I have two GStreamer instances : a sender and a receiver. I want to stream RTP / VP8 video. It works perfectly fine if I stream via UDP, like this :
gst-launch-0.10 -v videotestsrc ! vp8enc ! rtpvp8pay ! udpsink host=127.0.0.1 port=9001
gst-launch-0.10 udpsrc port=9001 caps = "application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)VP8-DRAFT-IETF-01, payload=(int)96" ! rtpvp8depay ! vp8dec ! ffmpegcolorspace ! autovideosink
That works fine. But when I try to stream throug a FIFO / named pipe (done with mkfifo()) with :
gst-launch-0.10 -v videotestsrc ! vp8enc ! rtpvp8pay ! filesink location = myPipe
gst-launch-0.10 filesrc location = myPipe ! capsfilter caps = application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)VP8-DRAFT-IETF-01, payload=(int)96 ! rtpvp8depay ! vp8dec ! ffmpegcolorspace ! autovideosink
It fails and my receiver continusouly outputs :
WARNING: from element /GstPipeline:pipeline0/GstRtpVP8Depay:rtpvp8depay0: Could not decode stream.
Additional debug info:
gstbasertpdepayload.c(387): gst_base_rtp_depayload_chain (): /GstPipeline:pipeline0/GstRtpVP8Depay:rtpvp8depay0:
Received invalid RTP payload, dropping
I think I read somewhere (but can't find it again) that it was because when using UDP, the RTP packets were separated properly, while using a named pipe like this, the packets being written are "chained" (not properly separated) and thus gstreamer doesn't know how much bytes to read to get a RTP packet.
Is this correct, and if yes, how can I change that ? Thanks in advance !
I finally solved my problem.
I did not manage the byte-stream thing through a pipe, but I managed to use an AppSrc to feed the gst pipeline.
So my whole pipeline (might be useful for other people) looks like this : appsrc -> rtpvp8depay -> vp8dec -> videoconvert -> videoscale -> appsink (I'm using Gstreamer1.0 on ArchLinux).
Hope this helps !