I'm afraid the problem is going to be more complex than the
toggle push solution would suggest. There is a lot of code around
detecting network state and managing things like keep-alive in
the app. I don't understand it very well, but it would definitely
need to be addressed.

There's also the issue with Android 8 Oreo restricting background
services even more than before.

Honestly, I don't think I really have the bandwidth to maintain a
fork long term myself. But, I would be happy to try to work on
problems like this though. I would just need some guidance to
know why some of the current design decisions were made.

Also, just to clarify, I don't mean to come across as hostile
towards the devs at all. I know they're working hard on an open
source project we all get to enjoy. :)

Greg Troxel <[email protected]> wrote:
> 
> You could maintain a friendly fork, where your changes are
> planned to go upstream someday. This could be a lot of git
> wrangling, but I wonder about cloning and leaving the
> development branch as is, and then making feature branches as
> necessary, rebasing them when convenient, and having an
> integration branch which is upstream development branch plus
> feature branches. I did something sort of like this once (in a
> different context, for a different reason), but it basically
> worked well git wise.
> 
> As for the change, if turning push on and then off makes
> background calls work, it seems that there is just some missing
> initialization to set up the background socket right on boot,
> and that's only done on deconfiguring push. Really if the push
> registration is not successful this should happen
> automatically, and the push checkbox should only be used to say
> "don't even try push, even if it seems to work".

-- 
Thank you,
Dominic
_______________________________________________
Linphone-users mailing list
[email protected]
https://lists.nongnu.org/mailman/listinfo/linphone-users

Reply via email to