#4599 input/regression/part-combine-solo-end.ly

[Dan Eble]

There doesn't seem to be a ticket tracking the defect exposed by input/regression/part-combine-solo-end.ly. [#913] and [#1677] are similar but not the same.

\header {

    texidoc = "SOLO is printed even if the solo voice ends before the
    other one. Unfortunately, the multi-rest of the 1st voice (which
    is 2 bars longer than the 2nd voice) does not get printed."

}

\version "2.16.0"
\layout { ragged-right = ##t }

<<
    \new Staff
    \partcombine
    { R1 * 2 }
    { c'8\> c'\! r2.  }
>>

[Dan Eble]

This example presents the difficulty that in solo passages, the part combiner routes the music events of the silent part into a NullVoice with no Multi_measure_rest_engraver; so making it possible for multi-measure rests to reappear within their voices [#4600] does not help this case.

In a real piece, it might be reasonable to work around this by dividing the rest (R1 R1 instead of R1*2) or adding rests after the end of the solo part.

I'm moving on to other problems.

[Dan Eble]

This fixes this issue. It also fixes one symptom of [#1677] but does not address the root problem.
https://codereview.appspot.com/265410043/

[Dan Eble]

I just updated the patch, so if you happen to be testing it already, I'm sorry, but it would be best to start over.
https://codereview.appspot.com/265410043/

[Anonymous]

Passes make, make check and a full make doc.

Reg test diffs are here:

https://drive.google.com/file/d/0B9nZ5LHV2Ds6dE95SWUtS3NWaVU/view?usp=sharing

[Anonymous]

Patch on countdown for Oct 11th

[Dan Eble]

Thanks for the review comments. I'll try to react to them tonight, but I'm not sure I'll have time.