Add do_snap parameter to PMP::autorefine_triangle_soup

[LeoValque]

Summary of Changes

The PR adds the do_snap parameter to autorefine_triangle_soup(). When set to true, the coordinates are rounded to fit in double with additional subdivisions, preventing any self-intersections from occurring.

Todo

• Add to change log

Release Management

• Affected package(s): PMP
• Issue(s) solved (if any):
• Feature/Small Feature (if any): link
• Link to compiled documentation (obligatory for small feature) wrong link name to be changed
• License and copyright ownership: GeometryFactory

[sloriot]

What do you think of apply_iterative_snap_rounding() ? cc @MaelRL

[LeoValque]

do you suggest to create a new function apply_iterative_snap_rounding() or simply to rename the parameter do_snap to apply_iterative_snap_rounding ?

[sloriot]

do you suggest to create a new function apply_iterative_snap_rounding() or simply to rename the parameter do_snap to apply_iterative_snap_rounding ?

renaming the named parameter, provided Mael thinks it is a better name too.

[afabri]

What do you think of apply_iterative_snap_rounding() ? cc @MaelRL

I do not like an apply_ prefix. Did you have a look at the existing function for 2D

[sloriot]

What do you think of apply_iterative_snap_rounding() ? cc @MaelRL

I do not like an apply_ prefix. Did you have a look at the existing function for 2D

note that's a named parameter, not a free function

[sloriot]

What do you think of apply_iterative_snap_rounding() ? cc @MaelRL

I do not like an apply_ prefix. Did you have a look at the existing function for 2D

note that's a named parameter, not a free function

there is already the named parameter : apply_per_connected_component()

[MaelRL]

What do you think of apply_iterative_snap_rounding() ? cc @MaelRL

OK, but I also meant the function snap_polygon_soup(): even if it's internal, best to still avoid naming similarities.

[sloriot]

What do you think of apply_iterative_snap_rounding() ? cc @MaelRL

OK, but I also meant the function snap_polygon_soup(): even if it's internal, best to still avoid naming similarities.

for that one it could be internal::polygon_soup_snap_rounding()

[sloriot]

/Users/magritte/cgal_root/CGAL-6.1-Ic-91/include/CGAL/Polygon_mesh_processing/internal/snap_polygon_soup.h:138:13: warning: unused variable 'pm' [-Wunused-variable]
/Users/magritte/cgal_root/CGAL-6.1-Ic-91/include/CGAL/Polygon_mesh_processing/internal/snap_polygon_soup.h:136:9: warning: unused type alias 'GT' [-Wunused-local-typedef]
/mnt/testsuite/include/CGAL/Polygon_mesh_processing/internal/snap_polygon_soup.h:223:5: warning: this 'else' clause does not guard... [-Wmisleading-indentation]

and PMP examples fully red https://cgal.geometryfactory.com/CGAL/testsuite/summary-6.1-Ic-91.html?package=Polygon_mesh_processing_Examples

[afabri]

Why does the function name end with an underscore ?

[LeoValque]

I need to find another name; I did that temporarily. I added a front function, autorefine_triangle_soup, that dispatches execution between this function and snap_triangle_soup because the mutual calling between them posed difficulties. Is autorefine_triangle_soup_internal a good name?

[afabri]

Putting it in an internal namespace?

[LeoValque]

It is already the case

[afabri]

Can we really have such a global variable?

[LeoValque]

I moved it inside the class that is using it