Grace notes shouldn't shorten the previous note unless they overlap with it.

Grace notes shouldn't shorten the previous note unless they overlap with it.

Previously a grace note after a rest shortened the last note before the rest.

This patch shortens the previous note to the start of the grace note rather than by the length of the grace note, so notes before rests aren't affected unless the grace notes are longer than the rest.

[Devon Schudy, 2013-11-20 08:17:53]

Here's a fix for a minor MIDI bug: grace notes after rests shorten the last note
before the rest.

[dak, 2013-11-21 09:00:49]

https://codereview.appspot.com/29550043/diff/1/lily/note-performer.cc
File lily/note-performer.cc (right):

https://codereview.appspot.com/29550043/diff/1/lily/note-performer.cc#newcode100
lily/note-performer.cc:100: tie_head->length_mom_ = min (tie_head->length_mom_,
Ok, I've brooded a lot over this line, and I'm still not sure what it does in
all cases.  Can you put a comment in here breaking down the possible scenarios
that this code can encounter?

The precondition apparently is now_mom ().grace_part < 0, and it's not clear to
me what grace parts tie_head->audio_column_->when () may have for what reason.

now_mom () - start only makes sense when now_mom and start only differ in
_either_ grace_part _or_ main_part, and it's not clear to me whether
tie_head->length_mom_ can be a grace duration.

So please break this down somewhat more.  It's ok if this significantly
decreases the original code/comment ratio of this file: in all deference to the
original LilyPond authors, that's one aspect of the original code style we are
not trying to copy slavishly.

It probably took you quite a while to figure this out, so it's just efficient to
save the next person running across this code the time.

Thanks

[Devon Schudy, 2013-11-21 15:44:11]

Clarify when notes are shortened.

[janek, 2013-11-25 13:39:30]

LGTM, thanks!