[Asterisk-Dev] The Almighty X-Lite DTMF Problem [PATCH]

[Christopher Heiser]

Here is my final patch.
I will be submitting this to bugs.digium.com

The patch includes the following fixes:

Recognize the "End" bit on an RFC2833 telephone-event packet as the end of a 
DTMF digit
Track the DTMF duration count.  If a DTMF event comes in with a smaller 
duration, recognize the previous DTMF event as complete.

This solves Xten's problems and will also still detect repeated DTMF digits 
without the "End" bit set (in case we lose that packet).

--Chris


>From: "Chris Heiser" <cheeseman00 at hotmail.com>
>Reply-To: asterisk-dev at lists.digium.com
>To: <asterisk-dev at lists.digium.com>
>Subject: RE: [Asterisk-Dev] The Almighty X-Lite DTMF Problem
>Date: Wed, 13 Aug 2003 16:19:26 -0400
>
>Pete,
>
>Try this patch below...  I noticed that eStara's softphone has the same
>problem as xten's softphone when it comes to DTMF.  Seems as though 
>Asterisk
>is not looking for the "end" bit per RFC2833.  So try this fix.  It should
>do the trick (at least... it fixed mine).
>
>--Chris
>
>Index: rtp.c
>===================================================================
>RCS file: /usr/cvsroot/asterisk/rtp.c,v
>retrieving revision 1.22
>diff -r1.22 rtp.c
>205a206
> >       unsigned int event_end;
>209a211,213
> >       event_end = ntohl(*((unsigned int *)(data)));
> >       event_end <<= 8;
> >       event_end >>= 24;
>224a229,234
> >       else if(event_end & 0x80)
> >       {
> >               f = send_dtmf(rtp);
> >               resp = 0;
> >       }
> >
>
> > -----Original Message-----
> > From: asterisk-dev-admin at lists.digium.com [mailto:asterisk-dev-
> > admin at lists.digium.com] On Behalf Of pgrace at fierymoon.com
> > Sent: Tuesday, August 12, 2003 2:35 PM
> > To: asterisk-dev at lists.digium.com
> > Subject: [Asterisk-Dev] The Almighty X-Lite DTMF Problem
> >
> > Hey guys,
> >
> > I just was told by Rob at xten that the timestamp problem is fixed in 
>the
> > rfc2833 implementation.  I'm still having the exact same problems with
> > voicemail(2) that I was before.  Can someone please un-resolve bug 14 
>and
> > maybe I can work with someone to help debug what's happening?
> >
> > Chris H, if you're still following this topic, fire me off an e-mail if
> > you want to see new debugs..
> >
> > Thanks,
> > Pete (km-)
> > _______________________________________________
> > Asterisk-Dev mailing list
> > Asterisk-Dev at lists.digium.com
> > http://lists.digium.com/mailman/listinfo/asterisk-dev
>_______________________________________________
>Asterisk-Dev mailing list
>Asterisk-Dev at lists.digium.com
>http://lists.digium.com/mailman/listinfo/asterisk-dev

_________________________________________________________________
STOP MORE SPAM with the new MSN 8 and get 2 months FREE*  
http://join.msn.com/?page=features/junkmail
-------------- next part --------------
Index: rtp.c
===================================================================
RCS file: /usr/cvsroot/asterisk/rtp.c,v
retrieving revision 1.22
diff -u -r1.22 rtp.c
--- rtp.c	13 Aug 2003 15:25:16 -0000	1.22
+++ rtp.c	14 Aug 2003 17:02:30 -0000
@@ -68,6 +68,7 @@
	int lasttxformat;
	int lastrxformat;
	int dtmfcount;
+	unsigned int dtmfduration;
	int nat;
	struct sockaddr_in us;
	struct sockaddr_in them;
@@ -169,6 +170,7 @@
	rtp->f.mallocd = 0;
	rtp->f.src = "RTP";
	rtp->resp = 0;
+	rtp->dtmfduration = 0;
	return &rtp->f;

}
@@ -203,10 +205,17 @@
static struct ast_frame *process_rfc2833(struct ast_rtp *rtp, unsigned char 
*data, int len)
{
	unsigned int event;
+	unsigned int event_end;
+	unsigned int duration;
	char resp = 0;
	struct ast_frame *f = NULL;
	event = ntohl(*((unsigned int *)(data)));
	event >>= 24;
+	event_end = ntohl(*((unsigned int *)(data)));
+	event_end <<= 8;
+	event_end >>= 24;
+	duration = ntohl(*((unsigned int *)(data)));
+	duration &= 0xFFFF;
#if 0
	printf("Event: %08x (len = %d)\n", event, len);
#endif
@@ -222,8 +231,22 @@
	if (rtp->resp && (rtp->resp != resp)) {
		f = send_dtmf(rtp);
	}
+	else if(event_end & 0x80)
+	{
+		f = send_dtmf(rtp);
+		resp = 0;
+		duration = 0;
+	}
+	else if(rtp->dtmfduration && (duration < rtp->dtmfduration))
+	{
+		f = send_dtmf(rtp);
+		resp = 0;
+		duration = 0;
+	}
+
	rtp->resp = resp;
	rtp->dtmfcount = dtmftimeout;
+	rtp->dtmfduration = duration;
	return f;
}