[Live-devel] Regarding- using '-4' option in openRTSP

Jegadish jegadishd at gmail.com
Wed Dec 5 11:56:15 PST 2007


Yes, this was the problem. I was ending with CTR+C. Now I see the ftyp box.
Then, I got a segmentaion fault in QuickTimeFileSink.cpp (line no. 1775)
  char* psets =
      strDup(fCurrentIOState->fOurSubsession.fmtp_spropparametersets());
  size_t comma_pos = strcspn(psets, ",");

here psets returns null, as I *do not* pass parameter set information in the
sdp file.
****sprop-parameter-sets (like
a=fmtp:99 profile-level-id=42A01E; packetization-mode=1;
sprop-parameter-sets=Z0IACpZTBYmI )

Later a sprop-parameter-sets value was copied from a sample sdp file in rfc
3984. This results in a *playable .mp4 file*. (although the quality was not
good..)

Based on rfc 3984, it appears that parameter set (sequence and picture are
sent in RTP packets, with different NAL type).

So, I plan to change the code, so as to pick the parameter set information
from RTP packets in addition to sdp file. ---> Please provide informantion
on this, if any. (like, if this would make sense)

And regarding picture width, height and frame rate information:
>From the available information, it appears that width, height information
can be retrieved from the parameter sets information. (but not sure about
frame rate). If this turns out to be true, then a valid file can be
generated without the -w, -h and -f options..Hope this turns to be true..

Thanks for your earlier quick reply...it has really put me on track..Thanks
a lot..
Please inform, if you have suggestions/comments on the above data..

Regards,
jegadish


On Dec 4, 2007 11:55 PM, Ross Finlayson <finlayson at live555.com> wrote:

> >Hi, Thanks for your quick reply..it was very helpful..
> >I looked in the file being produced by the QuickTimeFileSink.cpp. It
> >does not write some mandatory container boxes(like ftyp, moov etc)
> >that are needed to make it a playable mp4 file.
>
> Yes it does!  Look at the code again.
>
> Are you sure that you are ending "openRTSP" properly?  You can't just
> 'kill' the application (e.g., by typing <control>-C).  You *must* let
> "openRTSP" terminate normally (perhaps using the "-d <duration>"
> option), or (if you're on Unix) signalling it with SIGUSR1 or SIGHUP.
> If you do this, then the 'ftyp' and 'moov' atoms will be added to the
> file properly at the end.
> --
>
> Ross Finlayson
> Live Networks, Inc.
> http://www.live555.com/
> _______________________________________________
> live-devel mailing list
> live-devel at lists.live555.com
> http://lists.live555.com/mailman/listinfo/live-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.live555.com/pipermail/live-devel/attachments/20071205/c5ea038c/attachment-0001.html 


More information about the live-devel mailing list