Problem with OpenNetCF.Net adapter

  • Thread starter Thread starter Francesco
  • Start date Start date
F

Francesco

Hi everybody.
I have writed a small test application to understand the
functionalities offered by OpenNetCF.Net namespace.

My problem is that, when i do this call in the code:

AdapterCollection ac = OpenNETCF.Net.Networking.GetAdapters();

I get the following excpetion:ArgumentOutOfRangeException
it seems very strange to me, in fact i don't give any argument to the
method...

Anyone can help?
I tested it in a pocketPc2002, maybe the openNet framework is designed
for pocketPc 2003?

Thanks for your answers.
Francesco
 
There are two possible sources of the problem that I see, one more likely
than the other:

1. There might be no adapters in the list (less likely cause).

2. Rather than using the Next field of the current IP_ADAPTER_INFO structure
to find the next one, we're assuming that each one is the same size. A
similar problem showed up in one of the SSID-related enumerations.
Apparently, some network adapters return information differently and the
network stack simply passes that along. The problem, if this is the case,
is in IP_ADAPTER_INFO.NextAdapter(). A fix might be to keep, rather than
just an index value, the actual offset within the byte array of the current
adapter information. Alternatively, each time a NextAdapter() is done,
you'd have to step through the list from the beginning, counting each Next
as you went through it until you got to the index you wanted. The
NDIS_802_11_BSSID_LIST class does it this way (check the Item method).

Paul T.
 
I should have a fix for testing in the source today. Note that there may be
some minor changes to the API.

Paul T.
 
We have a fixed version to be tested. Someone will post here when the fix
is actually available publicly.

Paul T.
 
We have a fixed version to be tested. Someone will post here when the fix
is actually available publicly.

Hi!

Do you know when the fix will be published?
Is it possible to get a copy of the source code right now?

Kind regards,

Benjamin Lukner
 
As a general rule, no, we don't distribute random versions. Everything
drops at a certain time, giving us a reasonable level of confidence that
everything in that 'drop' works together.

Paul T.
 
Back
Top