mhostetter

Create a function to automate the type checking and raising of `TypeError`. This would standardize the error messages and reduced copy-pasted code. ### Before ```python @set_module("galois") def primitive_poly(order: int, degree:...

- Refactor `.encode(message, parity_only=False)` to `.ecnode(message, output="codeword" | "parity")`. - Add `.decode(codeword, output="message" | "codeword", errors=False)`.

When NumPy returns a scalar bool or integer, it's `np.bool_` or `np.int64`. This may cause issues if someone is expecting an `int`. Perhaps I should normalize these and convert them...

Perhaps for large fields `GF(p^m)`, that would normally use explicit calculation, a tower of `GF(p^e^k)` could be constructed. It would be such that `GF(p^e)` would use lookup tables. Then the...

Pure-Python arithmetic occasionally has issues

3

When using overridden ufuncs (like `np.bitwise_xor` for addition in binary fields) sometimes there are type errors. ```python In [1]: import galois In [2]: c = galois.irreducible_poly(2**100, 3); c Out[2]: Poly(x^3...

http://magma.maths.usyd.edu.au/calc/

`GF(p^m)` array structure

1

Consider making N-D `GF(p^m)` arrays actually (N + 1)-D arrays over `GF(p)` under the hood. This would require intercepting indexing calls. This would prevent the continual `int_to_poly()` and `poly_to_int()` calls...