Kernel#to_f

obj.to_f -> float

Returns: Float · Updated March 13, 2026 · Kernel Methods

conversion numbers float parsing

The to_f method converts a value to a floating-point number. It’s defined on many Ruby objects and provides a way to extract numeric values from strings and other objects.

Basic Usage with Numbers

# Integers to floats
42.to_f        # => 42.0
-100.to_f      # => -100.0

# Float to float (returns self)
3.14159.to_f   # => 3.14159

# Rational to float
Rational(1, 2).to_f  # => 0.5
Rational(22, 7).to_f # => 3.142857142857143

Converting Strings

# String to float
"3.14".to_f       # => 3.14
"123.45".to_f     # => 123.45
"-98.6".to_f     # => -98.6

# Leading/trailing whitespace is ignored
"  42.5  ".to_f  # => 42.5

# Scientific notation
"1.5e3".to_f     # => 1500.0
"1e-4".to_f      # => 0.0001

Edge Cases

# Invalid strings return 0.0
"hello".to_f     # => 0.0
"".to_f          # => 0.0

# Numbers embedded in strings
"price50".to_f   # => 0.0 (stops at non-numeric)
"50.5degrees".to_f  # => 50.5

Practical Examples

Processing User Input

def calculate_total(prices)
  prices.map(&:to_f).sum
end

calculate_total(["19.99", "5.50", "3.00"])  # => 28.49

Parsing Configuration

config = {
  "discount" => "0.15",
  "tax_rate" => "0.08",
  "shipping" => "5.99"
}

discount = config["discount"].to_f
tax_rate = config["tax_rate"].to_f
shipping = config["shipping"].to_f

Handling CSV Numeric Data

require 'csv'
data = CSV.read("numbers.csv")
# Assume data is [["3.14"], ["2.71"], ["1.41"]]
floats = data.map { |row| row[0].to_f }

Custom Objects

# Define to_f on your own class
class Temperature
  attr_reader :celsius
  
  def initialize(celsius)
    @celsius = celsius
  end
  
  def to_f
    @celsius * 9.0 / 5 + 32
  end
end

temp = Temperature.new(100)
temp.to_f  # => 212.0

Comparison with Integer#to_f

# Integer.to_f is equivalent
42.to_f       # => 42.0
Integer(42).to_f  # => 42.0
Float(42)     # => 42.0

The to_f method is the standard way to convert values to floats in Ruby, widely used in numeric processing and data parsing.