Re: Duration of contexts?

[Mats Bengtsson]

As I said, I didn't consider it a bug but a feature.
Short-lived contexts do die and I haven't objected against that.
The feature is that contexts that are completely empty at the
top of the score never die. Here are two illustrating examples.
The second one shows why I find this a useful feature. Without it,
I would need to find out how many empty bars to include in an
initial   \context Lyrics = "above" {s1*10}.

\version "2.5.20"

% Useless illustration of the feature

\score{ \relative c' <<
  \context Staff = "eternal" {}
  \context Staff = "ephemeral" {s1}
  \context Staff = "main" {
    c1 | c1
    << c1
      \context Staff = "ephemeral" { e1^"ephemeral" }
      \context Staff = "eternal" { g1^"eternal" }
    >>
    c1 | c1
    << c1
      \context Staff = "ephemeral" { e1^"ephemeral" }
      \context Staff = "eternal" { g1^"eternal" }
    >>
  }
>>
}


% Useful illustration of the feature

\score{ <<
  \context Lyrics = "above" {}
  \new Staff  \context Voice = "main" \relative c' {
      c1 d <<
        \context Voice = "upper" {\voiceOne e2 d2} \voiceTwo c1 >>
      \oneVoice e }
  \new Lyrics \lyricsto "main" {do re do mi }
  \context Lyrics = "above" \lyricsto "upper" { mi re }
>> }


   /Mats



Han-Wen Nienhuys wrote:
> Mats Bengtsson wrote:
>
> > Hi,
> >
> > In several examples and templates in the manual, we use constructs like
> >
> > \context Lyrics = "somename" {s1*20}
> >
> > to create an empty context that will live for at least 20 bars.
> > However, I noticed more or less by accident that if you instead do
> >
> > \context Lyrics = "somename" { }
> >
> > then you'll get a context that is automatically kept alive during the
> > full score. Of course, this feature is very nice since you don't have
> > to calculate how many empty bars are needed.
> > If this is a deliberate feature, then it should be properly documented
> > and advertised in the manual and all the corresponding examples should
> > be modified to use the feature. If it's just a coincidence, I strongly
> > recommend to make it a supported feature.
>
>
> Imagine the performance impact if contexts (eg. short-lived like << c \\ 
> d >> ) never die.
>
> Bugreport please.

--
=============================================
        Mats Bengtsson
        Signal Processing
        Signals, Sensors and Systems
        Royal Institute of Technology
        SE-100 44  STOCKHOLM
        Sweden
        Phone: (+46) 8 790 8463                         
        Fax:   (+46) 8 790 7260
        Email: address@hidden
        WWW: http://www.s3.kth.se/~mabe
=============================================