Mark Dickinson <mdickin...@enthought.com> writes: > Ben Bacarisse <ben.usenet <at> bsb.me.uk> writes: >> [1] Not being a Python expert I don't know how you show that actual >> value of a float. What is the Pythonic way to do that? > > I don't know about Pythonic, but here are some options. > > 1. Convert the float to Decimal, and print the result. This shows > the exact binary value that's stored, but displays it in decimal. > Be aware that the result will be hundreds of digits long for > very large or very small floats. > > >>> print(Decimal(pi)) > 3.141592653589793115997963468544185161590576171875 > > 2. If you're happy with a hexadecimal representation, use the > float.hex method. Again, this shows the exact value stored. > > >>> print(pi.hex()) > 0x1.921fb54442d18p+1 > > 3. To get an equivalent fraction, convert to the fractions.Fraction > type or use the as_integer_ratio method. > > >>> from fractions import Fraction > >>> print(Fraction(pi)) > 884279719003555/281474976710656 > >>> print(pi.as_integer_ratio()) > (884279719003555, 281474976710656)
Thanks. I worked out 1 and 3 after posting but I was not aware of how to get 2 in Python. It's the format I am most familiar with (and it's very compact for numbers that are anything but in decimal or as fractions) so that's handy. Once you told me that hex() does the trick I wondered if "%x" % 3.4 would give me 0x1.b333333333333p+1 but the float is converted to the expected integer. I doubt anyone relies on using %x for the integer part of a float so maybe %x could be provided for floats... -- Ben. -- https://mail.python.org/mailman/listinfo/python-list
