Re: [Jack-Devel] Client-Server models are just fine. Please?

PrevNext  Index
DateMon, 01 Feb 2016 11:47:09 -0500
From Paul Davis <[hidden] at linuxaudiosystems dot com>
ToKjetil Matheussen <[hidden] at notam02 dot no>
Cc"[hidden] at lists dot jackaudio dot org" <[hidden] at lists dot jackaudio dot org>
In-Reply-ToKjetil Matheussen Re: [Jack-Devel] Client-Server models are just fine. Please?
Follow-Up[hidden] at trellis dot ch Re: [Jack-Devel] Client-Server models are just fine. Please? 
On Mon, Feb 1, 2016 at 11:41 AM, Kjetil Matheussen <[hidden]
> wrote:

>
>
>
> I'm suggesting to add gui functions which are calling an external program
> to configure jack, where the communication between libjack and the gui
> happens by sockets, not by stdout+stdin.
>

this already exists. your issue with it is the communication is currently
asymmetric: there is a structured protocol to configure and start the
server, but not in the reverse direction. you want errors and status
returned as well. and you want a structured protocol for this direction,
rather than semi-random line-oriented text.


> This will not make libjack
> link with a graphics library, and it will most likely not increase the
> size of  libjack, and it will not make jackd disappear.
>

so how will an abritrary client become "the server" ?

it is just amazing to me, given all the other useful things that JACK could
be made to do, that anyone is talking about the mechanics of starting it.
PrevNext  Index

1454345235.32173_0.ltw:2, <CAFa_cKkgEW12AiMJte7XPodkhXzBMhicDU_akE0kCLPhYd=0GQ at mail dot gmail dot com>