[maemo-developers] Possible libosso bug?
From: Eero Tamminen eero.tamminen at nokia.comDate: Mon Sep 10 15:53:45 EEST 2007
- Previous message: Possible libosso bug?
- Next message: Possible libosso bug?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Hi, ext David Hazel wrote: > I encountered the following unexpected behaviour while testing my > application, and I wanted to run it past the people on this list to see > if this is a known "feature". > > My application opens the browser as part of its normal operation. On > application exit, the browser is closed. The call used to do this is: > > osso_rpc_run_with_defaults(mvarAppCtx, "osso_browser", > OSSO_BROWSER_EXIT, NULL, DBUS_TYPE_INVALID); > > Now, if the browser is open when this call runs, it closes down exactly > as expected. However, if it has already been closed by the user, the > above call OPENS it! > > Given that the call is telling the browser to exit, I would expect one > of two things to happen if the browser is not open: > > 1) The call is ignored, because it is asking for a browser state that > already exists The messages are answered by the co. processes, not D-BUS itself. If the message is sent with auto-activation flag, the co. process is started. Otherwise the message is lost (I think). I think all these libosso functions use the D-BUS auto-activation flag. > 2) The call fires up the browser and immediately closes it (because the > OSSO_BROWSER_EXIT request has to be dealt with by the browser itself). > > Instead, what actually happens is that the browser opens and stays open > (on its default home page). > > Is this a known "feature" of libosso or the Browser? If not, should I > report it as a bug? I would say that it's a bug in the Browser. You might also use the co. D-BUS message directly _without_ auto-activation. - Eero
- Previous message: Possible libosso bug?
- Next message: Possible libosso bug?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]