FiguredBass: Rewrite of the engraver to fix vertical position

FiguredBass: Rewrite of the engraver to fix vertical position

-) use list<..> instead of vector<..>
-) Don't reuse old figure storage and no longer simply fill in
   empty positions, which causes ordering to be lost in some
   extender cases.
-) Add description of engraver logic

[Colin Campbell, 2011-06-25 19:31:00]

Just tidying up a bit: has this been superceded by the fix to issue 1671?  If
so, the reitveld issue could be closed.

cheers,
Colin

[Reinhold, 2011-06-25 19:38:30]

On 2011/06/25 19:31:00, Colin Campbell wrote:
> Just tidying up a bit: has this been superceded by the fix to issue 1671?

No, that patch is for a totally unrelated problem, namely that for e.g.
  <6> <3 6 9>
the 3 (and of course the 9) will be placed BELOW instead of above the extender
for the 7. I'll continue on that patch during the summer, since I desparetely
need it for a score I'm publishing...

[pkx166h, 2011-08-27 08:14:28]

patching today's tree (27 Aug) I get

--snip--
jlowe@jlowe-lilybuntu2:~/lilypond-git$ patch -p1 <
../Desktop/issue224052_2001.diff 
patching file input/regression/figured-bass-continuation-center.ly
patching file input/regression/figured-bass-figureorder-position.ly
patching file input/regression/figured-bass-figureorder.ly
patching file lily/figured-bass-engraver.cc
Hunk #4 FAILED at 91.
Hunk #5 FAILED at 135.
Hunk #6 FAILED at 192.
Hunk #7 FAILED at 224.
Hunk #8 succeeded at 410 (offset 1 line).
Hunk #9 succeeded at 498 (offset 1 line).
Hunk #10 succeeded at 513 (offset 5 lines).
Hunk #11 FAILED at 528.
5 out of 11 hunks FAILED -- saving rejects to file
lily/figured-bass-engraver.cc.rej
jlowe@jlowe-lilybuntu2:~/lilypond-git$ 

--snip--

Make fails:

--snip--

make[1]: Entering directory `/home/jlowe/lilypond-git/build/lily'
rm -f ./out/figured-bass-engraver.dep;
DEPENDENCIES_OUTPUT="./out/figured-bass-engraver.dep
./out/figured-bass-engraver.o" g++ -c -Woverloaded-virtual 
-I/usr/include/python2.6 -I/usr/include/python2.6 -fno-strict-aliasing  -g
-fwrapv     -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/figured-bass-engraver.o
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc: In member function 'void
Figure_group::assign_from_event(Stream_event*)':
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc:94: error: 'item' was not
declared in this scope
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc: In member function 'void
Figured_bass_engraver::process_music()':
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc:328: error: 'struct
Figure_group' has no member named 'reset_figure'
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc:336: error: 'struct
Figure_group' has no member named 'is_continuation'
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc:338: error: 'struct
Figure_group' has no member named 'reset_figure'
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc:349: error: 'struct
Figure_group' has no member named 'is_continuation'
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc: In member function 'void
Figured_bass_engraver::create_grobs()':
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc:418: error: no match for
'operator=' in 'it =
((Figured_bass_engraver*)this)->Figured_bass_engraver::groups_.std::vector<_Tp,
_Alloc>::begin [with _Tp = Figure_group, _Alloc =
std::allocator<Figure_group>]()'
/usr/include/c++/4.4/bits/stl_list.h:114: note: candidates are:
std::_List_iterator<Figure_group>&
std::_List_iterator<Figure_group>::operator=(const
std::_List_iterator<Figure_group>&)
/home/jlowe/lilypond-git/lily/figured-bass-engraver.cc:418: error: no match for
'operator!=' in 'it !=
((Figured_bass_engraver*)this)->Figured_bass_engraver::groups_.std::vector<_Tp,
_Alloc>::end [with _Tp = Figure_group, _Alloc = std::allocator<Figure_group>]()'
/usr/include/c++/4.4/bits/stl_list.h:175: note: candidates are: bool
std::_List_iterator<_Tp>::operator!=(const std::_List_iterator<_Tp>&) const
[with _Tp = Figure_group]
/home/jlowe/lilypond-git/flower/include/std-string.hh:60: note:                
bool operator!=(const std::string&, const std::string&)
/home/jlowe/lilypond-git/flower/include/rational.hh:99: note:                
bool operator!=(const Rational&, const Rational&)
/home/jlowe/lilypond-git/lily/include/moment.hh:73: note:                 bool
operator!=(const Moment&, const Moment&)
/home/jlowe/lilypond-git/lily/include/duration.hh:56: note:                 bool
operator!=(Duration, Duration)
make[1]: *** [out/figured-bass-engraver.o] Error 1
make[1]: Leaving directory `/home/jlowe/lilypond-git/build/lily'
make: *** [all] Error 2
jlowe@jlowe-lilybuntu2:~/lilypond-git/build$

[reinhold_kainhofer.com, 2011-08-27 11:32:24]

Am Saturday, 27. August 2011, 10:14:28 schrieb pkx166h@gmail.com:
> patching today's tree (27 Aug) I get
> 
> --snip--
> jlowe@jlowe-lilybuntu2:~/lilypond-git$ patch -p1 <
> ../Desktop/issue224052_2001.diff
> patching file input/regression/figured-bass-continuation-center.ly
> patching file input/regression/figured-bass-figureorder-position.ly
> patching file input/regression/figured-bass-figureorder.ly
> patching file lily/figured-bass-engraver.cc
> Hunk #4 FAILED at 91.
> Hunk #5 FAILED at 135.
> Hunk #6 FAILED at 192.
> Hunk #7 FAILED at 224.

If any hunk fails to apply, you don't have to continue testing the patch. This 
means that when you compile lilypond you don't have all changes in the patch, 
but just some. Either lilypond will not compile at all, or it will compile, 
but the functionality that the patch should have provided will not be 
available, because they were never applied to the code.

Cheers,
Reinhold

-- 
------------------------------------------------------------------
Reinhold Kainhofer, reinhold@kainhofer.com, http://reinhold.kainhofer.com/
 * Financial & Actuarial Math., Vienna Univ. of Technology, Austria
 * http://www.fam.tuwien.ac.at/, DVR: 0005886
 * LilyPond, Music typesetting, http://www.lilypond.org

[Graham Percival (old account), 2011-09-24 09:51:54]

any chance of rebasing this patch?

[Reinhold, 2011-09-26 23:27:36]

On 2011/09/24 09:51:54, Graham Percival wrote:
> any chance of rebasing this patch?

New patchset uploaded, but that's not completely ready for consumption.

I'm still fighting with the issue that a child grob of the vertical alignment
spanner can only be on the same vertical position as its predecessor, but not
above (we have a check for max(0, dy) in there). If I make that max(0,...) check
optional, the bass figures are correctly shifted up, but unfotunately I loose
all padding :(