that fixes the warning. For the record, it was:
-------------------------------------------
Compiling linking_2d_and_3d ...
make[1]: Entering directory `/home/lrineau/CGAL/releases-and-testsuites/CGAL-3.4-I-407/test_i686_Linux-2.6_g++-4.3.1_CentOS-5.1-O3/Triangulation_3_Examples'
/usr/local/gcc43/bin/g++ -I../../include -Wall -frounding-math '-I/home/lrineau/CGAL/releases-and-testsuites/CGAL-3.4-I-407/include/CGAL/config/i686_Linux-2.6_g++-4.3.1_CentOS-5.1-O3' '-I/home/lrineau/CGAL/releases-and-testsuites/CGAL-3.4-I-407/include' -O3 -pipe -I/home/lrineau/CGAL/bimap -I/usr/lib/qt-3.3/include -DCGAL_USE_F2C -DCGAL_USE_F2C -c linking_2d_and_3d.cpp
../../include/CGAL/Compact_container.h: In member function 'typename CGAL::Triangulation_data_structure_2<Vb, Fb>::Vertex_handle CGAL::Triangulation_data_structure_2<Vb, Fb>::insert_dim_up(typename CGAL::Compact_container<typename Vb::Rebind_TDS<CGAL::Triangulation_data_structure_2<Vb, Fb> >::Other, CGAL::Default_argument>::iterator, bool) [with Vb = My_vertex_2<CGAL::Dummy_tds_3, CGAL::Triangulation_ds_vertex_base_2<void> >, Fb = CGAL::Triangulation_ds_face_base_2<void>]':
../../include/CGAL/Compact_container.h:772: warning: array subscript is above array bounds
/usr/local/gcc43/bin/g++ -o linking_2d_and_3d linking_2d_and_3d.o '-L/home/lrineau/CGAL/releases-and-testsuites/CGAL-3.4-I-407/lib/i686_Linux-2.6_g++-4.3.1_CentOS-5.1-O3' -L/usr/lib/qt-3.3/lib -Wl,-R/home/lrineau/CGAL/releases-and-testsuites/CGAL-3.4-I-407/lib/i686_Linux-2.6_g++-4.3.1_CentOS-5.1-O3:/usr/lib/qt-3.3/lib -lCGAL -lCGALcore++ -lCGALQt -lCGALimageIO -lCGALPDB -lqt-mt -llapack -lGL -lGLU -lblas -lz -lmpfr -lgmpxx -lgmp -lX11 -lboost_thread -lboost_program_options -lm
make[1]: Leaving directory `/home/lrineau/CGAL/releases-and-testsuites/CGAL-3.4-I-407/test_i686_Linux-2.6_g++-4.3.1_CentOS-5.1-O3/Triangulation_3_Examples'
Executing linking_2d_and_3d ...
-------------------------------------------
Maybe a gcc bug. Howerver, with that patch, the code seems easier to read
(the variable that contains the dimension is now named "dim", instead of
"i", and is filled as soon as the dimension() of the triangulation is
updated).
Things like "CGAL_assertion(denominator != 0)" produced assertion failures for no gain.
So now, the assertion is triggered only if the condition is certain, using CGAL::certainly().
That is, change the following in all assertion files :
((EX)?(static_cast<void>(0)): ::CGAL::assertion_fail( # EX , __FILE__, __LINE__))
to :
(CGAL::certainly(EX)?(static_cast<void>(0)): ::CGAL::assertion_fail( # EX , __FILE__, __LINE__))
det2x2_by_formula
det3x3_by_formula
det4x4_by_formula
det5x5_by_formula
det6x6_by_formula
to:
determinant
How cute... a name independent of the dimension, and even readable !
sign_of_determinant2x2
sign_of_determinant3x3
sign_of_determinant4x4
sign_of_determinant5x5
sign_of_determinant6x6
to:
sign_of_determinant
So that we have less dimension-dependent namings, at least internally...
as the default is now the empty string "".
It should fix the problem that we have lost the assertion messages
(seeing "what(): basic_string::_S_construct NULL not valid" instead),
for packages that use package-specific assertion macros.
Sylvain Pion
Laurent Rineau
Pedro Machado Manhaes de Castro
Trung Nguyen
Mariette Yvinec
Andreas Fabri