Alexander Grund

Actually this cannot be split. The bug/safety feature is a work-around for the problem described here. The issue **IS** that the policies cannot be combined freely as suggested by the...

@kumpera Would you be able to review and merge this? CI seems to be broken/unusable right now. @cdluminate If you want/can take a look if this works for you too?

So a 2D zero size is `(0, 0, 1)`? This looks rather strange... Is this a HDF5/Splash related requirement?

Really? What about e.g. writing particles when there are no particles for the current process? Or field slices where the current process has no data because it is outside the...

Well I did not know, that you can't write a zero-sized record and just did it. It actually writes a single value, but doesn't run into any problems besides that.

It is not solved as that function is still wrong for the reasons described. Another simple example: 3D size of 1: `(1, 1, 1)` for which `getDims` would return 1...

If you create a 3D object and it says it is 1D it is not inconsistent, it is plain wrong. Or would you use "inconsistent" for "works sometimes but fails...

> you document it reasonably well (what I have in mind is a paragraph or two to explain they why and how) What and where do you have in mind?...

Thanks a lot for your replies. However I run into a dead end for now in the use case I intended this for: Compiling TF with a preinstalled protobuf. After...

> the tarball created by `make dist` includes submodules, while the default files created by github does not. So yes, it is required. That's why I mentioned `actions/checkout` with the...