Excerpts from cal's message of 2010-06-09 06:11:04 +0200:
I've got yoshimi jack session support behaving
semi-sensibly with pyjacksm and
jsweeper, but there's one point I'm struggling with. Jack midi connections
aren't
being stored by session save -
<?xml version="1.0" ?>
<jacksession>
<jackclient cmdline="yoshimi -d ${SESSION_DIR} -u 4"
infra="False" jackname="yoshimi-yoshimi" uuid="4">
<port name="yoshimi-yoshimi:Left"
shortname="Left">
<conn dst="system:playback_1"/>
</port>
<port name="yoshimi-yoshimi:Right"
shortname="Right">
<conn dst="system:playback_2"/>
</port>
<port name="yoshimi-yoshimi:In"
shortname="In"/>
</jackclient>
</jacksession>
I figure it has to be some deficiency in my port registration, but that seems
straightforward enough -
midi.port = jack_port_register(jackClient, port_name, JACK_DEFAULT_MIDI_TYPE,
JackPortIsInput, 0);
I'm not asking for in depth debugging/analysis support here (I'll figure it out
eventually!) but I'm hopeful someone might have an 'off the top of the head'
suggestion.
cheers, Cal
I remember seeing some connection weirdness when I tested with pyjacksm,
but Torben said that's already fixed in another branch. It's called
"bpjack". I am currently unable to build it due to boost naming weirdness,
but depending on your distro you might have more luck. I think he said that
he changed the dbus protocol or something so that jsweeper will likely
not work.
I tried the master branch again and made two midi input connections.
Only one of those was restored. But I'm very glad to see that saving,
quitting and loading sessions seems to work with yoshimi, including
internal state, as it should. I'm reasonably sure what you experience is
a bug in pyjacksm master that might already be fixed in pyjacksm
"bpjack".
I'm very glad to see jack_session making headway.
Good work Cal :)
--
Regards,
Philipp
--
"Wir stehen selbst enttäuscht und sehn betroffen / Den Vorhang zu und alle Fragen
offen." Bertolt Brecht, Der gute Mensch von Sezuan