Reading MED results from Code_Aster : scrambled timesteps
Hi, I've been using Gmsh for several years to visualize FEM results from Code_Aster. This is quite convenient, and Gmsh is probably the best tool for this out there. Unfortunately I just found a strange bug (or is it?): when reading back post-processing views in MED 3.x format exported from Code_Aster (v11.3.0-2) with Gmsh (2.6.0 to 2.8.2), the displacement fields seem to be read in linear order from the file, while Code_Aster saves those fields in random order, associating some properties to the field to retrieve its correct timestep. I checked the result MED file with an external tool (ViTables) to be sure. As a result (at least that's what I guess happens) I end up with a correct post-processing view (i.e. for a each timestep taken individually the displacements are OK) but in a totally random order (i.e. timesteps are not ordered). I compared the results with a result export in the old v1.2 format, and in that later case I get correctly-ordered timesteps. That's how I now displacements are correct, only the timesteps are not. I couldn't find any mention of such behavior elsewhere. Are my versions of Gmsh or Code_Aster broken somehow? Thanks
- Jerome