Feature request: Support Audio ChannelMask with AviSynth+ r4001

[Jose Peirote]

With the last AviSynth+ r4001 the audio channelmask is stored in the VideoInfo.image_type variable (see details in the new ​https://github.com/AviSynth/AviSynthPlus/blob/master/avs_core/include/avisynth.h )

At least a decoder test (​https://github.com/HomeOfAviSynthPlusEvolution/L-SMASH-Works/issues/35) can store that info, or can be set in avs script with SetChannelMask().

Now only know VirtualDub2 to output WAVE_FORMAT_EXTENSIBLE with the correct channel map taken from that info.

Please add that feature to ffmpeg, now is ignored and a default channel mask, based in numchannels, is output.

Added a avs script and a eac3 sample to be load in VirtualDub2 and save audio:
MaskChannels : 22031 (FL FR FC LF SL SR TFL TFR).
If don't want install the test version of LWLibavAudioSource uncomment the SetChannelMask("22031")

Compare the output with the:

ffmpeg -i chmask.avs chbad.wav

MaskChannels : 1599 (FL FR FC LF BL BR SL SR)

[Francesco Bucciantini]

I can confirm that the new LWLibavAudioSource() is correctly populating the audio properties inside Avisynth (checked in AVSPmod mod and VirtualDub), so yeah, it would be nice to have them supported inside ffmpeg too.

Just as a reference for the ffmpeg devs, here's the stable build of Avisynth 3.7.3 which introduced support to the new ChannelMask (i.e SetChannelMask() ): ​https://github.com/AviSynth/AviSynthPlus/releases/tag/v3.7.3

[Jose Peirote]

Using other decoders can also obtain the same result with a workaround (until supported), see ​https://forum.doom9.org/showthread.php?p=1989675#post1989675

global OPT_UseWaveExtensible = true # Needed to use global OPT_dwChannelMask=
FFAudioSource("8w3D.ec3")
global OPT_dwChannelMask=FFCHANNEL_LAYOUT

or

global OPT_UseWaveExtensible = true # Needed to use global OPT_dwChannelMask=
BestAudioSource("8w3D.ec3")
global OPT_dwChannelMask=BASCHANNEL_LAYOUT