Add feature to create bottom topography from data

[andrewwinters5000]

This is a small step to add the capability of creating a bottom topography from data read in from a file. Then a 3D mesh is created using simple extrusion. To create the high-order boundary information a bicubic interpolation is used given the available data read in from the provided file.

Currently, this file data is assumed to be gridded data that has (presumably) been pre-processed by a user.

[codecov[bot]]

Codecov Report

Merging #27 (9e79ef2) into main (98e2631) will increase coverage by 9.34%. The diff coverage is 92.01%.

@@            Coverage Diff             @@
##             main      #27      +/-   ##
==========================================
+ Coverage   57.49%   66.83%   +9.34%     
==========================================
  Files          65       66       +1     
  Lines        9580     9927     +347     
==========================================
+ Hits         5508     6635    +1127     
+ Misses       4072     3292     -780     

Flag	Coverage Δ
unittests	66.83% <92.01%> (+9.34%)	:arrow_up:

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
Source/IO/MeshOutputMethods.f90	69.78% <ø> (ø)
Source/Mesh/LaplaceMeshSmoother.f90	0.00% <ø> (ø)
Source/Mesh/SpringMeshSmoother.f90	92.56% <ø> (+1.35%)	:arrow_up:
Source/Project/MeshProject.f90	72.36% <ø> (+4.27%)	:arrow_up:
...rce/Surfaces/ParametricEquationTopographyClass.f90	29.78% <ø> (+29.78%)	:arrow_up:
Source/HOHQMesh.f90	66.11% <6.25%> (+10.04%)	:arrow_up:
Source/Surfaces/DataFileTopographyClass.f90	88.46% <88.46%> (ø)
Source/Testing/MeshingTests.f90	90.81% <89.47%> (-2.78%)	:arrow_down:
Source/Testing/TestDataClass.f90	68.42% <94.44%> (+20.80%)	:arrow_up:
Source/3DSource/3DMeshController.f90	94.00% <100.00%> (+90.00%)	:arrow_up:
... and 29 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 98e2631...9e79ef2. Read the comment docs.

[fluidnumerics-joe]

@andrewwinters5000 - I just pushed up a patch that resolves the compiling issue with the exception handler call. I think we now want to settle on who to put as the poster

[fluidnumerics-joe]

@DavidAKopriva @andrewwinters5000 it looks like there are some coverage issues that can be resolved. I'll take a look at where we can improve coverage; I suspect we don't have tests running for the new example associated with this feature. It'd be great to get this merged in

[andrewwinters5000]

I'll take a look at where we can improve coverage; I suspect we don't have tests running for the new example associated with this feature. It'd be great to get this merged in

I agree it would be good to get this feature merged into main. Do we need to merge your spline_curve feature first @fluidnumerics-joe ? Or are the two features orthogonal in some sense?

[andrewwinters5000]

The coverage improved because the new 3D test also exercises the Abaqus output routines in 3DSource/Mesh3DOutputMethods.f90. It is strange that the destruct and release routines for the new SMTopographyFromFile are not covered. I believe that these should get invoked when lines 173-176 in destructModel are executed because that is when the topography pointer gets released. Do I understand this release call correctly @DavidAKopriva ?

[andrewwinters5000]

[DavidAKopriva]

Andrew,

1. The "how to add a test case" looks good. I'm almost wondering if the list of tests should be in a file rather than code, since there is actually no reason to have to edit code to add a test case. OTOH, there's one less file path to have to maintain.
2. In the docs for getting the topography from a file, there is no example showing how to invoke it. I presume it's "file =" in the TOPOGRAPHY block. Will start looking through the code next...

[andrewwinters5000]

• The "how to add a test case" looks good. I'm almost wondering if the list of tests should be in a file rather than code, since there is actually no reason to have to edit code to add a test case. OTOH, there's one less file path to have to maintain.

It might be easiest to move it out of the code and into a separate file. Having it directly in the MeshTesting.f90 file tends to regularly cause merge conflicts.

• In the docs for getting the topography from a file, there is no example showing how to invoke it. I presume it's "file =" in the TOPOGRAPHY block.

Whoops, totally forgot an example invoking it. I will add that.

[fluidnumerics-joe]

It might be easiest to move it out of the code and into a separate file. Having it directly in the MeshTesting.f90 file tends to regularly cause merge conflicts.

I agree with moving the list of tests to a file that just lists the set of control files under Benchmarks. Alternatively, if we don't want to maintain a list, we could modify the ci.yaml to create such a list on the fly by listing the control files under Benchmarks/ControlFiles.

If we need to require that a test file is available for comparison, we could modify the MeshingTests to throw an error if one is not found.

[DavidAKopriva]

"It might be easiest to move it out of the code and into a separate file. Having it directly in the MeshTesting.f90 file tends to regularly cause merge conflicts." --- That's my thinking, too. Let's finish up with this PR and think about that in the meantime. The issue will come up again when we merge the TopographyResolution branch in, will I will propose to look at after this one.

[andrewwinters5000]

That's my thinking, too. Let's finish up with this PR and think about that in the meantime. The issue will come up again when we merge the TopographyResolution branch in, will I will propose to look at after this one.

That's a good plan especially because this PR already has a lot in it.

[DavidAKopriva]

I'll go though the new stuff this morning, esp. the fact that the final routines are not being called.

[DavidAKopriva]

Thanks for noticing the fact the destructors weren't called. The bug was way up in the chain where the FINAL routine for the MeshProject class wasn't defined.