aligning lyrics to part-combined staves?

[Reinhold Kainhofer]

I'm currently creating many "piano reductions" of choral pieces (i.e. soprano 
and alto on one staff, tenor and bass on one staff), using part-combine for 
each 
of the two staves. 

The piano staff should of course also get the lyrics (the piece is homophonic, 
so all four voices have basically the same lyrics), but part-combine creates 
different voices for a2 and for split parts. 

So, I'm currently fighting with the Devnull context to use the soprano music to 
align the syllables to (see 
http://kainhofer.com/~lilypond/ajax/user/lilypond/Specific-uses-of-
lyrics.html#Lyrics-independent-of-notes)
Unfortunately, however, the Devnull context makes the lyrics ignore all 
melismata (ties, slurs, etc.) as described there, too. 

Is there any way to align lyrics to an invisible voice that also uses melisma 
correctly?

An example is attached. First, when the two voices are on separate staves, the 
lyrics are correct. However, when using part-combine, I'm using a Devnull 
context and all melismata are ignored, so the lyrics are positioned wrongly. 
(And no, in my case, there is no voice without melismata, that could be used 
for aligning).

Thanks a lot,
Reinhold
-- 
------------------------------------------------------------------
Reinhold Kainhofer, address@hidden, http://reinhold.kainhofer.com/
 * Financial & Actuarial Math., Vienna Univ. of Technology, Austria
 * http://www.fam.tuwien.ac.at/, DVR: 0005886
 * LilyPond, Music typesetting, http://www.lilypond.org

test.pdf
Description: Adobe PDF document

test.ly
Description: Text Data