On Tue, 2013-02-19 at 20:38 +0900, atsukada at users.sourceforge.net wrote: > From: Akihiro Tsukada <atsukada at users.sourceforge.net> > > --- > src/pavucontrol.glade | 19 ++++++++++++++++--- > src/pavucontrol.h | 1 + > src/sinkwidget.cc | 10 ++++++++++ > src/sinkwidget.h | 2 +- > 4 files changed, 28 insertions(+), 4 deletions(-) > > diff --git a/src/pavucontrol.glade b/src/pavucontrol.glade > index 6defb3d..4b3dd01 100644 > --- a/src/pavucontrol.glade > +++ b/src/pavucontrol.glade > @@ -427,9 +427,6 @@ > <property name="n_rows">2</property> > <property name="n_columns">3</property> > <child> > - <placeholder/> > - </child> > - <child> > <object class="GtkCheckButton" id="encodingFormatPCM"> > <property name="label" translatable="yes">PCM</property> > <property name="use_action_appearance">False</property> > @@ -499,6 +496,22 @@ > <property name="bottom_attach">2</property> > </packing> > </child> > + <child> > + <object class="GtkCheckButton" id="encodingFormatAAC"> > + <property name="label" translatable="yes">AAC</property> > + <property name="use_action_appearance">False</property> > + <property name="visible">True</property> > + <property name="can_focus">True</property> > + <property name="receives_default">False</property> > + <property name="draw_indicator">True</property> > + </object> > + <packing> > + <property name="left_attach">2</property> > + <property name="right_attach">3</property> > + <property name="top_attach">1</property> > + <property name="bottom_attach">2</property> > + </packing> > + </child> > </object> > <packing> > <property name="expand">False</property> > diff --git a/src/pavucontrol.h b/src/pavucontrol.h > index 65cb913..affaa89 100644 > --- a/src/pavucontrol.h > +++ b/src/pavucontrol.h > @@ -41,6 +41,7 @@ > > #define HAVE_SOURCE_OUTPUT_VOLUMES PA_CHECK_VERSION(0,99,0) > #define HAVE_EXT_DEVICE_RESTORE_API PA_CHECK_VERSION(0,99,0) > +#define HAVE_MPEG2_AAC_PASSTHROUGH PA_CHECK_VERSION(3,99,0) This is not necessary. You can just check whether PA_ENCODING_MPEG2_AAC_IEC61937 is defined. Or actually you can't, because format.h doesn't define that, but I'll fix that. Fixed: http://cgit.freedesktop.org/pulseaudio/pulseaudio/commit/?id=773ca81dea1a18f93793ebfab4242d9a9ff73585 > > enum SinkInputType { > SINK_INPUT_ALL, > diff --git a/src/sinkwidget.cc b/src/sinkwidget.cc > index 1726550..35b0aa7 100644 > --- a/src/sinkwidget.cc > +++ b/src/sinkwidget.cc > @@ -62,6 +62,16 @@ SinkWidget::SinkWidget(BaseObjectType* cobject, const Glib::RefPtr<Gtk::Builder> > encodings[i].encoding = PA_ENCODING_DTS_IEC61937; > x->get_widget("encodingFormatDTS", encodings[i].widget); > encodings[i].widget->signal_toggled().connect(sigc::mem_fun(*this, &SinkWidget::onEncodingsChange)); > + > + ++i; > + x->get_widget("encodingFormatAAC", encodings[i].widget); > +#if HAVE_MPEG2_AAC_PASSTHROUGH > + encodings[i].encoding = PA_ENCODING_MPEG2_AAC_IEC61937; > + encodings[i].widget->signal_toggled().connect(sigc::mem_fun(*this, &SinkWidget::onEncodingsChange)); > +#else > + encodings[i].encoding = PA_ENCODING_INVALID; > + encodings[i].widget->set_sensitive(false); > +#endif I'm afraid this is not good enough. One question is what happens if pavucontrol is compiled on a system with 4.0 headers, and run on a system with 3.0 libpulse? In that case you pass libpulse an encoding that it doesn't recognize. Well, it happens that libpulse doesn't validate the encoding values before sending them to the server, so there's no breakage. But it's still ugly to rely on the libpulse implementation details. I'd remove the check altogether and just depend on libpulse 4.0. Another question is what does the server do if it doesn't support the encoding? The answer is that saving the formats fails altogether, so the check is clearly not good enough. You need to check whether the server supports PA_ENCODING_MPEG2_AAC_IEC61937. How to do that? The check should be based on the protocol version that the server support. That can be queried with pa_context_get_server_protocol_version(). But there's a problem: the protocol version was not incremented when the new enumeration value was added. Adding the new encoding wasn't as simple as it seemed... Incrementing the protocol version is easy enough, but we also need to look at every case where encodings are passed between a client and a server, and decide the best way to handle situations where the receiving end doesn't support the encoding that the sender would like to send. I'll start working on this. -- Tanu
