basics/floats/methods_parens.rb

tests/_examples/basics/floats/methods_parens.rb in the metorex repository.

# The sign predicates, and the magnitude `abs` also answers.
p(0.0.zero?)
p(1.5.positive?)
p(-1.5.negative?)
p(-2.5.magnitude)

# `angle` is the direction the number points: zero for a positive one and Pi
# for a negative one, which a signed zero follows.
p(1.0.angle)
p(-1.0.angle == Math::PI)
p(-0.0.angle == Math::PI)

# The neighboring representable numbers, a single step away.
step = 0.0.next_float
p(step > 0.0)
p(step.prev_float)
p(Float::INFINITY.next_float)

# `coerce` answers the pair an operator is applied to, reading a String the
# way `Float()` reads one.
p(1.2.coerce(1))
p(1.0.coerce("2.5"))

# The parts of the exact fraction the number stands for.
p(0.5.numerator)
p(0.5.denominator)
p(12.0.to_r)

# A precision keeps that many digits after the point, and zero or less answers
# the whole number those digits sit in.
p(3.14.floor(1))
p(3.14.ceil(1))
p(34.56.truncate(1))
p(1234.5.floor(-2))

# Nothing compares against NaN, and `eql?` is equality without conversion.
nan = 0 / 0.0
p(nan > 1.0)
p(1.0.eql?(1))
p(1.0.eql?(1.0))

# `%` refuses a zero divisor whatever the receiver, where `/` answers an
# infinity.
begin
  1.0 % 0.0
rescue ZeroDivisionError => error
  puts(error.message)
end
p(7.0.divmod(3))
p(7.0.fdiv(2))

Output:

true
true
true
2.5
0
true
true
true
0.0
Infinity
[1.0, 1.2]
[2.5, 1.0]
1
2
(12/1)
3.1
3.2
34.5
1200
false
false
true
divided by 0
[2, 1.0]
3.5