Re: another spacing bug (\noBreak)

[Keith OHara]

Karol Majewski <karolm2 <at> wp.pl> writes:

> This is an excerpt from random piece. If you add \noBreak command at the 
end of measure 9 and 11, than the whole
> spacing becomes distorted! Bug?

Is the 'distorted' spacing better or worse than in the first case?  
If we want different behavior, we would need to describe what we want
in detail.

If this were a bug, somebody would have to create a minimized example in
order to isolate the error in the code.  The LilyPond project asks users
to minimize the example before putting the bug on the list, so that other
users have an easier time fixing the code.


This case is just a quirk of the design of LilyPond.

While minimizing this example, the behavior switches back and forth between
two styles of spacing.  Probably LilyPond is making a choice between two
distinct solutions; this seems strange because horizontal spacing is 
continuous, but at
<http://lilypond.org/doc/v2.18/Documentation/notation/
horizontal-spacing-overview> 
there is the description of the 'common-shortest-duration' that influences
line spacing.

When you add the \noBreak instructions, the common shortest duration changes
from an 8th note to a 16th note.  The rule to find the common shortest
duration finds the shortest note in each measure; for purposes of spacing,
LilyPond counts each non-breakable segment of music as a measure.

I recognized this only because we were working on spacing code a couple
months ago.  If we had not been familiar with the code, it would not
have been worth the time to figure this out.