Fix #189: Episema over single neume.

Fix #189: Episema over single neume.

* add Episema_engraver (based on Text_spanner_engraver) which listens to
  EpisemaEvent and creates Episema spanner

* set NoteHeads as 'side-support-elements so spanner can be quantized over
  neume/melisma

* ensure episema doesn't extend past bound items by setting NoteColumns
  as spanner bounds

* add regression test

[hanwenn, 2010-02-24 23:11:50]

lgtm. 

just a quick look at at the .cc

http://codereview.appspot.com/186189/diff/2001/2005
File lily/episema-engraver.cc (right):

http://codereview.appspot.com/186189/diff/2001/2005#newcode92
lily/episema-engraver.cc:92: announce_end_grob (finished_, SCM_EOL);
this should point to the STOP event

http://codereview.appspot.com/186189/diff/2001/2005#newcode101
lily/episema-engraver.cc:101: Episema_engraver::typeset_all ()
typeset_finished() ?

http://codereview.appspot.com/186189/diff/2001/2005#newcode114
lily/episema-engraver.cc:114: if (span_ && !span_->get_bound (LEFT))
what if note_cols is empty?

http://codereview.appspot.com/186189/diff/2001/2010
File scm/define-grobs.scm (right):

http://codereview.appspot.com/186189/diff/2001/2010#newcode746
scm/define-grobs.scm:746: line-interface
episema-interface ?

[Neil Puttock, 2010-02-26 20:19:03]

Hi Han-Wen,

Thanks for your review.

Cheers,
Neil

http://codereview.appspot.com/186189/diff/2001/2005
File lily/episema-engraver.cc (right):

http://codereview.appspot.com/186189/diff/2001/2005#newcode92
lily/episema-engraver.cc:92: announce_end_grob (finished_, SCM_EOL);
On 2010/02/24 23:11:50, hanwenn wrote:
> this should point to the STOP event

Done.

http://codereview.appspot.com/186189/diff/2001/2005#newcode101
lily/episema-engraver.cc:101: Episema_engraver::typeset_all ()
On 2010/02/24 23:11:50, hanwenn wrote:
> typeset_finished() ?

Done.

http://codereview.appspot.com/186189/diff/2001/2005#newcode114
lily/episema-engraver.cc:114: if (span_ && !span_->get_bound (LEFT))
On 2010/02/24 23:11:50, hanwenn wrote:
> what if note_cols is empty?

Hmm, that could be dangerous. :)

If I check for note_columns_.size (), is it worth using currentMusicalColumn
(which would allow an episema to be typeset over a skip)?

[hanwenn, 2010-02-26 22:04:46]

On Fri, Feb 26, 2010 at 5:19 PM,  <n.puttock@gmail.com> wrote:

>> what if note_cols is empty?
>
> Hmm, that could be dangerous. :)
>
> If I check for note_columns_.size (), is it worth using
> currentMusicalColumn (which would allow an episema to be typeset over a
> skip)?

that is the usual approach - it may not make sense, but at least it
does not crash.

Or at least, the crash will happen later on :)

-- 
Han-Wen Nienhuys - hanwen@xs4all.nl - http://www.xs4all.nl/~hanwen