Use `PetscRealPart` and `PetscImaginaryPart` in supermeshing.py

[ReubenHill]

Since we are using PetscScalar we should probably use PetscRealPart and PetscImaginaryPart.

This would be instead of

real_simplex[d*i+j] = creal(simplex[d*i+j]);
imag_simplex[d*i+j] = cimag(simplex[d*i+j]);

in static void seperate_real_and_imag(PetscScalar *simplex, double *real_simplex, double *imag_simplex, int d) - permalink here.

Originally posted by @djanekovic in #1696 (comment)

[ReubenHill]

See also (with reference to the use of _Complex_I at this line)

maybe use https://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/Sys/PETSC_i.html#PETSC_i
instead?

Originally posted by @djanekovic in #1696 (comment)

[ChinmayNadgouda]

Hello,

I am a Master's student in Computer Science from the University of Stuttgart, Germany, and currently am attending the course, Simulation Software Engineering (SSE). https://simulation-software-engineering.github.io/homepage/

For this course, we students were asked to select an open-source simulation framework, like Firedrake and make a small contribution to get acquainted with such open-source software development tool and their practices.

For this, I am interested in Firedrake and am now looking for something to work on. There are no clear expectations as to what this contribution should be. It can be one of the following, a new feature, a tutorial, some documentation or a bug fix.

I am quite new to this field of simulation softwares but eager to learn and would really like to contribute something useful, so is it possible if I work on this "good first issue"?

If yes,
I would like your input and guidance on how to get get started / acquainted with Firedrake and understand it better, so that I can give a meaningful contribution.

If no,
Can you point me to something else that I can contribute to in Firedrake?

Thanks 😄

[wence-]

Hi @ChinmayNadgouda, this issue is still relevant, and would be a (hopefully) good first issue to look at. The amount of code needed to be changed is quite small (restricted to the file supermeshing.py). If you want to have a go, the best place to start is to install Firedrake (see www.firedrakeproject.org/download.html). In this particular case, you'll eventually want the "complex-mode" install: how to get this should be possible to figure out from the installation instructions/installer (if not, there's another documentation bug to consider fixing!).

Thanks!