Bug #10578: Allow undefining methods in refinements

• Status: Closed
• Assignee: matz (Yukihiro Matsumoto)
• Target version: -
• ruby -v: ruby 2.1.1p76 (2014-02-24 revision 45161) [x86_64-linux]
• Backport: 2.0.0: UNKNOWN, 2.1: UNKNOWN, 2.2: UNKNOWN

[whitequark (whitequark *)]

Since refinements are locally scoped monkey patches, and it is possible and useful to undef a method globally, I think refinements should allow to undef methods as well as def them.

For example:

module NoPlus
  refine String do
    undef +
  end
end

using NoPlus
"a" + "b"

Expected behavior: undefined method `+' for "a":String
Actual behavior: undef ignored

[shugo (Shugo Maeda)]

• File undef_in_refine.diff undef_in_refine.diff added
• Status changed from Open to Assigned
• Assignee set to matz (Yukihiro Matsumoto)

Peter Zotov wrote:

Since refinements are locally scoped monkey patches, and it is possible and useful to undef a method globally, I think refinements should allow to undef methods as well as def them.

The attached patch allow refinements to undef original methods.

Matz, Can I commit it?
And, is it considered a bug fix?

[shugo (Shugo Maeda)]

• Status changed from Assigned to Closed
• % Done changed from 0 to 100

Applied in changeset r49554.

---

• vm_insnhelper.c (vm_call_method): stop method search when a method
  is not found in a refinement, to support undef in refinements.
  [ruby-core:66741] [Bug #10578]

[shugo (Shugo Maeda)]

• Backport changed from 2.0.0: UNKNOWN, 2.1: UNKNOWN to 2.0.0: UNKNOWN, 2.1: UNKNOWN, 2.2: UNKNOWN

I've committed the patch because I consider this issue a bug.

However, I'm not sure whether it should be backported.