How to raise an error when converting a non-numeric string to float

Question

I'm prompting the user to enter a float. Afterward I try to convert the received string to float. In case the user has entered an invalid value (string with special characters, etc.), which can't be converted to float, I expected an exception to be thrown.

But that isn't the case:

result = "XYZ§&()123!".to_f
#=> 0.0

Instead 0.0 is returned!

Actual I intended to put the conversion into a begin-rescue.

In case the user has entered an invalid value, it goes into the rescue-branch. An error-message is displayed and then the script terminated. At least, that was my intention.

How can I handle the situation, when the user entered an invalid value?

0.0 is a perfectly valid value, which can be entered by the user as well. How can I differentiate if it comes from .to_f as a result of a failed cast or was actually entered by the user? The both possibilities have to be handled differently.

The Ruby behavior differs from Python, where an exception is thrown.

float("XYZ$&()123!")
# Traceback (most recent call last):
#  File "<python-input-0>", line 1, in <module>
#    float("XYZ$&()123!")
#    ~~~~~^^^^^^^^^^^^^^^
# ValueError: could not convert string to float: 'XYZ$&()123!'

Weird or not, it is documented to work this way - apidock.com/ruby/v2_5_5/String/to_f — C3roe
– C3roe, Commented Jun 16 at 10:38

spickermann · Accepted Answer · 2025-06-16 12:37:43Z

4

You can use the Float() method when you want to raise an exception when the string contains anything other that the string representation of a float.

Float("XYZ$&()123!")
# raises 'Kernel#Float': invalid value for Float(): "XYZ$&()123!" (ArgumentError)

Float("123")
#=> 123.0

answered Jun 16 at 12:37

spickermann

108k9 gold badges115 silver badges147 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Stefan Jun 17 at 15:56

... not to forget hexadecimal floating point notation (0x...) which uses p instead of e. It might come as another surprise that Float("-0xc.90fep-2") parses perfectly fine ;)

Collectives™ on Stack Overflow

How to raise an error when converting a non-numeric string to float

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related