matthew1000 / gstreamer-cheat-sheet

Gstreamer command-line cheat sheet

GStreamer command-line cheat sheet

GStreamer is a powerful library for manipulating audio and video - including live streams. This repo provides:

• a cheat sheet for GStreamer on the command-line, and
• a few Python examples.

Whilst the command line is great, programmatic usage (in Python or another language) allows you to dynamically manipulate the A/V streams.

Contents

• Installing
• Basics
• Test streams
• RTMP
• Mixing video & audio
• Images
• Queues
• Writing to files
• Sending to multiple destinations (tee)
• Sharing and receiving pipelines (including sending/receiving video from shared memory)
• Network transfer (including how to send so that VLC can preview)
• RTP
• SRT

Sources and references

• Basic command line reference
• Pipeline examples
• List of all Gstreamer plugins
• Handy elements

Other cheat-sheets

• https://github.com/xmementoit/gstreamerCheatsheet/blob/master/README.md
• https://gist.github.com/nebgnahz/26a60cd28f671a8b7f522e80e75a9aa5

Interacting with the GStreamer pipeline

If you want to interact with GStreamer after it's started (e.g. respond to an event, or dynamically change a pipeline), the command-line GStreamer doesn't really cut it. Instead, here are some options:

• GStreamer Daemon (gstd) - allows setting and updating via a TCP connection
• Snowmix - an open-source live video mixer
• Develop using the GStreamer library, in either C, Python, or C#/.NET

Python with GStreamer

Python is an easy language, so it's no surprise that it's good way to develop using GStreamer.

Some example scripts can be found in the python_examples/ directory.

Other good GStreamer Python resources that I've found:

• Getting started with GStreamer with Python
• Python GStreamer Tutorial
• Function reference
• Including a useful mapping from C
• Nice example script

C/C++ with GStreamer

My favourite reference is Valadoc

Problems or suggestions with this guide?

If you spot anything incorrect or incomplete, reports are welcome, either using issues or pull requests

My GStreamer project

Creating this guide gave me enough GStreamer understanding to make a prototype Brave, a live video editor for the cloud. (Regrettably this prototype has not been updated in a few years.)

gstreamer-cheat-sheet's Issues

sending-to-an-rtmp-server needs location example

gst-launch-1.0 videotestsrc  is-live=true ! \
    queue ! x264enc ! flvmux name=muxer ! rtmpsink location="$RTMP_DEST live=1"

https://github.com/matthew1000/gstreamer-cheat-sheet/blob/master/rtmp.md#sending-to-an-rtmp-server

RTMP_DEST isn't defined.

it would be great if I knew how to make it work with an example from
https://github.com/arut/nginx-rtmp-module

I'm currently trying different things, all give me:
Could not connect to RTMP stream "rtmp://localhost/xyz/ live=1" for writing

for all xyz

Gstream rtmp to youtube broken?

Hello

Using gst-launch to try to stream to youtube via the documented command:

export RTMP_DEST="rtmp://a.rtmp.youtube.com/live2/xxxx-xxxx-xxxx"
gst-launch-1.0 \
    videotestsrc is-live=1 \
    ! videoconvert \
    ! "video/x-raw, width=1280, height=720, framerate=30/1" \
    ! queue \
    ! x264enc cabac=1 bframes=2 ref=1 \
    ! "video/x-h264,profile=main" \
    ! flvmux streamable=true name=mux \
    ! rtmpsink location="${RTMP_DEST} live=1" \
    audiotestsrc is-live=1 wave=ticks \
    ! voaacenc bitrate=128000 \
    ! mux.

Using version:

gst-launch-1.0 version 1.22.2
GStreamer 1.22.2
Unknown package origin

Seems this command does not work. Can you confirm that it still works on your side.

Thank you very much. ❤️

Question on using appsink on android

i looked through the implementation of gstreamer and it seems like there is a function that i need named Gst.parse_launch_full as well as pipeline.getElementByName("appsink") as AppSink

however, i cant find it in the android implementation of gstreamer.
I looked into the c code and could only find these functions

/* List of implemented native methods */
static JNINativeMethod native_methods[] = {
        {"nativeInit", "()V", (void *) gst_native_init},
        {"nativeFinalize", "()V", (void *) gst_native_finalize},
        {"nativePlay", "()V", (void *) gst_native_play},
        {"nativePause", "()V", (void *) gst_native_pause},
        {"nativeSurfaceInit", "(Ljava/lang/Object;)V",
                (void *) gst_native_surface_init},
        {"nativeSurfaceFinalize", "()V", (void *) gst_native_surface_finalize},
        {"nativeClassInit", "()Z", (void *) gst_native_class_init},
        {"nativeGetGStreamerInfo", "()Ljava/lang/String;", (void *) gst_native_get_gstreamer_info},
        {"nativeSetPipeline", "(Ljava/lang/String;)V", (void *) gst_native_set_pipeline}
};

is there something else equivalent to ```Gst.parse_launch_full```` ?

