http://codereview.appspot.com/5729057/diff/1/lily/note-collision.cc File lily/note-collision.cc (right):
http://codereview.appspot.com/5729057/diff/1/lily/note-collision.cc#newcode131 lily/note-collision.cc:131: min_downs = min (min_ups, dps[i][X_AXIS]); min (min_ups, ... Seriously? And it is nonsensical to compare the mins on either side, since for the stem-down configuration, the max is the part to merge. http://codereview.appspot.com/5729057/diff/1/lily/note-collision.cc#newcode139 lily/note-collision.cc:139: } This assumes that only mins/maxes need to be considered when merging when in reality it is every pairing that has a chance to overlap. It is just that the min/max pairing with the smallest distance determines which stem distance has a chance of succeeding, and once one has determined that, one needs to do a merge-like run through both columns to make sure that at this stem distance, all note heads either merge to identical positions, or don't touch one another. http://codereview.appspot.com/5729057/ _______________________________________________ lilypond-devel mailing list [email protected] https://lists.gnu.org/mailman/listinfo/lilypond-devel
