horizontal spacing regression

[Keith OHara]

Dear Bug Squad
  This is the only unreported regression I know about.
It change since 2.12.3, and although I did not get a huge response over on 
-user, we agree it got worse.
I believe this tiny example captures what we did not like in the real music 
examples from that thread.


% Note spacing should not be affected by accidentals unless necessary.
% Here, spacing is tightened near accidentals

\version "2.13.43"
{
  \repeat unfold 8 {f'8 bes' d'' f'' \noBreak }
  \break
  \repeat unfold 12 {f'8 bes' d'' f'' \noBreak }
}


------- Forwarded message -------
Date: Mon, 27 Sep 2010 21:30:03 +0100
From: Gerard McConnell <address@hidden>
Subject: Re: alpha test, horizontal spacing
To: address@hidden

I agree that in the examples you gave the accidentals look a lot better in
the 2.12 versions.

On Mon, Sep 27, 2010 at 7:56 PM, Keith E OHara <address@hidden> wrote:
> Well, opinions change over time. It is a subtle thing, but I think the old
> naive spacing around accidentals makes an easier-to-read page.  Here are two
> dense measures of Debussy that are a little difficult to space.
>
> Moving the notes over the clef-change is a good thing, of course.
>
> The stem-accidental collisions do not occur unless the beam crosses staffs,
> even if it is kneed, so they might be considered part of the cross-staff
> issues.  The spacing of the first three 16ths is probably a cross-staff
> artifact as well.  However, stems of cross-staff beams will always be
> special cases in collision resolution, so they can cross a long hairpin
> crescendo for example.  So I suggest that keeping accidentals clear of other
> note columns might be wisest.
>
> More simply, in the last three 16th notes (demisemiquavers) in the first
> measure, I want the accidental to give me a bit of extra space for
> readability.  I was able to create a small example showing a case where
> notes with an accidental in between were actually spaced closer together. (I
> imagined the new spacing engine getting a little too excited: "can I fit
> under the neighboring accidental?  Yes!  Oh boy lets slide these
> together!!")
>
> So I favor less aggressive tucking of noteheads under other noteheads and
> their attached accidentals, if that is an option. Other opinions?
> --
> Keith

tucking.png
Description: PNG image