Travis Oliphant <oliphant.travis <at> ieee.org> writes:
Don't lump those ideas together. Shapes and strides are necessary for N-dimensional array's (it's essentially what *defines* the N-dimensional array). I really don't want to sacrifice those in the extended buffer protocol. If you want to separate them into different functions then that is a possibility.
I don't understand. Do you want to discuss shapes and strides separately from the datatype or not? Note that in ctypes shape is a property of datatype (as in c_int*2*3). In your proposal, shapes and strides are communicated separately. This presents a unique memory management challenge: if the object does not contain shape information in a ready to be pointed to form, who is responsible for deallocating the shape array?
If we manage to agree on the standard way to pass primitive type information, it will be a big achievement and immediately useful because simple arrays are already in the standard library.
We could start there, I suppose. Especially if it helps us all get on the same page.
Let's start:
1. Should primitive types be associated with simple type codes (short, int, long, float, double) or type/size pairs [(int,16), (int, 32), (int, 64), (float, 32), (float, 64)]? - I prefer pairs
2. Should primitive type codes be characters or integers (from an enum) at C level? - I prefer integers
3. Should size be expressed in bits or bytes? - I prefer bits