[PATCH_v2] virtual-keyboard: Add new virtual keyboard protocol
dorota.czaplejewicz at puri.sm
Thu May 31 14:25:41 UTC 2018
On Wed, 30 May 2018 10:05:52 +0200
Jan Arne Petersen <janarne at gmail.com> wrote:
> On Wed, May 30, 2018 at 9:33 AM Peter Hutterer <peter.hutterer at who-t.net>
> > On Thu, May 24, 2018 at 08:27:29PM +0200, Dorota Czaplejewicz wrote:
> > > Provides the ability to emulate keyboards by applications.
> Complementary to input-method protocol.
> > >
> > > The interface is a mirror copy of wl_keyboard, with removed serials,
> and added seat binding.
> > > ---
> I do not really understand for what this protocol is useful, trying to map
> a virtual keyboard to a physical keyboard seems to be extremely limiting
> and I do not see really use-cases for it. If you want to be able to send
> non-text from a virtual keyboard to an application it makes more sense to
> have some protocol to send keysyms for example.
> Jan Arne Petersen
the immediate need is a virtual keyboard indeed. Keeping the protocol simple and general can make it useful for automating input for example, so I think it's worth to stick to the current form.
It's not a suboptimal form for a virtual keyboard either. I have considered keysyms, gathered opinions, and decided against them. The reason is simple. If we tried to send scan codes, then the compositor would have to do the scan code->keysym translation anyway, but with less information. Either that, or applications would have to support yet another protocol (in addition to wl_keyboard and text-input) to receive generic keyboard events, defeating the point of having a generic protocol.
On the other hand, raw scan codes can be trivially piped via wl_keyboard to any existing application that cares about keyboard input, even if it doesn't implement text-input.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 833 bytes
Desc: OpenPGP digital signature
More information about the wayland-devel