Original commit message from CVS:
fixed pipeline for vob files so we get sound (bad sound but still), Wim claims we only need 50 more rewrites to get it working well
Original commit message from CVS:
Use stereo2mono plugin for scopes which require mono input.
It would be better if esdmon and osssrc were able to give mono output:
they need caps negotiation.
Original commit message from CVS:
* get up-to-date with the gst_caps_debug api
* improved capsnego in mad
* improved capsnego in adder
* improved capsnego in intfloat plugins
* unbroke capsnego in stereomono plugins
* fix cothread stack allocation within the main thread in new cothreads
Original commit message from CVS:
* get up-to-date with the gst_caps_debug api
* improved capsnego in mad
* improved capsnego in adder
* improved capsnego in intfloat plugins
* unbroke capsnego in stereomono plugins
* fix cothread stack allocation within the main thread in new cothreads
Original commit message from CVS:
adding videosink
is there any good reason why the videosink plugin didn't just go in the xvideo
dir ? It's not like we *need* to have the most dirs we can get ;)
if no one objects then I guess I'll move it to the sensible place
Original commit message from CVS:
let's not update here, it's not necessary; ie, if you run cvs update in gst-plugins,
it will update common automatically, and we don't support half-updated trees
Original commit message from CVS:
adding 'videosink' element
It's basically a rewritten xvideosink, but is more modular and allows 'hooks'.
It is needed by GstVideoWidget in gst-player
Original commit message from CVS:
Fix up esdmon timestamps, so that they don't go completely wrong if
someone changes the frequency, depth or mono/stereo parameters.
Original commit message from CVS:
Added esdmon plugin.
Monitors output from the esd sound daemon.
Particularly useful for displaying a visualisation of the audio being
played on your system. eg:
gst-launch esdmon ! goom ! { queue leaky=yes ! colorspace ! sdlvideosink }
You can even have several of them going at once. :)
Original commit message from CVS:
Set the leaky parameter to 1 when using visualisers; if the queue gets
full, it is fine to drop data going to the visualiser. This solves the
problem I was having with playback freezing while playing some files with
visualisers.
Original commit message from CVS:
* on sink connect, check if the current pad is compatible with the given
caps
* cleaned up debug output
* change pad template to only accept allowed sample rates
if these changes are considered ok by others then the same should be
applied to other encoding plugins (notably the compatibility check)
Original commit message from CVS:
* fixed compiler warning in sink
* put debug output in proper categories
* don't set caps while opening device (in state change) since caps setting
is not allowed if element isn't ready yet
Original commit message from CVS:
somebody help me fix lame ;)
I commented out the state change function because it is called before lame has the right caps.
Is the state change function still necessary ?
in any case, at least now lame actually listens to osssrc re: rate and channels
