Re: Table of contents broken?

[Joe Neeman]

On Sun, 9 Dec 2007 22:23:00 Nicolas Sceaux wrote:
> Le 9 déc. 07 à 12:13, Nicolas Sceaux a écrit :
> > Le 9 déc. 07 à 11:58, Nicolas Sceaux a écrit :
> >> Le 8 déc. 07 à 23:41, Reinhold Kainhofer a écrit :
> >>> I'm trying to use \table-of-contents in a large mass (a full score
> >>> with 17
> >>> instruments/voices and 7 pieces). Now, I have several problems:
> >>>
> >>> -) All items added with \tocItem appear in the same line (i.e.
> >>> printed one
> >>> above the other)
> >>
> >> There's a regression indeed. It seems that it was OK in 2.11.35,
> >> but not
> >> anymore in 2.11.36.
> >
> > The culprit is
> > 
> > <http://git.sv.gnu.org/gitweb/?p=lilypond.git;a=commitdiff_plain;h=a0c6b9
> >7af8366c381263772aea1dfe1434a36c2c
>
> Joe, may I restore the previous definition on the function?
> Do you remember the rationale for this change?

I remember the rationale, but it's too late right now for me to find a proper 
fix. The problem is with spacing single-staff systems. Suppose you have two 
systems, the first has staff-extent '(-2.0 . -2.0) and the second has 
staff-extent '(-5.0 . -5.0). Then the original function would put too much 
space between the lines because of the inner "(max 0 ...)". Also, there is a 
problem in that line-minimum-distance uses the extents while 
line-ideal-distance uses the staff-extents, but these aren't directly 
comparable.

You're welcome to revert it; that will only cause some uneven spacing in a few 
situations and I'll have another look at fixing it later. Or if you want to 
figure out the correct solution, you're welcome to do that too :) It 
shouldn't need any special casing, it's just a matter of getting the 
arithmetic right.

Joe