Function to_ulps

Lib/test/test_math.py:41–55

Convert a non-NaN float x to an integer, in such a way that adjacent floats are converted to adjacent integers. Then abs(ulps(x) - ulps(y)) gives the difference in ulps between two floats. The results from this function will only make sense on platforms where native doubles are

(x)

def to_ulps(x):
    """Convert a non-NaN float x to an integer, in such a way that
    adjacent floats are converted to adjacent integers.  Then
    abs(ulps(x) - ulps(y)) gives the difference in ulps between two
    floats.

    The results from this function will only make sense on platforms
    where native doubles are represented in IEEE 754 binary64 format.

    Note: 0.0 and -0.0 are converted to 0 and -1, respectively.
    """
    n = struct.unpack('<q', struct.pack('<d', x))[0]
    if n < 0:
        n = ~(n+2**63)
    return n


def ulp_abs_check(expected, got, ulp_tol, abs_tol):