Predefine download filename with CSV downloads of dataframes

With the current version of streamlit v. 1.32.0, it is possible to download csv files directly:

How can I set the filename of the CSV file that gets downloaded? Per default, it creates <timestamp>_export.csv, however I did not find any option to set the download name in the API reference.

Thank you very much for your help!

Hi @johanneswerner,

Thanks for sharing this question!

The download feature is not customizable. You however get the opportunity to the name the file when you hit the download button. If you want a customized download name, you might consider using st.download_button.

@tonykip Thank you for your swift and dedicated reply. I already have an implementation with st.download_button, however I would like to implement the possiblity to use the same download name in both cases.

Is it possible to overload st.download_button? Or would it rather make sense to open a feature request in the streamlit github repo?

Thank you very much for your help.

What do you mean by “overload st.download_button”?

So you would like to use the same name you use with st.download_button with the downloader within the st.dataframe?

For reference, you’re using the st.download_button with custom file name like this right?

st.download_button(
    label="Download data as CSV",
    data=csv,
    file_name="custom_filename.csv",
    mime="text/csv"
)

Yes, correct, this is the way I am using st.download_button(). And I would like to use a naming schema for the download field in the top of the dataframe that (a) has the similar filename and (b) has a better understandable filename than <timestamp>_export.csv. Is it possible to redefine the st.dataframe() function? (of course, overload was the wrong term)

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.