[asterisk-commits] oej: branch group/pinefool-poor-mans-plc-1.4 r383440 - /team/group/pinefool-p...

[SVN commits to the Asterisk project]

Author: oej
Date: Wed Mar 20 13:03:31 2013
New Revision: 383440

URL: http://svnview.digium.com/svn/asterisk?view=rev&rev=383440
Log:
Ignore frames coming late to the party. Just ignore them.

Modified:
    team/group/pinefool-poor-mans-plc-1.4/main/rtp.c

Modified: team/group/pinefool-poor-mans-plc-1.4/main/rtp.c
URL: http://svnview.digium.com/svn/asterisk/team/group/pinefool-poor-mans-plc-1.4/main/rtp.c?view=diff&rev=383440&r1=383439&r2=383440
==============================================================================
--- team/group/pinefool-poor-mans-plc-1.4/main/rtp.c (original)
+++ team/group/pinefool-poor-mans-plc-1.4/main/rtp.c Wed Mar 20 13:03:31 2013
@@ -1333,6 +1333,15 @@
 		rtp->cycles += RTP_SEQ_MOD;
 
 	if (rtp->rxcount > 1) {
+		if (poormansplc && seqno < rtp->lastrxseqno)  {
+			/* This is a latecome we've already replaced. A jitter buffer would have handled this
+			   properly, but in many cases we can't afford a jitterbuffer and will have to live
+			   with the face that the poor man's PLC already has replaced this frame and we can't
+			   insert it AGAIN, because that would cause negative skew.
+			   Henry, just ignore this late visitor. Thank you.
+			*/
+			return AST_LIST_FIRST(&frames) ? AST_LIST_FIRST(&frames) : &ast_null_frame;
+		}
 		lostpackets = (int) seqno - (int) rtp->lastrxseqno - 1;
 		/* RTP sequence numbers are consecutive. Have we lost a packet? */
 		if (lostpackets) {