The wording was so insufficient that it made people think it's a fatal
error, and subsequently made them suspicious because it seemed like the
fatal error is ignored.
This is important in case the data aren't owned by the instance but
instead referencing something else, for example the importer, a
memory-mapped file or another instance. Will get increasingly
important for zero-copy data import.
To not have to duplicate these for each and every case, enlarging the
surface for potential bugs. Also changing the signatures to number +
identifier, instead of identifier repeated number of times. Means the
compiler won't be able to deduplicate / overlap the literals in the
binary, but is much more maintainable.
For Numpy this causes the slices to be reported as actual typed Numpy
arrays instead of typeless tuples, which is good I guess? Not sure why
that wasn't the case before, or what is the difference, and how it will
behave for sparse types such as aligned matrices.
It limits the support for CMake 3.12+, but it's much less verbose and I
don't expect people to use ancient CMake versions with IDEs like Xcode
or VS anyway, so this should be fine.
Only the double ones, exposed as floats, because the extra ALU required
by doubles is negligible to function call overhead. It'll be different
for non-scalar types, but here I use this.
Vladimír Vondruš