Solve TypeError in Python/IPython

Question

I am having trouble with the following function

def get_lexographically_next_bit_sequence(self, bits):
    """
    Bit hack from here:
    http://www-graphics.stanford.edu/~seander/bithacks.html#NextBitPermutation

    Generator even does this in poker order rank 
    so no need to sort when done! Perfect.
    """
    t = (bits | (bits - 1)) + 1 
    next = t | ((((t & -t) // (bits & -bits)) >> 1) - 1)  
    yield next
    while True:
        t = (next | (next - 1)) + 1 
        next = t | ((((t & -t) // (next & -next)) >> 1) - 1)
        yield next

This function returns the error:

TypeError: unsupported operand type(s) for >>: 'float' and 'int'

Notes: This python library was only supported in 2.7 and I used 2to3 in order to use it. Other parts of the library work as desired, so I'm generally confident that 2to3 worked.

I am trying to run this in IPython 3.5 and I have heard that some errors like this can happen specifically within IPython, so I'm wondering if it's related to that.

I tested the code with 10101 and it worked fine. what is the input in question? — Menachem Hornbacher, Apr 19 '17 at 03:49
my results: Windows 10 1607 python 2.7.3 and Python 3.6.0 your code returns no errors for 3 seconds — Menachem Hornbacher, Apr 19 '17 at 03:58

score 0 · Answer 1 · answered Apr 19 '17 at 03:46

0

The issue arises from the fact that you are trying to preform a Binary Right Shift (>>) between two different data types (float and int). Casting the float down to an int with (int(((t & -t) // (next & -next)) >> 1) - 1) should work to the best of my knowledge.

answered Apr 19 '17 at 03:46

Menachem Hornbacher

1,759
2
23
31

It shouldn't be *possible* for any of the inputs to `>>` to be floats, though. You can't get a float out of `&`, and `//` won't give floats unless you give it floats. – user2357112 supports Monica Apr 19 '17 at 16:20
I think he gave it floats. A Test case with 31 passed worked with the original code – Menachem Hornbacher Apr 19 '17 at 16:26
[If you gave this function a float, it would error out on `|`, not `>>`.](http://ideone.com/8mId00) There's no way for a float to reach any of the `>>` operators in this function. – user2357112 supports Monica Apr 19 '17 at 16:30
Then I guess the error is being thrown elsewhere. Since DMS has said nothing I assume it's fixed. – Menachem Hornbacher Apr 19 '17 at 16:34

Solve TypeError in Python/IPython

1 Answers1