Dash Data Table Virtualization completely broken with an editable Table

Suppose this very simple example of an editable Data Table with 1000 rows and virtualization enabled:

app.layout = html.Div([
    dash_table.DataTable(
        id='table',
        columns=(
                [{'id': 'col1', 'name': 'col1', 'editable': False}] +
                [{'id': 'col2', 'name': 'col2', 'type': 'numeric'}]
        ),
        data=[
            {'col1': i, 'col2': i}
            for i in range(1000)
        ],
        style_table={
            'height': 240,
            'width': 290,
            'overflowY': 'auto'
        },
        style_cell_conditional=[
            {'if': {'column_id': 'col1'},
             'width': '55%'},
            {'if': {'column_id': 'col2'},
             'width': '45%'}
        ],
        editable=True,
        page_action='none',
        virtualization=True
    )
]
)

Now if you actually try to edit the table, it is going to break. The first few rows are fine, no problem.
But then suddenly if you try to change a value from like row number 8 on the table does really weird stuff and then breaks with an error or just outright breaks immediately.
I dont even get errors in the console, just an error popup saying something went wrong in Javascript.

Here is a short video of it (sorry for low quality) but given the simple code you can reproduce it easily yourself:
https://i.imgur.com/A2YNqZQ.mp4

As my Data Table is going to have a lot of rows in it I kinda rely on virtualization.

So if anyone has an idea how to fix this I would really appreciate help

An issue has been opened and this being worked on: https://github.com/plotly/dash-table/pull/584

I’m hopeful it will get included in the upcoming dash==1.3.0 release.

There’s no workaround for this issue.

three reasons why you should buy plotly pro: support open source, get great support, host your plots and dashboards online