Creates a glissando stem grob that uses stems' functionality.

Creates a glissando stem grob that uses stems' functionality.

[MikeSol, 2011-07-20 12:32:54]

Tries to minimize overlap with stem.cc - Glissando_stem::has_interface now
appears four (indispensable) times in stem.cc.  I think it is pretty
modular.  If I get run over by a truck and no one knows how to maintain this
thing and people decide to delete glissando stems from LilyPond, I think it
is much easier to do with this version than previous ones.

A lot of files change because Stem::head_positions is now a callback.  I may
even rename the property - it should really be called extremal-positions, as
it describes the position of the extreme two heads (stems) or glissandi
(glissando stems).

This implementation, while long, is more lilypond-esque.

The input syntax is HORRIBLE - it really needs to access the parser somehow.
Otherwise, users will have to build event chords from scratch.  Users need
to be able to do:

gs4:32-.^"foo"

Where gs is glissando stem (or, alternatively, mikeSolomonIsANinja).  Otherwise,
they'll have to roll their own event chords to get the markup and
articulation in there, and that seems like a ferocious waste of time.  If
anyone could help me figure out how to do that, I'd be much obliged!

For now, the input syntax is shown in the regtest.

Passes regtests, although I had to create a dummy property called "dimension" to
get it to pass (I have no clue why) - I'll remove this once I figure that out.

I'll eventually get around to making real extent functions for
GlissandoStemTremolo.

Cheers,
MS

[hanwenn, 2011-07-20 15:27:34]

On Wed, Jul 20, 2011 at 9:32 AM,  <mtsolo@gmail.com> wrote:
> A lot of files change because Stem::head_positions is now a callback.  I
> may
> even rename the property - it should really be called
> extremal-positions, as
> it describes the position of the extreme two heads (stems) or glissandi
> (glissando stems).

Could you put this in a separate patch?

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

[hanwenn, 2011-07-20 15:31:57]

http://codereview.appspot.com/4777044/diff/7001/lily/beam.cc
File lily/beam.cc (right):

http://codereview.appspot.com/4777044/diff/7001/lily/beam.cc#newcode706
lily/beam.cc:706: Interval positions = robust_scm2interval ((*s)->get_property
("head-positions"), Interval (0,0));
rather than changing all callers, why don't you put the get_property in
Stem::head_positions() ?

[mike_apollinemike.com, 2011-07-20 16:05:17]

On Jul 20, 2011, at 5:31 PM, hanwenn@gmail.com wrote:

> 
> http://codereview.appspot.com/4777044/diff/7001/lily/beam.cc
> File lily/beam.cc (right):
> 
> http://codereview.appspot.com/4777044/diff/7001/lily/beam.cc#newcode706
> lily/beam.cc:706: Interval positions = robust_scm2interval
> ((*s)->get_property ("head-positions"), Interval (0,0));
> rather than changing all callers, why don't you put the get_property in
> Stem::head_positions() ?
> 

Whatever is more LilyPondish I'll do - your suggestion would require creating
another callback for the property, which seems like it may create confusion for
future programmers (i.e. why is head_positions called in place X but the
property head-positions called in place Y?).  That's the reason I swapped out
the function with property calls.  But, again, I'm a relative newb - whatever
you think is best I'll do.

Cheers,
MS

[Neil Puttock, 2011-07-20 16:27:34]

http://codereview.appspot.com/4777044/diff/7001/input/regression/glissando-st...
File input/regression/glissando-stem.ly (right):

http://codereview.appspot.com/4777044/diff/7001/input/regression/glissando-st...
input/regression/glissando-stem.ly:10: \xenakisStem #(ly:make-duration 3) #32
This should be a music function, possibly incorporating the \glissando for the
initial note.

Something like this, perhaps:

#(define-public (note-to-gliss-stem music)
  "Replace @code{NoteEvent}s and @code{SkipEvent}s by
@code{GlissandoStemEvent}s."
  (if (or (music-has-type music 'skip-event)
          (music-has-type music 'note-event))
      (make-music 'GlissandoStemEvent
                  'origin (ly:music-property music 'origin)
                  'duration (ly:music-property music 'duration))
      music))

xenakisStems =
#(define-music-function (parser location music) (ly:music?)
   (music-map note-to-gliss-stem music))

