Suggested changes for put_datatable()

[amks1]

The put_datatable() function is a really great addition in v1.8! I have a few suggestions:

1. I am making a table which the user will populate with rows. Currently there is no easy way to add a new row in an empty table because the actions toolbar doesn't show unless any row is selected. So, we can have an optional 3rd element in the actions tuple which specifies whether a row needs to be selected for the button to be visible.
   Example:

   put_datatable(data, actions=[('Add Row', lambda: datatable_insert('user', test_data), False),], instance_id='user')

   The False here would signify the button available even when there are no rows, and it does not return the row_id to the function.

2. Add a get_datatable(instance_id='instance_id') type function which will return the current data in the datatable.
   This is helpful when we have done many changes to the datatable (adds, edits, deletes) and we want to get the current data of the datatable. It will be much easier compared to keeping separate variables with the data we updated.

[wang0618]

To add buttons that don't need row to be selected for showing, I think you can use the put_buttons().

As for get_datatable(), I think it'a a good idea, I will add this function on this weekend.

[amks1]

To add buttons that don't need row to be selected for showing, I think you can use the put_buttons().

Yes, this is what I'm doing now. But the beauty of the interface is a bit lacking. :)

As for get_datatable(), I think it'a a good idea, I will add this function on this weekend.

Great, thanks!

[TM0088]

Pywebio provide great documentaton for creating usable feature.
I am now getting the data from datatable by function below

def get_datatable_csv(instance_id):
    from pywebio.session import eval_js
    instance_id = f"ag_grid_{instance_id}_promise"
    datatable_csv_string=eval_js("""window[instance_id] ? window[instance_id].then((grid) => {
        var allRows = grid.api.exportDataAsCsv();
        return allRows;
    }) : console.error(Datatable instance [${instance_id}] not found);
    """, instance_id=instance_id)
    return datatable_csv_string

[wang0618]

Here are another example to read data of datatable. It can work when row record is nested dict.

def read_datatable(instance_id):
    """return the current data in the datatable"""
    return eval_js("""
    window[`ag_grid_${instance_id}_promise`].then(function(gridOptions) {
        function unflatten_row(row) {
            let result = {};
            Object.keys(row).forEach((field) => {
                let path = gridOptions.field2path(field);
                let val = row[field];
                let current = result;
                for (let i = 0; i < path.length - 1; i++) {
                    if (!(path[i] in current)) {
                        current[path[i]] = {};
                    }
                    current = current[path[i]];
                }
                current[path[path.length - 1]] = val;
            });
            return result;
        }
        let rows = []
        gridOptions.api.forEachLeafNode((row)=> rows.push(unflatten_row(row.data)))
        return rows;
    })""", instance_id=instance_id)

demo

Need note that: according to the aggrid doc, the grid provides simple string editing and stores the result as a string by default.

@TM0088

[amks1]

Thanks a lot for this, it has been helpful so far. When I add colID parameters to the column_args, is there a way to get the return value to give the column IDs instead of the labels?

[toben123]

hi, did i miss something, was the function "get_datatable" added? would be really great, its also what i was looking for!

[amks1]

It wasn't added to the library, but you can use the code given in this reply to read the datatable.