DataTable returns selected_rows out of list index

Describe your context
Please provide us your environment, so we can easily reproduce the issue.

• replace the result of pip list | grep dash below

dash                 2.5.1
dash-core-components 2.0.0
dash-html-components 2.0.0
dash-table           5.0.0

• if frontend related, tell us your Browser, Version and OS

  • OS: CentOS
  • Browser chrome
  • Version Version 103.0.5060.114 (Official Build) (64-bit)

Describe the bug
I have a DataTable in which I can select rows. At the start no row is selected.

If I click selected_rows returns None as it should.

Now I click on Bernd and remove the name and get this:

If I click now without selecting anything the selected_rows value is 1, even though there is only one element (with index 0). If I repeat remove Anna, I have list with no entries and clicking again return selected_rows= 0.

Expected behavior

If no option is (visibly) selected it should return None.

This is a minimal working example:

import dash
from dash import Dash, html, dcc, dash_table, Input, Output, State, ctx
import pandas as pd

def nameTable():
    filtered_df = df.loc[df['Name'].isin(enabledName)]
    return html.Div([
                dash_table.DataTable(filtered_df.to_dict('records'), [{"name": i, "id": i} for i in df.columns], 
                                     id='nameTable',
                                     sort_action='native',
                                     sort_mode="multi",
                                     row_selectable="single",)
            ], style={'width': '5%'})

name_list = ["Anna", "Bernd"]

df = pd.DataFrame({'Name':name_list
                   }, index=name_list)

enabledName = ["Anna", "Bernd"]

app = Dash(__name__)
app.layout = html.Div(children=[
    nameTable(),
    html.Button("remove", id="removeButton")
])

@app.callback(
    Output('nameTable', 'data'),
    Input('removeButton', 'n_clicks'),
    State('nameTable', 'selected_rows'),
    State('nameTable', 'data'),
)
def updateEnable(n_clicks, selected_rows, data):
    if(selected_rows is not None):
        if(selected_rows[0] < len(data)):
            enabledName.remove(data[selected_rows[0]]['Name'])
        else:
            print(f"Tried to enable Name that's not in list: index= {selected_rows[0] }")
    else:
        print("is none")

    enabled_df = df.loc[df['Name'].isin(enabledName)]
    return enabled_df.to_dict('records')

if __name__ == '__main__':
    app.run_server(debug=True)

Start with the DataTable and the provided minimal Python example, focusing on how selected_rows behaves after the updateEnable callback changes the data. Reproduce the row-removal sequence and verify the selection value after each update. Done means an unselected table returns None rather than an out-of-range or stale row index.