Re: beam breaking in tuplets

[Erik Sandberg]

On Monday 06 November 2006 12:59, Mats Bengtsson wrote:
> Erik Sandberg wrote:
> > On Monday 06 November 2006 09:25, Mats Bengtsson wrote:
> >
> > Mats, do you think it would be useful with an operator \newClone to clone
> > the current context? E.g.
> > \new Staff \with {\consists Foo_engraver bar=#'baz}
> >  { c d \tiny  e f \newClone Staff { g f e d} c d e }
> > would be equivalent to:
> > \new Staff \with {\consists Foo_engraver bar=#'baz}
> >  { c d \tiny e f \new Staff \with {\consists Foo_engraver bar=#'baz} {
> > \tiny g f e d} c d e }
> >
> > (I think such operator can come out as a side-effect of some work I'm
> > doing)
>
> Would it even make sense to make this the default behaviour of \new?
> At least, it would probably cause less surprise to a newbie than the
> current implementation, in examples like:
>  \relative c'{ \tiny c d <<{e d} \\ {c b}>> }

You have a point here. Perhaps we could have a context property 
cloneNewContexts, defaulted to #t for voices, which makes sure that contexts 
creations in the current context clone the current context, if the context 
types match. (so below, the context hierarchy would be Global-Score-Lyrics 
where the inner Voice is created, hence no settings would be inherited; this 
also means that the inner \new Voice should create a new staff implicitly).

I think the difficult thing is that the syntactical hierarchy of our input 
language doesn't match the hierarchy of contexts.

> On the other hand, it would probably make it less obvious for the advanced
> user to predict exactly what LilyPond does, for example in situations like
>
> \relative c' \new Voice{
>   c d e f
>   <<
>     c1
>     \new Lyrics \lyricmode {
>       Some4 text
>       << \new Voice \notemode { c d e2 }
>          {some more text }
>
>     }
>
> }
> where it's not obvious if the inner Voice context would inherit anything
> from the outer one (maybe this is already crystal clear to you).

-- 
Erik