Re: [PATCH 0/7] RFC: proximity cleanup

[Johan Hedberg <johan.hedberg@gmail.com>]

Hi Andrzej,

On Mon, Sep 03, 2012, Andrzej Kaczmarek wrote:
> On 08/31/2012 06:17 PM, Anderson Lizardo wrote:
> >Hi Andrzej,
> >
> >On Fri, Aug 31, 2012 at 8:48 AM, Andrzej Kaczmarek
> ><andrzej.kaczmarek@tieto.com> wrote:
> >>Hi,
> >>
> >>Here are few patches to cleanup proximity profile code a bit which seem to
> >>keep track of some data which are redundant and has minor issue with device
> >>driver registration.
> >>
> >>Most cleanup is done in patch #3 which removes tracking of DBusConnection
> >>across whole plugin and simply uses get_dbus_connection() wherever necessary.
> >>This is different than other profile plugins do, but reduces code size and
> >>memory footprint a bit so can be done for other plugins as well if accepted.
> >
> >While some cleanups here are very welcome, I'm worried that you are
> >cleaning up things before finishing the Proximity Reporter
> >implementation. Power Level characteristic for instance is currently a
> >stub, how do you plan to store it per adapter, without the struct
> >reporter_adapter?
> 
> Ah, I didn't find it on TODO so assumed it's complete. But you're
> right, it probably should not be removed.
> 
> >Therefore I suggest you split the "trivial" cleanups from the removal
> >of struct reporter_adapter.
> 
> Please just consider patches #6 and #7 as invalid for now, I think
> there's no point in sending #1-5 once more until there are some
> comments to fix.
> 
> >BTW, do you have plans to add more features to the Proximity Reporter
> >implementation?
> 
> No plans at the moment, but perhaps I'll take a look on this in more
> details after HRP is finished (v2 will be sent soon).

I've applied patches 1-4. Patch 5 didn't really make sense anymore after
the btd_profile refactoring that went in just a moment ago, and as you
said patch 6 and 7 can be considered invalid for now.

Johan