_______________________________________________On Mon, Jul 15, 2013 at 1:59 PM, TadeasKriz <tkriz@redhat.com> wrote:
I've created the JIRA subtask:You said some push networks recommend not to unregister from device. Well, I was more thinking about it like the device will send information to the unified push server and the server will then remove the device from the list of available tokens.Well - Apple is pretty strict about "not following" best practices.In iOS there is a notification centre. There the user will disable "notification" per app (and users know that). And that's exactly why Apple has a Feedback Service to query for "invalid" tokens.Regardless if we all love things or not, I guess for the iOS SDK we will not see a "unregister" hook (for those above reasons).Kinda sucks if a developer's app got's rejected because AeroGear-iOS, and its developers :-)Also the device would unregister from the push network if it should (I think it's valid way with GCM, don't know about others).From the API side, I think it might be added as a method of Registrar. So there would be Registrar#register and Registrar#unregister. What do you think?yes, I guess that's what the JIRA ticket is for, right ?The UnifiedPush Server does have this API endpoint already:On Jul 15, 2013, at 12:57 PM, Matthias Wessendorf <matzew@apache.org> wrote:it is opional, since some push networks recommend not performing "in app unregister" themselves. (eg APNs). There are (APNs)hooks to query for invalid tokens (aka feedback service).For Android SDK, can u file JIRA sub task of AGDROID-35 ?Thanks,MatthiasPS: there is no real harm, if payload to push network contains "outdated" tokens.
On Monday, July 15, 2013, TadeasKriz wrote:Hey,
I'm now trying the unified push with Android and I've noticed, that there is no way (or I didn't find it?) to unregister the device. Example real usage: in application settings there might be a checkbox to disable notifications, so the user probably won't like receiving push messages which would then be discarded by the device (it'd be a waste of his data package). It seems to me like it wasn't left out on purpose, but rather forgotten to be added. Any thoughts on this?
Thanks
_______________________________________________
aerogear-dev mailing list
aerogear-dev@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/aerogear-dev
--
Matthias Wessendorf
blog: http://matthiaswessendorf.wordpress.com/
sessions: http://www.slideshare.net/mwessendorf
twitter: http://twitter.com/mwessendorf
_______________________________________________
aerogear-dev mailing list
aerogear-dev@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/aerogear-dev
_______________________________________________
aerogear-dev mailing list
aerogear-dev@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/aerogear-dev--
Matthias Wessendorf
blog: http://matthiaswessendorf.wordpress.com/
sessions: http://www.slideshare.net/mwessendorf
twitter: http://twitter.com/mwessendorf
aerogear-dev mailing list
aerogear-dev@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/aerogear-dev