my goal is to be able to capture individual frames from the video stream so that i can save screenshots (not of the whole screen), save the video,, and also perform object detection (which implies i need to be able to extract the bitmap)

Pulling 2 RTMP Sources and creating a PiP on a Jetson Nano

I'm trying to create a pipeline that will pull in 2 separate local RTMP sources, make them Picture-in-picture, mux it all together and send to a SRTLA server, these are the pipelines I tried but failed;

First Try:

rtmpsrc name=cam1 location=rtmp://127.0.0.1/live/cam1 ! flvdemux name=demux0 ! queue ! demux0.video ! identity name=v_delay signal-handoffs=TRUE ! h264parse ! nvv4l2decoder ! nvvidconv ! queue ! comp.sink_0 
rtmpsrc name=cam2 location=rtmp://127.0.0.1/live/cam2 ! flvdemux name=demux1 ! queue ! demux1.video ! identity signal-handoffs=TRUE ! h264parse ! nvv4l2decoder ! nvvidconv ! queue ! comp.sink_1 

nvcompositor name=comp sink_0::xpos=0 sink_0::ypos=0 sink_0::width=1920 sink_0::height=1080 sink_1::xpos=0 sink_1::ypos=240 sink_1::width=320 sink_1::height=240 ! 
videorate ! video/x-raw,framerate=60/1 ! 
nvvidconv interpolation-method=5 ! 

nvv4l2h265enc control-rate=1 qp-range="28,50:0,38:0,50" iframeinterval=60 preset-level=4 maxperf-enable=true EnableTwopassCBR=true insert-sps-pps=true name=venc_bps ! 
h265parse config-interval=-1 ! queue max-size-time=10000000000 max-size-buffers=1000 max-size-bytes=41943040 ! mux. 

demux0. ! queue ! audio/mpeg ! decodebin ! audioconvert ! audioresample ! autoaudiosink 

mpegtsmux name=mux ! 
appsink name=appsink

Second Try:

rtmpsrc name=cam1 location=rtmp://127.0.0.1/live/cam1 ! 
flvdemux name=demux0
demux0.video ! identity name=v_delay0 signal-handoffs=TRUE ! h264parse ! nvv4l2decoder ! nvvidconv ! queue ! comp.sink_0

rtmpsrc name=cam2 location=rtmp://127.0.0.1/live/cam2 ! 
flvdemux name=demux1
demux1.video ! identity name=v_delay1 signal-handoffs=TRUE ! h264parse ! nvv4l2decoder ! videobox left=-4 right=-4 top=-4 bottom=-4 ! nvvidconv ! queue ! comp.sink_1

nvcompositor name=comp sink_0::width=1920 sink_0::height=1080 sink_1::width=640 sink_1::height=360 sink_1::xpos=10 sink_1::ypos=10 ! 
queue ! identity name=v_delay0 signal-handoffs=TRUE ! nvvidconv interpolation-method=5 ! 
queue ! identity name=v_delay1 signal-handoffs=TRUE ! nvvidconv interpolation-method=5 ! 
nvv4l2h265enc control-rate=1 qp-range="28,50:0,38:0,50" iframeinterval=60 preset-level=4 maxperf-enable=true EnableTwopassCBR=true insert-sps-pps=true name=venc_bps ! 
h265parse config-interval=-1 ! queue max-size-time=10000000000 max-size-buffers=1000 max-size-bytes=41943040 ! 
mpegtsmux name=mux ! appsink name=appsink

demux0.audio ! aacparse ! avdec_aac ! identity name=a_delay signal-handoffs=TRUE ! volume volume=1.0 ! 
audioconvert ! opusenc bitrate=320000 ! opusparse ! queue max-size-time=10000000000 max-size-buffers=1000 ! mux. 

Please help if at all possible, thank you.

Mixing audio sources from n participant and sending n-1 sources as output to each participant

Hi,

This is actually not an issue, but I was looking for ways to implement how with gstreamer( can be audiomixer) with audio sources mixed from n participants can be used and later from output each participant listens to only audio from other n-1, not the self voice

Its same as in one of the GStreamer issues at http://gstreamer-devel.966125.n4.nabble.com/audiomixer-implementation-for-audio-conference-cancel-user-1-contribution-when-streaming-to-user-1-td4659790.html#a4693565

I came across this https://github.com/matthew1000/gstreamer-cheat-sheet/blob/master/mixing.md#mixing-audio section where I was curious if this above issue can be asked in this forum

Any suggestions on this regard will really help

Thanks

Question on splitting pipelines

I need to read rtsp and after that to split pipeline to several branches. Smth like tee plugin does, but tee copy the frame. I need that first frame to go to one branch of pipeline and another one to go to another branch the third frame to the first branch and so on... Is it possible to do in Gstreamer?

v4l2loopback for "Sharing and splitting pipelines" too?

Thanks for this project, very useful!