roystgnr

Well, I can replicate this one too, so that's a good start!

And it's exactly what I was fearing might be a failure case: our ray from one hole point is "intersecting" (within tolerance) at one vertex of the other hole, but...

I think a random shift is a terrifying idea - it won't actually fix the bug, it'll just make the bug harder to replicate. My vague intuition here is that...

> Does "do not merge" also mean "do not review"? It's already bad enough to need fixing, huh? ;-) Nah, please jump in as you see fit. I've got more...

`vec_dim` isn't `_system_vars.size()`; it would be the dimensionality of the vector variables, not the total number of variables. `_system_vars` is a vector of variable indices; vector variables with any of...

Variables are added dynamically to a `System`; `DofMap::n_variables()` returns the total count of them for its associated system. Each variable's information is available from `DofMap::variable(i)`; that information includes a `FEType`,...

I didn't think we'd turned on whitespace requirements for libMesh, even on just "spaces vs tabs"... The fact that the error message is giving MOOSE formatting rather than libMesh formatting...

Make sure it doesn't fail (even if only because it's `#ifdef`ed out) on non-PETSc builds; if so, that'd be a very good solution. IIRC we use those options on examples...

We were having Civet issues, so perhaps the trigger never triggered? Try rewording a commit message or rebasing on master or something and then force push to see if it'll...

They're public members, but they're members wrapped in `unique_ptr`, so one could be left unallocated, if we were careful about not actually trying to dereference it in library code or...