Update BoundaryErrors.f90

[DavidAKopriva]

Change the error output to be easy to plot the error by segment.

[codecov]

Codecov Report

❌ Patch coverage is 94.11765% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 78.25%. Comparing base (bed3d30) to head (7017b47).

Files with missing lines	Patch %	Lines
Source/BoundaryOptimization/BoundaryErrors.f90	90.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #171      +/-   ##
==========================================
- Coverage   78.25%   78.25%   -0.01%     
==========================================
  Files          83       83              
  Lines       12387    12387              
  Branches        2        2              
==========================================
- Hits         9694     9693       -1     
- Misses       2693     2694       +1     

Flag	Coverage Δ
unittests	78.25% <94.11%> (-0.01%)	⬇️

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

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[andrewwinters5000]

Having access to the error values is good for visualization. Are the gStart and gEnd values of the parametrization variable t? So, for plotting one needs access to the boundary parametric curves and/or polynomials?

[DavidAKopriva]

Yes, those are the parametrization values. One can then use the model to compute the physical space locations. Alternatively, if you think it would be better, I could write out the physical space locations. (Or both.) That would eliminate a visualization step.

[andrewwinters5000]

Adding the physical space locations would save a step and make the visualization easier.

[DavidAKopriva]

Will do.

[andrewwinters5000]

In the docs for the error controlled meshing it says that

Two files can be written prepended with the name of the control file. One contains the point-wise error along each boundary as a function of the chain curve parameter.

but one the *_Norms.txt file is currently written. The point-wise is computed on the fly in the ComputeBoundaryErrors function but not saved. Did we want both outputs? Or should I update the docs?

[DavidAKopriva]

I guess I removed the point wise error output. I was using that primarily for development, and don't see the need for it now. It can always be added back later if anyone sees the need. So I say just remove the reference from the docs.

[andrewwinters5000]

I guess I removed the point wise error output. I was using that primarily for development, and don't see the need for it now. It can always be added back later if anyone sees the need. So I say just remove the reference from the docs.

Okay, I assumed it was the case that the pointwise error was useful in the development. I think the integrated value along a segment is sufficient to highlight the "problem" elements along a boundary in terms of the L2 or H1 errors. I will update the docs.

[DavidAKopriva]

I agree. I think that is sufficient.

[DavidAKopriva]

Looks fine. Maybe replace "Stats and boundary errors not written" with "Stats and boundary errors are not written"