On Wed, 2016-02-17 at 15:57 +0200, Tanu Kaskinen wrote: > On Wed, 2016-02-17 at 18:47 +0530, Arun Raghavan wrote: > > On Mon, 2016-02-15 at 13:12 +0200, Tanu Kaskinen wrote: > > > On Mon, 2016-01-18 at 13:06 +0530, [email protected] wrote: > > > > + pa_log("The beamformer currently only supports > > > > targeting along the azimuth"); > > > > + goto fail; > > > > + } > > > > + > > > > + direction.s[0] = f[0]; > > > > + direction.s[1] = f[1]; > > > > + direction.s[2] = f[2]; > > > > + } > > > > + > > > > + /* The beamformer gives us a single channel */ > > > > + out_ss->channels = 1; > > > > + pa_channel_map_init_mono(out_map); > > > > > > Could this be moved to fixate_spec()? I don't like the audio > > > format > > > parameters being calculated all over the place. > > > > I wanted to do that, but we can't because we need the capture > > channel > > count for parsing the mic geometry (there is a comment explaining > > that > > where if (beamforming) { ... } is). > > I meant moving only the out_ss and out_map assignments. Those don't > depend on the capture channel count. fixate_spec() will then need to > take the beamforming bool as a parameter, but I think that's fine.
Ah, sure, that's doable. -- Arun _______________________________________________ pulseaudio-discuss mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
