Good Morning Tomasz, On 30.01.2012 08:07, Tomasz Bursztyka wrote: > Hi Daniel, > >>> Due to that, I propose in patch 2, 3, 4, 5, 6 several changes in >>> notifications: >>> - session does not notify anything about Name/Bearer/Interface/Ipv4/Ipv6 >>> until it reaches a relevant state: connected and/or online. >> We should clear this list when we go offline as well. > My proposal is about the contrary: there will be dozens of application > using session api, so why not limiting the information flow from dbus to > the strictly necessary only.
Of course we should not send a new message just to clear the settings, this can be as is it now send in the same message. My reasoning is here, that for UI developers has to take care about resetting/showing the settings depending on the state of the session. Sure, this can be done, but I think resetting all the values when going into disconnected state is what people would expect (or at least I would). And as I said this doesn't involve an new message. > In this case: application should assume that > Name/Bearer/Interface/Ipv4/Ipv6 become invalid information (and can be > forgotten) when State become "disconnected". > So, we are asking the application to interpret the value of State a bit > more here. > The bogus info you had in your ui was due to that: it did not forget > about those info when state become "disconnected". I think this is breaking the behaviour pattern of the ConnMan API we have so far. >>> It's a mistake right? I would rather change that so ready/online service >>> state on local type will set session state to connected. >> I am fine with this. >> >> One of the glitches I see during my testing is when we have two session >> one connected to wifi and the other to ethernet, the first one will be >> online, the second one only connected. > Actually that bug exists already. The current problem is that session > relies on service's state to know if it can be online or not. However: > only one service can be online at a time . Yep, this needs to be fixed as well :) > I have an idea how to fix that, though this require a bit of change in > service also. > > I will check the issue you found about this 2 session/2 different > AllowedBearers/session.connect(). If it helps, I have a test UI for the sessions. With this UI you find easily bugs :) cheers, daniel _______________________________________________ connman mailing list [email protected] http://lists.connman.net/listinfo/connman
