Re: setting the number of pages for a score

[Joe Neeman]

Werner LEMBERG wrote:

> > > It is necessary that both the left and the right page are treated
> > > as a pair, having the same number of bars.  A good place for a page
> > > break is a larger rest in either the left or right hand of one of
> > > the two players.
> > >
> > > Hopefully, the used algorithm can be prepared to handle this --
> > > it's not necessary to implement it right now (it sounds like a
> > > sponsored feature :-) but it would be a pity if the new stuff makes
> > > it impossible to do that.
> > >      
> > I think the current plan is to include a page-turn-weight variable
> > that, if set to zero will cause the breaker to ignore page turns
> > entirely. You could then force breaks with \pageBreak and
> > \noPageBreak and do it manually that way. I would like to continue
> > to allow manual overrides as much as possible.
> >    
>
> Hmm, I don't understand your answer.  My main concern is to make
> lilypond handle the above described situation automatically --
>  
Ah, sorry. I was just pointing out that it could still be done manually 
(ie. we aren't losing functionality (right? currently it must be done 
manually?)). I'm sure the page breaking code can be extended easily to 
cover this. I think the task of splitting the (simultaneous) music into 
two printable piano-staves might be more tricky but I don't know that 
part of the code at all well.

Maybe it would be a good idea to make a Page_break_algorithm class (like 
the Break_algorithm class) to contain various page-spacing and utility 
functions and make other algorithms subclasses. Or would this just be 
messy -- how many separate algorithms would we actually need?