Re: Fixes issue 39 by raising stems (issue3934041)

[k-ohara5a5a]

http://codereview.appspot.com/3934041/diff/2001/lily/note-collision.cc
File lily/note-collision.cc (right):

http://codereview.appspot.com/3934041/diff/2001/lily/note-collision.cc#newcode276
lily/note-collision.cc:276: // magic numbers...
already been said, referring to this block.  I wanted assurance from the
comment that these numbers are intentionally non-monotonic, and reminder
that they are in half-spaces (right?)

http://codereview.appspot.com/3934041/diff/2001/lily/note-collision.cc#newcode278
lily/note-collision.cc:278: Real rv = raise_vals[durlog - 3];
Do we know 2 < durlog < 7 here? Forever?

http://codereview.appspot.com/3934041/diff/2001/lily/note-collision.cc#newcode280
lily/note-collision.cc:280: Real xtra = (distant_half_collide ||
close_half_collide) ? 0.0 : 1.3;
I take you to mean,  if (full_collide) rv += 1.3;

http://codereview.appspot.com/3934041/diff/2001/lily/stem-engraver.cc
File lily/stem-engraver.cc (right):

http://codereview.appspot.com/3934041/diff/2001/lily/stem-engraver.cc#newcode68
lily/stem-engraver.cc:68: stem_->set_property ("note-collision-bump",
scm_from_double (0.0));
Tell us why you use a new property, instead of adjusting 'length.   I
would need a better name to understand this after I forget about this
bug; for the way it acts now, something like extra-raise-tip would help
the future me.

http://codereview.appspot.com/3934041/diff/2001/lily/stem.cc
File lily/stem.cc (right):

http://codereview.appspot.com/3934041/diff/2001/lily/stem.cc#newcode321
lily/stem.cc:321: return scm_from_double (stem_end + xtra);
Today I remmber that xtra<>0 only for upstems, but someday we will
wonder you /raise/ the flag by note-collision-bump,
even if the stem points down?

http://codereview.appspot.com/3934041/