Open cornhundred opened 8 months ago
This is related to https://github.com/developmentseed/lonboard/issues/322.
I think that View
is always passed in to Deck
directly, not the Layer
class, right? In that case it should be relatively simple.
As a general statement, much of lonboard so far has geospatial assumptions baked in, but I think nothing should preclude using non-geospatial data
Thanks, that is my understanding too that it is passed into Deck directly.
Here's how I'm setting the view on an ObservableHQ notebook
const view = new deck.OrthographicView({id: 'ortho'})
deck.DeckGL({
container,
views:[view],
initialViewState:INITIAL_VIEW_STATE,
controller: true,
});
I can have a look at lonboard to see if I can figure out how to make a pull request.
A PR would be great. In general, models are defined first in TS https://github.com/developmentseed/lonboard/blob/main/src/model/layer.ts (maybe create a new file for view.ts
?), then create wrapper classes (subclassing from ipywidgets.Widget
) that map to the deck.gl api: https://github.com/developmentseed/lonboard/blob/4daca9bb36013a56a998dc1507d03058f582b141/lonboard/_layer.py#L288-L357
Context
I would like to visualize non geographic data (e.g., spatial multi-omic biological data) using lonboard. I belive that we would have the add the option for the OrthographicView class in deck.gl.
Issue
I think that we just need to expose the view classes to lonboard to allow users to use the OrthograhicView.
Acceptance-Criteria
I'm not sure what these would be.