This WIP currently consists of two patches:
- "waylandsink: add basic support for pointer devices" (which handles the pointer devices, but has nothing to do with navigation interface)
- "waylandsink: implement navigation interface" on top of the first one.
It is a simple implementation with the one from glimagesink used as the example. Tested with
gst-launch-1.0 -v videotestsrc ! navigationtest ! waylandsink only, and this particular test seems to work OK under wayland/weston.
Currently the navigation interface support breaks the ability to move the waylandsink window introduced by "waylandsink: add basic support for pointer devices" (as it simply comments out the code which starts moving the window), and I'd like to fix that, but don't have the ideas yet.
I've borrowed some code from xvimagesink, and planned to let to move the window when gst_pad_push_event() returns FALSE. But what I currently see is that it always return FALSE, so I can't tell if the navigation events are handled by the upstream element or not.