Provide a streamlit component for ITables

[mwouts]

The purpose of this PR is to provide a component that would display a DataFrame in a streamlit app.

The first version of the component could have the following limitations:

• complex columns would not be supported
• custom JS code would not be supported
• the component does not return any data from the JS side (rows selected etc won't be accessible yet)

TODO:

• extract a function from to_html_datatable that, given a dataframe and the usual arguments, returns the arguments to be passed to DataFrame on the JS end
• In addition, for the present use case, that function would have to set a columns argument too (since we don't pass a HTML header for the table)
• The frontend for our streamlit component should depend on dt_for_itables, in order to provide the same datatables extensions in Streamlit as in Jupyter Notebooks.
• Test the light and dark mode

[github-actions]

Thank you for making this pull request.

Did you know? You can try it on Binder: .

Also, the version of ITables developed in this PR can be installed with pip:

pip install git+https://github.com/mwouts/itables.git@streamlit_component

(this requires nodejs, see more at Developing ITables)

[mwouts]

The latest version of itables includes a streamlit component.

Install it with

pip install 'itables>=2.1.0'

To display a dataframe in a streamlit app, just type:

from itables.streamlit import interactive_table

interactive_table(df)

The interactive_table function takes the same parameter as the show function, up to certain limitations (in short: multi indexed columns will look a bit different, passing javascript code is not supported, pandas Style objects are not supported).

See the documentation at https://mwouts.github.io/itables/streamlit.html

A sample application is available at https://itables.streamlit.app/, see https://github.com/mwouts/demo_itables_in_streamlit for the source code.