Re: [Lilypond-auto] Issue 2584 in lilypond: please make partcombine merge slurs

[lilypond]

Comment #15 on issue 2584 by address@hidden: please make partcombine  
merge slurs
http://code.google.com/p/lilypond/issues/detail?id=2584

Re: comment 14 first paragraph, \partcombine does not want to generate  
concurrent slurs.

When inner chair violins play  {e'2( f'4 g') g'2( f')}
and outer chair violins play   {c'2( f'4 g') e'2( d')}
a score often shows both divisi parts on a single staff, and when the  
rhythms match, as a single voice.  Formerly Lilypond analyzed the combined  
part as : a slur starting on e' and the same slur starting on c', and later  
that same slur ending on g'.

Commit b986b38f14 causes wrong output for this common case, and makes the  
relevant regressions test disagree with its self-description.  That commit  
avoided a warning from partcombine, in response to issue 1967, but the  
resulting incorrect output is worse than a warning, so we need to revert  
b986b38f14

The patch of comment 1 addresses issue 1967 by implementing rules:
  warn when we see ( but "already have a slur" that started earlier
  warn when we see ) but there is no slur to end
  treat multiple simultaneous (s or )s as redundant requests.

If there is good reason to reject the patch of comment 1, we can simply  
revert b986b38f14

Attachments:
        2584.png  4.3 KB