For builtin/stdlib types we have
int:
Return integer ratio.
Return a pair of integers, whose ratio is exactly equal to the original int
and with a positive denominator.
Fraction:
as_integer_ratio(self)
Return the integer ratio as a tuple.
Return a tuple of two integers, whose ratio is equal to the
Fraction and with a positive denominator.
float:
Return integer ratio.
Return a pair of integers, whose ratio is exactly equal to the original float
and with a positive denominator.
Decimal:
Return a pair of integers, whose ratio is exactly equal to the original
Decimal and with a positive denominator. The ratio is in lowest terms.
Raise OverflowError on infinities and a ValueError on NaNs.
All methods actually return a pair, whose ratio is in lowest terms, but only the last docstring says this explicitly. Shouldn't other methods do same? (Taken from this.)
Edit: Ditto for some sphinx docs, e.g. for the float.as_integer_ratio(): "Return a pair of integers whose ratio is exactly equal to the original float and with a positive denominator. Raises OverflowError on infinities and a ValueError on NaNs."
Linked PRs
For builtin/stdlib types we have
int:
Fraction:
float:
Decimal:
All methods actually return a pair, whose ratio is in lowest terms, but only the last docstring says this explicitly. Shouldn't other methods do same? (Taken from this.)
Edit: Ditto for some sphinx docs, e.g. for the float.as_integer_ratio(): "Return a pair of integers whose ratio is exactly equal to the original float and with a positive denominator. Raises OverflowError on infinities and a ValueError on NaNs."
Linked PRs