Re: [sipxtapi-dev] Detecting RTP DTMF in sipXtapi
Carsten Avenhaus wrote:
Removing the "continue;" on line 266 of sipXmediaLib/src/mp/MprDecode.cpp
seems to fix the problem at first glance. Maybe someone who knows the
code better than me could check if this would cause any other problems?
...Carsten
Index: sipXtapi/sipXmediaLib/src/mp/MprDecode.cpp
===================================================================
--- sipXtapi/sipXmediaLib/src/mp/MprDecode.cpp (revision 10627)
+++ sipXtapi/sipXmediaLib/src/mp/MprDecode.cpp (working copy)
@@ -263,7 +263,14 @@
// If signaling - decode it now.
if (tryDecodeAsSignalling(tmpRtp))
{
- continue;
+ /* Fixed Bug: This 'continue' prevents the DTMF RTP packet from
+ being decoded. => codecs/PlgTone.c:PLG_DECODE_V1() is never
+ called, and so mpTones->mHaveValidData always stays FALSE.
+ codecs/PlgTone.c:PLG_SIGNALING_V1() never has data to work with.
+ MpDecoderBase::getSignalingData() always returns NO_MORE_DATA.
+
+ continue;
+ */
}
// Accept packet, if it is later then one we already have, else drop
it.
if ( !rtp.isValid()
_______________________________________________
I had the same problem and this fix seemed to sort out the problem.
I however have a different problem with DTMF "events". I have an
application that talks to a device that detects incoming DTMF and
converts to DTMF / Telephone event packets, without any audio.
However the tone events are getting buffered somewhere in the stack
and then require additional packets to "push" them out.
I.e. the application sends
#12345* (No audio, just the DTMF events).
But I only see #1234, sending audio or more DTMF tones from the device
will "flush" the 5* events out of the queue.
The Events seen with wireshark look like
DTMF Event # (Start)
DTMF Event # (End x 3 )
DTMF Event 1 (start)
DTMF Event 1 (End x 3)
DTMF Event 2 (start)
DTMF Event 2 (End x 3)
DTMF Event 3 (start)
DTMF Event 3 (End x 3)
DTMF Event 4 (start)
DTMF Event 4 (End x 3)
DTMF Event 5 (start)
DTMF Event 5 (End x 3)
DTMF Event * (start)
DTMF Event * (End x 3)
I have been working through the code but I have not (yet) been able to
figure out where the events are being buffered.
Has anyone got any pointers?
Regards
Paul
