David Kastrup wrote:
>>   `\magnifyStaff mag' will do these overrides:
>>      for each shrinkable prop S,
>>        S = (* S (/ mag old-mag))
>>      for each unshrinkable prop U,
>>        U =
>>            (if (< mag 1)
>>              (if (< old-mag 1)
>>                  U
>>                  (/ U old-mag))
>>              (if (< old-mag 1)
>>                (* U mag))
>>                (* U (/ mag old-mag)))
>
> This assumes access to old-mag.  It also assumes that
> floating point arithmetic is perfect and has no
> accumulative errors.

`old-mag' would be the context-property
Staff.magnifyStaffValue (which defaults to 1 if not found),
accessed through \applyContext.  Also, I had thought about
the loss of accuracy with floating point numbers, but I just
assumed that such effects would be negligible.  You are
leading me to believe that I am foolish to assume such
things...  :)

> So I don't see that this approach will fly.

Darn, and I felt so clever, too!  Okay.

>> It makes sense to me that, at least initially, the
>> default size should be the size of the staff when
>> magnifyStaff is first used.
>
> I repeat: that seems like a total nightmare in situations
> using quotes or similar.  The question is where we could
> access a reasonably "default" setting that relative
> references could be based on.

I guess I'm having trouble visualizing the problem.  Can you
provide a minimal example (abbreviated if need be)
demonstrating what you're referring to?

> Perhaps the easiest way out would be to have a staff-size
> relative setting in analogy to the current font-size
> setting.  It would have to be factored in "at the last
> moment" just like font-size is and should cater for ossia
> staves and similar reasonably gracefully.
>
> Apropos gracefully: maybe it might be worth checking how
> grace-settings et al work and take a hint from them.

Okay, but at this point, this may have to wait at least 2
weeks as I am coming up on a possible (short) hiatus to my
LilyPond work.  But answers to my current questions will
still help me no doubt.

Anyway, thanks for all of your help so far,
Mark
_______________________________________________
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Reply via email to