Re: Scheme for rhythmic transposition

[Phil Holmes]

----- Original Message ----- 
From: "David Kastrup" <address@hidden>
To: "Phil Holmes" <address@hidden>
Cc: <address@hidden>
Sent: Saturday, November 19, 2011 11:22 AM
Subject: Re: Scheme for rhythmic transposition


> "Phil Holmes" <address@hidden> writes:
>
> > \shiftDurations seems to be remaining poorly documented.  I'd be happy
> > to update the LSR with that function rather than the way it is now,
> > but it seems to me that durations can only be made shorter, not
> > longer?  If I use:
> >
> >    \shiftDurations 1 0 { \music }
> >
> > all is OK.  However, with
> >
> >    \shiftDurations -1 0 { \music }
> >
> > I get:
> >
> > ShiftDur.ly:11:20: error: wrong type for argument 1.  Expecting
> > integer, found #<Prob: Music C++: Music((digit . 1) (origin
> > . #<location ShiftDur.ly:11:21>))((display-methods #<procedure #f
> > (event parser)>) (name . FingeringEvent) (types general-music
> > fingering-event event)) >
> >
> >    \shiftDurations
> >                    -1 0 { \music }
> >
> > Is this the correct usage, and does this mean that \shiftDurations
> > does only shorted music?
>
> 1 is an UNSIGNED.  -1 is a fingering.  Hooray.  Try #-1 instead.  I'll
> eventually get around to getting the "interpret in the correct manner by
> magic" behavior for negative numbers as well, but it has never been in
> Lilypond so far.  Actually, historically Lilypond would have balked at 1
> as well, calling for #1 instead.  So in a way, your impression is partly
> my fault.

http://lsr.dsi.unimi.it/LSR/Item?id=305

The simpler syntax doesn't work with the LSR, since it's still at 2.12. 
FWIW, with 2.15.18, using a negative number as the number of dots to add (I 
know, it makes no sense) appears to cause lily to loop and grab memory.  I 
killed the process at 40 seconds and 1.8 Gigs of memory.

--
Phil Holmes