Re: [g-a-devel] Cross desktop environment event capture for assistive technologies.

[Alejandro Piñeiro <apinheiro igalia com>]

On 08/12/16 03:36, Luke Yelavich wrote:
> On Mon, Dec 05, 2016 at 10:31:19PM AEDT, Alejandro Piñeiro wrote:
> > Yes, the same applies for Wayland. It is one of our oldest TODOs. We
> > even had some email threads about the topic on wayland-dev some years
> > ago, without too much success. The thing is that they are open to
> > discuss it, but they didn't give too much hints about how to implement
> > it, and we should be the ones doing it.
> >
> > FWIW, the issue is not only about snoop input events, but also about
> > synthesize input events.
> Yes, I am aware of that, and forgot to mention it.
>
> > Yes, the first step, as far as I remember the chats, would be solving it
> > first on the compositor. For Wayland-GNOME, would be gnome-shell.
> This is what I was thinking. You mention going via at-spi. Are you referring 
> to the registry daemon itself, or are you referring to libatspi only?

libatspi. In that sense the idea is libatspi working as a kind of
wrapper of what wayland/mir would provide (or any wayland/mir extension).

> I was thinking that it doesn't make sense for the registry daemon to deal 
> with input event capture and synthesizing input events. As such, my thought 
> was to change libatspi to work with the shell/compositor directly, bypassing 
> one level of indirection, and reducing the amount of back and forth IPC 
> required. Of course for pure X environments things would stay as they are.

That was the idea behind using XEvIE on X. But as I mentioned, that
module was not properly maintained (and in fact, it was not the purpose
on it). So we relied on the snoop at the application side, that was
suboptimal. FWIW, in theory there are ways on X to avoid this, based on
a proof-of-concept that one X developer showed after one GUADEC. We
didn't have time to try to integrate it on at-spi though.

Or in other words: yes I agree, ideally it would be better to not need
to use the daemon.

> Ok reading your reply again, you stated reusing the at-spi API, which I agree 
> with, but as above, I do think the registry daemon shouldn't be needed in the 
> longer term, and libatspi should deal with the shells itself.

Yes.

> > Finally, other place to take a look to are screen on keyboards. They
> > face similar problems, and I bet that there are Mir-based platforms with
> > some kind of screen on keyboard (Ubuntu Touch?)
> Ubuntu touch uses Maliit as the keyboard framework, and I think it uses Qt in 
> some way to directly inject keyboard events, and may support other systems as 
> well, my understanding is not that great for this.

But it would be good to confirm it ;)

> Thanks to Chris as well, both of you have provided a bit more reading 
> material which I will have a look over.

Thanks to you

Best regards