Looks good to me.
https://codereview.appspot.com/226430043/diff/20001/input/regression/display-lily-tests.ly File input/regression/display-lily-tests.ly (right): https://codereview.appspot.com/226430043/diff/20001/input/regression/display-lily-tests.ly#newcode224 input/regression/display-lily-tests.ly:224: \test "NOT A BUG" ##[ \oneVoice #] % resetting a bunch of properties Given that \displayLilyMusic does not reverse-engineer \voiceXxx, it seems it would take some work to teach it to reverse-engineer \partcombineUp/Down https://codereview.appspot.com/226430043/diff/40001/lily/part-combine-iterator.cc File lily/part-combine-iterator.cc (left): https://codereview.appspot.com/226430043/diff/40001/lily/part-combine-iterator.cc#oldcode396 lily/part-combine-iterator.cc:396: SCM lst = get_music ()->get_property ("elements"); Later, it would be good to have PartCombineMusic contain a ContextSpecdMusic for each of the four output Voices and the Devnull in its 'elements. Instead of the loop above, we would step through the list of 'elements and link handles_[] to the contexts defined by \partcombine. Then instead of the strings in outlet_names_[] we just use the context-id of each of the 'elements. Then we can combine two output Voices into one, such as CONTEXT_SHARED and CONTEXT_SOLO, by giving the corresponding ContextSpecedMusics the same context-id. https://codereview.appspot.com/226430043/ _______________________________________________ lilypond-devel mailing list [email protected] https://lists.gnu.org/mailman/listinfo/lilypond-devel
