[patch] wireless patch, take 2

[Kristian Høgsberg]

David Zeuthen wrote:
> On Wed, 2004-05-26 at 15:56 -0400, Joe Shaw wrote:
> 
>>On Wed, 2004-05-26 at 21:17 +0200, David Zeuthen wrote:
>>
>>>This is extremely cool. Please run, don't walk, to the terminal and
>>>commit it! Regarding the property.d callouts I've added an item to the
>>>TODO Wiki page so we don't forget to fix it.
>>
>>Thanks!  This is committed.  I just noticed a small bug in the access
>>point scanning code (it gets the channel instead of the frequency for
>>the scanned APs), so I'll fix that as well.  Next up is support for
>>setting the ESSID via HAL.  After that I'll take a look into doing the
>>callout limiting stuff, at least in the daemon.  Probably won't get to
>>all that today, though. :)
> 
> I wonder, would setting the ESSID via HAL be implemented like a D-BUS
> method? One solution might be to expose an interface on the object with
> the UDI, say,
> 
>  org.freedesktop.Hal.Device.Net.Ethernet.80211 
> 
> with the single method void SetESSID(string ess_id) or something? This
> would throw an exception if the device doesn't have the capability
> net.ethernet.80211.
> 
> Security-wise this could be locked down to only trusted binaries using
> the D-BUS security framework (IIRC you can at least restrict access to
> certain users, dunno about specific processes). Just a thought.

Why should you be able to set the ESSID through HAL?  I thought the 
primary purpose of HAL was discovery, device specific control happens 
through direct access to the device file or similar out-of-band mechanism.

Kristian

_______________________________________________
hal mailing list
hal at freedesktop.org
http://freedesktop.org/mailman/listinfo/hal