Andrew Cohen <acohen@ust.hk> writes:

Hi,

>     >> I have not had much time for Emacs outside of my regular
>     >> activities, but I made the changes you suggested and have been
>     >> using the package for the past year. Unless I hear objections,
>     >> I'll push the package to master sometime soon.
>
>     EZ> Thanks, but could we perhaps implement this in a way that is
>     EZ> less modeled on D-Bus, thus allowing different back-ends on
>     EZ> other systems?  For example, could we have the D-Bus backend
>     EZ> inject an input event, which could then be bound to a command,
>     EZ> instead of directly calling the hooks from a D-Bus event
>     EZ> handler?  That would allow adding other backend, which will
>     EZ> generate the same input event "by other means".
>
>     EZ> Stefan and Michael, I'd appreciate your comments, both on the
>     EZ> patch and on my proposal.  Maybe you will have other, better
>     EZ> design ideas.  (FWIW, I've read Michael's original comments in
>     EZ> this bug report as saying things similar to what I say above.
>     EZ> But either I misunderstood or the patch as posted doesn't do
>     EZ> what Michael suggested back then.)
>
>
> Thanks, I'm happy to make changes. I think I understand what you are
> suggesting (and what Michael suggested earlier) and it is certainly
> desirable to make this more independent of D-Bus. My programming skills
> are extremely rudimentary so it might take some time (or I might fail
> altogether).

Yes, I still believe my proposal stands. OTOH, nobody else has taken the
stab over the last months, so I don't know whether such changes will
happen.

FTR, the majority of my comments on the code are integrated. What's left
could be done later. What I'm missing is the documentation in the Elisp
manual, if we integrate it into Emacs core.

> Best,
> Andy

Best regards Michael.