bitshift typing on unsigned operands is tedious to work with #8002

Activity

1. pitrou commented on Sep 1, 2016

   

   pitrou

   on Sep 1, 2016

   MemberAuthor

   More actions

   In Numba we are considering typing based only on the first operand, see numba/numba#2072
2. sklam commented on Sep 1, 2016

   

   sklam

   on Sep 1, 2016

   More actions

   Numpy is probably casting (unsigned, signed) to (float64, float64), which is not defined. Then, it tries to downcast it back to integer, but it is unsafe.
3. ahaldane commented on Sep 22, 2016

   

   ahaldane

   on Sep 22, 2016

   Last edited by ahaldane

   Member

   More actions

   I've hit the same problem. I have thought it would have been better to copy C typing as closely as possible. Then for an algorithm already written in C I can just paste it into python, and not have to worry about wrapping all the constants in uint64.

   Here's something I had to write last week that demonstrates the annoyance:

       assert(type(crc) is uint16 and type(onech) is uint8)
   
       ans = uint32(crc ^ onech << uint16(8))
       for ind in range(8):
           if ans & uint32(0x8000):
             ans <<= uint32(1)
             ans = ans ^ uint32(4129)
           else:
             ans <<= uint32(1)
4. seberg commented on Jan 28, 2024

   

   seberg

   on Jan 28, 2024

   Member

   More actions

   Effectively fixed by NEP 50 in main/NumPy 2.0. Closing.
5. seberg

   closed this as completedon Jan 28, 2024