Re: stem.cc - issue 5303 - misplaced notehead (issue 570830043 by address@hidden)

[Carl Sorensen]

On 7/11/19, 11:49 AM, "Lukas-Fabian Moser" <address@hidden on behalf of 
address@hidden> wrote:

    Hi Carl,
    
    > Perhaps this patch should also revert
    > 87eb2f9fe1be3a532675fe4b7322bbba5a60ba5c
    >
    > since that patch was a workaround, rather than a real fix, as
    > demonstrated during this troubleshooting thread.
    >
    > Carl
    >
    >
    > https://codereview.appspot.com/570830043/
    
    I'm afraid I do not understand the connection between the 
    overdone-notehead patch you refer to and the rounding-issue patch 
    regarding chord-notehead-placing. They do not seem to be linked to me.
    
    But I noticed that your old patch 87eb2f9fe, which was supposed to fix 
    issue https://sourceforge.net/p/testlilyissues/issues/4693/, seems to 
    only fix that issue at "normal" staff sizes. See for instance:
    
    \version "2.18"
    
    #(set-global-staff-size 90)
    
    \layout {
       \override Score.StaffSymbol.layer = 100
    }
    
    \new Voice \with {
       \override NoteHead.color = #red
       } {
       \relative {
         f'4 a c e \bar "|."
       }
    }
    
This can never work properly, because overdone-noteheads is a metafont 
parameter used in the creation of the feta noteheads.  LilyPond creates feta in 
only a few sizes: 11, 13, 14, 16, 18, 20, 23, and 26.  All other staff sizes 
are scaled from these fonts.  And remember that the staff-space/staff-line 
ratio varies with staff sizes, so the scaled notehead sizes will be 
approximately, but not exactly, correct.

When scaling to sizes outsize the range of the generated fonts, the user is 
likely to be required to override the NoteHead.font-size property to get the 
noteheads to fit exactly.

I would consider this patch to be appropriate if it worked correctly for all 
the given feta font staff sizes.  And my testing indicates that it does.


    (And, for the record, 87eb2f9fe introduced 
    https://sourceforge.net/p/testlilyissues/issues/5530/, as Malte reminded 
    us.)
    o   be 
    So it seems to me that that patch should probably be reconsidered, but I 
    think not without being related to issues 5303/5522 (which are what the 
    current patch is about).

I had assumed (without checking) that resolving the stem rounding problem in 
5303/5522 would solve the problem in 5530, because Harm showed us[1] that 
slightly reducing the notehead size would solve the problem of the slur 
spacing.  If it doesn't solve the problem then something else needs to be done 
with 5530.  But I can't solve it, because 5530 doesn't manifest itself on my 
system, as I mentioned earlier.

Thanks,

Carl



1. https://lists.gnu.org/archive/html/lilypond-user/2019-06/msg00170.html