Bug #6274: Float addition incorrect

• Status: Rejected
• Assignee: -
• Target version: -
• ruby -v: ruby 1.9.3p21 (2012-01-07 revision 34224) [x86_64-linux]

[swanboy (Michael Swan)]

Adding decimals with repeating values appears to lead to an error in returned value.

For example if you inspect:
10 + 1.0 / 3.0
or
33 + 1.0 / 3.0

the values are not what they should be. You will need to test this in IRB.

[marcandre (Marc-Andre Lafortune)]

• Status changed from Open to Rejected

This is due to the fact that floats are approximate. The results follow the IEEE standard. Try it in Javascript, you'll get the same result.

Issue #6274 has been updated by marcandre (Marc-Andre Lafortune).

Status changed from Open to Rejected

This is due to the fact that floats are approximate. The results follow the IEEE standard. Try it in Javascript, you'll get the same result.

I am almost happy that Ruby shows the rounding point error.
Apparently many programmers just don't realize just how imprecise
those underlying floats are, and now get to learn :)

-roger-