[PATCH] dbus: Expose connected stations on D-Bus

[Andrej Shadura]

On 12/05/18 15:57, Slava Monich wrote:
> On 10/05/18 20:07, Andrej Shadura wrote:
>> On 08/05/18 18:53, Dan Williams wrote:
>>> On Mon, 2018-05-07 at 20:19 +0200, Andrej Shadura wrote:
>>>> From: Mathieu Trudel-Lapierre <mathieu.trudel-lapierre at canonical.com>
>>>>
>>>> Make it possible to list connected stations in AP mode over D-Bus,
>>>> along
>>>> with some of their properties: rx/tx packets, bytes, capabilities,
>>>> etc.
>>> It looks like this patch changes the signal signature of
>>> StaAuthorized/StaDeauthorized, is that correct?  That won't be
>>> backwards compatible with existing D-Bus clients and we should probably
>>> choose new signal names to allow old clients to continue working.
>> Hmm, indeed it does, and it probably shouldn’t. What would you call a
>> new signal with the new signature?
>>
>> Mathieu, what applications rely on the proposed new signature of this
>> signal?
> 
> Sorry for the spam, I accidentally hit the send button.
> 
> I don't think that Mathieu (or anyone else) knows the full list of
> applications relying on the current signature of this particular D-Bus
> signal. Here is an example of code which I'm sure Mathieu wasn't aware of:
> 
> https://git.merproject.org/mer-core/libgsupplicant/blob/master/src/gsupplicant_interface.c#L1987
> 
> Please don't break backward compatibility. It's not cool.

Sure. My question was rather what relies on the new signature (Ubuntu
was shipping that patch for about two years I think).

I think Mathieu’s message didn’t reach the mailing list, so let me copy
his response here:

On 10/05/18 23:07, Mathieu Trudel-Lapierre wrote:
> I had written this a long while ago to use to inform GPS software,
> that would help triagulate using APs in range.
>
> FWIW, I won't need it at this point; feel free to rework the code to
> your heart's content if you can make use of it. :)

-- 
Cheers,
  Andrej