On Sun, 20 Nov 2011 23:57:22 -0800, <m...@apollinemike.com> wrote:
On Mon, 21 Nov 2011 07:06:53 +0000, k-ohara5...@oco.net wrote:
Well, any overestimation in your blocking heights also becomes an
overestimation in the tentative space between Staves/Lyrics/Dynamics
at the horizontal spacing step.
I think that you're right for most cases, but for a piece with lots of
accidentals that could potentially hang over barlines, this complexity
in estimation seems to help.
Have an example where pure-from-neighbor would do better than
axis-group-interface:height ?
As for the extra spacing, I'd be curious
to see if a gimungous score gets looser with this patch applied.
You might have understood me backwards. An overestimation in tentative
(vertical) space between Staves/Lyrics/Dynamics will let the spacing
spanner do sometimes a /tighter/ horizontal spacing.
The overestimated tentative vertical space is forgotten when the
real vertical spacing is done.
I didn't see, and don't see any risk of, loose spacing from this patch.
I expect that occasionally, this patch will let a collision with a span
bar leak through, but it would be something crazy like an \espressivo atop
a \downbow on a note connected to a cross-staff slur. The \espressivo
isn't pure-relevant because of the cross-staff poisoning.
_______________________________________________
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel
