That would be pretty cool
Jonas Vautherin (41ccc731) at 22 Mar 15:28
Some default_
Jonas Vautherin (0715a97e) at 22 Mar 15:08
Also media factory
Jonas Vautherin (ab3c527c) at 22 Mar 13:42
Debug rtsp-media init
Jonas Vautherin (91919618) at 21 Mar 12:09
blah
Jonas Vautherin (6b02d959) at 19 Mar 16:09
Update subprojects/gst-plugins-bad/gst-libs/gst/codecparsers/gsth26...
Jonas Vautherin (cd346056) at 19 Mar 00:12
Add some debug output
Jonas Vautherin (6ad2836a) at 18 Mar 23:31
Update gsth264parser.c
Jonas Vautherin (f875214f) at 18 Oct 08:40
What should I do then? Keep ANDROID_NDK
, or leave it out? I'm fine with both, just not sure which one you want
Sure I can, but I'm skeptical ANDROID_NDK
is a thing at all.
ANDROID_NDK_HOME
, and also the deprecated ANDROID_NDK_ROOT
, but not ANDROID_NDK
.Is there a way to run the CI and check if something breaks without ANDROID_NDK
?
It seems to me that it can be removed, but it's your call of course
Following up on !684. This fixes the openssl build on my machine.
I am not sure if ANDROID_NDK is a thing at all, but replacing it worked for me.
One note: when building Android projects with gradle, I get the warning:
Support for ANDROID_NDK_HOME is deprecated and will be removed in the future. Use android.ndkVersion in build.gradle instead.
Not sure what it implies for Cerbero, but I wanted to mention it. Happy to make more testing if that's useful.
Resolves #317
Jonas Vautherin (96d0ee2f) at 13 Oct 21:08
android: set ANDROID_NDK_HOME instead of ANDROID_NDK
I am creating an RTSP proxy using gst-rtsp-server. My code is pretty short, but still I removed some lines to make it clearer (see below).
I essentially do what the examples do:
GstRTSPMediaFactory
, give it a launch string (in my case I connect to a remote RTSP server)GstRTSPServer
And it works really well, except when I restart the remote RTSP server while clients are connected. In that case I can see that my code receives the client-connected
callback in a loop. If I kill the client responsible for those connections and restart it, then it works again.
I do not really understand yet what happens when the remote RTSP server is restarted and why that gets me into this infinite client-connected
loop, and I am still reading the codebase. But I would be happy to get some directions here: does that sound like a bug in gst-rtsp-server, or does that sound like I am doing something wrong in my (very small) implementation?
My (almost) entire code is:
{
gst_init(NULL, NULL);
GMainLoop* main_loop = g_main_loop_new(NULL, FALSE);
GstRTSPServer* server = gst_rtsp_server_new();
g_object_set(server, "service", 8554, NULL);
char launch_str[2048] = "rtspsrc latency=40 udp-reconnect=true location=rtsp://192.168.1.12:8554/stream ! rtph264depay ! rtph264pay name = pay0";
GstRTSPMediaFactory* factory = gst_rtsp_media_factory_new();
gst_rtsp_media_factory_set_launch(factory, launch_str);
gst_rtsp_media_factory_set_shared(factory, TRUE);
gst_rtsp_media_factory_set_latency(factory, 0);
GstRTSPMountPoints* mount_points = gst_rtsp_server_get_mount_points(server);
gst_rtsp_mount_points_add_factory(mount_points, served_stream_name, factory);
g_object_unref(mount_points);
gst_rtsp_server_attach(server, NULL);
g_signal_connect(server, "client-connected", (GCallback)client_connected, NULL);
g_main_loop_run(main_loop);
}