\relative c' {
  c4:32\glissando
  \xenakisStems {
    s8:32\f\> s s16 c s8:64
  }
  a''\p
}

http://codereview.appspot.com/4777044/diff/7001/scm/define-grobs.scm
File scm/define-grobs.scm (right):

http://codereview.appspot.com/4777044/diff/7001/scm/define-grobs.scm#newcode960
scm/define-grobs.scm:960: (duration . ,(grob::calc-property-by-copy
'duration-log))
remove

http://codereview.appspot.com/4777044/diff/7001/scm/define-grobs.scm#newcode961
scm/define-grobs.scm:961: (duration-log . ,(grob::calc-property-by-copy
'duration-log))
,stem::calc-duration-log

[mike_apollinemike.com, 2011-07-20 20:04:46]

On Jul 20, 2011, at 6:27 PM, n.puttock@gmail.com wrote:

> 
>
http://codereview.appspot.com/4777044/diff/7001/input/regression/glissando-st...
> File input/regression/glissando-stem.ly (right):
> 
>
http://codereview.appspot.com/4777044/diff/7001/input/regression/glissando-st...
> input/regression/glissando-stem.ly:10: \xenakisStem #(ly:make-duration
> 3) #32
> This should be a music function, possibly incorporating the \glissando
> for the initial note.
> 
> Something like this, perhaps:
> 
> #(define-public (note-to-gliss-stem music)
>  "Replace @code{NoteEvent}s and @code{SkipEvent}s by
> @code{GlissandoStemEvent}s."
>  (if (or (music-has-type music 'skip-event)
>          (music-has-type music 'note-event))
>      (make-music 'GlissandoStemEvent
>                  'origin (ly:music-property music 'origin)
>                  'duration (ly:music-property music 'duration))
>      music))
> 
> xenakisStems =
> #(define-music-function (parser location music) (ly:music?)
>   (music-map note-to-gliss-stem music))
> 
> \relative c' {
>  c4:32\glissando
>  \xenakisStems {
>    s8:32\f\> s s16 c s8:64
>  }
>  a''\p
> }
> 

Great idea!  Done.

> http://codereview.appspot.com/4777044/diff/7001/scm/define-grobs.scm
> File scm/define-grobs.scm (right):
> 
>
http://codereview.appspot.com/4777044/diff/7001/scm/define-grobs.scm#newcode960
> scm/define-grobs.scm:960: (duration . ,(grob::calc-property-by-copy
> 'duration-log))
> remove
> 

Done (have no clue why that was in there...).

>
http://codereview.appspot.com/4777044/diff/7001/scm/define-grobs.scm#newcode961
> scm/define-grobs.scm:961: (duration-log . ,(grob::calc-property-by-copy
> 'duration-log))
> ,stem::calc-duration-log

Done.

New patch set uploaded.  The new patch has script support.  I'll be tinkering
with beams in the next few days.

Cheers,
MS

[hanwenn, 2011-07-21 02:30:52]

On Wed, Jul 20, 2011 at 1:05 PM, mike@apollinemike.com
<mike@apollinemike.com> wrote:

>> http://codereview.appspot.com/4777044/diff/7001/lily/beam.cc#newcode706
>> lily/beam.cc:706: Interval positions = robust_scm2interval
>> ((*s)->get_property ("head-positions"), Interval (0,0));
>> rather than changing all callers, why don't you put the get_property in
>> Stem::head_positions() ?
>>
>
> Whatever is more LilyPondish I'll do - your suggestion would require creating
another callback for the property, which seems like it may create confusion for
future programmers (i.e. why is head_positions called in place X but the
property head-positions called in place Y?).  That's the reason I swapped out
the function with property calls.  But, again, I'm a relative newb - whatever
you think is best I'll do.

You'd move the current code for head_positions() into a
calc_head_positions callback.

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

[hanwenn, 2011-07-21 02:35:51]

On Wed, Jul 20, 2011 at 11:30 PM, Han-Wen Nienhuys <hanwenn@gmail.com> wrote:
>>> http://codereview.appspot.com/4777044/diff/7001/lily/beam.cc#newcode706
>>> lily/beam.cc:706: Interval positions = robust_scm2interval
>>> ((*s)->get_property ("head-positions"), Interval (0,0));
>>> rather than changing all callers, why don't you put the get_property in
>>> Stem::head_positions() ?
>>>
>>
>> Whatever is more LilyPondish I'll do - your suggestion would require creating
another callback for the property, which seems like it may create confusion for
future programmers (i.e. why is head_positions called in place X but the
property head-positions called in place Y?).  That's the reason I swapped out
the function with property calls.  But, again, I'm a relative newb - whatever
you think is best I'll do.
>
> You'd move the current code for head_positions() into a
> calc_head_positions callback.

