Dear Giovanni, Its good to hear that you managed to solve / work around this issue! Would it be possible for you to post the reduced (problematic) mesh so that we can see if there's anything that we can learn from it. I'm guessing that you expected lines with entries such as 6 0.1999999999991087 0 0 to be parsed as 6 0.2 0 0 ? Did you ultimately have to output a grid that had the coordinates recorded with a greater precision?
Thanks, Jean-Paul On Thursday, July 20, 2017 at 7:47:04 PM UTC+2, Giovanni Di Ilio wrote: > > Dear Wolfgang, > > thanks for your quick reply and for your suggestion. > I simplified the case to only two cells, the one that was supposed to be > "wrong" and one of its neighbours, that was supposed to be ok, for > comparison. > By watching at the coordinates of the vertices in the file .msh I found > out that the problem was related to some rounding error. I guess there is a > bug in the parser. I just converted the coordinate values into double and > now the mapping works fine everywhere. > > Thank you very much again, > > Best, > Giovanni > > On Thursday, July 20, 2017 at 4:32:43 PM UTC+2, Wolfgang Bangerth wrote: >> >> On 07/20/2017 05:53 AM, Giovanni Di Ilio wrote: >> > >> > I provide a small code that I am testing as well as the .msh and .geo >> files I >> > am using. >> > The code computes the mapped point of a generic real point which is >> placed in >> > the center of a "sick" cell (id=354). I would expect that the >> coordinates of >> > the unit point are x=0.5, y=0.5. However, what I get is x=0.5, y=0.0. >> The same >> > apply if you pick an other real point within this cell. >> >> It's often difficult to debug these cases with such a large mesh. But >> since >> you know which cell the problem appears in, take the test.msh file and >> remove >> (by hand) all other cells and all of the vertices not needed. This way >> you >> should end up with a testcase that has only one cell and that should be >> much >> easier to figure out. >> >> Best >> W. >> >> -- >> ------------------------------------------------------------------------ >> Wolfgang Bangerth email: bang...@colostate.edu >> www: http://www.math.colostate.edu/~bangerth/ >> >> -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
