Bug #11736: Float(arg) not consistent with documentation

• Status: Closed
• Priority: Normal
• Assignee: hsbt (Hiroshi SHIBATA)
• Target version: -
• ruby -v: ruby 2.1.5p273
• Backport: 2.1: UNKNOWN, 2.2: DONE, 2.3: DONE

[bwheeler96 (Brian Wheeler)]

Float(arg) → float
Returns arg converted to a float. Numeric types are converted directly, the rest are converted using arg.to_f. Converting nil generates a TypeError.

Strings are NOT converted using arg.to_f

'123_INVALID'.to_f 
# => 123.0

Float('123_INVALID') 
# => ArgumentError

This could be fixed by changing object.c like so:

-       return DBL2NUM(rb_str_to_dbl(val, TRUE));
+       return DBL2NUM(rb_str_to_dbl(val, FALSE));

But it appears that this is desired behavior as there are many specs to ensure this strict validity checking. Please update the documentation to reflect this.

Thanks,

Brian

[ko1 (Koichi Sasada)]

• Description updated (diff)
• Assignee set to zzak (Zachary Scott)

Document is wrong.
Zack, could you fix it?

[dpulliam (Dylan Pulliam)]

• File 0001-Improve-documentation-on-Float-Arg.patch 0001-Improve-documentation-on-Float-Arg.patch added
• ruby -v set to ruby 2.1.5p273

I have prepared a patch for this. Let me know what you think.

[hsbt (Hiroshi SHIBATA)]

• Status changed from Open to Closed

Applied in changeset r56421.

---

• object.c: Improve documentation for Integer conversion. [ruby-core:71661][Bug #11736][ci skip]

[hsbt (Hiroshi SHIBATA)]

• Backport changed from 2.0.0: UNKNOWN, 2.1: UNKNOWN, 2.2: UNKNOWN to 2.1: UNKNOWN, 2.2: REQUIRED, 2.3: REQUIRED
• Assignee changed from zzak (Zachary Scott) to hsbt (Hiroshi SHIBATA)

[usa (Usaku NAKAMURA)]

• Backport changed from 2.1: UNKNOWN, 2.2: REQUIRED, 2.3: REQUIRED to 2.1: UNKNOWN, 2.2: DONE, 2.3: REQUIRED

ruby_2_2 r56501 merged revision(s) 56421,56422.

[nagachika (Tomoyuki Chikanaga)]

• Backport changed from 2.1: UNKNOWN, 2.2: DONE, 2.3: REQUIRED to 2.1: UNKNOWN, 2.2: DONE, 2.3: DONE

ruby_2_3 r56722 merged revision(s) 56421,56422.

[jeremyevans0 (Jeremy Evans)]

• Related to Bug #9574: Inconsistent behavior between Kernel#Float and to_f added