Re: [Live-devel] How to repair the abnormally interrupted openRTSP recorded file?

2010-06-11 Thread Ross Finlayson
Look at the function "signalHandlerShutdown()" in "testProgs/playCommon.cpp". Remember, You Have Complete Source Code. -- Ross Finlayson Live Networks, Inc. http://www.live555.com/ ___ live-devel mailing list live-devel@lists.live555.com http://lists.l

[Live-devel] How to repair the abnormally interrupted openRTSP recorded file?

2010-06-11 Thread Jim Cao
Hi, All: I'm using openRTSP to receive live video/audio from remote web camera, and I redirect the stdout to a local file. So I can play the recorded file later. As we all know, we need use "kill -HUP" or "kill -USR1" to terminate the openRTSP to make the file written properly. But in my cases,

Re: [Live-devel] memory leak using tcp

2010-06-11 Thread Ross Finlayson
Our simple solution for this is to prevent recreating the socket hash table in lookupSocketDescriptor when createIfNotFound is set to false and exiting with NULL before socketHashTable is called. OK, I think I have fixed this now in the latest release of the code (2010.06.11). -- Ross Finlay

Re: [Live-devel] segmentation fault with latest version

2010-06-11 Thread Ross Finlayson
I'd like to add another (once this one is patched) in RTSPClient::constructSubsessionURL() on line 1169. if prefix is NULL, it's set to "" so strlen(prefix) will return 0 so you try to look at prefix[-1] OK, thanks. This has now been fixed in the most recent release (2010.06.11). -- Ross F

Re: [Live-devel] segmentation fault with latest version

2010-06-11 Thread Denis Charmet
Ross Finlayson wrote: I tried to use vlc with the latest version. I thought it would work as you say it is backward compatible, but I have a segmentation fault. I think the line 136 in RTSPClient.cpp is buggy because if parameterName is NULL, sprintf segfaults. Oops, yes - my mistake. Thi

Re: [Live-devel] MPEG2TransportStreamIndexer still fails on some files

2010-06-11 Thread Melms, Kenneth
Imagine my chagrin... I had downloaded, built and tested the updated MPEG2TransportStreamIndexer you made on 4/10 - but the one that was residing in my system path was an older version. Please accept my most humble apologies for wasting your time on this one Ross - and thanks! K From: live-d

Re: [Live-devel] memory leak using tcp

2010-06-11 Thread Alexandr Němec
Dear Ross & all, first of all, sorry for the broken link in my last response to this topic, the correct link follows below. My e-mail client has been playing around with the links. http://lists.live555.com/pipermail/live-devel/2005-September/003174.html Now. I can confirm, that this old prob

Re: [Live-devel] MPEG2TransportStreamIndexer with H264

2010-06-11 Thread Marco Amadori
In data venerdì 11 giugno 2010 10:27:35, Ross Finlayson ha scritto: > Unfortunately I don't know how H.264 video is supposed to be carried > within MPEG Transport Streams. (Could someone please point me at the > official document that is supposed to define this?) http://en.wikipedia.org/wiki/H.2

Re: [Live-devel] MPEG2TransportStreamIndexer with H264

2010-06-11 Thread Ross Finlayson
MPEG2IndexFromTransportStream.cpp, line 220 if (payload_unit_start_indicator) { // Note: The following works only for MPEG-2 data # u_int8_t PES_header_data_length = fInputBuffer[totalHeaderSize+8]; //fprintf(stderr, "PES_header_data_length: %d\n", PES_header_data_length);//###

Re: [Live-devel] memory leak using tcp

2010-06-11 Thread Alexandr Němec
Dear Ross & all, I do not believe that these memory leaks still exist in the current release, some of them are apperently solved in the code. As it seems to me, only one of the memory leaks was not "adequately addressed" at that time. For this http://lists.live555.com/pipermail/live-devel/200