Standardised replacement for pathbar widget

I noticed that there are some rumblings from the design team around wanting to standardise the pathbar with libadwaita at some point in the future, maybe post 1.0.

I couldn’t help but wonder whether gtk tagged-entry might be a more appropriate starting point for inspiration moving forward.

Here is an example of a tagged entry using a something called tagify for those unfamiliar.
tagify1

If using it for a path I would probably get rid of x on the tags and have them trim the subsequent tags when you click on the body of the preceding ones.

choosing / or ▹ to seperate paths from one another could be an implementation detail and catered for with the underlying api of the path-entry widget.

Here is an example of user defined strings separating tags which isn’t exactly what I have described above but should help illustrate the point nonetheless.
tagify2

I would encourage the @tbernard and the design team to consider using / instead of the ▹ though as it is clearer for users when editing paths.

One thing which hasn’t been covered on that gitlab issue is that depending on the application some paths are going to be editable and some are going to be readonly. That is supported in other tagged entry implementations and should also be supported for a path-entry.

A user might have an expectation to edit a path in nautilus but that expectation may not hold for an application like secrets.

Having a flat button within the entry to surface the fact that it is indeed a path which is editable might be a good idea, having another next to it so that the user can easily copy the path might also be useful functionality to surface on the top level of the application.

Overall I wasn’t the biggest fan of the pathbar that nautilus switched to in the GNOME3 cycle and I think now that GTK4 is a thing that we can do much better from a usability perspective.

I don’t think that the situation with editing paths right now is as good as it could be. The discoverability of it isn’t great tbh.

I think that you could achieve everything that has been proposed in the latest mockup in an entry and would make big wins from a usability perspective as a result.

Thanks for taking the time to read this.

1 Like