Bug #2420

super call may use wrong receiver object

Added by Nobuyoshi Nakada over 5 years ago. Updated almost 4 years ago.

[ruby-core:27000]
Status:Closed
Priority:Normal
Assignee:Koichi Sasada
ruby -v:ruby 1.9.2dev (2009-12-01 trunk 25970) [i386-darwin9.0] Backport:

Description

=begin
The code below prints:
C
main
Object::m

which means that the super calls the instance method B#m on a receiver that is not an instance of B. It's not what makes the super call to pick the main object as a receiver in this case. I would expect it to use the "self" object that the block captures, i.e. the class object C, and thus an exception "no superclass method `m' (NoMethodError)" should be raised.

class B
def m
p self
puts self.class.to_s + '::m'
end
end

class C < B
q = Proc.new do
p self
super()
end

mq = define_method :m, &q

mq.call
end

BTW: The behavior is even worse in Ruby 1.9:
ruby 1.9.1p129 (2009-05-12 revision 23412) [i386-mswin32]
C
a.rb:11:in ==': wrong number of arguments(0 for 1) (ArgumentError)
from a.rb:11:in
block in class:C'
from a.rb:16:in call'
from a.rb:16:in
class:C'
from a.rb:8:in `'
=end


Related issues

Duplicated by Ruby trunk - Bug #3136: reuse of singleton method definition causes SEGV Closed 04/12/2010

Associated revisions

Revision 29063
Added by _ wanabe almost 5 years ago

  • vm.c (vm_define_method): copy iseq to avoid overwriting iseq->klass. #2502, #3136. see #2420.

Revision 29063
Added by _ wanabe almost 5 years ago

  • vm.c (vm_define_method): copy iseq to avoid overwriting iseq->klass. #2502, #3136. see #2420.

History

#1 Updated by Yusuke Endoh almost 5 years ago

  • Assignee changed from Nobuyoshi Nakada to Koichi Sasada
  • Target version set to 2.0.0

=begin
Hi,

I committed r28043 as temporary measure for 1.9.2 release.
This will be fixed correctly at 1.9.3 or later.
See [Bug #2502] [Bug #3136] in detail.

--
Yusuke Endoh mame@tsg.ne.jp
=end

#2 Updated by _ wanabe almost 5 years ago

  • Status changed from Open to Closed

=begin
This issue was solved with changeset r29063.
Yusuke, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.

=end

#3 Updated by _ wanabe almost 5 years ago

  • Status changed from Closed to Open

=begin

=end

#4 Updated by Shyouhei Urabe over 4 years ago

=begin
Hi, can I close this?
=end

#5 Updated by Koichi Sasada almost 4 years ago

wanabe-san,

Should we keep it open or not?

#6 Updated by Nobuyoshi Nakada almost 4 years ago

  • Status changed from Open to Closed

This issue has gone away.

Also available in: Atom PDF