styler is currently not supported because of technical limitations (it would seriously slow down the data editor). But we are exploring alternative ways to apply styling and modify display values.

2 Likes
9

As far as I can recall, there was an issue like this soon after st.dataframe first came out, I’m pretty sure to have seen it on github issues… hope you guys find some front-end workaround, this experimental_data_editor is a game changer!

1 Like