Fix MIDI output for Kievan

Fix MIDI output for Kievan

Adding KievanStaff and KievanVoice for MIDI output of Kievan notation.

[aleksandr.andreev, 2012-05-04 02:39:16]

Here is a patch that attempts to fix issues with MIDI playback of Kievan
notation.

[Carl, 2012-05-04 02:59:06]

Looks like it works, but I think it can (and should) be simplified.  If it
can't, then LGTM.

http://codereview.appspot.com/6193043/diff/1/ly/performer-init.ly
File ly/performer-init.ly (right):

http://codereview.appspot.com/6193043/diff/1/ly/performer-init.ly#newcode49
ly/performer-init.ly:49: \consists "Staff_performer"
Do these \consists need to be added?  Does the \alias Staff bring in the
Staff_performer and the Key_performer from the Staff Performer_group?

http://codereview.appspot.com/6193043/diff/1/ly/performer-init.ly#newcode88
ly/performer-init.ly:88: \consists "Dynamic_performer"
Again, don't the performers from the Voice performer group get added
automatically by the \alias Voice?

[aleksandr.andreev, 2012-05-04 12:59:32]

For whatever reason, \alias Staff and \alias Voice do not bring in the
performers. If I comment out the line \consists "Note_performer", I get no notes
in the MIDI file, only the appropriate number of rests.

I think this is also why Vaticana does not appear to work properly with MIDI.
Try this example:

\include "gregorian.ly"
\score {
  <<
    \new VaticanaVoice = "cantus" {
 	c' d' e'
     }
    \new Lyrics \lyricsto "cantus" {
      ut re mi
    }
  >>
  \midi { }
}

On 2012/05/04 02:59:06, Carl wrote:
> Looks like it works, but I think it can (and should) be simplified.  If it
> can't, then LGTM.
>

[dak, 2012-05-04 15:34:17]

http://codereview.appspot.com/6193043/diff/1/ly/performer-init.ly
File ly/performer-init.ly (right):

http://codereview.appspot.com/6193043/diff/1/ly/performer-init.ly#newcode88
ly/performer-init.ly:88: \consists "Dynamic_performer"
On 2012/05/04 02:59:06, Carl wrote:
> Again, don't the performers from the Voice performer group get added
> automatically by the \alias Voice?

Aliases are only relevant for properties, I think.  The initial values get
inherited, and any use of context property setting commands in the music or
context mods will accept changes for Voice into this context.

But translators are independent.

[aleksandr.andreev, 2012-05-04 16:48:11]

On 2012/05/04 15:34:17, dak wrote:
> Aliases are only relevant for properties, I think.  The initial values get
> inherited, and any use of context property setting commands in the music or
> context mods will accept changes for Voice into this context.
> 
> But translators are independent.

Sorry, I'm a little green. Does this mean I need to change something in this
patch?

[dak, 2012-05-04 16:49:01]

On 2012/05/04 16:48:11, aleksandr.andreev wrote:
> On 2012/05/04 15:34:17, dak wrote:
> > Aliases are only relevant for properties, I think.  The initial values get
> > inherited, and any use of context property setting commands in the music or
> > context mods will accept changes for Voice into this context.
> > 
> > But translators are independent.
> 
> Sorry, I'm a little green. Does this mean I need to change something in this
> patch?

No.

[Carl, 2012-05-04 17:27:23]

On 2012/05/04 16:48:11, aleksandr.andreev wrote:

> Sorry, I'm a little green. Does this mean I need to change something in this
> patch?

No.  David was explaining why the \alias didn't cause the performers to be
brought in automatically.

My comment was invalid.  Your patch is good.

Thanks,

Carl

[benko.pal, 2012-05-04 19:33:06]

hi Aleksandr,

2012/5/4  <aleksandr.andreev@gmail.com>:
> For whatever reason, \alias Staff and \alias Voice do not bring in the
> performers. If I comment out the line \consists "Note_performer", I get
> no notes in the MIDI file, only the appropriate number of rests.
>
> I think this is also why Vaticana does not appear to work properly with
> MIDI. Try this example:
>
> \include "gregorian.ly"
> \score {
>  <<
>    \new VaticanaVoice = "cantus" {
>        c' d' e'
>     }
>    \new Lyrics \lyricsto "cantus" {
>      ut re mi
>    }
>  >>
>  \midi { }
>
> }

would you fix Vaticana and Mensural too?

p

[aleksandr.andreev, 2012-05-05 01:00:28]

On 2012/05/04 19:33:06, benko.pal wrote:
> 
> would you fix Vaticana and Mensural too?
> 

I've uploaded a patch that gets at Vaticana and Mensural as well. It seems that
"Beam_performer" would not be necessary for the two, but I don't know about
"Tie_performer" and "Slur_performer". It would probably be good for someone who
knows more than me about Gregorian chant to double check that it works right.

In addition to 2512, I think this addresses 2004.

[benko.pal, 2012-05-05 16:58:22]

hi Aleksandr,

> I've uploaded a patch that gets at Vaticana and Mensural as well. It seems
that
> "Beam_performer" would not be necessary for the two, but I don't know about
> "Tie_performer" and "Slur_performer". It would probably be good for someone
who
> knows more than me about Gregorian chant to double check that it works right.
> 
> In addition to 2512, I think this addresses 2004.

thank you very much!  please add Beam_performer too to ensure consistency with
Voice (may be needed when doing ancient and modern edition from a common
source).

p

[aleksandr.andreev, 2012-05-05 19:27:04]

On 2012/05/05 16:58:22, benko.pal wrote:
> thank you very much!  please add Beam_performer too to ensure consistency with
> Voice (may be needed when doing ancient and modern edition from a common
> source).

Added Beam_performer for Vaticana and Mensural based on this comment.

[Graham Percival, 2012-05-07 18:27:13]

LGTM