Re: Don't wrap EventChord around rhythmic events by default. (issue 5440084)

[dak]

On 2012/01/20 17:33:44, dak wrote:
> mailto:address@hidden writes:

> > Hi David,
> >
> > Should I wait for a new patch or can I test using the latest one
here?
> >
> > I've tried it out briefly on a real music example, and have a
problem
> > with identifiers:
> >
> > foo = \mark \default
> >
> > \relative c' {
> >  \foo
> >   c1
> > }

> You have the newest.  The problem is the definition of ly:event? (for
> recognizing postevents).  It is currently (in lily/music-scheme.cc)
> defined as the equivalent of
> (define (ly:event? m)
>          (and (ly:music? m)
>               (ly:is-music-of-type? m 'event)
>               (not (ly:is-music-of-type? m 'rhythmic-event))))

> That seemed to be more or less right.  Apparently you found a "less
> right" case.

> Suggestions?

define-music-display-methods has a rather crude post-event? definition
but I am not all too sure whether it is early enough in the load-order.

I think I will go the tiresome way of explicitly putting post-event into
the types of each suitable music event.  That puts the information to an
obvious place, and it does not look like the current event types
discriminate correctly.

And either post-event? or ly:event? should eventually be eradicated.

http://codereview.appspot.com/5440084/