Le 08/07/2022 à 15:19, Wes McKinney a écrit :
* I believe that having a Type::RLE is the right approach in C++ and it makes dynamic dispatch everywhere in the library pretty straightforward.
+1 on this, as it will raise a nice NotImplemented error for existing code rather than crash or corrupt memory.
* It's unclear to me whether it is a good idea for RLE to replace the use of Scalar values in the compute kernels sublibrary, especially after the refactoring that I've just done to drop the scalar-specific code paths from most kernels (and about to remove the rest once I finish refactoring the aggregate kernels). It feels complicated (as opposed to letting scalars and RLE co-exist peacefully) but I could be convinced otherwise -- it may be hard to arrive at a conclusion without some prototyping / exploration
Indeed, sounds rather complicated to me. Regards Antoine.
