Re: Chord changes over repeat alternatives

[Paul Scott]

Yitz Gale wrote:
> Paul, thanks for your attention to this issue.
>
> Paul Scott wrote:
>
>   
> > If you can carefully describe what you want
> > different than the current behavior I can
> > help you get it.
> >     
>
> A chord at the beginning of a second or
> subsequent repeat alternative should print
> unconditionally, without checking whether
> it is a change.
>   
I see your point and can see that you can't use the current version of 
chordChange = ##t if you have a case where an ending begins with the 
same chord as the last chord in the previous ending.  This is the only 
case where your feature is needed.  I have used this mode a few times 
and in each of those times the last chord in the first ending was a 
tonic chord and the first chord in the 2nd ending was some kind of 
resolving chord therefore it worked for my needs.  Otherwise I use 
whatever length I need on each chord name.  Here are some examples of 
chords of varying lengths:

c1*3 d4:m7 g4*3:7 c1 a4*3:7.5+ a4:7

> This is the same as the current behavior
> at the beginning of a line.
>   
Yes and this is documented but not the behavior you would like.
> Let me try to explain a little bit better why
> I think that this is an important issue:
>
> Much of the power of chord change mode comes
> from the fact that you specify it at the
> score level. For example, you can combine
> many fragments of music without having to worry
> about sewing them all together at the edges to
> get the chord changes right.
>   
I do appreciate your point even though I consider such care necessary in 
using LilyPond.  I have to take the same care when combining music 
fragments to be sure that octaves or note lengths don't change in an 
unintended way for example.  To get the result you want you will need to 
turn chordChanges off and set the length of each chord.  I see that the 
solution I gave you earlier does not give you the result you want.  This 
one should:

{
\new ChordNames \with{ voltaOnThisStaff = ##t } {
  \set chordChanges = ##f
  \chordmode {
    \repeat volta 2 {
      c1
    } \alternative { {
      f
    } {
      f2 g
    } }
    c1
  }
}
}


> If you need to turn it on and off manually inside
> the music even once, that power is lost. So
> this issue is a major blow to the usefulness
> of chord change mode.
>   
I do understand and am convinced that this feature is not in Lily's 
current design.  The issue you describe is similar to the issue of slurs 
or ties into endings.  These are not handled automatically by LilyPond 
even though many of us would love this to be true.  This would involve 
the same special logic.

I believe what you want is a feature and you will either need to 
convince the developers of the need for this feature or sponsor it 
yourself (or convince a group of users to co-sponsor it).

It is clear to me that this is not a bug but a desired feature since 
chordChanges = ##t behaves exactly as documented (2.9.14) in


       7.2.3 Printing chord names

...

"You can make the chord changes stand out by setting ChordNames 
<http://lilypond.org/doc/v2.9/Documentation/user/lilypond-internals/ChordNames.html#ChordNames>.|chordChanges| 
to true. This will only display chord names when there is a change in 
the chords scheme and at the start of a new line"


> Thanks,
>   
You're welcome,

Paul