Fedora Tour

[Jeff Spaleta]

On 12/14/05, Karsten Wade <kwade at redhat.com> wrote:
> Is the work to make such an mplayer package and get it accepted upstream
> something we can handle ourselves?  What can we offer the Mplayer
> developers to get them to do or accept any necessary changes?

Oh, I very much doubt that mplayer upstream is interested in helping
us functionally cripple an mplayer package to the extent necessary to
get distributed as part of Extras.
This page doesn't give me warm fuzzies http://rpm.greysector.net/yum.html

But I'm reading over the wikipage again and I'm not sure exactly why
richard requires the use of mplayer.  To use theora_encode at the
moment you need a wav input for audio
and a yuv for video.  I'm not an expert in using gst-launch, but can
gst as shipped in fedora produce yuv format instead of having to go
through mplayer?   And can we just use a gst pipeline construction to
mix in the wav file audio with the original theora video without
having to decode to yuv video first and avoid using theora_encode all
together?
I know exactly why Richard wrote the instructions has he did, he's
following the example instructions that come with the libtheora
package documentation.. but there has to be a way to do all this
inside the existing gstreamer framework.

For example the theoradec element in gstreamer should be able to take
the place of the mplayer step. I just don't know how to use gstreamer
pipelines well enough to write up sample instructions.



-jef