Re: would this qualify as a bug (in partcombine)?

[David Boersma]

Hello Arie,

I tried out your example, with lily 1.6.10, and changed the octaves. I
agree that it looks like a bug. But in my view the problem is not so much
with the stems but more with the flags/beams of the last notes (eighths)
which are printed double, making them look like a sixteenth note (maybe
this is what you actually meant to say).

I did some more experiments:

If I change one of the d notes (the second notes of both threads) into a b
then the problem goes away, and also the warning messages vanish.

If, in the code with relative c' (and the second notes both on "d" again),
I stepwise lower both last notes, then I find that the problem is still
there with "a" in threadOne and "c" in threadTwo, but when I go one step
lower, to "g," and "b" respectively, then the problem goes away, although
the warning messages are still there.

Then I did the same game with different clefs (G_8 and F). I found that
when I lowered the note values of the third notes such that the highest
one of the two was beneath the first ledger line, then the problem goes
away. Now to the opposite side of the staff: increase the note values and
also when you choose note values such that the lowest note is above the
first ledger line (above the staff), the problem also goes away.

So the actual note value is not determining the error behaviour, it is the
vertical position w.r.t. the staff. If both notes are at least one ledger
line away from the staff, then they are correctly printed. If one of them
is closer then the flags/beams are printed doubly (*if* the previous note
was identical in both threads).

Now I hope that these descriptions are clear enough and that these are
somehow useful diagnostics.

David Boersma

On 23 Jun 2003, arie-lily wrote:

> I extracted the following strange behaviour in \partcombine, with this
> snippet it works well:
>
> \version "1.6.10"
> \score {
>     \context Voice = voiceOne {
>         \partcombine Voice
>         \context Thread = threadOne \notes \relative c { c8 d e }
>         \context Thread = threadTwo \notes \relative c { e8 d g }
>     }
> }
>
> That is to say: ly2dvi complains about (and consecutively ignores them)
> clashing notecolumns, but displays the music correctly.
> However, when I change
> \relative c
> into
> \relative c'
> or
> \relative c''
> (of course both of them have to be changed at the same time with the
> same amount), then the stems of the third notes are both printed.
> I think this must be a bug, not?
>
> arie - just asking before filing a bug report for nothing...
>
>
>
>
> _______________________________________________
> Lilypond-user mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/lilypond-user
>

--
David Boersma

 /"\  ASCII ribbon campaign
 \ /  ---------------------
  X     against HTML mail
 / \      and postings