Oh, you already did that.

In order to have as little moving pieces as possible, put the
get_property() and the conversion in head_positions(), and leave the
callers alone.  We can later decide to drop head_positions(), but it's
not really necessary.  There are many such C++ functions that are
shortcuts for a get_property, eg., Beam::get_beam_thickness

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

[MikeSol, 2011-08-06 16:01:03]

I've merged my glissando stem work with current master.  It passes regtests and
can currently handle tremoli and scripts (see the regtest).  However, these
scripts just kinda float in the ether - they are prone to bump into anything
outside the staff (for now).  If anyone has any suggestions for how to make:

a'4 \glissando
\xenakisStems { s4-.^"foo" }
a''4

such that the staccato does not bump into foo, I'd be much obliged.  It likely
has something to do with outside staff skylines and such, but I'm not sure yet
how to negotiate this with pure heights.

Once this gets pushed, I'll work on beams.

Cheers,
MS

[pkx166h, 2011-08-06 20:56:14]

make fails.

--snip--

 -DHAVE_CONFIG_H  -DNDEBUG -I/home/jlowe/lilypond-git/lily/include -I./out
-I/home/jlowe/lilypond-git/flower/include -I../flower/./out -I../flower/include
-I/home/jlowe/lilypond-git/lily/out  -O2 -finline-functions -g -pipe -pthread
-I/usr/include/freetype2   -pthread -I/usr/include/pango-1.0
-I/usr/include/freetype2 -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include  
-Wno-pmf-conversions  -W -Wall -Wconversion -o out/note-head-line-engraver.o
/home/jlowe/lilypond-git/lily/note-head-line-engraver.cc
/home/jlowe/lilypond-git/lily/glissando-stem.cc:134: warning: unused parameter
'smob'
/home/jlowe/lilypond-git/lily/glissando-stem.cc: In static member function
'static scm_unused_struct*
Glissando_stem::calc_head_positions(scm_unused_struct*)':
/home/jlowe/lilypond-git/lily/glissando-stem.cc:154: error:
'VALGRIND_PRINTF_BACKTRACE' was not declared in this scope
make[1]: *** [out/glissando-stem.o] Error 1
make[1]: *** Waiting for unfinished jobs....
make[1]: Leaving directory `/home/jlowe/lilypond-git/build/lily'
make: *** [all] Error 2
jlowe@jlowe-lilybuntu2:~/lilypond-git/build$ 


sorry

James

[mike_apollinemike.com, 2011-08-07 09:17:09]

On Aug 6, 2011, at 10:56 PM, pkx166h@gmail.com wrote:

> make fails.
> 
> --snip--
> 
> -DHAVE_CONFIG_H  -DNDEBUG -I/home/jlowe/lilypond-git/lily/include
> -I./out -I/home/jlowe/lilypond-git/flower/include -I../flower/./out
> -I../flower/include -I/home/jlowe/lilypond-git/lily/out  -O2
> -finline-functions -g -pipe -pthread -I/usr/include/freetype2   -pthread
> -I/usr/include/pango-1.0 -I/usr/include/freetype2
> -I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include
> -Wno-pmf-conversions  -W -Wall -Wconversion -o
> out/note-head-line-engraver.o
> /home/jlowe/lilypond-git/lily/note-head-line-engraver.cc
> /home/jlowe/lilypond-git/lily/glissando-stem.cc:134: warning: unused
> parameter 'smob'
> /home/jlowe/lilypond-git/lily/glissando-stem.cc: In static member
> function 'static scm_unused_struct*
> Glissando_stem::calc_head_positions(scm_unused_struct*)':
> /home/jlowe/lilypond-git/lily/glissando-stem.cc:154: error:
> 'VALGRIND_PRINTF_BACKTRACE' was not declared in this scope
> make[1]: *** [out/glissando-stem.o] Error 1
> make[1]: *** Waiting for unfinished jobs....
> make[1]: Leaving directory `/home/jlowe/lilypond-git/build/lily'
> make: *** [all] Error 2
> jlowe@jlowe-lilybuntu2:~/lilypond-git/build$
> 
> 
> sorry
> 
> James
> 
> 

Sorry about that!  Long story short - I had done a git grep to get rid of all my
valgrinding, but before I remembered to add new files (like glissando-stem.cc),
so valgrind stuck around in some of them.  It's gone now and should still pass
regtests.

Cheers,
MS

[MikeSol, 2011-08-10 16:18:44]

New patch set up that gets Y-offset right for glissando stem scripts.

Cheers,
MS