On Wed, 2010-09-29 at 13:04 +0200, IOhannes m zmoelnig wrote: > On 2010-09-28 15:20, Sebastian Dröge wrote: > > On Tue, 2010-09-28 at 14:49 +0200, IOhannes m zmoelnig wrote: > >> Package: wnpp > >> Severity: wishlist > >> Owner: IOhannes m zmoelnig <zmoelnig@iem.at> > >> > >> > >> * Package name : gstreamer0.10-v4l2loopback > >> Version : 0.2.0 > >> Upstream Author : IOhannes m zmoelnig > >> * URL : http://github.com/umlaeute/gst-v4l2loopback > >> * License : LGPL > >> Programming Lang: C > >> Description : GStreamer sink to write into v4l2 loopback devices > >> > >> provides a GStreamer sink element "v4l2loopback" that takes a video-stream and > >> writes it into a V4L2 loopback device. > >> ordinary v4l2-enabled applications can then access the video as if it was coming > >> from an ordinary capture device. > >> among other things, this allows to add stream-receiving and support for non-v4l2 > >> capture devices capabilities to V4L2-only applications. > > > > How is this different from the v4l2sink element from gst-plugins-good? > > That element can output a video stream to any v4l2 device that supports > > this, including a loopback device. > > > the difference is, that the v4l2sink is currently unable to send data to > the v4l2loopback device. > the problem seems to be, that the v4l2loopback device accepts any > format, whereas the v4l2sink expects the device to return the one single > format accepted... > > on the long run, i would prefer if v4l2sink could write to the loopback > device as well. Well, instead of writing something new wouldn't it be better to fix v4lsink instead? :) I'd be happy to review patches and commit them upstream :P But IIRC v4l2sink only expects the v4l2 device to list all supported formats and then can use any of these formats. Also IIRC it was written by someone who wanted to use it for a loopback device so it actually worked at some point ;)
Attachment:
signature.asc
Description: This is a digitally signed message part