Custom Query (2683 matches)

It would be useful, if Xpra would show the currently used video encoding and video encoder in {System Tray > Session Info > Connection}, similar to how the codec is displayed for "Speaker".

xpra-2.1.1 fails to compile with ffmpeg-3.1.7 on Alpine Linux 3.5:

gcc -fno-strict-aliasing -Os -fomit-frame-pointer -g -DNDEBUG -Os -fomit-frame-pointer -g -fPIC -I/usr/include/python2.7 -c xpra/codecs/enc_ffmpeg/encoder.c -o build/temp.linux-x86_64-2.7/xpra/codecs/enc_ffmpeg/encoder.o -Wall -Werror -fPIC
xpra/codecs/enc_ffmpeg/encoder.c: In function 'initencoder':
xpra/codecs/enc_ffmpeg/encoder.c:18128:36: error: 'FF_PROFILE_H264_MULTIVIEW_HIGH' undeclared (first use in this function)
   __pyx_t_3 = __Pyx_PyInt_From_int(FF_PROFILE_H264_MULTIVIEW_HIGH); if (unlikely(!__pyx_t_3)) __PYX_ERR(0, 448, __pyx_L1_error)
                                    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
xpra/codecs/enc_ffmpeg/encoder.c:18128:36: note: each undeclared identifier is reported only once for each function it appears in
xpra/codecs/enc_ffmpeg/encoder.c:18164:36: error: 'FF_PROFILE_H264_STEREO_HIGH' undeclared (first use in this function)
   __pyx_t_3 = __Pyx_PyInt_From_int(FF_PROFILE_H264_STEREO_HIGH); if (unlikely(!__pyx_t_3)) __PYX_ERR(0, 451, __pyx_L1_error)
                                    ^~~~~~~~~~~~~~~~~~~~~~~~~~~
error: command 'gcc' failed with exit status 1

The build error should be perfectly reproducible, if you take ​my Xpra Dockerfile for Alpine, replace XPRA_VERSION=2.0.2 with XPRA_VERSION=2.1.1 and remove the invocation of patch.

Seems whatever server side 'speaker-codec' configuration value I place in:

/etc/xpra/conf.d/20_sound.conf 

is ignored or superseded by some other configuration somewhere.

Using:

speaker-codec = wav 

or

speaker-codec = flac

does not set the codec for the session - it always starts with opus / ogg:

2019-02-28 17:20:59,365 sound output using audio codec opus
2019-02-28 17:20:59,365 sound output using container format ogg

Setting this value client side however applies the codec okay for the session (ie: --speaker-codec=flac+ogg)