Urs Liska <li...@openlilylib.org> writes:
> It doesn't even work correclty when the slurs are \omit-ted:
>
> %%%
>
> \relative{
> \key a \major
> \omit Slur
> cis'''16 -3 [ \( ( b -2 ) b ( a -1 ) \) a8 \strich ]
> }
>
> \relative{
> \key a \major
> cis'''16 -3 [ \( b -2 b a -1 \) a8 \strich ]
> }
> %%%
>
> In the second example the phrasing slur correctly avoids the
> fingerings while in the first one it prints right through them.
>
> I thought this might happen when the starting points of slur and
> phrasing slur coincide but the collision handling of the phrasing slur
> seems disabled when there is a slur *at any time*:
>
> %%%
>
> \relative{
> \key a \major
> cis'''16 -3 [ \( b -2 ( b )a -1 \) a8 \strich ]
> }
>
> %%%
>
> If it is true that collision handling freaks out when slurs and
> phrasing slurs exist at the same time it would be explainable that the
> end points of both fall into one.
Taking a look at the collision handling I freaked out. The way I read
the code, extra objects avoid at most exactly _one_ slur (including
phrasing slurs) which tends to be the first in the group of the last
encountered kind, basically the one entered into its grob object "slur"
at the time it is being typeset.
This is disturbing, yes.
--
David Kastrup
_______________________________________________
bug-lilypond mailing list
bug-lilypond@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-lilypond
