Wireshark-bugs: [Wireshark-bugs] [Bug 6770] Add dissectors for GSM and UMTS Cell Broadcast proto

Date: Wed, 22 Feb 2012 22:13:30 -0800 (PST)
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6770

--- Comment #19 from Mike Morrin <wireshark@xxxxxxxxxxxxxxx> 2012-02-22 22:13:27 PST ---
1. Can you provide (or point me to) a capture file (or files) containing the 
   GSM_CELL_BROADCAST protocol which would exercise the various code paths
   code in dissect_gsm_cell_broadcast() ?

   It would be helpful if the capture contained a single "page" as well as
   multiple pages (which would need to be concatenated (defragmented).
[mm] I don't currently have access to equipment to generate more traces, but if
I get a chance I will add any which show interesting code paths.

   Also: can you provide a capture file with data encoded as SMS_ENCODING_UCS2
?
[mm] I will see if I can find one.

2. In dissect_cbs_data():
   The 'SMS_ENCODING_7BIT' case returns a UTF8 tvb (after unpacking, etc)

   The 'SMS_ENCODING_UCS2' case uses iconv to convert to UTF8 but then
   discards the converted buffer.

   Is there some reason why the UTF8 buffer isn't returned in this case 
   in a manner similar to that for the SMS_ENCODING_7BIT case ?
[mm] This is a bug.  I will fix it when I find a UCS2 trace to test with.

3. While testing with the attached capture file I noticed that 
   the list of 'protocols in the frame' (shown in the details pane) did not
   include 'gsm_cell_broadcast'.
   After a little poking around I found that this is because
   dissect_sabp_Broadcast_Message_Content() in packet-sabp.c calls
   dissect_umts_cell_broadcast_message() directly thus bypassing the
   Wireshark code which keeps track of the protocols in a frame.

   I'm guessing that this case should also be treated as a 'protocol'....
   (in a manner similar to that used to call dissect_gsm_cell_broadcast() ?)
[mm] I was not sure what to do here.  The problem is that when UMTS Cell
Broadcast is transported by SABP or BMC, there are Cell Broadcast IEs (Serial
Number and Message Identifier) which cannot be passed in the same tvb as the
CBS message that they describe.  It seemed overkill to define separate
protocols for each of these IEs, so I just used direct function calls for these
and for the UMTS CBS message.  I now see that there is an advantage in using an
indirect call (for the main message dissector), and will patch that when I get
a chance.

-- 
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.