Re: Shape the angle of a broken hairpin's first part

[Urs Liska]

Am 18.09.2018 um 11:49 schrieb Thomas Morley:
> 2018-09-18 11:13 GMT+02:00 Urs Liska <address@hidden>:
> > Am 18.09.2018 um 11:09 schrieb Urs Liska:
> > > Hi Harm,
> > >
> > > thank you very much!
> > >
> > >
> > > Am 17.09.2018 um 22:35 schrieb Thomas Morley:
> > > > 2018-09-17 12:05 GMT+02:00 Thomas Morley <address@hidden>:
> > > > > 2018-09-17 11:55 GMT+02:00 Urs Liska <address@hidden>:
> > > > > > thanks for this, which gives me a handle to the first part of the
> > > > > > hairpin at
> > > > > > least.
> > > > > >
> > > > > > But my problem is not solved with the 'height property.
> > > > > [...]
> > > > >
> > > > > > See my updated example
> > > > > [...]
> > > > >
> > > > > I'll have a look in the evening, now I have to run for my regular job
> > > > > ...
> > > > >
> > > > > Cheers,
> > > > >     Harm
> > > > Hi Urs,
> > > >
> > > > as said before you'll need to go for the 'stencil.
> > > > Though, ly:hairpin::print, the default-stencil, is coded in C++. There
> > > > the defaults-heights for broken hairpins are set to 1/3 and 2/3 of
> > > > 'height.
> > > > Thus we need a scheme-rewrite of the stencil-procedure to reset them.
> > > > Below I slightly extend a coding by David Nalesnik from
> > > > https://lists.gnu.org/archive/html/lilypond-user/2018-02/msg00181.html
> > > >
> > > > Btw, the rest of the linked thread is very interesting as well:
> > > > Once you have a scheme-rewrite it's possible to add a plethora of
> > > > other features...
> > > >
> > > > Back on topic.
> > > > Attached you'll find David's coding. My addition is to add a
> > > > 'broken-heights-property.
> > > > Now you can adjust the values and call the scheme-stencil afterwards,
> > > > instead of the default-stencil.
> > >
> > > This will work for my case - although it's somewhat ridiculous to include
> > > such a massive file just to fix one graphic issue.
> > > I think this is something that should be user-settable. Maybe a property
> > > like yours as a native Hairpin property and/or a setting of the maximum
> > > angle hairpin lines may have. The latter would also be a viable setting (as
> > > opposed to minimum-length) to prevent those crippled too-short hairpins that
> > > are often hard to spot. I think I'll post a feature request.
> > >
> > > Best
> > > Urs
> >
> > Ah, I forgot to ask one thing: 'broken-heights is overridden with two
> > values, but I don't see the second one making any difference. What is that
> > about?
> >
> > Urs
> For decrescendo the first value determines the height for all but the
> last broken hairpin-end.
> For decrescendo the second value determines the height for all but the
> first broken hairpin-start.
>
> For crescendo the first value determines the height for all but first
> broken hairpin-start.
> For crescendo the second value determines the height for all but last
> broken hairpin-end.
>
> Internally these values are multiplied with the 'height-value.
>
> See:
> {
>    \override Hairpin.broken-heights = #'(0.2 5)
>    \override Hairpin.stencil = #hairpin::print-scheme
>    c'1\>
>    \break
>    c'1 \break
>    s2 c'\!
> }
>
> {
>    \override Hairpin.broken-heights = #'(0.2 5)
>    \override Hairpin.stencil = #hairpin::print-scheme
>    c'1\<
>    \break
>    c'1 \break
>    s2 c'\!
> }
>
> Cheers,
>    Harm

Ah, OK. But that didn't take any effect in the earlier example since 
that only applied the override to the first sibling. If I use a 
crescendo instead then only the second value takes effect.

Thanks again
Urs