roystgnr
roystgnr
## Reason MOOSE has had support for `SIDE_DISCONTINUOUS` variables for a while, but no way to visualize them easily. This should fix that. ## Design By setting side_discontinuous=true in Exodus...
I want to add one more unit test before merging this, but first I want to see just how inflexible the PETSCVIEWERBINARY format is - it breaks when I move...
This isn't in working order (I'm sure it'll fail installcheck, most of the features aren't there...) but it's passing the most basic test now and probably good enough to let...
I'm still trying to figure out how to make this easy to use, but I'd like to at least test that what I've got so far still passes CI.
A Cardinal+OpenMC user ran into this, on a system where libtirpc is necessary for XDR support but where the headers are still installed in `/usr/include/rpc/`, not `/usr/include/tirpc/rpc/`. Our first attempt...
We have `is_prepared()` to say that *every* bit of batch-computed information is up to date, and we have `set_isnt_prepared()` to manually warn future users that a mesh change has been...
And then it's only copying over integers that existed on an old `DofObject`, so when copying from meshes that don't have some of our own extra integers, we might find...
I'm not sure this is the right interface (there's at least some PetscDiffSolver encapsulation that I've broken with `public:` but should redo properly with `friend`), and I'm not sure this...
This may be leading to problems in apps using MOOSE, which treats names as the "first-class" identifier rather than numeric ids, in the case of complex compound-constructed meshes.
## Reason This lets us read IsoGeometric Analysis meshes where the extracted `RATIONAL_BERNSTEIN` elements don't form a conforming C0 mesh, via the trick of not bothering to try to construct...