[vlc-devel] [PATCH 09/39] vout: request the display with a decoder device

[Rémi Denis-Courmont]

Le perjantaina 4. lokakuuta 2019, 12.47.01 EEST Alexandre Janniaux a écrit :
> Hi,
> 
> That's indeed a pivotal point to tackle, but it's probably not an
> issue in this design.
> 
> The main point of window providing is that you don't have to actually
> close the native window when the vout_window is closed by the core.
> 
> Audio visualization are creating a vout_thread, which creates a window
> so this doesn't conflict with this AFAIK.

An audio visualisation may create a video output as a helper to output 
bitmaps, but that is an implementation detail of the audio visualisation plug-
in and indeed some (GL-based) don't create a video output already now.

Obviously if the window or the window allocator would have to be supplied to 
both the audio side for visualisation and the video side for video output. But 
the window can be (and already sometimes is) used without a video output. So 
it seems unavoidable that the video output is created from the window and not 
vice-versa.

Back to the decoder device, it comes down to maximizing commonality between 
the different decoder use cases. As far as I can guess, there's more 
opportunity for sharing code if the decoder device comes from the window to 
the decoder, and then *from* the decoder to the video output rather than from 
the video output to the decoder. Simply because we will eventually have 
decoder device without video output.

-- 
レミ・デニ-クールモン
http://www.remlab.net/