lilypond-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: updates on branch jneeman


From: Joe Neeman
Subject: Re: updates on branch jneeman
Date: Mon, 18 Dec 2006 17:30:10 +0200

On 12/18/06, Han-Wen Nienhuys <address@hidden> wrote:
Joe Neeman escreveu:
> 2) use skylines for horizontal spacing even when one of the columns is
> non-musical.
>   - as part of this change, items in the spacing problem read the
> 'padding property (should I create a new property with a different
> name?) and expand the space they require by that amount. This allows us
> to set BarLine #'padding = 0.1 and fix issue #10

I'm not sure if this is the right solution; it might be that adjusting the
springs is a better approach. There is already quite elaborate tuning going
here for springs.

I think the problem is that, if there is a chord with several accidentals and an arpeggio, the spring is going to be quite long. If it is long, it can't be very stiff (or we will get uneven spacing when we stretch). Therefore it will compress easily until its blocking point, causing a near-collision. I think the ideal solution would be to somehow stick a spring between the left side of the arpeggio and the barline but I don't see how this is possible currently.

In any event, the name padding should be changed; it's already used a lot for
(mostly vertical) padding in side-position-interface. Maybe it's possible to
integrate the functionality of no-spacing-rods into the property as well.

What if I call it extra-width and make it a number-pair? The car will be the padding on the left, the cdr will be the padding on the right. Then "no-spacing-rods = ##t" is equivalent to "extra-width = #'(+inf.0 . - inf.0)".

>   - this change also removes a fair chunk of unused code. If it's OK,
> I'll try to do something similar for note-spacing


which code is that?

Separation_item::width, Separation_item::conditional_width and Separation_item::set_distance.  Separation_item::width can be almost removed even without this change.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]