[ff3d-dev] bug ??

[olivier Gipouloux]

i want to solve a simple laplacian problem in a box  (<0,0,0> , 
<1,1,1>)with a jump of flux on an internal surface
at the middle of the box (i.e of eqn: x=.5)
for to do that, i generate to "half domains" and i make the union 
(computational domain) and the intersection
(the "jump surface")
*************************the pov 
file*********************************************
union {
   box {
      <0,0,0>, <1,1,0.5>
       }
   box {
      <1,1,0.5>, <1,1,1>
       }
   pigment { color rgb <1,0,0> }
      }
intersection {
   box {
      <0,0,0>, <1,1,0.5>
       }
   box {
      <1,1,0.5>, <1,1,1>
       }
   pigment { color rgb <0,1,0> }
            }

*************************the ff 
file*********************************************
// Definition de la boite //
vector n = (21,21,21);
vector a = (0,0,0);
vector b = (1,1,1);
scene S = pov("essai.pov");
mesh M = structured(n,a,b);

solve (u) in M krylov(precond=ichol)
         {
          test(v)
             int(dx(u)*dx(v) +dy(u)*dy(v) + dz(u)*dz(v) )
//         I am interested about incuding this term to
//            + int[<0,1,0>](dx(u)*dx(v))
//         The following term generat one error
         =  int(<0,1,0>)(3.*v);
             u = 0 on M xmin;
             u = 1. on M xmax;
         }           

save(medit,"essai",M);
save(medit,"essai",u,M);


*************************** the error 
file********************************************

Computing degrees of freedom positions...
Computing degrees of freedom positions: done
/home/delpino/src/ff3d/solver/DegreeOfFreedomSetBuilder.cpp:117: NOT 
FINISHED (PERIODIC BC)
Finite element method: disretization...
- assembling matrix
- assembling matrix: done [cost: time: 1s - CPU time: 0.57s]
- assembling second member
- assembling second member: done [cost: time: 0s - CPU time: 0s]
- discretizing boundary conditions

/home/delpino/src/ff3d/solver/BoundaryConditionCommonFEMDiscretization.hpp:92: 
The mesh of boundary  POVref : <0,1,0> was not generated

UNEXPECTED ERROR: this should not occure, please report it


I use ff3d-2005-12-03 because i cant find the version 1.0pre8
perhaps is it too old?

--
Olivier Gipouloux
Institut Camille Jordan UMR 5208
Batiment ISTIL
15 Boulevard Latarjet
69622 Villeurbanne Cedex
T� :  (33) 04 72 44 82 47 / Fax : (33) 04 72 43 11 45
https://mcs.univ-lyon1.fr/membres/gipouloux.html

Olivier.Gipouloux.vcf
Description: Vcard