Uncaught exception in IS05 test_25

[rainey]

on master, currently commit f47247a
Traceback:

Traceback (most recent call last):
  File "/mnt/raid88/srainey/nmos-testing/nmos-testing/nmos-testing/nmostesting/GenericTest.py", line 170, in execute_test
    self.result.append(method(test))
                       ~~~~~~^^^^^^
  File "/mnt/raid88/srainey/nmos-testing/nmos-testing/nmos-testing/nmostesting/suites/IS0501Test.py", line 615, in test_25
    valid2, response2 = self.is05_utils.check_sdp_matches_params(sender)
                        ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^
  File "/mnt/raid88/srainey/nmos-testing/nmos-testing/nmos-testing/nmostesting/IS05Utils.py", line 556, in check_sdp_matches_params
    if connection_line.group(1) != transport_params["destination_ip"]:
       ^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'group'

This is probably the offending SDP file - the last thing captured by wireshark before the exception

v=0
o=- 1750778293 1750778293 IN IP4 10.10.0.10
s=
c=IN IP4 10.10.0.11
t=0 0
a=recvonly
m=video 44243 RTP/AVP 112
b=AS:0
a=fmtp:112 packetmode=0; width=0; height=0; exactframerate=0/0; interlace=1
a=source-filter: incl IN IP4 10.10.0.11 10.10.0.10
a=mediaclk:direct=0
a=ts-refclk:localmac=00-0F-5B-06-B0-00

It is relatively trivial to fix this, but it's questionable if the test might need more work. According to RFC8866, S5.7:

A session description MUST contain either at least one "c=" field in
each media description or a single "c=" field at the session level...

Nothing in the NMOS spec that I see specifically prohibits the latter option, which is where this is crashing. I may submit a PR regarding this soon.

[garethsb]

I think you're correct per RFC 4566/8866. I think it's not seen more often because all the examples in SMPTE ST 2110 standards and AMWA NMOS specs use the media-level connection line.

Probably the NMOS Testing Tool needs test cases that check Receivers handle less common SDP variants...

(I note that sony/nmos-cpp OSS does support SDP files with a session-level connection line (nmos/sdp_utils.cpp) but doesn't handle a session-level source-filter attribute...)

[garethsb]

FWIW, that SDP file does have issues...

a=fmtp:112 packetmode=0; width=0; height=0; exactframerate=0/0; interlace=1

The interlace format-specific parameter takes no value (=1 needs to be deleted) and clearly the width, height and frame rate aren't right. Zero in the bandwidth line also...

And it almost certainly needs an rtpmap media-level attribute!