David Kastrup <dak <at> gnu.org> writes:

> 
> Dan Eble <nine.fierce.ballads <at> gmail.com> writes:
> 
> >>I'm not top posting.
> > \version "2.19.0"
> >
> > music = \relative b' {
> >   \time 3/4
> >   b4 b b | b2 \bar "||" \break
> >   
> >   % This adjustment to measurePosition is required to suppress
> >   % the warning "\time in mid-measure...."
> >   % 
> >   % The notation reference says,
> >   %
> >   %     \partial duration can also be written as:
> >   %       \set Timing.measurePosition -duration
> >   %
> >   % Assuming that the warning should remain because it is helpful
> >   % more often than not, can this workaround be improved upon?
> >   % I tried \partial 1*0, but it doesn't seem to be equivalent
> >   % as the notation reference would have me believe.
> >   \set Timing.measurePosition = #(ly:make-moment 0) % suppress warning
> 
> \partial now only does action at the end of its timestep.  You can
> rewind time _before_ the b2.  Since you reset with another \partial
> later, something stupid like \partial 1*3 will move far enough into the
> past that \time 4/4 does not dare to complain.  And \partial 4 will then
> rewind time.
> 
> >   \time 4/4
> >   \partial 4
> >   f4 | f f f f |
> > }
> 
> I'm not convinced enough of that to recommend it in snippet/manual
> though.  Behavior of \partial changed quite a bit over the last
> versions.  Frankly, that code seems weird anyway...

Thank you.  Your suggestion also has the advantage of printing a parenthesized
bar number at the mid-measure line break.  Setting measurePosition to zero
printed a regular bar number, which I didn't notice at first.

It's too bad this requires a nonsensical parameter and has to be so far 
separated
from the \time that causes the warning, but I'll use it.

It would be best if Lilypond could recognize that
    \time 3/4 s2 \time 4/4 \partial 4
is logical and not worthy of a warning.  Maybe that's hard to accomplish.
-- 
Dan


_______________________________________________
bug-lilypond mailing list
bug-lilypond@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-lilypond

Reply via email to