GitLab

#499 The number of processor-wise dofs per numbering subset is now computed and stored; this is what is displayed in the output files.

#499 Introduce a struct to hold the count of dofs. Currently it still doesn't know the number per numbering subset; it is just equivalent to previous code.

#480 Introduce in Node a new data that tracks the numbering subsets to which the node belongs to. At first sight it didn't seem to work as processor 0 was empty for my small numbering subsets, but in fact it's logical: numbering subsets related to a given small domain are likely to be handled by the same processor due to Parmetis partitioning.

#480 Problem is corrected and the code is working... but not prettily; some clean-up is needed (in progress).

#480 Order a bit the writing of outputs: now there are three levels of folders: base, god of dof, numbering subset.

#480 [WIP] Give more responsibilities about outputs to GodOfDof (and less to Model, VariationalFormulation...).

#480 Create a FormulationSolver library and modify VariationalFormulation so that output directory depends on the numbering subset id. Update XCode project to new XCode version.

#480 Abusive assert has been removed; Stokes works well with 3 numbering subsets. Last test before closing the ticket is to see whether the patterns are all correct.

#480 Stokes: introduce two other numbering subsets to check whether it works. It doesn't at the time being.

#480 VariationalFormulation: add a dependancy on the numbering subset (through a const reference). So far only hyperelasticity model has been updated accordingly.

#480 An assert I removed earlier was in fact perfectly valid, once other bugs are corrected of course...

#480 List of numbering subset in a GodOfDof really needs to be public; remove the comment that suggest otherwise.

#480 Write dof output taking into account the numbering subset. PostProcessing however has been quickly patched by hardcoding the fact there is only one numbering subset which index is 1; ticket #495 has been issued to fix properly PostProcessing.

#480 Remove hardcoded line from OndomaticLikePrint(). Rewrite the main of this test to use the InitHappyHeart recently introduced in the code (otherwise TimeKeep was not properly initialized).

#480 GodOfDof: remove one of the temporary hardcoded line and compute properly the processor-wise index for each numbering subset.

#480 Previous commit was actually incomplete: the step in which a dof not present in a given numbering subset was called was missing.