On segunda-feira, 4 de novembro de 2013 18:12:15, Jussi Laako wrote: > On 4.11.2013 17:49, Thiago Macieira wrote: > > That's a very narrow point of view. It doesn't include: > The original posting was about what glib uses internally to implement > dbus support and I responded that libdbus is obsolete for that since any > remotely recent version of glib doesn't depend on libdbus.
Ok, then that's probably because you replied to a narrow part of the original post. This thread has expanded to a system-wide discussion, which is why I replied what I replied. > > * older glib-based applications that use dbus-glib and haven't been > > ported > > They should do the port to avoid becoming obsolete... > > > * most system daemons, which are bindingless > > I don't know what you mean by this. Part of the system-wide discussion. System daemons often don't use any binding (dbus-glib, gio, QtDBus, whatever), but instead use libdbus-1 directly to avoid bringing in too much in dependencies. > Nice thing with GIO's dbus implementation is that is can be easily used > for peer-to-peer dbus without dbus-daemon. That's not a GIO feature. It's been part of the D-Bus spec since I joined the project, in 2006. > In case auto-invocation is > needed, it is convenient to implement dummy session-bus interface that > just fires up the service and responds with socket address for p2p > communication. This improves communication efficiency and avoids many of > the performance and security problems caused by dbus-daemon without > requiring support for kdbus... > > As an example, we support p2p dbus in gSSO, in addition to traditional > system and session bus. -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ Dev mailing list [email protected] https://lists.tizen.org/listinfo/dev
