mirror of https://github.com/CGAL/cgal
68 lines
2.6 KiB
Plaintext
68 lines
2.6 KiB
Plaintext
- Remove old file of Data_structure_using_octree
|
||
-- Laurent Rineau 2006/02/23
|
||
|
||
- Verify that FT can be from an exact kernel...
|
||
- Make depends for applications/makefile
|
||
-- Laurent Rineau, 2005/11/14
|
||
|
||
- add an assertion in compute_critical_radius() of Sliver_exuder.h
|
||
- clean global functions in include/CGAL/Mesh_3/Sliver_exuder.h
|
||
- clean include/CGAL/IO/File_medit.h Surface_mesher/include/CGAL/IO/
|
||
and include/CGAL/IO.h
|
||
-- Laurent Rineau, 2005/11/09
|
||
|
||
- error messages, if reading file does not work, in applications/
|
||
- add a "writing..." in combined_spheres.C
|
||
-- Laurent Rineau, 2005/11/04
|
||
|
||
- examples/Mesh_3/polyhedral_surface_mesher.C is broken because:
|
||
- output_pslg_to_medit() in include/CGAL/IO/File_medit.h need
|
||
surface_index() in points. :-(
|
||
- Weighted_point wp = Weighted_point(p, best_weight, index);
|
||
in include/CGAL/Mesh_3/Slivers_exuder.h
|
||
- add_in_graph() & o in vertices (this sucks!)
|
||
- should use Weighted_point_with_surface_index_geom_traits
|
||
- test/Mesh_3/multi_spheres.C is broken because
|
||
../../../Surface_mesher/ \
|
||
include/CGAL/Surface_mesher/Oracles/Multi_implicit_oracle.h
|
||
*is* broken.
|
||
- File_medit.h should not require systematicaly surface_index()
|
||
[DONE]- provide .cin files for tests (read Dev'Manual)
|
||
[DONE]- remove info()s everywhere
|
||
Mesh_3/include/CGAL/Implicit_surfaces_mesher_3.h
|
||
Mesh_3/include/CGAL/Mesh_3/Facet_on_surface_criterion.h
|
||
Mesh_3/include/CGAL/Mesh_3/Implicit_surface_mesher_visitor.h
|
||
Mesh_3/include/CGAL/Mesh_3/Refine_tets.h
|
||
Mesh_3/include/CGAL/Mesh_3/Slivers_exuder.h
|
||
|
||
-- Laurent Rineau, 2005/10/02
|
||
|
||
|
||
Sliver_exuder.h should contain only functions, I think.
|
||
-- Laurent Rineau, 2005/04/06
|
||
|
||
include/CGAL/Mesh_3/ should only contain private files, but examples .C
|
||
files use some of them.
|
||
Renommer Mesh_3::Complex_2_in_triangulation_cell_base_3 parce qu'il
|
||
est très moche!!
|
||
Est-ce que les visiteurs doivent etre chaines par des pointeurs ou
|
||
des ref?
|
||
-- Laurent Rineau, 2050/05/10
|
||
|
||
As for potential robustess problems, in Sliver_exuder...
|
||
Problems come from compute_critical_squared_radius, that is a constructor.
|
||
cricical radii are compared.
|
||
|
||
Perhaps should be templatize Sliver_exuder with two traits, like
|
||
Filtered_kernel: an exact one, and an inexact one, as well as a
|
||
converter.
|
||
|
||
With meta-programming, we could provide defaults:
|
||
- for Tr whose traits is a Filtered_kernel, use it
|
||
- defaults: use the same traits classes.
|
||
|
||
Then, in Slivers_exuder, the prestart could save exact
|
||
squared_radii.
|
||
|
||
-- Laurent Rineau, 2005/05/21
|