Re: Padding paper setting not honored

[Jean Abou Samra]

Le 31/08/2022 à 01:33, Knute Snortum a écrit :
> Hi everyone,
>
> I have a problem that is unfortunately very difficult to reproduce.
> It seems in some circumstances that the padding setting in the paper
> block is not honored.  I've attached a screenshot.  In the middle you
> can see that the slur from the bottom stave is going through the pedal
> marking of the top stave.  It looks like the padding is measured from
> the ottava bracket and not the slur, but that's just a guess.
>
> I have tried and failed to create a MWE.  I've attached my attempt but
> unfortunately it does not reproduce the problem. (I couldn't get the
> ottava bracket below the slur which may be why the MWE fails.)
>
> For the brave, I have uploaded my source files to GitHub:
> https://github.com/ksnortum/chopin-opus-25/tree/main
>
> If anyone is willing to tackle this problem, it would be much appreciated.



Well, here is a reasonably minimal example:

\version "2.23.12"

veryHigh =
  \shape #'((0 . 0) (0 . 10) (0 . 10) (0 . 0)) \etc

<<
  \new Staff {
    R1
    \break
    % OK
    c'''2\veryHigh ( 2)
    \break
    % KO
    c'''2\veryHigh ( \change Staff = down 2)
  }
  \new Staff = down {
    s1*3
  }
>>


Unfortunately, this is hard to fix. The second is cross-staff. Because of
this, its shape, and consequently its height, depend on the exact distance
between the staves (inside one system). But that distance depends itself
on everything else that is on the page (other systems), as LilyPond will
try to find a good fit for the page globally. So LilyPond needs to space
the staves before it can know how tall the slur it is.

(*Maybe* it would be possible to let LilyPond reuse the pure height 
estimation
in this case. I'm not sure, and I haven't tried.)

For now, I am afraid you need to resort to manual system positioning,
as described on

https://lilypond.org/doc/v2.22/Documentation/notation/explicit-staff-and-system-positioning

Best,
Jean