CGAL users discussion list

[francoisv2 <>]

I get the assertion violation:

Expr: e->incident_sface() != SFace_const_handle()
File: /usr/local/include/CGAL/Nef_S2/SM_const_decorator.h
Line: 326

when converting the following polyhedron (in OFF) format to a
Nef_polyhedron:

OFF
32 31 0

24955.6 18330.3 -8856.45
24745.5 18145.5 -8856.45
24896.4 18034.6 -8856.45
24907.2 18402.6 -8856.45
24645.6 18304 -8856.45
25027.9 18281.9 -8856.45
25077.5 17986.7 -8856.45
24890.3 18488 -8856.45
24633.4 18637.2 -8856.45
24610.7 18488 -8856.45
25113.3 18265 -8856.45
25263.5 18008.4 -8856.45
25198.6 18281.9 -8856.45
24909.5 18578.7 -8856.45
24699.4 18773 -8856.45
25428.7 18096.7 -8856.45
25271 18330.3 -8856.45
24964.1 18653.7 -8856.45
24946.8 18823 -8856.45
24802.7 18883.1 -8856.45
25044.4 18700.1 -8856.45
25550 18239.4 -8856.45
25319.3 18402.6 -8856.45
25087.6 18755.6 -8856.45
25136.6 18709.8 -8856.45
25224.8 18681.1 -8856.45
25357.9 18599.6 -8856.45
25322.8 18564.2 -8856.45
25336.3 18488 -8856.45
25284.1 18631.3 -8856.45
25486.7 18511.3 -8856.45
25610.7 18416.6 -8856.45
3  0 1 2
3  3 4 1
3  3 1 0
3  5 2 6
3  5 0 2
3  7 8 9
3  7 9 4
3  7 4 3
3  10 6 11
3  10 5 6
3  12 10 11
3  13 14 8
3  13 8 7
3  15 12 11
3  16 12 15
3  17 14 13
3  18 19 14
3  18 17 20
3  18 14 17
3  21 22 16
3  21 16 15
3  23 18 20
3  24 23 20
3  25 23 24
3  26 27 28
3  26 29 27
3  26 25 29
3  30 28 22
3  30 22 21
3  30 26 28
3  31 30 21

If the order of facet indices in the last face is flipped to 21 30 31,
everything seems to work fine... But I believe this is the wrong orientation
for the last face.

Note that all facets are triangles and on the same plane. I've tried GMP
exact number types as well as CORE::Expr, with the same result. I suspect
that it is not a robustness issue, but rather a topological one incurred
during construction.

Also, enabling TRACEN gives a segmentation fault at line 289 in
polyhedron_3_to_nef_3.h.



--
View this message in context: 
http://cgal-discuss.949826.n4.nabble.com/Converting-from-Polyhedron-3-to-Nef-polyhedron-3-error-tp4659766.html
Sent from the cgal-discuss mailing list archive at Nabble.com.