Hi David,

thanks for your reply!

Am 19.06.2012 um 19:20 schrieb David Nalesnik:

> Hi Jan-Peter,
> 
> On Tue, Jun 19, 2012 at 7:51 AM, Jan-Peter Voigt <[email protected]> wrote:
> Hello list, hello David,
> 
> I have two questions concerning the \shape command (2.15.40) created by David 
> Nalesnik:
> 
> 1. Is it possible to shape ties in a chord?
> 
> { \shape Tie #'(((0 . 0)(0 . 1)(0 . -1)(0 . 0))((0 . 0)(0 . 1)(0 . -1)(0 . 
> 0))) <c' e' g' b'>1 ~ | q4 }
> 
> Unfortunately, no.  It should be fairly straightforward to define a command 
> to operate on unbroken/broken stacks of ties, and I can look into this.
that would be great! :-) I think two problems have to be solved:
1. Where is the stack of control-points found and how is it modified?
2. Howto define the offsets on this extra-dimension. If we have a stack of 
broken ties, we have to define N x M x 4 offset-pairs:
{ <c' e' g' b'>1 ~ \break q4 } => 4 x 2 x 4 = 32 x '(dx . dy)
Earlier today I had a case, were I would have modified all 3 ties with the same 
values. So this might be one option. But in many cases tie-direction is 
opposite for a two-note-chord.
Hm, I think about this.
>  
> 
> 2. In the development version warnings were given and a color was used to 
> indicate a mismatch of offset count and number of curves in a broken one. 
> What is the reason to not include that into the release?
> 
> My thinking here was to get the basic functionality into the code base, and 
> these warnings are helpful but not strictly necessary.   (They were designed 
> to be part of a slur-editing tool and might be more suitable for now as an 
> enhancement on the LSR--opinions welcome!)
That sounds reasonable. But probably there is room for a "backdoor" to switch 
it on or plug in some kind of callback - a context property(?) - so there's no 
need to copy the whole function to LSR - now that it's already in the codebase 
;-) ... just a thought.

For now, thanks again for this amazing helper :-)

Best Jan-Peter

_______________________________________________
lilypond-user mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/lilypond-user

Reply via email to