Maxi, there are multiple reasons why you get errors when trying to compile the current version of step-17 with complex PETSc scalar type. 1. We are in te process of updating the example steps and didn't consider this one yet. (This why you have a lot of the deprecations warnings regarding teh use of ConstraintMatrix) 2. The number type for the remaining objects are not matching PetscScalar. We only allow AffineConstraints to bes used when both the corresponding vector and matrix classes use std::complex<> as underlying numer type as well. You should be able to work your way through this just changing the number type at the appropriate places to PetscScalar. Sometimes, this is quite tedious especially if you are really only considering real values. Comparing values for example is not uniquely defined. This is also a reason why it might not make sense to change every program to allow for both real and complex values. This directly leads to the last point: 3. This tutorial program doesn't need complex values, so having the check for complex support is intentional.
There are multiple tests using PETSc with complex scalar. E.g., have a look at tests/petsc_complex/parallel_sparse_matrix_01.cc. Best, Daniel -- 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 [email protected]. For more options, visit https://groups.google.com/d/optout.
