Issues with new D-Bus API

Marcel Holtmann marcel at holtmann.org
Thu Dec 31 17:14:15 EST 2009


Hi Witold,

> >>> so I have been hacking on a client using the new D-Bus API and found a
> >>> couple of things that are sort of weird and could be done way simpler.
> >>> Some of them seems just copied from the old API and I think it is time
> >>> to fix this now before we actual get real users of this API.
> >>>
> >> I started to work on some patches to address some of those issues. You
> >> can pull it from my git: git://repo.or.cz/hostap-gosc2009.git dbus
> >> branch. I haven't tested it yet and I haven't modified python test scrip
> >> to fit API changes too, but the documentation is up to date. There is
> >> some duplicated code cleaning too.
> > 
> > so far so good, but I did notice some issues.
> > 
> > 1) BSS are now wrongly identified as ad-hoc. Seems there is a mixup.
> > 
> Fixed. There was logical AND (&&) instead of bit AND (&).

works perfectly fine for me now.

> > 2) The Signal property of a BSS needs to be int16_t (n) since the dBm
> > values can be negative.
> > 
> Fixed.

Looks good now.

> > 3) For some reason the new debug properties don't work in d-feet.
> > 
> Works for me. Take note that d-feet doesn't show properties which value
> is zero or false. If you run wpa_supplicant with -dd option then all
> debug properties are zero or false. Try using other DBus debugger like
> qdbusviewer or get these properties some other way then with d-feet.

You are right, d-feet is utterly stupid in this area.

However we always include RSNIE, WPAIE and WPSIE all the time now. At
least it seems they are in the introspection data. However the Get call
then fails. Or something similar. I haven't really debugged this yet.

I also have no problem to always include these IE, but then at least we
have to have them as empty byte array instead of an error. Otherwise it
is better to not include them at all.

In theory clients could just check for the availability to determine if
RSN, WPA or WPS is supported, but that is always a little bit short
sighted since you need to parse these anyway.

> > 4) The extra dict properties got added to InterfaceRemoved signal.
> > 
> Fixed. The sent signal was correct. The introspection data was wrong.

Okay.

Jouni, I am okay if the current changes get merged back into your tree.

Regards

Marcel




More information about the HostAP mailing list