Re: Bug when using acciaccatura and polyphony?

[Erik Sandberg]

On Saturday 09 July 2005 02.32, Ismael Cortes wrote:
> Hello, I've been writing a python script to transform drum tablatures
> into lilypond (and then to PDF). After implementing polyphony to the
> script and testing some tablatures I realized that in the log it showed
> the text:
>
> "programming error: moving backwards in time
> continuing, cross fingers"
>
> (no quotes, of course)
>
> and the resulting PDF started looking as if the (Auto_)Beam_engraver has
> been removed (ie no auto Beams where shown), but only after the fault
> measure. It just seems to happen when the voice starts with a
> \acciaccatura grace note. It doesn't matters if it's the first or the
> second voice and adding \voiceOne nor \voiceTwo after the grace note
> doesn't helps.
>
> I append a partial file transformed by my script which shows the error
> (I'm not sure if I can legally append the whole file, but it's too big
> for a mailing list anyway). Note that if I manually beam the notes
> (using '[' and ']') they look as they should (but it's something i'd
> prefer not to have to do).
>
> Also any temporary workaround is appreciated as well as an explanation
> of what the heck is the error text suppoused to mean.

The error text is just a notification that the program behaves in a way that 
it didn't expect itself to behave. The problem is a variant of the grace note 
bug. The problem is that the time signature etc. happen at time 0, while the 
following grace notes happen at time 0-1/8g.

This can be worked around by adding a dummy grace skip before the \time etc, 
something like:
<<
{ \grace s16 \time 4/4 s1 }
<< { music } \\
{ music } >>
>>

-- 
Erik Sandberg
Maintainer of the Lilypond bug CVS archive,
http://savannah.gnu.org/cgi-bin/viewcvs/lilypond/lily-bugs/bugs/
http://lilypond.org/doc/v2.5/bugs/out/index.html