Re: [Lilypond-auto] Issue 3307 in lilypond: Rest positions incorrect with non-standard line count

[lilypond]

Comment #30 on issue 3307 by address@hidden: Rest positions incorrect  
with non-standard line count
http://code.google.com/p/lilypond/issues/detail?id=3307

This patch (output attached) makes no net change to the position of short  
rests in separated voices, \voiceOne r4, on the standard 5-line staff.   
These positions are within common practice in engraved scores.

Longer rests, \voiceOne r1, are moved to the staff line if they would  
otherwise be one line out, which brings them in line with common practice.   
In the lower voice, I see Breitkopf engravers hanging the whole (semibreve)  
rests from the second line (as this patch does) whereas other engravers  
move them down to the first line, as in the scan posted above from Ted  
Ross' book.

Full-measure rests, \voiceTwo R1, are in the same locations as the r1  
version.  The problem is that the low-voice R1 would move relative to  
version 2.16, and LilyPond does not yet resolve collisions between  
full-measure rests and notes (these rests span the measure after it is laid  
out). Therefore multi-voice scores will have collisions appear with R1  
rests if these scores are upgraded to version 2.18, which collisions need  
to be resolved by hand, if and when discovered.

That backward-compatibility consideration might be an argument for placing  
R differently than r, but maybe there are better ways to give  
backward-compatibility.

You might resolve both your questions by rounding-down the multiple-voice  
shift for r4 and shorter, to the nearest staff line, while rounding-up the  
magnitude of the multiple-voice shift for R and for r2 and longer.

Attachments:
        test.pdf  22.7 KB

--
You received this message because this project is configured to send all  
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings