Wireshark-bugs: [Wireshark-bugs] [Bug 2368] Wireshark decodes all G.711 packets as T.38 after a
Michael Mann
changed
bug 2368
What |
Removed |
Added |
Attachment #9936 Flags |
|
review_for_checkin?
|
Comment # 14
on bug 2368
from Michael Mann
Created attachment 9936 [details]
SDP with context for SIP
Attached is a patch that fixes this bug and starts down the path of providing a
transport layer for SDP. Because SDP doesn't have a transport layer, it's up
to the higher layer protocols to use the new SDP API (setup_sdp_transport) to
provide OFFER/ANSWER context.
I attempted to keep backwards compatibility until all higher layer protocols
set up OFFER/ANSWER context. If OFFER/ANSWER (request/response) can't be found
(which would be the case for higher layer protocols without context), Wireshark
should resort to the old behavior of applying SDP information as it sees
packets.
This patch only implements the higher layer in SIP. I'm not sure how many
other higher layer protocols need the same functionality, but some may be
exposed in other existing bugs.
This patch does not fix bug 2992, as I haven't figured out enough of the "SDP
state machine" (now that it has some context) to correctly implement RFC 3264
(and discard a negotiated port)
I'm also curious of the desire to backport. To get it "completely right" (with
all higher layer dissectors involved) will certainly take a few revisions, so
my vote is to not backport it (and just make it ready for v1.10), under the
direction of just "getting better dissection" (ie enhancement), not necessarily
a bug.
You are receiving this mail because:
- You are the assignee for the bug.
- You are watching all bug changes.