Re: [PATCH] Fix 524 (pitched trill accidental forgetfulness)

[Neil Puttock]

2008/6/12 Han-Wen Nienhuys <address@hidden>:

> the inefficiency of computing things twice is not that important.  I'm
> a bit more concerned with duplication of logic, if there is any.  Can
> you fold things together into a common function?

As it stands the only duplication would be the code for checking the
bar number, so that could quite easily be folded into a common
function.

> I think there are a
> couple more utility functions for contexts/engravers, so you could add
> to that.

Do you mean something like the functions in general-scheme.cc?

Of course, it would be desirable for the accidental behaviour for
pitched trills to follow that of normal accidentals (excepting the
'natural-always-visible' rule); but that would require a much more
complicated rewrite of both engravers.

> it should happen automatically if you set force-accidental on the
> corresponding music event in the parser.

Well, I can see that force-accidental is set on the trill pitch in the
parser, but the stumbling block is that there's no music event passed
by the parser for a pitched trill - as far as my limited scheme
knowledge goes, it appears that the \pitchedTrill function filters out
the trill-span-event from the main note, then adds the trill pitch to
the list, which the Pitched_trill_engraver notices.

I've tried reading the force-accidental property from the music event,
then adding it to the trill-span-event list, but it causes a segfault.
:(

Regards,
Neil