CARTAvis / carta-frontend

Source code repository for the frontend component of CARTA, a new visualization tool designed for the ALMA, the VLA and the SKA pathfinders.
https://cartavis.github.io/
GNU General Public License v3.0
20 stars 10 forks source link

preference for showing contours from active image only #806

Open veggiesaurus opened 4 years ago

veggiesaurus commented 4 years ago

Should we always show contours of all spatially matched images, or should we have a preference setting for showing only those contours from the active image?

Our current approach is to show all contours of all images that are spatially matched to the active image. This may be confusing to DS9 users, as they would probably be used to contours being independent.

arusstaylor commented 4 years ago

we should be able to turn contours from any spatially matched image off or on and be able to configure the contours from each matched image independently.

veggiesaurus commented 4 years ago

we should be able to turn contours from any spatially matched image off or on and be able to configure the contours from each matched image independently.

We can already do that (via the layer widget). But currently it shows all by default. I'm not sure whether this default behaviour is confusing

arusstaylor commented 4 years ago

Ok I see. I think it should only be the active image by default. Then the user can add additional contours from other images manually.

veggiesaurus commented 4 years ago

One of the issues with that approach is how to handle changing the active image. If the user as the "active image only" setting enabled, then only the active image would be shown. If the user switches to image A, the contours for image A would be shown. If they then switch to image B, the contours for image B would be shown (and not A). If they then explicitly enable contours from image A, the contours for both images A and B would be shown.

Possibly we need a separate setting in the layer widget for contour visibility: Off, On, Only when active. That's reasonably easy to implement, but may add another layer of complexity

The simplest and most consistent way to handle this would be to just hide all but the active contours each time the user changes images, and allow them to then re-enable each set one by one. But that could become very frustrating to users.