On 05/02/2010 07:14, Michael Hunter wrote:
> On Thu, 04 Feb 2010 15:02:35 -0800
> Renee Danson Sommerfeld<renee.sommerfeld at sun.com> wrote:
>
>
>> I found this small bug while doing manual test 3, AP selection. I've
>> tested this fix and verified that this test now passes. Please let me
>> know what you think!
>>
>> http://jurassic.sfbay/~okie/webrev.14377/
>>
> ncu_phys.c:908 I'm not sure I follow this. In this past the code was
> such that if the essid didn't match we continued. Now the essid might
> not match but the bssid could and we would get through.
>
>
Good point - I'm not sure if BSSIDs are guaranteed to be unique
between different WLANs (they are mostly the AP's MAC addresses
in practice, but there are exceptions). Checking if
cur_wlan->nww_essid is blank (indicating a non-ESSID-broadcasting
WLAN) as well might be worthwhile to be on the safe side, i.e.
if (strcmp(cur_wlan->nww_essid, name) != 0 &&
!(cur_wlan->nww_essid[0] == '\0' && b_match))
Alan
