New tutorial

[andrewwinters5000]

This adds an additional tutorial and discussion of the newly released feature in HOHQMesh that one can define a straight about which a mesh can be reflected.

[DavidAKopriva]

A couple of comments:

1. HOHQMesh (rather than fortran) is case insensitive for the "symmetry" keyword. (Hmm, but all other keywords are case sensitive.
2. You should only need to change the curve names rather than re-construct the model. e.g. setCurveName!(line9,"B9") and setCurveName!(line2,"symmetry") and then regenerate.
3. In the second tutorial, the boundaries O.4 and O.8 will be B3 and B3_R and similarly for O.8. They become internal boundaries, rather than internal parts of the mesh. To do a fully symmetric mesh without internal boundaries, O.4 and O.8 need their names to also be set to "symmetry". This flexibility lets one define internal boundaries, as requested in issue 49 of HOHQMesh.
4. You need to delete the mesh before re-generating. (I think.) I suppose the default behavior should be that when generate_mesh is called the mesh, if it exists, should be deleted. Does it do that? I don't remember.

That's all I can think of at the moment.

[andrewwinters5000]

Thanks for the comments @DavidAKopriva ! I will update and simplify the second part of the tutorial from your suggestions. I also need to investigate the case sensitivity (or seemingly the lack of it) for the boundary name "symmetry". I tried a the three capitalization strategies for "Symmetry", "SYMMETRY" and sYmMeTrY" and a valid mesh was generated for all of them.

[DavidAKopriva]

Thanks for writing the tutorial. Yes, HOHQMesh calls ToLower(str) on the name before testing if it is equal to "symmetry", so it doesn't matter what case one uses. We should think if that should be done for all keywords, but it doesn't matter using HOHQmesh.jl because the control file written out will have the correct case.

[andrewwinters5000]

Thanks for writing the tutorial. Yes, HOHQMesh calls ToLower(str) on the name before testing if it is equal to "symmetry", so it doesn't matter what case one uses. We should think if that should be done for all keywords, but it doesn't matter using HOHQmesh.jl because the control file written out will have the correct case.

Thanks for the information. I assumed that there was something under hood helping with the case sensitivity, but I was only checking inside the Julia code. Avoiding case sensitivity on all the keywords could help make the control files more " idiot proof", but we can discuss this further in a different HOHQMesh PR.

[andrewwinters5000]

Better boundary name figures

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.81%. Comparing base (e044f3b) to head (526a29c).
Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #57      +/-   ##
==========================================
+ Coverage   98.80%   98.81%   +0.01%     
==========================================
  Files          21       21              
  Lines        1840     1863      +23     
==========================================
+ Hits         1818     1841      +23     
  Misses         22       22              

Flag	Coverage Δ
unittests	98.81% <100.00%> (+0.01%)	⬆️

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

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

[DavidAKopriva]

Changes as is are fine, but I think the renameCurve! procedure description should be added to the API documentation, probably in the model API since it is a model operation. I don't think that it is in there yet.

[andrewwinters5000]

Changes as is are fine, but I think the renameCurve! procedure description should be added to the API documentation, probably in the model API since it is a model operation. I don't think that it is in there yet.

I added the renameCurve! to the API and updated the remover_mesh! function to also delete the plot, control, and stats files.

[sloede]

Just a few comments/suggestions that came up when perusing this PR

[andrewwinters5000]

New figures now that that symmetry name has changed to :symmetry

[DavidAKopriva]

I think the changes are fine, and I approve modulo the windows fix.

[sloede]

LGTM

[DavidAKopriva]

I see there is no check on whether or not a file exists when rm() is called in the remove_mesh! procedure. I would guess that an error message will be written, and after thinking about it, that's OK.

[andrewwinters5000]

@sloede Are we good to merge? Once this is done I can release a new version of HOHQMesh.jl