Navigation:
Title Page,
Overview,
Installation,
Usage,
Algorithms,
Geometry Kernels,
Bibliography
Installation
In addition to the source code
you need
- CGAL, version 1.2. (January 99).
Later versions won't work, because of the introduction of namespace CGAL.
- LEDA; the version I used to run the
experiments was 3.7.2
Optionally, you also need
- LCEP Number Types
(for CGAL_Expanded_double), if you compile with -DCGAL_USE_LCEP_NUMBER_TYPES
- doubledouble, if you compile with -DCGAL_USE_DOUBLEDOUBLE
Once you have installed CGAL and LEDA and unpacked the
source code package,
you can use the makefile to compile the experiments.
You have to change the top lines according to your installation of
CGAL and LEDA.
# CGAL + LEDA :
CGAL_MAKEFILE = /KM/projects/CGAL/CGAL-1.2/make/makefile_sparc_SunOS-5.6_eg++-egcs-2.91.66_LEDA
If you want to use doubledouble and/or CGAL_Expanded_double and have
these packages installed, you also need to modify the lines
#DOUBLEDOUBLE_INCL =
#DOUBLEDOUBLE_LIBDIR =
#DOUBLEDOUBLE_LIB = -ldoubledouble
#DOUBLEDOUBLE_FLAGS = -DCGAL_USE_DOUBLEDOUBLE -DDD_INLINE
#LCEP_NUMBER_TYPES_INCL =
#LCEP_NUMBER_TYPES_LIBDIR =
#LCEP_NUMBER_TYPES_LIB = -lLCEP_Number_types
#LCEP_NUMBER_TYPES_FLAGS = -DCGAL_USE_LCEP_NUMBER_TYPES
Finally, type make and have fun (and a lot of patience. Since you are
compiling more than 300 convex hull algorithm variations
(see src/makefile),
compilation may take 2 to 4 hours!)