On Mon, 2007-08-06 at 22:25 +1000, Josh Stewart wrote: > On 8/6/07, iain <[EMAIL PROTECTED]> wrote:> > > Are you using GStreamer by itself or are you using it through the > > clutter-gst stuff? Are you reading the TV streams from the socket and > > writing to a file, and then telling GStreamer to play the file? > > > > I was hoping to use as much of the clutter-gst stuff as possible. Yes, > currently reading the stream from the socket into a file and getting a > VideoTexture to read from that.
What I think you'll have to do is to build the pipeline by yourself and use the ClutterGstVideoSink as the output for the pipeline. "fdsrc ! queue ! decodebin" would be the start of the pipeline and listen for the new-decoded-pad signal from decode bin. On that signal you'd connect a ClutterGstVideo to the video pad and a standard gstreamer audio sink to any audio pads. The queue will then control the buffering and you change the properties on it to set the length of the buffer. > Yep I think that's exactly it. Creating a buffer / queue is easy > enough, is it possible to use GstFDSrc just using clutter-gst (Maybe > through a URI), or is this something I need to be interfacing GST > directly for? It looks like at the moment you'll need to get dirty with some raw GStreamer coding, but the ClutterGstVideoSink is available for getting Clutter to display the data at least. iain -- To unsubscribe send a mail to [EMAIL PROTECTED]
