Re: Issue 650 in lilypond: Broken hairpin placed too high after a line break

[lilypond]

Comment #1 on issue 650 by address@hidden: Broken hairpin placed too  
high after a line break
http://code.google.com/p/lilypond/issues/detail?id=650

Neil's response on lilypond-devel  
(http://lists.gnu.org/archive/html/lilypond-devel/2010-11/msg00594.html):

On 26 November 2010 20:22, Reinhold Kainhofer <address@hidden>  
wrote:
> BTW, the same problem also happens with ottava brackets, volta brackets,  
> etc.
> but NOT with text cresc spanners, trill spanners, normal text spanners or
> tuplet brackets.

DynamicTextSpanner uses the line-spanner-interface, so it's possible
to set the left-bound accurately (via attach-dir and padding).  The
same applies to TextSpanner/TrillSpanner, though these also support
end-on-note (which means they're usually anchored on the first note
after the break).

> As Hairpin and DynamicTextSpanner grobs are treated completely equal in  
> the
> New_dynamic_engraver, I assume it must be some problem with grob settings  
> of
> the Hairpin grob. Any idea?

(hairpin.cc)

111 x_points[d] = b->extent (common, X_AXIS)[RIGHT];

Cheers,
Neil