[PATCH weston 1/4] Handle requests using outputs that have been unplugged

[Neil Roberts]

Jason Ekstrand <jason at jlekstrand.net> writes:

> Yeah, these are the wrong semantics.  If they specify an output and it
> turns out to be a zombie, we should do nothing.  A null output in
> wl_fullscreen_shell.present_surface means "put it somewhere".  In the case
> of weston's implementation, it goes on all outputs.  We don't want a
> surface to accidentally end up everywhere when it was intended for a
> particular output.

Ok yes, that makes sense.

> This is a bit ugly. In my original snippit, I pulled a similar stunt
> and stashed the opcode in the implementation implementation pointer.
> Is there some particular reason why you opted for the internal wl_list
> link instead?

Well, the main reason was that I wanted to avoid having to add a
wl_resource_get_implementation function, but I agree it would be cleaner
to just add that as you say.

Thanks for the review.

- Neil