Different (Text)Renderers based on column dtype? about ipydatagrid

[ibdafna]

Hello again @sachin-smart and thank you for the feature suggestions!

ipydatagrid was designed to be extended using the widget ecosystem, and therefore it is probably not within scope to incorporate rendering logic based on data type - that logic is something users can implement using the existing Python language constructs.

To your point on using a single, default renderer for the grid - this exists and is actually well documented in the examples folder. All of ipydatagrid's documentation is example based. We can also use basic Python to implement logic for selecting a different header renderer for each column based on its data type. See below for a short example which achieves that:

import pandas as pd
import numpy as np
import ipydatagrid as datagrid

data = pd.DataFrame({'Integer':[1,2,3,4,5], 
                     'Float':[1.1, 2.2, 3.3, 4.4, 5.5],
                     'Text':'A B C D E'.split(), 
                     'Dates':pd.date_range(start='2021-06-01', end='2021-06-05'),
                     'Mixed':[1, 2.2, 'C', np.datetime64('2021-05-05'), 5]})

default_renderer = datagrid.TextRenderer(background_color="salmon")
string_renderer = datagrid.TextRenderer(background_color="limegreen")
int_renderer = datagrid.TextRenderer(background_color="yellow")

dtype_renderers = {
    np.dtype('int64'): int_renderer,
    np.dtype('O'): string_renderer
}

column_renderers = {name:dtype_renderers[dtype] for (name, dtype) in data.dtypes.items() if dtype in dtype_renderers}
grid = datagrid.DataGrid(data, 
                         default_renderer=default_renderer, 
                         renderers=column_renderers,
                         layout={"height":"150px"})
grid

[sachin-smart]

That's great. Much appreciated! And apologies, I did not see the full extent of examples or the usage of default_renderer. Cheers!